uprobes, mm, x86: Add the ability to install and remove uprobes breakpoints
Add uprobes support to the core kernel, with x86 support.
This commit adds the kernel facilities, the actual uprobes
user-space ABI and perf probe support comes in later commits.
General design:
Uprobes are maintained in an rb-tree indexed by inode and offset
(the offset here is from the start of the mapping). For a unique
(inode, offset) tuple, there can be at most one uprobe in the
rb-tree.
Since the (inode, offset) tuple identifies a unique uprobe, more
than one user may be interested in the same uprobe. This provides
the ability to connect multiple 'consumers' to the same uprobe.
Each consumer defines a handler and a filter (optional). The
'handler' is run every time the uprobe is hit, if it matches the
'filter' criteria.
The first consumer of a uprobe causes the breakpoint to be
inserted at the specified address and subsequent consumers are
appended to this list. On subsequent probes, the consumer gets
appended to the existing list of consumers. The breakpoint is
removed when the last consumer unregisters. For all other
unregisterations, the consumer is removed from the list of
consumers.
Given a inode, we get a list of the mms that have mapped the
inode. Do the actual registration if mm maps the page where a
probe needs to be inserted/removed.
We use a temporary list to walk through the vmas that map the
inode.
- The number of maps that map the inode, is not known before we
walk the rmap and keeps changing.
- extending vm_area_struct wasn't recommended, it's a
size-critical data structure.
- There can be more than one maps of the inode in the same mm.
We add callbacks to the mmap methods to keep an eye on text vmas
that are of interest to uprobes. When a vma of interest is mapped,
we insert the breakpoint at the right address.
Uprobe works by replacing the instruction at the address defined
by (inode, offset) with the arch specific breakpoint
instruction. We save a copy of the original instruction at the
uprobed address.
This is needed for:
a. executing the instruction out-of-line (xol).
b. instruction analysis for any subsequent fixups.
c. restoring the instruction back when the uprobe is unregistered.
We insert or delete a breakpoint instruction, and this
breakpoint instruction is assumed to be the smallest instruction
available on the platform. For fixed size instruction platforms
this is trivially true, for variable size instruction platforms
the breakpoint instruction is typically the smallest (often a
single byte).
Writing the instruction is done by COWing the page and changing
the instruction during the copy, this even though most platforms
allow atomic writes of the breakpoint instruction. This also
mirrors the behaviour of a ptrace() memory write to a PRIVATE
file map.
The core worker is derived from KSM's replace_page() logic.
In essence, similar to KSM:
a. allocate a new page and copy over contents of the page that
has the uprobed vaddr
b. modify the copy and insert the breakpoint at the required
address
c. switch the original page with the copy containing the
breakpoint
d. flush page tables.
replace_page() is being replicated here because of some minor
changes in the type of pages and also because Hugh Dickins had
plans to improve replace_page() for KSM specific work.
Instruction analysis on x86 is based on instruction decoder and
determines if an instruction can be probed and determines the
necessary fixups after singlestep. Instruction analysis is done
at probe insertion time so that we avoid having to repeat the
same analysis every time a probe is hit.
A lot of code here is due to the improvement/suggestions/inputs
from Peter Zijlstra.
Changelog:
(v10):
- Add code to clear REX.B prefix as suggested by Denys Vlasenko
and Masami Hiramatsu.
(v9):
- Use insn_offset_modrm as suggested by Masami Hiramatsu.
(v7):
Handle comments from Peter Zijlstra:
- Dont take reference to inode. (expect inode to uprobe_register to be sane).
- Use PTR_ERR to set the return value.
- No need to take reference to inode.
- use PTR_ERR to return error value.
- register and uprobe_unregister share code.
(v5):
- Modified del_consumer as per comments from Peter.
- Drop reference to inode before dropping reference to uprobe.
- Use i_size_read(inode) instead of inode->i_size.
- Ensure uprobe->consumers is NULL, before __uprobe_unregister() is called.
- Includes errno.h as recommended by Stephen Rothwell to fix a build issue
on sparc defconfig
- Remove restrictions while unregistering.
- Earlier code leaked inode references under some conditions while
registering/unregistering.
- Continue the vma-rmap walk even if the intermediate vma doesnt
meet the requirements.
- Validate the vma found by find_vma before inserting/removing the
breakpoint
- Call del_consumer under mutex_lock.
- Use hash locks.
- Handle mremap.
- Introduce find_least_offset_node() instead of close match logic in
find_uprobe
- Uprobes no more depends on MM_OWNER; No reference to task_structs
while inserting/removing a probe.
- Uses read_mapping_page instead of grab_cache_page so that the pages
have valid content.
- pass NULL to get_user_pages for the task parameter.
- call SetPageUptodate on the new page allocated in write_opcode.
- fix leaking a reference to the new page under certain conditions.
- Include Instruction Decoder if Uprobes gets defined.
- Remove const attributes for instruction prefix arrays.
- Uses mm_context to know if the application is 32 bit.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Also-written-by: Jim Keniston <jkenisto@us.ibm.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Roland McGrath <roland@hack.frob.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Anton Arapov <anton@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Denys Vlasenko <vda.linux@googlemail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linux-mm <linux-mm@kvack.org>
Link: http://lkml.kernel.org/r/20120209092642.GE16600@linux.vnet.ibm.com
[ Made various small edits to the commit log ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-02-09 17:26:42 +08:00
|
|
|
/*
|
2012-02-17 16:27:41 +08:00
|
|
|
* User-space Probes (UProbes)
|
uprobes, mm, x86: Add the ability to install and remove uprobes breakpoints
Add uprobes support to the core kernel, with x86 support.
This commit adds the kernel facilities, the actual uprobes
user-space ABI and perf probe support comes in later commits.
General design:
Uprobes are maintained in an rb-tree indexed by inode and offset
(the offset here is from the start of the mapping). For a unique
(inode, offset) tuple, there can be at most one uprobe in the
rb-tree.
Since the (inode, offset) tuple identifies a unique uprobe, more
than one user may be interested in the same uprobe. This provides
the ability to connect multiple 'consumers' to the same uprobe.
Each consumer defines a handler and a filter (optional). The
'handler' is run every time the uprobe is hit, if it matches the
'filter' criteria.
The first consumer of a uprobe causes the breakpoint to be
inserted at the specified address and subsequent consumers are
appended to this list. On subsequent probes, the consumer gets
appended to the existing list of consumers. The breakpoint is
removed when the last consumer unregisters. For all other
unregisterations, the consumer is removed from the list of
consumers.
Given a inode, we get a list of the mms that have mapped the
inode. Do the actual registration if mm maps the page where a
probe needs to be inserted/removed.
We use a temporary list to walk through the vmas that map the
inode.
- The number of maps that map the inode, is not known before we
walk the rmap and keeps changing.
- extending vm_area_struct wasn't recommended, it's a
size-critical data structure.
- There can be more than one maps of the inode in the same mm.
We add callbacks to the mmap methods to keep an eye on text vmas
that are of interest to uprobes. When a vma of interest is mapped,
we insert the breakpoint at the right address.
Uprobe works by replacing the instruction at the address defined
by (inode, offset) with the arch specific breakpoint
instruction. We save a copy of the original instruction at the
uprobed address.
This is needed for:
a. executing the instruction out-of-line (xol).
b. instruction analysis for any subsequent fixups.
c. restoring the instruction back when the uprobe is unregistered.
We insert or delete a breakpoint instruction, and this
breakpoint instruction is assumed to be the smallest instruction
available on the platform. For fixed size instruction platforms
this is trivially true, for variable size instruction platforms
the breakpoint instruction is typically the smallest (often a
single byte).
Writing the instruction is done by COWing the page and changing
the instruction during the copy, this even though most platforms
allow atomic writes of the breakpoint instruction. This also
mirrors the behaviour of a ptrace() memory write to a PRIVATE
file map.
The core worker is derived from KSM's replace_page() logic.
In essence, similar to KSM:
a. allocate a new page and copy over contents of the page that
has the uprobed vaddr
b. modify the copy and insert the breakpoint at the required
address
c. switch the original page with the copy containing the
breakpoint
d. flush page tables.
replace_page() is being replicated here because of some minor
changes in the type of pages and also because Hugh Dickins had
plans to improve replace_page() for KSM specific work.
Instruction analysis on x86 is based on instruction decoder and
determines if an instruction can be probed and determines the
necessary fixups after singlestep. Instruction analysis is done
at probe insertion time so that we avoid having to repeat the
same analysis every time a probe is hit.
A lot of code here is due to the improvement/suggestions/inputs
from Peter Zijlstra.
Changelog:
(v10):
- Add code to clear REX.B prefix as suggested by Denys Vlasenko
and Masami Hiramatsu.
(v9):
- Use insn_offset_modrm as suggested by Masami Hiramatsu.
(v7):
Handle comments from Peter Zijlstra:
- Dont take reference to inode. (expect inode to uprobe_register to be sane).
- Use PTR_ERR to set the return value.
- No need to take reference to inode.
- use PTR_ERR to return error value.
- register and uprobe_unregister share code.
(v5):
- Modified del_consumer as per comments from Peter.
- Drop reference to inode before dropping reference to uprobe.
- Use i_size_read(inode) instead of inode->i_size.
- Ensure uprobe->consumers is NULL, before __uprobe_unregister() is called.
- Includes errno.h as recommended by Stephen Rothwell to fix a build issue
on sparc defconfig
- Remove restrictions while unregistering.
- Earlier code leaked inode references under some conditions while
registering/unregistering.
- Continue the vma-rmap walk even if the intermediate vma doesnt
meet the requirements.
- Validate the vma found by find_vma before inserting/removing the
breakpoint
- Call del_consumer under mutex_lock.
- Use hash locks.
- Handle mremap.
- Introduce find_least_offset_node() instead of close match logic in
find_uprobe
- Uprobes no more depends on MM_OWNER; No reference to task_structs
while inserting/removing a probe.
- Uses read_mapping_page instead of grab_cache_page so that the pages
have valid content.
- pass NULL to get_user_pages for the task parameter.
- call SetPageUptodate on the new page allocated in write_opcode.
- fix leaking a reference to the new page under certain conditions.
- Include Instruction Decoder if Uprobes gets defined.
- Remove const attributes for instruction prefix arrays.
- Uses mm_context to know if the application is 32 bit.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Also-written-by: Jim Keniston <jkenisto@us.ibm.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Roland McGrath <roland@hack.frob.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Anton Arapov <anton@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Denys Vlasenko <vda.linux@googlemail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linux-mm <linux-mm@kvack.org>
Link: http://lkml.kernel.org/r/20120209092642.GE16600@linux.vnet.ibm.com
[ Made various small edits to the commit log ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-02-09 17:26:42 +08:00
|
|
|
*
|
|
|
|
* This program is free software; you can redistribute it and/or modify
|
|
|
|
* it under the terms of the GNU General Public License as published by
|
|
|
|
* the Free Software Foundation; either version 2 of the License, or
|
|
|
|
* (at your option) any later version.
|
|
|
|
*
|
|
|
|
* This program is distributed in the hope that it will be useful,
|
|
|
|
* but WITHOUT ANY WARRANTY; without even the implied warranty of
|
|
|
|
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
|
|
|
|
* GNU General Public License for more details.
|
|
|
|
*
|
|
|
|
* You should have received a copy of the GNU General Public License
|
|
|
|
* along with this program; if not, write to the Free Software
|
|
|
|
* Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
|
|
|
|
*
|
2012-02-22 18:37:29 +08:00
|
|
|
* Copyright (C) IBM Corporation, 2008-2012
|
uprobes, mm, x86: Add the ability to install and remove uprobes breakpoints
Add uprobes support to the core kernel, with x86 support.
This commit adds the kernel facilities, the actual uprobes
user-space ABI and perf probe support comes in later commits.
General design:
Uprobes are maintained in an rb-tree indexed by inode and offset
(the offset here is from the start of the mapping). For a unique
(inode, offset) tuple, there can be at most one uprobe in the
rb-tree.
Since the (inode, offset) tuple identifies a unique uprobe, more
than one user may be interested in the same uprobe. This provides
the ability to connect multiple 'consumers' to the same uprobe.
Each consumer defines a handler and a filter (optional). The
'handler' is run every time the uprobe is hit, if it matches the
'filter' criteria.
The first consumer of a uprobe causes the breakpoint to be
inserted at the specified address and subsequent consumers are
appended to this list. On subsequent probes, the consumer gets
appended to the existing list of consumers. The breakpoint is
removed when the last consumer unregisters. For all other
unregisterations, the consumer is removed from the list of
consumers.
Given a inode, we get a list of the mms that have mapped the
inode. Do the actual registration if mm maps the page where a
probe needs to be inserted/removed.
We use a temporary list to walk through the vmas that map the
inode.
- The number of maps that map the inode, is not known before we
walk the rmap and keeps changing.
- extending vm_area_struct wasn't recommended, it's a
size-critical data structure.
- There can be more than one maps of the inode in the same mm.
We add callbacks to the mmap methods to keep an eye on text vmas
that are of interest to uprobes. When a vma of interest is mapped,
we insert the breakpoint at the right address.
Uprobe works by replacing the instruction at the address defined
by (inode, offset) with the arch specific breakpoint
instruction. We save a copy of the original instruction at the
uprobed address.
This is needed for:
a. executing the instruction out-of-line (xol).
b. instruction analysis for any subsequent fixups.
c. restoring the instruction back when the uprobe is unregistered.
We insert or delete a breakpoint instruction, and this
breakpoint instruction is assumed to be the smallest instruction
available on the platform. For fixed size instruction platforms
this is trivially true, for variable size instruction platforms
the breakpoint instruction is typically the smallest (often a
single byte).
Writing the instruction is done by COWing the page and changing
the instruction during the copy, this even though most platforms
allow atomic writes of the breakpoint instruction. This also
mirrors the behaviour of a ptrace() memory write to a PRIVATE
file map.
The core worker is derived from KSM's replace_page() logic.
In essence, similar to KSM:
a. allocate a new page and copy over contents of the page that
has the uprobed vaddr
b. modify the copy and insert the breakpoint at the required
address
c. switch the original page with the copy containing the
breakpoint
d. flush page tables.
replace_page() is being replicated here because of some minor
changes in the type of pages and also because Hugh Dickins had
plans to improve replace_page() for KSM specific work.
Instruction analysis on x86 is based on instruction decoder and
determines if an instruction can be probed and determines the
necessary fixups after singlestep. Instruction analysis is done
at probe insertion time so that we avoid having to repeat the
same analysis every time a probe is hit.
A lot of code here is due to the improvement/suggestions/inputs
from Peter Zijlstra.
Changelog:
(v10):
- Add code to clear REX.B prefix as suggested by Denys Vlasenko
and Masami Hiramatsu.
(v9):
- Use insn_offset_modrm as suggested by Masami Hiramatsu.
(v7):
Handle comments from Peter Zijlstra:
- Dont take reference to inode. (expect inode to uprobe_register to be sane).
- Use PTR_ERR to set the return value.
- No need to take reference to inode.
- use PTR_ERR to return error value.
- register and uprobe_unregister share code.
(v5):
- Modified del_consumer as per comments from Peter.
- Drop reference to inode before dropping reference to uprobe.
- Use i_size_read(inode) instead of inode->i_size.
- Ensure uprobe->consumers is NULL, before __uprobe_unregister() is called.
- Includes errno.h as recommended by Stephen Rothwell to fix a build issue
on sparc defconfig
- Remove restrictions while unregistering.
- Earlier code leaked inode references under some conditions while
registering/unregistering.
- Continue the vma-rmap walk even if the intermediate vma doesnt
meet the requirements.
- Validate the vma found by find_vma before inserting/removing the
breakpoint
- Call del_consumer under mutex_lock.
- Use hash locks.
- Handle mremap.
- Introduce find_least_offset_node() instead of close match logic in
find_uprobe
- Uprobes no more depends on MM_OWNER; No reference to task_structs
while inserting/removing a probe.
- Uses read_mapping_page instead of grab_cache_page so that the pages
have valid content.
- pass NULL to get_user_pages for the task parameter.
- call SetPageUptodate on the new page allocated in write_opcode.
- fix leaking a reference to the new page under certain conditions.
- Include Instruction Decoder if Uprobes gets defined.
- Remove const attributes for instruction prefix arrays.
- Uses mm_context to know if the application is 32 bit.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Also-written-by: Jim Keniston <jkenisto@us.ibm.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Roland McGrath <roland@hack.frob.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Anton Arapov <anton@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Denys Vlasenko <vda.linux@googlemail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linux-mm <linux-mm@kvack.org>
Link: http://lkml.kernel.org/r/20120209092642.GE16600@linux.vnet.ibm.com
[ Made various small edits to the commit log ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-02-09 17:26:42 +08:00
|
|
|
* Authors:
|
|
|
|
* Srikar Dronamraju
|
|
|
|
* Jim Keniston
|
2015-11-16 18:08:45 +08:00
|
|
|
* Copyright (C) 2011-2012 Red Hat, Inc., Peter Zijlstra
|
uprobes, mm, x86: Add the ability to install and remove uprobes breakpoints
Add uprobes support to the core kernel, with x86 support.
This commit adds the kernel facilities, the actual uprobes
user-space ABI and perf probe support comes in later commits.
General design:
Uprobes are maintained in an rb-tree indexed by inode and offset
(the offset here is from the start of the mapping). For a unique
(inode, offset) tuple, there can be at most one uprobe in the
rb-tree.
Since the (inode, offset) tuple identifies a unique uprobe, more
than one user may be interested in the same uprobe. This provides
the ability to connect multiple 'consumers' to the same uprobe.
Each consumer defines a handler and a filter (optional). The
'handler' is run every time the uprobe is hit, if it matches the
'filter' criteria.
The first consumer of a uprobe causes the breakpoint to be
inserted at the specified address and subsequent consumers are
appended to this list. On subsequent probes, the consumer gets
appended to the existing list of consumers. The breakpoint is
removed when the last consumer unregisters. For all other
unregisterations, the consumer is removed from the list of
consumers.
Given a inode, we get a list of the mms that have mapped the
inode. Do the actual registration if mm maps the page where a
probe needs to be inserted/removed.
We use a temporary list to walk through the vmas that map the
inode.
- The number of maps that map the inode, is not known before we
walk the rmap and keeps changing.
- extending vm_area_struct wasn't recommended, it's a
size-critical data structure.
- There can be more than one maps of the inode in the same mm.
We add callbacks to the mmap methods to keep an eye on text vmas
that are of interest to uprobes. When a vma of interest is mapped,
we insert the breakpoint at the right address.
Uprobe works by replacing the instruction at the address defined
by (inode, offset) with the arch specific breakpoint
instruction. We save a copy of the original instruction at the
uprobed address.
This is needed for:
a. executing the instruction out-of-line (xol).
b. instruction analysis for any subsequent fixups.
c. restoring the instruction back when the uprobe is unregistered.
We insert or delete a breakpoint instruction, and this
breakpoint instruction is assumed to be the smallest instruction
available on the platform. For fixed size instruction platforms
this is trivially true, for variable size instruction platforms
the breakpoint instruction is typically the smallest (often a
single byte).
Writing the instruction is done by COWing the page and changing
the instruction during the copy, this even though most platforms
allow atomic writes of the breakpoint instruction. This also
mirrors the behaviour of a ptrace() memory write to a PRIVATE
file map.
The core worker is derived from KSM's replace_page() logic.
In essence, similar to KSM:
a. allocate a new page and copy over contents of the page that
has the uprobed vaddr
b. modify the copy and insert the breakpoint at the required
address
c. switch the original page with the copy containing the
breakpoint
d. flush page tables.
replace_page() is being replicated here because of some minor
changes in the type of pages and also because Hugh Dickins had
plans to improve replace_page() for KSM specific work.
Instruction analysis on x86 is based on instruction decoder and
determines if an instruction can be probed and determines the
necessary fixups after singlestep. Instruction analysis is done
at probe insertion time so that we avoid having to repeat the
same analysis every time a probe is hit.
A lot of code here is due to the improvement/suggestions/inputs
from Peter Zijlstra.
Changelog:
(v10):
- Add code to clear REX.B prefix as suggested by Denys Vlasenko
and Masami Hiramatsu.
(v9):
- Use insn_offset_modrm as suggested by Masami Hiramatsu.
(v7):
Handle comments from Peter Zijlstra:
- Dont take reference to inode. (expect inode to uprobe_register to be sane).
- Use PTR_ERR to set the return value.
- No need to take reference to inode.
- use PTR_ERR to return error value.
- register and uprobe_unregister share code.
(v5):
- Modified del_consumer as per comments from Peter.
- Drop reference to inode before dropping reference to uprobe.
- Use i_size_read(inode) instead of inode->i_size.
- Ensure uprobe->consumers is NULL, before __uprobe_unregister() is called.
- Includes errno.h as recommended by Stephen Rothwell to fix a build issue
on sparc defconfig
- Remove restrictions while unregistering.
- Earlier code leaked inode references under some conditions while
registering/unregistering.
- Continue the vma-rmap walk even if the intermediate vma doesnt
meet the requirements.
- Validate the vma found by find_vma before inserting/removing the
breakpoint
- Call del_consumer under mutex_lock.
- Use hash locks.
- Handle mremap.
- Introduce find_least_offset_node() instead of close match logic in
find_uprobe
- Uprobes no more depends on MM_OWNER; No reference to task_structs
while inserting/removing a probe.
- Uses read_mapping_page instead of grab_cache_page so that the pages
have valid content.
- pass NULL to get_user_pages for the task parameter.
- call SetPageUptodate on the new page allocated in write_opcode.
- fix leaking a reference to the new page under certain conditions.
- Include Instruction Decoder if Uprobes gets defined.
- Remove const attributes for instruction prefix arrays.
- Uses mm_context to know if the application is 32 bit.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Also-written-by: Jim Keniston <jkenisto@us.ibm.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Roland McGrath <roland@hack.frob.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Anton Arapov <anton@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Denys Vlasenko <vda.linux@googlemail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linux-mm <linux-mm@kvack.org>
Link: http://lkml.kernel.org/r/20120209092642.GE16600@linux.vnet.ibm.com
[ Made various small edits to the commit log ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-02-09 17:26:42 +08:00
|
|
|
*/
|
|
|
|
|
|
|
|
#include <linux/kernel.h>
|
|
|
|
#include <linux/highmem.h>
|
|
|
|
#include <linux/pagemap.h> /* read_mapping_page */
|
|
|
|
#include <linux/slab.h>
|
|
|
|
#include <linux/sched.h>
|
2017-02-09 01:51:29 +08:00
|
|
|
#include <linux/sched/mm.h>
|
2017-02-09 01:51:30 +08:00
|
|
|
#include <linux/sched/coredump.h>
|
2013-01-14 02:03:34 +08:00
|
|
|
#include <linux/export.h>
|
uprobes, mm, x86: Add the ability to install and remove uprobes breakpoints
Add uprobes support to the core kernel, with x86 support.
This commit adds the kernel facilities, the actual uprobes
user-space ABI and perf probe support comes in later commits.
General design:
Uprobes are maintained in an rb-tree indexed by inode and offset
(the offset here is from the start of the mapping). For a unique
(inode, offset) tuple, there can be at most one uprobe in the
rb-tree.
Since the (inode, offset) tuple identifies a unique uprobe, more
than one user may be interested in the same uprobe. This provides
the ability to connect multiple 'consumers' to the same uprobe.
Each consumer defines a handler and a filter (optional). The
'handler' is run every time the uprobe is hit, if it matches the
'filter' criteria.
The first consumer of a uprobe causes the breakpoint to be
inserted at the specified address and subsequent consumers are
appended to this list. On subsequent probes, the consumer gets
appended to the existing list of consumers. The breakpoint is
removed when the last consumer unregisters. For all other
unregisterations, the consumer is removed from the list of
consumers.
Given a inode, we get a list of the mms that have mapped the
inode. Do the actual registration if mm maps the page where a
probe needs to be inserted/removed.
We use a temporary list to walk through the vmas that map the
inode.
- The number of maps that map the inode, is not known before we
walk the rmap and keeps changing.
- extending vm_area_struct wasn't recommended, it's a
size-critical data structure.
- There can be more than one maps of the inode in the same mm.
We add callbacks to the mmap methods to keep an eye on text vmas
that are of interest to uprobes. When a vma of interest is mapped,
we insert the breakpoint at the right address.
Uprobe works by replacing the instruction at the address defined
by (inode, offset) with the arch specific breakpoint
instruction. We save a copy of the original instruction at the
uprobed address.
This is needed for:
a. executing the instruction out-of-line (xol).
b. instruction analysis for any subsequent fixups.
c. restoring the instruction back when the uprobe is unregistered.
We insert or delete a breakpoint instruction, and this
breakpoint instruction is assumed to be the smallest instruction
available on the platform. For fixed size instruction platforms
this is trivially true, for variable size instruction platforms
the breakpoint instruction is typically the smallest (often a
single byte).
Writing the instruction is done by COWing the page and changing
the instruction during the copy, this even though most platforms
allow atomic writes of the breakpoint instruction. This also
mirrors the behaviour of a ptrace() memory write to a PRIVATE
file map.
The core worker is derived from KSM's replace_page() logic.
In essence, similar to KSM:
a. allocate a new page and copy over contents of the page that
has the uprobed vaddr
b. modify the copy and insert the breakpoint at the required
address
c. switch the original page with the copy containing the
breakpoint
d. flush page tables.
replace_page() is being replicated here because of some minor
changes in the type of pages and also because Hugh Dickins had
plans to improve replace_page() for KSM specific work.
Instruction analysis on x86 is based on instruction decoder and
determines if an instruction can be probed and determines the
necessary fixups after singlestep. Instruction analysis is done
at probe insertion time so that we avoid having to repeat the
same analysis every time a probe is hit.
A lot of code here is due to the improvement/suggestions/inputs
from Peter Zijlstra.
Changelog:
(v10):
- Add code to clear REX.B prefix as suggested by Denys Vlasenko
and Masami Hiramatsu.
(v9):
- Use insn_offset_modrm as suggested by Masami Hiramatsu.
(v7):
Handle comments from Peter Zijlstra:
- Dont take reference to inode. (expect inode to uprobe_register to be sane).
- Use PTR_ERR to set the return value.
- No need to take reference to inode.
- use PTR_ERR to return error value.
- register and uprobe_unregister share code.
(v5):
- Modified del_consumer as per comments from Peter.
- Drop reference to inode before dropping reference to uprobe.
- Use i_size_read(inode) instead of inode->i_size.
- Ensure uprobe->consumers is NULL, before __uprobe_unregister() is called.
- Includes errno.h as recommended by Stephen Rothwell to fix a build issue
on sparc defconfig
- Remove restrictions while unregistering.
- Earlier code leaked inode references under some conditions while
registering/unregistering.
- Continue the vma-rmap walk even if the intermediate vma doesnt
meet the requirements.
- Validate the vma found by find_vma before inserting/removing the
breakpoint
- Call del_consumer under mutex_lock.
- Use hash locks.
- Handle mremap.
- Introduce find_least_offset_node() instead of close match logic in
find_uprobe
- Uprobes no more depends on MM_OWNER; No reference to task_structs
while inserting/removing a probe.
- Uses read_mapping_page instead of grab_cache_page so that the pages
have valid content.
- pass NULL to get_user_pages for the task parameter.
- call SetPageUptodate on the new page allocated in write_opcode.
- fix leaking a reference to the new page under certain conditions.
- Include Instruction Decoder if Uprobes gets defined.
- Remove const attributes for instruction prefix arrays.
- Uses mm_context to know if the application is 32 bit.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Also-written-by: Jim Keniston <jkenisto@us.ibm.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Roland McGrath <roland@hack.frob.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Anton Arapov <anton@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Denys Vlasenko <vda.linux@googlemail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linux-mm <linux-mm@kvack.org>
Link: http://lkml.kernel.org/r/20120209092642.GE16600@linux.vnet.ibm.com
[ Made various small edits to the commit log ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-02-09 17:26:42 +08:00
|
|
|
#include <linux/rmap.h> /* anon_vma_prepare */
|
|
|
|
#include <linux/mmu_notifier.h> /* set_pte_at_notify */
|
|
|
|
#include <linux/swap.h> /* try_to_free_swap */
|
uprobes/core: Handle breakpoint and singlestep exceptions
Uprobes uses exception notifiers to get to know if a thread hit
a breakpoint or a singlestep exception.
When a thread hits a uprobe or is singlestepping post a uprobe
hit, the uprobe exception notifier sets its TIF_UPROBE bit,
which will then be checked on its return to userspace path
(do_notify_resume() ->uprobe_notify_resume()), where the
consumers handlers are run (in task context) based on the
defined filters.
Uprobe hits are thread specific and hence we need to maintain
information about if a task hit a uprobe, what uprobe was hit,
the slot where the original instruction was copied for xol so
that it can be singlestepped with appropriate fixups.
In some cases, special care is needed for instructions that are
executed out of line (xol). These are architecture specific
artefacts, such as handling RIP relative instructions on x86_64.
Since the instruction at which the uprobe was inserted is
executed out of line, architecture specific fixups are added so
that the thread continues normal execution in the presence of a
uprobe.
Postpone the signals until we execute the probed insn.
post_xol() path does a recalc_sigpending() before return to
user-mode, this ensures the signal can't be lost.
Uprobes relies on DIE_DEBUG notification to notify if a
singlestep is complete.
Adds x86 specific uprobe exception notifiers and appropriate
hooks needed to determine a uprobe hit and subsequent post
processing.
Add requisite x86 fixups for xol for uprobes. Specific cases
needing fixups include relative jumps (x86_64), calls, etc.
Where possible, we check and skip singlestepping the
breakpointed instructions. For now we skip single byte as well
as few multibyte nop instructions. However this can be extended
to other instructions too.
Credits to Oleg Nesterov for suggestions/patches related to
signal, breakpoint, singlestep handling code.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Jim Keniston <jkenisto@linux.vnet.ibm.com>
Cc: Linux-mm <linux-mm@kvack.org>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/20120313180011.29771.89027.sendpatchset@srdronam.in.ibm.com
[ Performed various cleanliness edits ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-03-14 02:00:11 +08:00
|
|
|
#include <linux/ptrace.h> /* user_enable_single_step */
|
|
|
|
#include <linux/kdebug.h> /* notifier mechanism */
|
2012-07-30 02:22:49 +08:00
|
|
|
#include "../../mm/internal.h" /* munlock_vma_page */
|
2012-11-15 02:03:42 +08:00
|
|
|
#include <linux/percpu-rwsem.h>
|
uprobes: Change uprobe_copy_process() to dup xol_area
This finally fixes the serious bug in uretprobes: a forked child
crashes if the parent called fork() with the pending ret probe.
Trivial test-case:
# perf probe -x /lib/libc.so.6 __fork%return
# perf record -e probe_libc:__fork perl -le 'fork || print "OK"'
(the child doesn't print "OK", it is killed by SIGSEGV)
If the child returns from the probed function it actually returns
to trampoline_vaddr, because it got the copy of parent's stack
mangled by prepare_uretprobe() when the parent entered this func.
It crashes because a) this address is not mapped and b) until the
previous change it doesn't have the proper->return_instances info.
This means that uprobe_copy_process() has to create xol_area which
has the trampoline slot, and its vaddr should be equal to parent's
xol_area->vaddr.
Unfortunately, uprobe_copy_process() can not simply do
__create_xol_area(child, xol_area->vaddr). This could actually work
but perf_event_mmap() doesn't expect the usage of foreign ->mm. So
we offload this to task_work_run(), and pass the argument via not
yet used utask->vaddr.
We know that this vaddr is fine for install_special_mapping(), the
necessary hole was recently "created" by dup_mmap() which skips the
parent's VM_DONTCOPY area, and nobody else could use the new mm.
Unfortunately, this also means that we can not handle the errors
properly, we obviously can not abort the already completed fork().
So we simply print the warning if GFP_KERNEL allocation (the only
possible reason) fails.
Reported-by: Martin Cermak <mcermak@redhat.com>
Reported-by: David Smith <dsmith@redhat.com>
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Acked-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
2013-10-14 03:18:44 +08:00
|
|
|
#include <linux/task_work.h>
|
2014-05-20 02:41:36 +08:00
|
|
|
#include <linux/shmem_fs.h>
|
2012-02-17 16:27:41 +08:00
|
|
|
|
uprobes, mm, x86: Add the ability to install and remove uprobes breakpoints
Add uprobes support to the core kernel, with x86 support.
This commit adds the kernel facilities, the actual uprobes
user-space ABI and perf probe support comes in later commits.
General design:
Uprobes are maintained in an rb-tree indexed by inode and offset
(the offset here is from the start of the mapping). For a unique
(inode, offset) tuple, there can be at most one uprobe in the
rb-tree.
Since the (inode, offset) tuple identifies a unique uprobe, more
than one user may be interested in the same uprobe. This provides
the ability to connect multiple 'consumers' to the same uprobe.
Each consumer defines a handler and a filter (optional). The
'handler' is run every time the uprobe is hit, if it matches the
'filter' criteria.
The first consumer of a uprobe causes the breakpoint to be
inserted at the specified address and subsequent consumers are
appended to this list. On subsequent probes, the consumer gets
appended to the existing list of consumers. The breakpoint is
removed when the last consumer unregisters. For all other
unregisterations, the consumer is removed from the list of
consumers.
Given a inode, we get a list of the mms that have mapped the
inode. Do the actual registration if mm maps the page where a
probe needs to be inserted/removed.
We use a temporary list to walk through the vmas that map the
inode.
- The number of maps that map the inode, is not known before we
walk the rmap and keeps changing.
- extending vm_area_struct wasn't recommended, it's a
size-critical data structure.
- There can be more than one maps of the inode in the same mm.
We add callbacks to the mmap methods to keep an eye on text vmas
that are of interest to uprobes. When a vma of interest is mapped,
we insert the breakpoint at the right address.
Uprobe works by replacing the instruction at the address defined
by (inode, offset) with the arch specific breakpoint
instruction. We save a copy of the original instruction at the
uprobed address.
This is needed for:
a. executing the instruction out-of-line (xol).
b. instruction analysis for any subsequent fixups.
c. restoring the instruction back when the uprobe is unregistered.
We insert or delete a breakpoint instruction, and this
breakpoint instruction is assumed to be the smallest instruction
available on the platform. For fixed size instruction platforms
this is trivially true, for variable size instruction platforms
the breakpoint instruction is typically the smallest (often a
single byte).
Writing the instruction is done by COWing the page and changing
the instruction during the copy, this even though most platforms
allow atomic writes of the breakpoint instruction. This also
mirrors the behaviour of a ptrace() memory write to a PRIVATE
file map.
The core worker is derived from KSM's replace_page() logic.
In essence, similar to KSM:
a. allocate a new page and copy over contents of the page that
has the uprobed vaddr
b. modify the copy and insert the breakpoint at the required
address
c. switch the original page with the copy containing the
breakpoint
d. flush page tables.
replace_page() is being replicated here because of some minor
changes in the type of pages and also because Hugh Dickins had
plans to improve replace_page() for KSM specific work.
Instruction analysis on x86 is based on instruction decoder and
determines if an instruction can be probed and determines the
necessary fixups after singlestep. Instruction analysis is done
at probe insertion time so that we avoid having to repeat the
same analysis every time a probe is hit.
A lot of code here is due to the improvement/suggestions/inputs
from Peter Zijlstra.
Changelog:
(v10):
- Add code to clear REX.B prefix as suggested by Denys Vlasenko
and Masami Hiramatsu.
(v9):
- Use insn_offset_modrm as suggested by Masami Hiramatsu.
(v7):
Handle comments from Peter Zijlstra:
- Dont take reference to inode. (expect inode to uprobe_register to be sane).
- Use PTR_ERR to set the return value.
- No need to take reference to inode.
- use PTR_ERR to return error value.
- register and uprobe_unregister share code.
(v5):
- Modified del_consumer as per comments from Peter.
- Drop reference to inode before dropping reference to uprobe.
- Use i_size_read(inode) instead of inode->i_size.
- Ensure uprobe->consumers is NULL, before __uprobe_unregister() is called.
- Includes errno.h as recommended by Stephen Rothwell to fix a build issue
on sparc defconfig
- Remove restrictions while unregistering.
- Earlier code leaked inode references under some conditions while
registering/unregistering.
- Continue the vma-rmap walk even if the intermediate vma doesnt
meet the requirements.
- Validate the vma found by find_vma before inserting/removing the
breakpoint
- Call del_consumer under mutex_lock.
- Use hash locks.
- Handle mremap.
- Introduce find_least_offset_node() instead of close match logic in
find_uprobe
- Uprobes no more depends on MM_OWNER; No reference to task_structs
while inserting/removing a probe.
- Uses read_mapping_page instead of grab_cache_page so that the pages
have valid content.
- pass NULL to get_user_pages for the task parameter.
- call SetPageUptodate on the new page allocated in write_opcode.
- fix leaking a reference to the new page under certain conditions.
- Include Instruction Decoder if Uprobes gets defined.
- Remove const attributes for instruction prefix arrays.
- Uses mm_context to know if the application is 32 bit.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Also-written-by: Jim Keniston <jkenisto@us.ibm.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Roland McGrath <roland@hack.frob.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Anton Arapov <anton@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Denys Vlasenko <vda.linux@googlemail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linux-mm <linux-mm@kvack.org>
Link: http://lkml.kernel.org/r/20120209092642.GE16600@linux.vnet.ibm.com
[ Made various small edits to the commit log ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-02-09 17:26:42 +08:00
|
|
|
#include <linux/uprobes.h>
|
|
|
|
|
2012-03-31 02:26:31 +08:00
|
|
|
#define UINSNS_PER_PAGE (PAGE_SIZE/UPROBE_XOL_SLOT_BYTES)
|
|
|
|
#define MAX_UPROBE_XOL_SLOTS UINSNS_PER_PAGE
|
|
|
|
|
uprobes, mm, x86: Add the ability to install and remove uprobes breakpoints
Add uprobes support to the core kernel, with x86 support.
This commit adds the kernel facilities, the actual uprobes
user-space ABI and perf probe support comes in later commits.
General design:
Uprobes are maintained in an rb-tree indexed by inode and offset
(the offset here is from the start of the mapping). For a unique
(inode, offset) tuple, there can be at most one uprobe in the
rb-tree.
Since the (inode, offset) tuple identifies a unique uprobe, more
than one user may be interested in the same uprobe. This provides
the ability to connect multiple 'consumers' to the same uprobe.
Each consumer defines a handler and a filter (optional). The
'handler' is run every time the uprobe is hit, if it matches the
'filter' criteria.
The first consumer of a uprobe causes the breakpoint to be
inserted at the specified address and subsequent consumers are
appended to this list. On subsequent probes, the consumer gets
appended to the existing list of consumers. The breakpoint is
removed when the last consumer unregisters. For all other
unregisterations, the consumer is removed from the list of
consumers.
Given a inode, we get a list of the mms that have mapped the
inode. Do the actual registration if mm maps the page where a
probe needs to be inserted/removed.
We use a temporary list to walk through the vmas that map the
inode.
- The number of maps that map the inode, is not known before we
walk the rmap and keeps changing.
- extending vm_area_struct wasn't recommended, it's a
size-critical data structure.
- There can be more than one maps of the inode in the same mm.
We add callbacks to the mmap methods to keep an eye on text vmas
that are of interest to uprobes. When a vma of interest is mapped,
we insert the breakpoint at the right address.
Uprobe works by replacing the instruction at the address defined
by (inode, offset) with the arch specific breakpoint
instruction. We save a copy of the original instruction at the
uprobed address.
This is needed for:
a. executing the instruction out-of-line (xol).
b. instruction analysis for any subsequent fixups.
c. restoring the instruction back when the uprobe is unregistered.
We insert or delete a breakpoint instruction, and this
breakpoint instruction is assumed to be the smallest instruction
available on the platform. For fixed size instruction platforms
this is trivially true, for variable size instruction platforms
the breakpoint instruction is typically the smallest (often a
single byte).
Writing the instruction is done by COWing the page and changing
the instruction during the copy, this even though most platforms
allow atomic writes of the breakpoint instruction. This also
mirrors the behaviour of a ptrace() memory write to a PRIVATE
file map.
The core worker is derived from KSM's replace_page() logic.
In essence, similar to KSM:
a. allocate a new page and copy over contents of the page that
has the uprobed vaddr
b. modify the copy and insert the breakpoint at the required
address
c. switch the original page with the copy containing the
breakpoint
d. flush page tables.
replace_page() is being replicated here because of some minor
changes in the type of pages and also because Hugh Dickins had
plans to improve replace_page() for KSM specific work.
Instruction analysis on x86 is based on instruction decoder and
determines if an instruction can be probed and determines the
necessary fixups after singlestep. Instruction analysis is done
at probe insertion time so that we avoid having to repeat the
same analysis every time a probe is hit.
A lot of code here is due to the improvement/suggestions/inputs
from Peter Zijlstra.
Changelog:
(v10):
- Add code to clear REX.B prefix as suggested by Denys Vlasenko
and Masami Hiramatsu.
(v9):
- Use insn_offset_modrm as suggested by Masami Hiramatsu.
(v7):
Handle comments from Peter Zijlstra:
- Dont take reference to inode. (expect inode to uprobe_register to be sane).
- Use PTR_ERR to set the return value.
- No need to take reference to inode.
- use PTR_ERR to return error value.
- register and uprobe_unregister share code.
(v5):
- Modified del_consumer as per comments from Peter.
- Drop reference to inode before dropping reference to uprobe.
- Use i_size_read(inode) instead of inode->i_size.
- Ensure uprobe->consumers is NULL, before __uprobe_unregister() is called.
- Includes errno.h as recommended by Stephen Rothwell to fix a build issue
on sparc defconfig
- Remove restrictions while unregistering.
- Earlier code leaked inode references under some conditions while
registering/unregistering.
- Continue the vma-rmap walk even if the intermediate vma doesnt
meet the requirements.
- Validate the vma found by find_vma before inserting/removing the
breakpoint
- Call del_consumer under mutex_lock.
- Use hash locks.
- Handle mremap.
- Introduce find_least_offset_node() instead of close match logic in
find_uprobe
- Uprobes no more depends on MM_OWNER; No reference to task_structs
while inserting/removing a probe.
- Uses read_mapping_page instead of grab_cache_page so that the pages
have valid content.
- pass NULL to get_user_pages for the task parameter.
- call SetPageUptodate on the new page allocated in write_opcode.
- fix leaking a reference to the new page under certain conditions.
- Include Instruction Decoder if Uprobes gets defined.
- Remove const attributes for instruction prefix arrays.
- Uses mm_context to know if the application is 32 bit.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Also-written-by: Jim Keniston <jkenisto@us.ibm.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Roland McGrath <roland@hack.frob.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Anton Arapov <anton@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Denys Vlasenko <vda.linux@googlemail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linux-mm <linux-mm@kvack.org>
Link: http://lkml.kernel.org/r/20120209092642.GE16600@linux.vnet.ibm.com
[ Made various small edits to the commit log ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-02-09 17:26:42 +08:00
|
|
|
static struct rb_root uprobes_tree = RB_ROOT;
|
2012-11-26 02:54:29 +08:00
|
|
|
/*
|
|
|
|
* allows us to skip the uprobe_mmap if there are no uprobe events active
|
|
|
|
* at this time. Probably a fine grained per inode count is better?
|
|
|
|
*/
|
|
|
|
#define no_uprobe_events() RB_EMPTY_ROOT(&uprobes_tree)
|
2012-02-17 16:27:41 +08:00
|
|
|
|
uprobes, mm, x86: Add the ability to install and remove uprobes breakpoints
Add uprobes support to the core kernel, with x86 support.
This commit adds the kernel facilities, the actual uprobes
user-space ABI and perf probe support comes in later commits.
General design:
Uprobes are maintained in an rb-tree indexed by inode and offset
(the offset here is from the start of the mapping). For a unique
(inode, offset) tuple, there can be at most one uprobe in the
rb-tree.
Since the (inode, offset) tuple identifies a unique uprobe, more
than one user may be interested in the same uprobe. This provides
the ability to connect multiple 'consumers' to the same uprobe.
Each consumer defines a handler and a filter (optional). The
'handler' is run every time the uprobe is hit, if it matches the
'filter' criteria.
The first consumer of a uprobe causes the breakpoint to be
inserted at the specified address and subsequent consumers are
appended to this list. On subsequent probes, the consumer gets
appended to the existing list of consumers. The breakpoint is
removed when the last consumer unregisters. For all other
unregisterations, the consumer is removed from the list of
consumers.
Given a inode, we get a list of the mms that have mapped the
inode. Do the actual registration if mm maps the page where a
probe needs to be inserted/removed.
We use a temporary list to walk through the vmas that map the
inode.
- The number of maps that map the inode, is not known before we
walk the rmap and keeps changing.
- extending vm_area_struct wasn't recommended, it's a
size-critical data structure.
- There can be more than one maps of the inode in the same mm.
We add callbacks to the mmap methods to keep an eye on text vmas
that are of interest to uprobes. When a vma of interest is mapped,
we insert the breakpoint at the right address.
Uprobe works by replacing the instruction at the address defined
by (inode, offset) with the arch specific breakpoint
instruction. We save a copy of the original instruction at the
uprobed address.
This is needed for:
a. executing the instruction out-of-line (xol).
b. instruction analysis for any subsequent fixups.
c. restoring the instruction back when the uprobe is unregistered.
We insert or delete a breakpoint instruction, and this
breakpoint instruction is assumed to be the smallest instruction
available on the platform. For fixed size instruction platforms
this is trivially true, for variable size instruction platforms
the breakpoint instruction is typically the smallest (often a
single byte).
Writing the instruction is done by COWing the page and changing
the instruction during the copy, this even though most platforms
allow atomic writes of the breakpoint instruction. This also
mirrors the behaviour of a ptrace() memory write to a PRIVATE
file map.
The core worker is derived from KSM's replace_page() logic.
In essence, similar to KSM:
a. allocate a new page and copy over contents of the page that
has the uprobed vaddr
b. modify the copy and insert the breakpoint at the required
address
c. switch the original page with the copy containing the
breakpoint
d. flush page tables.
replace_page() is being replicated here because of some minor
changes in the type of pages and also because Hugh Dickins had
plans to improve replace_page() for KSM specific work.
Instruction analysis on x86 is based on instruction decoder and
determines if an instruction can be probed and determines the
necessary fixups after singlestep. Instruction analysis is done
at probe insertion time so that we avoid having to repeat the
same analysis every time a probe is hit.
A lot of code here is due to the improvement/suggestions/inputs
from Peter Zijlstra.
Changelog:
(v10):
- Add code to clear REX.B prefix as suggested by Denys Vlasenko
and Masami Hiramatsu.
(v9):
- Use insn_offset_modrm as suggested by Masami Hiramatsu.
(v7):
Handle comments from Peter Zijlstra:
- Dont take reference to inode. (expect inode to uprobe_register to be sane).
- Use PTR_ERR to set the return value.
- No need to take reference to inode.
- use PTR_ERR to return error value.
- register and uprobe_unregister share code.
(v5):
- Modified del_consumer as per comments from Peter.
- Drop reference to inode before dropping reference to uprobe.
- Use i_size_read(inode) instead of inode->i_size.
- Ensure uprobe->consumers is NULL, before __uprobe_unregister() is called.
- Includes errno.h as recommended by Stephen Rothwell to fix a build issue
on sparc defconfig
- Remove restrictions while unregistering.
- Earlier code leaked inode references under some conditions while
registering/unregistering.
- Continue the vma-rmap walk even if the intermediate vma doesnt
meet the requirements.
- Validate the vma found by find_vma before inserting/removing the
breakpoint
- Call del_consumer under mutex_lock.
- Use hash locks.
- Handle mremap.
- Introduce find_least_offset_node() instead of close match logic in
find_uprobe
- Uprobes no more depends on MM_OWNER; No reference to task_structs
while inserting/removing a probe.
- Uses read_mapping_page instead of grab_cache_page so that the pages
have valid content.
- pass NULL to get_user_pages for the task parameter.
- call SetPageUptodate on the new page allocated in write_opcode.
- fix leaking a reference to the new page under certain conditions.
- Include Instruction Decoder if Uprobes gets defined.
- Remove const attributes for instruction prefix arrays.
- Uses mm_context to know if the application is 32 bit.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Also-written-by: Jim Keniston <jkenisto@us.ibm.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Roland McGrath <roland@hack.frob.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Anton Arapov <anton@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Denys Vlasenko <vda.linux@googlemail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linux-mm <linux-mm@kvack.org>
Link: http://lkml.kernel.org/r/20120209092642.GE16600@linux.vnet.ibm.com
[ Made various small edits to the commit log ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-02-09 17:26:42 +08:00
|
|
|
static DEFINE_SPINLOCK(uprobes_treelock); /* serialize rbtree access */
|
|
|
|
|
|
|
|
#define UPROBES_HASH_SZ 13
|
|
|
|
/* serialize uprobe->pending_list */
|
|
|
|
static struct mutex uprobes_mmap_mutex[UPROBES_HASH_SZ];
|
2012-02-17 16:27:41 +08:00
|
|
|
#define uprobes_mmap_hash(v) (&uprobes_mmap_mutex[((unsigned long)(v)) % UPROBES_HASH_SZ])
|
uprobes, mm, x86: Add the ability to install and remove uprobes breakpoints
Add uprobes support to the core kernel, with x86 support.
This commit adds the kernel facilities, the actual uprobes
user-space ABI and perf probe support comes in later commits.
General design:
Uprobes are maintained in an rb-tree indexed by inode and offset
(the offset here is from the start of the mapping). For a unique
(inode, offset) tuple, there can be at most one uprobe in the
rb-tree.
Since the (inode, offset) tuple identifies a unique uprobe, more
than one user may be interested in the same uprobe. This provides
the ability to connect multiple 'consumers' to the same uprobe.
Each consumer defines a handler and a filter (optional). The
'handler' is run every time the uprobe is hit, if it matches the
'filter' criteria.
The first consumer of a uprobe causes the breakpoint to be
inserted at the specified address and subsequent consumers are
appended to this list. On subsequent probes, the consumer gets
appended to the existing list of consumers. The breakpoint is
removed when the last consumer unregisters. For all other
unregisterations, the consumer is removed from the list of
consumers.
Given a inode, we get a list of the mms that have mapped the
inode. Do the actual registration if mm maps the page where a
probe needs to be inserted/removed.
We use a temporary list to walk through the vmas that map the
inode.
- The number of maps that map the inode, is not known before we
walk the rmap and keeps changing.
- extending vm_area_struct wasn't recommended, it's a
size-critical data structure.
- There can be more than one maps of the inode in the same mm.
We add callbacks to the mmap methods to keep an eye on text vmas
that are of interest to uprobes. When a vma of interest is mapped,
we insert the breakpoint at the right address.
Uprobe works by replacing the instruction at the address defined
by (inode, offset) with the arch specific breakpoint
instruction. We save a copy of the original instruction at the
uprobed address.
This is needed for:
a. executing the instruction out-of-line (xol).
b. instruction analysis for any subsequent fixups.
c. restoring the instruction back when the uprobe is unregistered.
We insert or delete a breakpoint instruction, and this
breakpoint instruction is assumed to be the smallest instruction
available on the platform. For fixed size instruction platforms
this is trivially true, for variable size instruction platforms
the breakpoint instruction is typically the smallest (often a
single byte).
Writing the instruction is done by COWing the page and changing
the instruction during the copy, this even though most platforms
allow atomic writes of the breakpoint instruction. This also
mirrors the behaviour of a ptrace() memory write to a PRIVATE
file map.
The core worker is derived from KSM's replace_page() logic.
In essence, similar to KSM:
a. allocate a new page and copy over contents of the page that
has the uprobed vaddr
b. modify the copy and insert the breakpoint at the required
address
c. switch the original page with the copy containing the
breakpoint
d. flush page tables.
replace_page() is being replicated here because of some minor
changes in the type of pages and also because Hugh Dickins had
plans to improve replace_page() for KSM specific work.
Instruction analysis on x86 is based on instruction decoder and
determines if an instruction can be probed and determines the
necessary fixups after singlestep. Instruction analysis is done
at probe insertion time so that we avoid having to repeat the
same analysis every time a probe is hit.
A lot of code here is due to the improvement/suggestions/inputs
from Peter Zijlstra.
Changelog:
(v10):
- Add code to clear REX.B prefix as suggested by Denys Vlasenko
and Masami Hiramatsu.
(v9):
- Use insn_offset_modrm as suggested by Masami Hiramatsu.
(v7):
Handle comments from Peter Zijlstra:
- Dont take reference to inode. (expect inode to uprobe_register to be sane).
- Use PTR_ERR to set the return value.
- No need to take reference to inode.
- use PTR_ERR to return error value.
- register and uprobe_unregister share code.
(v5):
- Modified del_consumer as per comments from Peter.
- Drop reference to inode before dropping reference to uprobe.
- Use i_size_read(inode) instead of inode->i_size.
- Ensure uprobe->consumers is NULL, before __uprobe_unregister() is called.
- Includes errno.h as recommended by Stephen Rothwell to fix a build issue
on sparc defconfig
- Remove restrictions while unregistering.
- Earlier code leaked inode references under some conditions while
registering/unregistering.
- Continue the vma-rmap walk even if the intermediate vma doesnt
meet the requirements.
- Validate the vma found by find_vma before inserting/removing the
breakpoint
- Call del_consumer under mutex_lock.
- Use hash locks.
- Handle mremap.
- Introduce find_least_offset_node() instead of close match logic in
find_uprobe
- Uprobes no more depends on MM_OWNER; No reference to task_structs
while inserting/removing a probe.
- Uses read_mapping_page instead of grab_cache_page so that the pages
have valid content.
- pass NULL to get_user_pages for the task parameter.
- call SetPageUptodate on the new page allocated in write_opcode.
- fix leaking a reference to the new page under certain conditions.
- Include Instruction Decoder if Uprobes gets defined.
- Remove const attributes for instruction prefix arrays.
- Uses mm_context to know if the application is 32 bit.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Also-written-by: Jim Keniston <jkenisto@us.ibm.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Roland McGrath <roland@hack.frob.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Anton Arapov <anton@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Denys Vlasenko <vda.linux@googlemail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linux-mm <linux-mm@kvack.org>
Link: http://lkml.kernel.org/r/20120209092642.GE16600@linux.vnet.ibm.com
[ Made various small edits to the commit log ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-02-09 17:26:42 +08:00
|
|
|
|
2012-11-15 02:03:42 +08:00
|
|
|
static struct percpu_rw_semaphore dup_mmap_sem;
|
|
|
|
|
2012-10-01 02:11:45 +08:00
|
|
|
/* Have a copy of original instruction */
|
2012-10-01 03:12:44 +08:00
|
|
|
#define UPROBE_COPY_INSN 0
|
2012-10-01 02:11:45 +08:00
|
|
|
|
2012-02-22 17:16:02 +08:00
|
|
|
struct uprobe {
|
|
|
|
struct rb_node rb_node; /* node in the rb tree */
|
|
|
|
atomic_t ref;
|
2012-11-25 00:29:40 +08:00
|
|
|
struct rw_semaphore register_rwsem;
|
2012-02-22 17:16:02 +08:00
|
|
|
struct rw_semaphore consumer_rwsem;
|
|
|
|
struct list_head pending_list;
|
|
|
|
struct uprobe_consumer *consumers;
|
|
|
|
struct inode *inode; /* Also hold a ref to inode */
|
|
|
|
loff_t offset;
|
uprobes: Support SDT markers having reference count (semaphore)
Userspace Statically Defined Tracepoints[1] are dtrace style markers
inside userspace applications. Applications like PostgreSQL, MySQL,
Pthread, Perl, Python, Java, Ruby, Node.js, libvirt, QEMU, glib etc
have these markers embedded in them. These markers are added by developer
at important places in the code. Each marker source expands to a single
nop instruction in the compiled code but there may be additional
overhead for computing the marker arguments which expands to couple of
instructions. In case the overhead is more, execution of it can be
omitted by runtime if() condition when no one is tracing on the marker:
if (reference_counter > 0) {
Execute marker instructions;
}
Default value of reference counter is 0. Tracer has to increment the
reference counter before tracing on a marker and decrement it when
done with the tracing.
Implement the reference counter logic in core uprobe. User will be
able to use it from trace_uprobe as well as from kernel module. New
trace_uprobe definition with reference counter will now be:
<path>:<offset>[(ref_ctr_offset)]
where ref_ctr_offset is an optional field. For kernel module, new
variant of uprobe_register() has been introduced:
uprobe_register_refctr(inode, offset, ref_ctr_offset, consumer)
No new variant for uprobe_unregister() because it's assumed to have
only one reference counter for one uprobe.
[1] https://sourceware.org/systemtap/wiki/UserSpaceProbeImplementation
Note: 'reference counter' is called as 'semaphore' in original Dtrace
(or Systemtap, bcc and even in ELF) documentation and code. But the
term 'semaphore' is misleading in this context. This is just a counter
used to hold number of tracers tracing on a marker. This is not really
used for any synchronization. So we are calling it a 'reference counter'
in kernel / perf code.
Link: http://lkml.kernel.org/r/20180820044250.11659-2-ravi.bangoria@linux.ibm.com
Reviewed-by: Masami Hiramatsu <mhiramat@kernel.org>
[Only trace_uprobe.c]
Reviewed-by: Oleg Nesterov <oleg@redhat.com>
Reviewed-by: Song Liu <songliubraving@fb.com>
Tested-by: Song Liu <songliubraving@fb.com>
Signed-off-by: Ravi Bangoria <ravi.bangoria@linux.ibm.com>
Signed-off-by: Steven Rostedt (VMware) <rostedt@goodmis.org>
2018-08-20 12:42:47 +08:00
|
|
|
loff_t ref_ctr_offset;
|
2012-10-01 03:12:44 +08:00
|
|
|
unsigned long flags;
|
2013-11-20 00:20:21 +08:00
|
|
|
|
|
|
|
/*
|
|
|
|
* The generic code assumes that it has two members of unknown type
|
|
|
|
* owned by the arch-specific code:
|
|
|
|
*
|
|
|
|
* insn - copy_insn() saves the original instruction here for
|
|
|
|
* arch_uprobe_analyze_insn().
|
|
|
|
*
|
|
|
|
* ixol - potentially modified instruction to execute out of
|
|
|
|
* line, copied to xol_area by xol_get_insn_slot().
|
|
|
|
*/
|
2012-02-22 17:16:02 +08:00
|
|
|
struct arch_uprobe arch;
|
|
|
|
};
|
|
|
|
|
uprobes: Support SDT markers having reference count (semaphore)
Userspace Statically Defined Tracepoints[1] are dtrace style markers
inside userspace applications. Applications like PostgreSQL, MySQL,
Pthread, Perl, Python, Java, Ruby, Node.js, libvirt, QEMU, glib etc
have these markers embedded in them. These markers are added by developer
at important places in the code. Each marker source expands to a single
nop instruction in the compiled code but there may be additional
overhead for computing the marker arguments which expands to couple of
instructions. In case the overhead is more, execution of it can be
omitted by runtime if() condition when no one is tracing on the marker:
if (reference_counter > 0) {
Execute marker instructions;
}
Default value of reference counter is 0. Tracer has to increment the
reference counter before tracing on a marker and decrement it when
done with the tracing.
Implement the reference counter logic in core uprobe. User will be
able to use it from trace_uprobe as well as from kernel module. New
trace_uprobe definition with reference counter will now be:
<path>:<offset>[(ref_ctr_offset)]
where ref_ctr_offset is an optional field. For kernel module, new
variant of uprobe_register() has been introduced:
uprobe_register_refctr(inode, offset, ref_ctr_offset, consumer)
No new variant for uprobe_unregister() because it's assumed to have
only one reference counter for one uprobe.
[1] https://sourceware.org/systemtap/wiki/UserSpaceProbeImplementation
Note: 'reference counter' is called as 'semaphore' in original Dtrace
(or Systemtap, bcc and even in ELF) documentation and code. But the
term 'semaphore' is misleading in this context. This is just a counter
used to hold number of tracers tracing on a marker. This is not really
used for any synchronization. So we are calling it a 'reference counter'
in kernel / perf code.
Link: http://lkml.kernel.org/r/20180820044250.11659-2-ravi.bangoria@linux.ibm.com
Reviewed-by: Masami Hiramatsu <mhiramat@kernel.org>
[Only trace_uprobe.c]
Reviewed-by: Oleg Nesterov <oleg@redhat.com>
Reviewed-by: Song Liu <songliubraving@fb.com>
Tested-by: Song Liu <songliubraving@fb.com>
Signed-off-by: Ravi Bangoria <ravi.bangoria@linux.ibm.com>
Signed-off-by: Steven Rostedt (VMware) <rostedt@goodmis.org>
2018-08-20 12:42:47 +08:00
|
|
|
struct delayed_uprobe {
|
|
|
|
struct list_head list;
|
|
|
|
struct uprobe *uprobe;
|
|
|
|
struct mm_struct *mm;
|
|
|
|
};
|
|
|
|
|
|
|
|
static DEFINE_MUTEX(delayed_uprobe_lock);
|
|
|
|
static LIST_HEAD(delayed_uprobe_list);
|
|
|
|
|
2013-11-10 02:49:39 +08:00
|
|
|
/*
|
2013-11-20 00:20:21 +08:00
|
|
|
* Execute out of line area: anonymous executable mapping installed
|
|
|
|
* by the probed task to execute the copy of the original instruction
|
|
|
|
* mangled by set_swbp().
|
|
|
|
*
|
2013-11-10 02:49:39 +08:00
|
|
|
* On a breakpoint hit, thread contests for a slot. It frees the
|
|
|
|
* slot after singlestep. Currently a fixed number of slots are
|
|
|
|
* allocated.
|
|
|
|
*/
|
|
|
|
struct xol_area {
|
2015-07-21 21:40:33 +08:00
|
|
|
wait_queue_head_t wq; /* if all slots are busy */
|
|
|
|
atomic_t slot_count; /* number of in-use slots */
|
|
|
|
unsigned long *bitmap; /* 0 = free slot */
|
2013-11-10 02:49:39 +08:00
|
|
|
|
2015-07-21 21:40:33 +08:00
|
|
|
struct vm_special_mapping xol_mapping;
|
|
|
|
struct page *pages[2];
|
2013-11-10 02:49:39 +08:00
|
|
|
/*
|
|
|
|
* We keep the vma's vm_start rather than a pointer to the vma
|
|
|
|
* itself. The probed process or a naughty kernel module could make
|
|
|
|
* the vma go away, and we must handle that reasonably gracefully.
|
|
|
|
*/
|
2015-07-21 21:40:33 +08:00
|
|
|
unsigned long vaddr; /* Page(s) of instruction slots */
|
2013-11-10 02:49:39 +08:00
|
|
|
};
|
|
|
|
|
uprobes, mm, x86: Add the ability to install and remove uprobes breakpoints
Add uprobes support to the core kernel, with x86 support.
This commit adds the kernel facilities, the actual uprobes
user-space ABI and perf probe support comes in later commits.
General design:
Uprobes are maintained in an rb-tree indexed by inode and offset
(the offset here is from the start of the mapping). For a unique
(inode, offset) tuple, there can be at most one uprobe in the
rb-tree.
Since the (inode, offset) tuple identifies a unique uprobe, more
than one user may be interested in the same uprobe. This provides
the ability to connect multiple 'consumers' to the same uprobe.
Each consumer defines a handler and a filter (optional). The
'handler' is run every time the uprobe is hit, if it matches the
'filter' criteria.
The first consumer of a uprobe causes the breakpoint to be
inserted at the specified address and subsequent consumers are
appended to this list. On subsequent probes, the consumer gets
appended to the existing list of consumers. The breakpoint is
removed when the last consumer unregisters. For all other
unregisterations, the consumer is removed from the list of
consumers.
Given a inode, we get a list of the mms that have mapped the
inode. Do the actual registration if mm maps the page where a
probe needs to be inserted/removed.
We use a temporary list to walk through the vmas that map the
inode.
- The number of maps that map the inode, is not known before we
walk the rmap and keeps changing.
- extending vm_area_struct wasn't recommended, it's a
size-critical data structure.
- There can be more than one maps of the inode in the same mm.
We add callbacks to the mmap methods to keep an eye on text vmas
that are of interest to uprobes. When a vma of interest is mapped,
we insert the breakpoint at the right address.
Uprobe works by replacing the instruction at the address defined
by (inode, offset) with the arch specific breakpoint
instruction. We save a copy of the original instruction at the
uprobed address.
This is needed for:
a. executing the instruction out-of-line (xol).
b. instruction analysis for any subsequent fixups.
c. restoring the instruction back when the uprobe is unregistered.
We insert or delete a breakpoint instruction, and this
breakpoint instruction is assumed to be the smallest instruction
available on the platform. For fixed size instruction platforms
this is trivially true, for variable size instruction platforms
the breakpoint instruction is typically the smallest (often a
single byte).
Writing the instruction is done by COWing the page and changing
the instruction during the copy, this even though most platforms
allow atomic writes of the breakpoint instruction. This also
mirrors the behaviour of a ptrace() memory write to a PRIVATE
file map.
The core worker is derived from KSM's replace_page() logic.
In essence, similar to KSM:
a. allocate a new page and copy over contents of the page that
has the uprobed vaddr
b. modify the copy and insert the breakpoint at the required
address
c. switch the original page with the copy containing the
breakpoint
d. flush page tables.
replace_page() is being replicated here because of some minor
changes in the type of pages and also because Hugh Dickins had
plans to improve replace_page() for KSM specific work.
Instruction analysis on x86 is based on instruction decoder and
determines if an instruction can be probed and determines the
necessary fixups after singlestep. Instruction analysis is done
at probe insertion time so that we avoid having to repeat the
same analysis every time a probe is hit.
A lot of code here is due to the improvement/suggestions/inputs
from Peter Zijlstra.
Changelog:
(v10):
- Add code to clear REX.B prefix as suggested by Denys Vlasenko
and Masami Hiramatsu.
(v9):
- Use insn_offset_modrm as suggested by Masami Hiramatsu.
(v7):
Handle comments from Peter Zijlstra:
- Dont take reference to inode. (expect inode to uprobe_register to be sane).
- Use PTR_ERR to set the return value.
- No need to take reference to inode.
- use PTR_ERR to return error value.
- register and uprobe_unregister share code.
(v5):
- Modified del_consumer as per comments from Peter.
- Drop reference to inode before dropping reference to uprobe.
- Use i_size_read(inode) instead of inode->i_size.
- Ensure uprobe->consumers is NULL, before __uprobe_unregister() is called.
- Includes errno.h as recommended by Stephen Rothwell to fix a build issue
on sparc defconfig
- Remove restrictions while unregistering.
- Earlier code leaked inode references under some conditions while
registering/unregistering.
- Continue the vma-rmap walk even if the intermediate vma doesnt
meet the requirements.
- Validate the vma found by find_vma before inserting/removing the
breakpoint
- Call del_consumer under mutex_lock.
- Use hash locks.
- Handle mremap.
- Introduce find_least_offset_node() instead of close match logic in
find_uprobe
- Uprobes no more depends on MM_OWNER; No reference to task_structs
while inserting/removing a probe.
- Uses read_mapping_page instead of grab_cache_page so that the pages
have valid content.
- pass NULL to get_user_pages for the task parameter.
- call SetPageUptodate on the new page allocated in write_opcode.
- fix leaking a reference to the new page under certain conditions.
- Include Instruction Decoder if Uprobes gets defined.
- Remove const attributes for instruction prefix arrays.
- Uses mm_context to know if the application is 32 bit.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Also-written-by: Jim Keniston <jkenisto@us.ibm.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Roland McGrath <roland@hack.frob.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Anton Arapov <anton@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Denys Vlasenko <vda.linux@googlemail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linux-mm <linux-mm@kvack.org>
Link: http://lkml.kernel.org/r/20120209092642.GE16600@linux.vnet.ibm.com
[ Made various small edits to the commit log ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-02-09 17:26:42 +08:00
|
|
|
/*
|
|
|
|
* valid_vma: Verify if the specified vma is an executable vma
|
|
|
|
* Relax restrictions while unregistering: vm_flags might have
|
|
|
|
* changed after breakpoint was inserted.
|
|
|
|
* - is_register: indicates if we are in register context.
|
|
|
|
* - Return 1 if the specified virtual address is in an
|
|
|
|
* executable vma.
|
|
|
|
*/
|
|
|
|
static bool valid_vma(struct vm_area_struct *vma, bool is_register)
|
|
|
|
{
|
2014-04-29 02:15:43 +08:00
|
|
|
vm_flags_t flags = VM_HUGETLB | VM_MAYEXEC | VM_MAYSHARE;
|
uprobes, mm, x86: Add the ability to install and remove uprobes breakpoints
Add uprobes support to the core kernel, with x86 support.
This commit adds the kernel facilities, the actual uprobes
user-space ABI and perf probe support comes in later commits.
General design:
Uprobes are maintained in an rb-tree indexed by inode and offset
(the offset here is from the start of the mapping). For a unique
(inode, offset) tuple, there can be at most one uprobe in the
rb-tree.
Since the (inode, offset) tuple identifies a unique uprobe, more
than one user may be interested in the same uprobe. This provides
the ability to connect multiple 'consumers' to the same uprobe.
Each consumer defines a handler and a filter (optional). The
'handler' is run every time the uprobe is hit, if it matches the
'filter' criteria.
The first consumer of a uprobe causes the breakpoint to be
inserted at the specified address and subsequent consumers are
appended to this list. On subsequent probes, the consumer gets
appended to the existing list of consumers. The breakpoint is
removed when the last consumer unregisters. For all other
unregisterations, the consumer is removed from the list of
consumers.
Given a inode, we get a list of the mms that have mapped the
inode. Do the actual registration if mm maps the page where a
probe needs to be inserted/removed.
We use a temporary list to walk through the vmas that map the
inode.
- The number of maps that map the inode, is not known before we
walk the rmap and keeps changing.
- extending vm_area_struct wasn't recommended, it's a
size-critical data structure.
- There can be more than one maps of the inode in the same mm.
We add callbacks to the mmap methods to keep an eye on text vmas
that are of interest to uprobes. When a vma of interest is mapped,
we insert the breakpoint at the right address.
Uprobe works by replacing the instruction at the address defined
by (inode, offset) with the arch specific breakpoint
instruction. We save a copy of the original instruction at the
uprobed address.
This is needed for:
a. executing the instruction out-of-line (xol).
b. instruction analysis for any subsequent fixups.
c. restoring the instruction back when the uprobe is unregistered.
We insert or delete a breakpoint instruction, and this
breakpoint instruction is assumed to be the smallest instruction
available on the platform. For fixed size instruction platforms
this is trivially true, for variable size instruction platforms
the breakpoint instruction is typically the smallest (often a
single byte).
Writing the instruction is done by COWing the page and changing
the instruction during the copy, this even though most platforms
allow atomic writes of the breakpoint instruction. This also
mirrors the behaviour of a ptrace() memory write to a PRIVATE
file map.
The core worker is derived from KSM's replace_page() logic.
In essence, similar to KSM:
a. allocate a new page and copy over contents of the page that
has the uprobed vaddr
b. modify the copy and insert the breakpoint at the required
address
c. switch the original page with the copy containing the
breakpoint
d. flush page tables.
replace_page() is being replicated here because of some minor
changes in the type of pages and also because Hugh Dickins had
plans to improve replace_page() for KSM specific work.
Instruction analysis on x86 is based on instruction decoder and
determines if an instruction can be probed and determines the
necessary fixups after singlestep. Instruction analysis is done
at probe insertion time so that we avoid having to repeat the
same analysis every time a probe is hit.
A lot of code here is due to the improvement/suggestions/inputs
from Peter Zijlstra.
Changelog:
(v10):
- Add code to clear REX.B prefix as suggested by Denys Vlasenko
and Masami Hiramatsu.
(v9):
- Use insn_offset_modrm as suggested by Masami Hiramatsu.
(v7):
Handle comments from Peter Zijlstra:
- Dont take reference to inode. (expect inode to uprobe_register to be sane).
- Use PTR_ERR to set the return value.
- No need to take reference to inode.
- use PTR_ERR to return error value.
- register and uprobe_unregister share code.
(v5):
- Modified del_consumer as per comments from Peter.
- Drop reference to inode before dropping reference to uprobe.
- Use i_size_read(inode) instead of inode->i_size.
- Ensure uprobe->consumers is NULL, before __uprobe_unregister() is called.
- Includes errno.h as recommended by Stephen Rothwell to fix a build issue
on sparc defconfig
- Remove restrictions while unregistering.
- Earlier code leaked inode references under some conditions while
registering/unregistering.
- Continue the vma-rmap walk even if the intermediate vma doesnt
meet the requirements.
- Validate the vma found by find_vma before inserting/removing the
breakpoint
- Call del_consumer under mutex_lock.
- Use hash locks.
- Handle mremap.
- Introduce find_least_offset_node() instead of close match logic in
find_uprobe
- Uprobes no more depends on MM_OWNER; No reference to task_structs
while inserting/removing a probe.
- Uses read_mapping_page instead of grab_cache_page so that the pages
have valid content.
- pass NULL to get_user_pages for the task parameter.
- call SetPageUptodate on the new page allocated in write_opcode.
- fix leaking a reference to the new page under certain conditions.
- Include Instruction Decoder if Uprobes gets defined.
- Remove const attributes for instruction prefix arrays.
- Uses mm_context to know if the application is 32 bit.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Also-written-by: Jim Keniston <jkenisto@us.ibm.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Roland McGrath <roland@hack.frob.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Anton Arapov <anton@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Denys Vlasenko <vda.linux@googlemail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linux-mm <linux-mm@kvack.org>
Link: http://lkml.kernel.org/r/20120209092642.GE16600@linux.vnet.ibm.com
[ Made various small edits to the commit log ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-02-09 17:26:42 +08:00
|
|
|
|
2012-09-17 01:31:39 +08:00
|
|
|
if (is_register)
|
|
|
|
flags |= VM_WRITE;
|
uprobes, mm, x86: Add the ability to install and remove uprobes breakpoints
Add uprobes support to the core kernel, with x86 support.
This commit adds the kernel facilities, the actual uprobes
user-space ABI and perf probe support comes in later commits.
General design:
Uprobes are maintained in an rb-tree indexed by inode and offset
(the offset here is from the start of the mapping). For a unique
(inode, offset) tuple, there can be at most one uprobe in the
rb-tree.
Since the (inode, offset) tuple identifies a unique uprobe, more
than one user may be interested in the same uprobe. This provides
the ability to connect multiple 'consumers' to the same uprobe.
Each consumer defines a handler and a filter (optional). The
'handler' is run every time the uprobe is hit, if it matches the
'filter' criteria.
The first consumer of a uprobe causes the breakpoint to be
inserted at the specified address and subsequent consumers are
appended to this list. On subsequent probes, the consumer gets
appended to the existing list of consumers. The breakpoint is
removed when the last consumer unregisters. For all other
unregisterations, the consumer is removed from the list of
consumers.
Given a inode, we get a list of the mms that have mapped the
inode. Do the actual registration if mm maps the page where a
probe needs to be inserted/removed.
We use a temporary list to walk through the vmas that map the
inode.
- The number of maps that map the inode, is not known before we
walk the rmap and keeps changing.
- extending vm_area_struct wasn't recommended, it's a
size-critical data structure.
- There can be more than one maps of the inode in the same mm.
We add callbacks to the mmap methods to keep an eye on text vmas
that are of interest to uprobes. When a vma of interest is mapped,
we insert the breakpoint at the right address.
Uprobe works by replacing the instruction at the address defined
by (inode, offset) with the arch specific breakpoint
instruction. We save a copy of the original instruction at the
uprobed address.
This is needed for:
a. executing the instruction out-of-line (xol).
b. instruction analysis for any subsequent fixups.
c. restoring the instruction back when the uprobe is unregistered.
We insert or delete a breakpoint instruction, and this
breakpoint instruction is assumed to be the smallest instruction
available on the platform. For fixed size instruction platforms
this is trivially true, for variable size instruction platforms
the breakpoint instruction is typically the smallest (often a
single byte).
Writing the instruction is done by COWing the page and changing
the instruction during the copy, this even though most platforms
allow atomic writes of the breakpoint instruction. This also
mirrors the behaviour of a ptrace() memory write to a PRIVATE
file map.
The core worker is derived from KSM's replace_page() logic.
In essence, similar to KSM:
a. allocate a new page and copy over contents of the page that
has the uprobed vaddr
b. modify the copy and insert the breakpoint at the required
address
c. switch the original page with the copy containing the
breakpoint
d. flush page tables.
replace_page() is being replicated here because of some minor
changes in the type of pages and also because Hugh Dickins had
plans to improve replace_page() for KSM specific work.
Instruction analysis on x86 is based on instruction decoder and
determines if an instruction can be probed and determines the
necessary fixups after singlestep. Instruction analysis is done
at probe insertion time so that we avoid having to repeat the
same analysis every time a probe is hit.
A lot of code here is due to the improvement/suggestions/inputs
from Peter Zijlstra.
Changelog:
(v10):
- Add code to clear REX.B prefix as suggested by Denys Vlasenko
and Masami Hiramatsu.
(v9):
- Use insn_offset_modrm as suggested by Masami Hiramatsu.
(v7):
Handle comments from Peter Zijlstra:
- Dont take reference to inode. (expect inode to uprobe_register to be sane).
- Use PTR_ERR to set the return value.
- No need to take reference to inode.
- use PTR_ERR to return error value.
- register and uprobe_unregister share code.
(v5):
- Modified del_consumer as per comments from Peter.
- Drop reference to inode before dropping reference to uprobe.
- Use i_size_read(inode) instead of inode->i_size.
- Ensure uprobe->consumers is NULL, before __uprobe_unregister() is called.
- Includes errno.h as recommended by Stephen Rothwell to fix a build issue
on sparc defconfig
- Remove restrictions while unregistering.
- Earlier code leaked inode references under some conditions while
registering/unregistering.
- Continue the vma-rmap walk even if the intermediate vma doesnt
meet the requirements.
- Validate the vma found by find_vma before inserting/removing the
breakpoint
- Call del_consumer under mutex_lock.
- Use hash locks.
- Handle mremap.
- Introduce find_least_offset_node() instead of close match logic in
find_uprobe
- Uprobes no more depends on MM_OWNER; No reference to task_structs
while inserting/removing a probe.
- Uses read_mapping_page instead of grab_cache_page so that the pages
have valid content.
- pass NULL to get_user_pages for the task parameter.
- call SetPageUptodate on the new page allocated in write_opcode.
- fix leaking a reference to the new page under certain conditions.
- Include Instruction Decoder if Uprobes gets defined.
- Remove const attributes for instruction prefix arrays.
- Uses mm_context to know if the application is 32 bit.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Also-written-by: Jim Keniston <jkenisto@us.ibm.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Roland McGrath <roland@hack.frob.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Anton Arapov <anton@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Denys Vlasenko <vda.linux@googlemail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linux-mm <linux-mm@kvack.org>
Link: http://lkml.kernel.org/r/20120209092642.GE16600@linux.vnet.ibm.com
[ Made various small edits to the commit log ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-02-09 17:26:42 +08:00
|
|
|
|
2012-09-17 01:31:39 +08:00
|
|
|
return vma->vm_file && (vma->vm_flags & flags) == VM_MAYEXEC;
|
uprobes, mm, x86: Add the ability to install and remove uprobes breakpoints
Add uprobes support to the core kernel, with x86 support.
This commit adds the kernel facilities, the actual uprobes
user-space ABI and perf probe support comes in later commits.
General design:
Uprobes are maintained in an rb-tree indexed by inode and offset
(the offset here is from the start of the mapping). For a unique
(inode, offset) tuple, there can be at most one uprobe in the
rb-tree.
Since the (inode, offset) tuple identifies a unique uprobe, more
than one user may be interested in the same uprobe. This provides
the ability to connect multiple 'consumers' to the same uprobe.
Each consumer defines a handler and a filter (optional). The
'handler' is run every time the uprobe is hit, if it matches the
'filter' criteria.
The first consumer of a uprobe causes the breakpoint to be
inserted at the specified address and subsequent consumers are
appended to this list. On subsequent probes, the consumer gets
appended to the existing list of consumers. The breakpoint is
removed when the last consumer unregisters. For all other
unregisterations, the consumer is removed from the list of
consumers.
Given a inode, we get a list of the mms that have mapped the
inode. Do the actual registration if mm maps the page where a
probe needs to be inserted/removed.
We use a temporary list to walk through the vmas that map the
inode.
- The number of maps that map the inode, is not known before we
walk the rmap and keeps changing.
- extending vm_area_struct wasn't recommended, it's a
size-critical data structure.
- There can be more than one maps of the inode in the same mm.
We add callbacks to the mmap methods to keep an eye on text vmas
that are of interest to uprobes. When a vma of interest is mapped,
we insert the breakpoint at the right address.
Uprobe works by replacing the instruction at the address defined
by (inode, offset) with the arch specific breakpoint
instruction. We save a copy of the original instruction at the
uprobed address.
This is needed for:
a. executing the instruction out-of-line (xol).
b. instruction analysis for any subsequent fixups.
c. restoring the instruction back when the uprobe is unregistered.
We insert or delete a breakpoint instruction, and this
breakpoint instruction is assumed to be the smallest instruction
available on the platform. For fixed size instruction platforms
this is trivially true, for variable size instruction platforms
the breakpoint instruction is typically the smallest (often a
single byte).
Writing the instruction is done by COWing the page and changing
the instruction during the copy, this even though most platforms
allow atomic writes of the breakpoint instruction. This also
mirrors the behaviour of a ptrace() memory write to a PRIVATE
file map.
The core worker is derived from KSM's replace_page() logic.
In essence, similar to KSM:
a. allocate a new page and copy over contents of the page that
has the uprobed vaddr
b. modify the copy and insert the breakpoint at the required
address
c. switch the original page with the copy containing the
breakpoint
d. flush page tables.
replace_page() is being replicated here because of some minor
changes in the type of pages and also because Hugh Dickins had
plans to improve replace_page() for KSM specific work.
Instruction analysis on x86 is based on instruction decoder and
determines if an instruction can be probed and determines the
necessary fixups after singlestep. Instruction analysis is done
at probe insertion time so that we avoid having to repeat the
same analysis every time a probe is hit.
A lot of code here is due to the improvement/suggestions/inputs
from Peter Zijlstra.
Changelog:
(v10):
- Add code to clear REX.B prefix as suggested by Denys Vlasenko
and Masami Hiramatsu.
(v9):
- Use insn_offset_modrm as suggested by Masami Hiramatsu.
(v7):
Handle comments from Peter Zijlstra:
- Dont take reference to inode. (expect inode to uprobe_register to be sane).
- Use PTR_ERR to set the return value.
- No need to take reference to inode.
- use PTR_ERR to return error value.
- register and uprobe_unregister share code.
(v5):
- Modified del_consumer as per comments from Peter.
- Drop reference to inode before dropping reference to uprobe.
- Use i_size_read(inode) instead of inode->i_size.
- Ensure uprobe->consumers is NULL, before __uprobe_unregister() is called.
- Includes errno.h as recommended by Stephen Rothwell to fix a build issue
on sparc defconfig
- Remove restrictions while unregistering.
- Earlier code leaked inode references under some conditions while
registering/unregistering.
- Continue the vma-rmap walk even if the intermediate vma doesnt
meet the requirements.
- Validate the vma found by find_vma before inserting/removing the
breakpoint
- Call del_consumer under mutex_lock.
- Use hash locks.
- Handle mremap.
- Introduce find_least_offset_node() instead of close match logic in
find_uprobe
- Uprobes no more depends on MM_OWNER; No reference to task_structs
while inserting/removing a probe.
- Uses read_mapping_page instead of grab_cache_page so that the pages
have valid content.
- pass NULL to get_user_pages for the task parameter.
- call SetPageUptodate on the new page allocated in write_opcode.
- fix leaking a reference to the new page under certain conditions.
- Include Instruction Decoder if Uprobes gets defined.
- Remove const attributes for instruction prefix arrays.
- Uses mm_context to know if the application is 32 bit.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Also-written-by: Jim Keniston <jkenisto@us.ibm.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Roland McGrath <roland@hack.frob.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Anton Arapov <anton@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Denys Vlasenko <vda.linux@googlemail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linux-mm <linux-mm@kvack.org>
Link: http://lkml.kernel.org/r/20120209092642.GE16600@linux.vnet.ibm.com
[ Made various small edits to the commit log ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-02-09 17:26:42 +08:00
|
|
|
}
|
|
|
|
|
2012-07-30 02:22:47 +08:00
|
|
|
static unsigned long offset_to_vaddr(struct vm_area_struct *vma, loff_t offset)
|
uprobes, mm, x86: Add the ability to install and remove uprobes breakpoints
Add uprobes support to the core kernel, with x86 support.
This commit adds the kernel facilities, the actual uprobes
user-space ABI and perf probe support comes in later commits.
General design:
Uprobes are maintained in an rb-tree indexed by inode and offset
(the offset here is from the start of the mapping). For a unique
(inode, offset) tuple, there can be at most one uprobe in the
rb-tree.
Since the (inode, offset) tuple identifies a unique uprobe, more
than one user may be interested in the same uprobe. This provides
the ability to connect multiple 'consumers' to the same uprobe.
Each consumer defines a handler and a filter (optional). The
'handler' is run every time the uprobe is hit, if it matches the
'filter' criteria.
The first consumer of a uprobe causes the breakpoint to be
inserted at the specified address and subsequent consumers are
appended to this list. On subsequent probes, the consumer gets
appended to the existing list of consumers. The breakpoint is
removed when the last consumer unregisters. For all other
unregisterations, the consumer is removed from the list of
consumers.
Given a inode, we get a list of the mms that have mapped the
inode. Do the actual registration if mm maps the page where a
probe needs to be inserted/removed.
We use a temporary list to walk through the vmas that map the
inode.
- The number of maps that map the inode, is not known before we
walk the rmap and keeps changing.
- extending vm_area_struct wasn't recommended, it's a
size-critical data structure.
- There can be more than one maps of the inode in the same mm.
We add callbacks to the mmap methods to keep an eye on text vmas
that are of interest to uprobes. When a vma of interest is mapped,
we insert the breakpoint at the right address.
Uprobe works by replacing the instruction at the address defined
by (inode, offset) with the arch specific breakpoint
instruction. We save a copy of the original instruction at the
uprobed address.
This is needed for:
a. executing the instruction out-of-line (xol).
b. instruction analysis for any subsequent fixups.
c. restoring the instruction back when the uprobe is unregistered.
We insert or delete a breakpoint instruction, and this
breakpoint instruction is assumed to be the smallest instruction
available on the platform. For fixed size instruction platforms
this is trivially true, for variable size instruction platforms
the breakpoint instruction is typically the smallest (often a
single byte).
Writing the instruction is done by COWing the page and changing
the instruction during the copy, this even though most platforms
allow atomic writes of the breakpoint instruction. This also
mirrors the behaviour of a ptrace() memory write to a PRIVATE
file map.
The core worker is derived from KSM's replace_page() logic.
In essence, similar to KSM:
a. allocate a new page and copy over contents of the page that
has the uprobed vaddr
b. modify the copy and insert the breakpoint at the required
address
c. switch the original page with the copy containing the
breakpoint
d. flush page tables.
replace_page() is being replicated here because of some minor
changes in the type of pages and also because Hugh Dickins had
plans to improve replace_page() for KSM specific work.
Instruction analysis on x86 is based on instruction decoder and
determines if an instruction can be probed and determines the
necessary fixups after singlestep. Instruction analysis is done
at probe insertion time so that we avoid having to repeat the
same analysis every time a probe is hit.
A lot of code here is due to the improvement/suggestions/inputs
from Peter Zijlstra.
Changelog:
(v10):
- Add code to clear REX.B prefix as suggested by Denys Vlasenko
and Masami Hiramatsu.
(v9):
- Use insn_offset_modrm as suggested by Masami Hiramatsu.
(v7):
Handle comments from Peter Zijlstra:
- Dont take reference to inode. (expect inode to uprobe_register to be sane).
- Use PTR_ERR to set the return value.
- No need to take reference to inode.
- use PTR_ERR to return error value.
- register and uprobe_unregister share code.
(v5):
- Modified del_consumer as per comments from Peter.
- Drop reference to inode before dropping reference to uprobe.
- Use i_size_read(inode) instead of inode->i_size.
- Ensure uprobe->consumers is NULL, before __uprobe_unregister() is called.
- Includes errno.h as recommended by Stephen Rothwell to fix a build issue
on sparc defconfig
- Remove restrictions while unregistering.
- Earlier code leaked inode references under some conditions while
registering/unregistering.
- Continue the vma-rmap walk even if the intermediate vma doesnt
meet the requirements.
- Validate the vma found by find_vma before inserting/removing the
breakpoint
- Call del_consumer under mutex_lock.
- Use hash locks.
- Handle mremap.
- Introduce find_least_offset_node() instead of close match logic in
find_uprobe
- Uprobes no more depends on MM_OWNER; No reference to task_structs
while inserting/removing a probe.
- Uses read_mapping_page instead of grab_cache_page so that the pages
have valid content.
- pass NULL to get_user_pages for the task parameter.
- call SetPageUptodate on the new page allocated in write_opcode.
- fix leaking a reference to the new page under certain conditions.
- Include Instruction Decoder if Uprobes gets defined.
- Remove const attributes for instruction prefix arrays.
- Uses mm_context to know if the application is 32 bit.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Also-written-by: Jim Keniston <jkenisto@us.ibm.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Roland McGrath <roland@hack.frob.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Anton Arapov <anton@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Denys Vlasenko <vda.linux@googlemail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linux-mm <linux-mm@kvack.org>
Link: http://lkml.kernel.org/r/20120209092642.GE16600@linux.vnet.ibm.com
[ Made various small edits to the commit log ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-02-09 17:26:42 +08:00
|
|
|
{
|
2012-07-30 02:22:47 +08:00
|
|
|
return vma->vm_start + offset - ((loff_t)vma->vm_pgoff << PAGE_SHIFT);
|
uprobes, mm, x86: Add the ability to install and remove uprobes breakpoints
Add uprobes support to the core kernel, with x86 support.
This commit adds the kernel facilities, the actual uprobes
user-space ABI and perf probe support comes in later commits.
General design:
Uprobes are maintained in an rb-tree indexed by inode and offset
(the offset here is from the start of the mapping). For a unique
(inode, offset) tuple, there can be at most one uprobe in the
rb-tree.
Since the (inode, offset) tuple identifies a unique uprobe, more
than one user may be interested in the same uprobe. This provides
the ability to connect multiple 'consumers' to the same uprobe.
Each consumer defines a handler and a filter (optional). The
'handler' is run every time the uprobe is hit, if it matches the
'filter' criteria.
The first consumer of a uprobe causes the breakpoint to be
inserted at the specified address and subsequent consumers are
appended to this list. On subsequent probes, the consumer gets
appended to the existing list of consumers. The breakpoint is
removed when the last consumer unregisters. For all other
unregisterations, the consumer is removed from the list of
consumers.
Given a inode, we get a list of the mms that have mapped the
inode. Do the actual registration if mm maps the page where a
probe needs to be inserted/removed.
We use a temporary list to walk through the vmas that map the
inode.
- The number of maps that map the inode, is not known before we
walk the rmap and keeps changing.
- extending vm_area_struct wasn't recommended, it's a
size-critical data structure.
- There can be more than one maps of the inode in the same mm.
We add callbacks to the mmap methods to keep an eye on text vmas
that are of interest to uprobes. When a vma of interest is mapped,
we insert the breakpoint at the right address.
Uprobe works by replacing the instruction at the address defined
by (inode, offset) with the arch specific breakpoint
instruction. We save a copy of the original instruction at the
uprobed address.
This is needed for:
a. executing the instruction out-of-line (xol).
b. instruction analysis for any subsequent fixups.
c. restoring the instruction back when the uprobe is unregistered.
We insert or delete a breakpoint instruction, and this
breakpoint instruction is assumed to be the smallest instruction
available on the platform. For fixed size instruction platforms
this is trivially true, for variable size instruction platforms
the breakpoint instruction is typically the smallest (often a
single byte).
Writing the instruction is done by COWing the page and changing
the instruction during the copy, this even though most platforms
allow atomic writes of the breakpoint instruction. This also
mirrors the behaviour of a ptrace() memory write to a PRIVATE
file map.
The core worker is derived from KSM's replace_page() logic.
In essence, similar to KSM:
a. allocate a new page and copy over contents of the page that
has the uprobed vaddr
b. modify the copy and insert the breakpoint at the required
address
c. switch the original page with the copy containing the
breakpoint
d. flush page tables.
replace_page() is being replicated here because of some minor
changes in the type of pages and also because Hugh Dickins had
plans to improve replace_page() for KSM specific work.
Instruction analysis on x86 is based on instruction decoder and
determines if an instruction can be probed and determines the
necessary fixups after singlestep. Instruction analysis is done
at probe insertion time so that we avoid having to repeat the
same analysis every time a probe is hit.
A lot of code here is due to the improvement/suggestions/inputs
from Peter Zijlstra.
Changelog:
(v10):
- Add code to clear REX.B prefix as suggested by Denys Vlasenko
and Masami Hiramatsu.
(v9):
- Use insn_offset_modrm as suggested by Masami Hiramatsu.
(v7):
Handle comments from Peter Zijlstra:
- Dont take reference to inode. (expect inode to uprobe_register to be sane).
- Use PTR_ERR to set the return value.
- No need to take reference to inode.
- use PTR_ERR to return error value.
- register and uprobe_unregister share code.
(v5):
- Modified del_consumer as per comments from Peter.
- Drop reference to inode before dropping reference to uprobe.
- Use i_size_read(inode) instead of inode->i_size.
- Ensure uprobe->consumers is NULL, before __uprobe_unregister() is called.
- Includes errno.h as recommended by Stephen Rothwell to fix a build issue
on sparc defconfig
- Remove restrictions while unregistering.
- Earlier code leaked inode references under some conditions while
registering/unregistering.
- Continue the vma-rmap walk even if the intermediate vma doesnt
meet the requirements.
- Validate the vma found by find_vma before inserting/removing the
breakpoint
- Call del_consumer under mutex_lock.
- Use hash locks.
- Handle mremap.
- Introduce find_least_offset_node() instead of close match logic in
find_uprobe
- Uprobes no more depends on MM_OWNER; No reference to task_structs
while inserting/removing a probe.
- Uses read_mapping_page instead of grab_cache_page so that the pages
have valid content.
- pass NULL to get_user_pages for the task parameter.
- call SetPageUptodate on the new page allocated in write_opcode.
- fix leaking a reference to the new page under certain conditions.
- Include Instruction Decoder if Uprobes gets defined.
- Remove const attributes for instruction prefix arrays.
- Uses mm_context to know if the application is 32 bit.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Also-written-by: Jim Keniston <jkenisto@us.ibm.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Roland McGrath <roland@hack.frob.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Anton Arapov <anton@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Denys Vlasenko <vda.linux@googlemail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linux-mm <linux-mm@kvack.org>
Link: http://lkml.kernel.org/r/20120209092642.GE16600@linux.vnet.ibm.com
[ Made various small edits to the commit log ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-02-09 17:26:42 +08:00
|
|
|
}
|
|
|
|
|
2012-07-30 02:22:42 +08:00
|
|
|
static loff_t vaddr_to_offset(struct vm_area_struct *vma, unsigned long vaddr)
|
|
|
|
{
|
|
|
|
return ((loff_t)vma->vm_pgoff << PAGE_SHIFT) + (vaddr - vma->vm_start);
|
|
|
|
}
|
|
|
|
|
uprobes, mm, x86: Add the ability to install and remove uprobes breakpoints
Add uprobes support to the core kernel, with x86 support.
This commit adds the kernel facilities, the actual uprobes
user-space ABI and perf probe support comes in later commits.
General design:
Uprobes are maintained in an rb-tree indexed by inode and offset
(the offset here is from the start of the mapping). For a unique
(inode, offset) tuple, there can be at most one uprobe in the
rb-tree.
Since the (inode, offset) tuple identifies a unique uprobe, more
than one user may be interested in the same uprobe. This provides
the ability to connect multiple 'consumers' to the same uprobe.
Each consumer defines a handler and a filter (optional). The
'handler' is run every time the uprobe is hit, if it matches the
'filter' criteria.
The first consumer of a uprobe causes the breakpoint to be
inserted at the specified address and subsequent consumers are
appended to this list. On subsequent probes, the consumer gets
appended to the existing list of consumers. The breakpoint is
removed when the last consumer unregisters. For all other
unregisterations, the consumer is removed from the list of
consumers.
Given a inode, we get a list of the mms that have mapped the
inode. Do the actual registration if mm maps the page where a
probe needs to be inserted/removed.
We use a temporary list to walk through the vmas that map the
inode.
- The number of maps that map the inode, is not known before we
walk the rmap and keeps changing.
- extending vm_area_struct wasn't recommended, it's a
size-critical data structure.
- There can be more than one maps of the inode in the same mm.
We add callbacks to the mmap methods to keep an eye on text vmas
that are of interest to uprobes. When a vma of interest is mapped,
we insert the breakpoint at the right address.
Uprobe works by replacing the instruction at the address defined
by (inode, offset) with the arch specific breakpoint
instruction. We save a copy of the original instruction at the
uprobed address.
This is needed for:
a. executing the instruction out-of-line (xol).
b. instruction analysis for any subsequent fixups.
c. restoring the instruction back when the uprobe is unregistered.
We insert or delete a breakpoint instruction, and this
breakpoint instruction is assumed to be the smallest instruction
available on the platform. For fixed size instruction platforms
this is trivially true, for variable size instruction platforms
the breakpoint instruction is typically the smallest (often a
single byte).
Writing the instruction is done by COWing the page and changing
the instruction during the copy, this even though most platforms
allow atomic writes of the breakpoint instruction. This also
mirrors the behaviour of a ptrace() memory write to a PRIVATE
file map.
The core worker is derived from KSM's replace_page() logic.
In essence, similar to KSM:
a. allocate a new page and copy over contents of the page that
has the uprobed vaddr
b. modify the copy and insert the breakpoint at the required
address
c. switch the original page with the copy containing the
breakpoint
d. flush page tables.
replace_page() is being replicated here because of some minor
changes in the type of pages and also because Hugh Dickins had
plans to improve replace_page() for KSM specific work.
Instruction analysis on x86 is based on instruction decoder and
determines if an instruction can be probed and determines the
necessary fixups after singlestep. Instruction analysis is done
at probe insertion time so that we avoid having to repeat the
same analysis every time a probe is hit.
A lot of code here is due to the improvement/suggestions/inputs
from Peter Zijlstra.
Changelog:
(v10):
- Add code to clear REX.B prefix as suggested by Denys Vlasenko
and Masami Hiramatsu.
(v9):
- Use insn_offset_modrm as suggested by Masami Hiramatsu.
(v7):
Handle comments from Peter Zijlstra:
- Dont take reference to inode. (expect inode to uprobe_register to be sane).
- Use PTR_ERR to set the return value.
- No need to take reference to inode.
- use PTR_ERR to return error value.
- register and uprobe_unregister share code.
(v5):
- Modified del_consumer as per comments from Peter.
- Drop reference to inode before dropping reference to uprobe.
- Use i_size_read(inode) instead of inode->i_size.
- Ensure uprobe->consumers is NULL, before __uprobe_unregister() is called.
- Includes errno.h as recommended by Stephen Rothwell to fix a build issue
on sparc defconfig
- Remove restrictions while unregistering.
- Earlier code leaked inode references under some conditions while
registering/unregistering.
- Continue the vma-rmap walk even if the intermediate vma doesnt
meet the requirements.
- Validate the vma found by find_vma before inserting/removing the
breakpoint
- Call del_consumer under mutex_lock.
- Use hash locks.
- Handle mremap.
- Introduce find_least_offset_node() instead of close match logic in
find_uprobe
- Uprobes no more depends on MM_OWNER; No reference to task_structs
while inserting/removing a probe.
- Uses read_mapping_page instead of grab_cache_page so that the pages
have valid content.
- pass NULL to get_user_pages for the task parameter.
- call SetPageUptodate on the new page allocated in write_opcode.
- fix leaking a reference to the new page under certain conditions.
- Include Instruction Decoder if Uprobes gets defined.
- Remove const attributes for instruction prefix arrays.
- Uses mm_context to know if the application is 32 bit.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Also-written-by: Jim Keniston <jkenisto@us.ibm.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Roland McGrath <roland@hack.frob.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Anton Arapov <anton@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Denys Vlasenko <vda.linux@googlemail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linux-mm <linux-mm@kvack.org>
Link: http://lkml.kernel.org/r/20120209092642.GE16600@linux.vnet.ibm.com
[ Made various small edits to the commit log ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-02-09 17:26:42 +08:00
|
|
|
/**
|
|
|
|
* __replace_page - replace page in vma by new page.
|
|
|
|
* based on replace_page in mm/ksm.c
|
|
|
|
*
|
|
|
|
* @vma: vma that holds the pte pointing to page
|
2012-07-30 02:22:16 +08:00
|
|
|
* @addr: address the old @page is mapped at
|
uprobes, mm, x86: Add the ability to install and remove uprobes breakpoints
Add uprobes support to the core kernel, with x86 support.
This commit adds the kernel facilities, the actual uprobes
user-space ABI and perf probe support comes in later commits.
General design:
Uprobes are maintained in an rb-tree indexed by inode and offset
(the offset here is from the start of the mapping). For a unique
(inode, offset) tuple, there can be at most one uprobe in the
rb-tree.
Since the (inode, offset) tuple identifies a unique uprobe, more
than one user may be interested in the same uprobe. This provides
the ability to connect multiple 'consumers' to the same uprobe.
Each consumer defines a handler and a filter (optional). The
'handler' is run every time the uprobe is hit, if it matches the
'filter' criteria.
The first consumer of a uprobe causes the breakpoint to be
inserted at the specified address and subsequent consumers are
appended to this list. On subsequent probes, the consumer gets
appended to the existing list of consumers. The breakpoint is
removed when the last consumer unregisters. For all other
unregisterations, the consumer is removed from the list of
consumers.
Given a inode, we get a list of the mms that have mapped the
inode. Do the actual registration if mm maps the page where a
probe needs to be inserted/removed.
We use a temporary list to walk through the vmas that map the
inode.
- The number of maps that map the inode, is not known before we
walk the rmap and keeps changing.
- extending vm_area_struct wasn't recommended, it's a
size-critical data structure.
- There can be more than one maps of the inode in the same mm.
We add callbacks to the mmap methods to keep an eye on text vmas
that are of interest to uprobes. When a vma of interest is mapped,
we insert the breakpoint at the right address.
Uprobe works by replacing the instruction at the address defined
by (inode, offset) with the arch specific breakpoint
instruction. We save a copy of the original instruction at the
uprobed address.
This is needed for:
a. executing the instruction out-of-line (xol).
b. instruction analysis for any subsequent fixups.
c. restoring the instruction back when the uprobe is unregistered.
We insert or delete a breakpoint instruction, and this
breakpoint instruction is assumed to be the smallest instruction
available on the platform. For fixed size instruction platforms
this is trivially true, for variable size instruction platforms
the breakpoint instruction is typically the smallest (often a
single byte).
Writing the instruction is done by COWing the page and changing
the instruction during the copy, this even though most platforms
allow atomic writes of the breakpoint instruction. This also
mirrors the behaviour of a ptrace() memory write to a PRIVATE
file map.
The core worker is derived from KSM's replace_page() logic.
In essence, similar to KSM:
a. allocate a new page and copy over contents of the page that
has the uprobed vaddr
b. modify the copy and insert the breakpoint at the required
address
c. switch the original page with the copy containing the
breakpoint
d. flush page tables.
replace_page() is being replicated here because of some minor
changes in the type of pages and also because Hugh Dickins had
plans to improve replace_page() for KSM specific work.
Instruction analysis on x86 is based on instruction decoder and
determines if an instruction can be probed and determines the
necessary fixups after singlestep. Instruction analysis is done
at probe insertion time so that we avoid having to repeat the
same analysis every time a probe is hit.
A lot of code here is due to the improvement/suggestions/inputs
from Peter Zijlstra.
Changelog:
(v10):
- Add code to clear REX.B prefix as suggested by Denys Vlasenko
and Masami Hiramatsu.
(v9):
- Use insn_offset_modrm as suggested by Masami Hiramatsu.
(v7):
Handle comments from Peter Zijlstra:
- Dont take reference to inode. (expect inode to uprobe_register to be sane).
- Use PTR_ERR to set the return value.
- No need to take reference to inode.
- use PTR_ERR to return error value.
- register and uprobe_unregister share code.
(v5):
- Modified del_consumer as per comments from Peter.
- Drop reference to inode before dropping reference to uprobe.
- Use i_size_read(inode) instead of inode->i_size.
- Ensure uprobe->consumers is NULL, before __uprobe_unregister() is called.
- Includes errno.h as recommended by Stephen Rothwell to fix a build issue
on sparc defconfig
- Remove restrictions while unregistering.
- Earlier code leaked inode references under some conditions while
registering/unregistering.
- Continue the vma-rmap walk even if the intermediate vma doesnt
meet the requirements.
- Validate the vma found by find_vma before inserting/removing the
breakpoint
- Call del_consumer under mutex_lock.
- Use hash locks.
- Handle mremap.
- Introduce find_least_offset_node() instead of close match logic in
find_uprobe
- Uprobes no more depends on MM_OWNER; No reference to task_structs
while inserting/removing a probe.
- Uses read_mapping_page instead of grab_cache_page so that the pages
have valid content.
- pass NULL to get_user_pages for the task parameter.
- call SetPageUptodate on the new page allocated in write_opcode.
- fix leaking a reference to the new page under certain conditions.
- Include Instruction Decoder if Uprobes gets defined.
- Remove const attributes for instruction prefix arrays.
- Uses mm_context to know if the application is 32 bit.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Also-written-by: Jim Keniston <jkenisto@us.ibm.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Roland McGrath <roland@hack.frob.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Anton Arapov <anton@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Denys Vlasenko <vda.linux@googlemail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linux-mm <linux-mm@kvack.org>
Link: http://lkml.kernel.org/r/20120209092642.GE16600@linux.vnet.ibm.com
[ Made various small edits to the commit log ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-02-09 17:26:42 +08:00
|
|
|
* @page: the cowed page we are replacing by kpage
|
|
|
|
* @kpage: the modified page we replace page by
|
|
|
|
*
|
|
|
|
* Returns 0 on success, -EFAULT on failure.
|
|
|
|
*/
|
2012-07-30 02:22:16 +08:00
|
|
|
static int __replace_page(struct vm_area_struct *vma, unsigned long addr,
|
2016-08-17 23:37:04 +08:00
|
|
|
struct page *old_page, struct page *new_page)
|
uprobes, mm, x86: Add the ability to install and remove uprobes breakpoints
Add uprobes support to the core kernel, with x86 support.
This commit adds the kernel facilities, the actual uprobes
user-space ABI and perf probe support comes in later commits.
General design:
Uprobes are maintained in an rb-tree indexed by inode and offset
(the offset here is from the start of the mapping). For a unique
(inode, offset) tuple, there can be at most one uprobe in the
rb-tree.
Since the (inode, offset) tuple identifies a unique uprobe, more
than one user may be interested in the same uprobe. This provides
the ability to connect multiple 'consumers' to the same uprobe.
Each consumer defines a handler and a filter (optional). The
'handler' is run every time the uprobe is hit, if it matches the
'filter' criteria.
The first consumer of a uprobe causes the breakpoint to be
inserted at the specified address and subsequent consumers are
appended to this list. On subsequent probes, the consumer gets
appended to the existing list of consumers. The breakpoint is
removed when the last consumer unregisters. For all other
unregisterations, the consumer is removed from the list of
consumers.
Given a inode, we get a list of the mms that have mapped the
inode. Do the actual registration if mm maps the page where a
probe needs to be inserted/removed.
We use a temporary list to walk through the vmas that map the
inode.
- The number of maps that map the inode, is not known before we
walk the rmap and keeps changing.
- extending vm_area_struct wasn't recommended, it's a
size-critical data structure.
- There can be more than one maps of the inode in the same mm.
We add callbacks to the mmap methods to keep an eye on text vmas
that are of interest to uprobes. When a vma of interest is mapped,
we insert the breakpoint at the right address.
Uprobe works by replacing the instruction at the address defined
by (inode, offset) with the arch specific breakpoint
instruction. We save a copy of the original instruction at the
uprobed address.
This is needed for:
a. executing the instruction out-of-line (xol).
b. instruction analysis for any subsequent fixups.
c. restoring the instruction back when the uprobe is unregistered.
We insert or delete a breakpoint instruction, and this
breakpoint instruction is assumed to be the smallest instruction
available on the platform. For fixed size instruction platforms
this is trivially true, for variable size instruction platforms
the breakpoint instruction is typically the smallest (often a
single byte).
Writing the instruction is done by COWing the page and changing
the instruction during the copy, this even though most platforms
allow atomic writes of the breakpoint instruction. This also
mirrors the behaviour of a ptrace() memory write to a PRIVATE
file map.
The core worker is derived from KSM's replace_page() logic.
In essence, similar to KSM:
a. allocate a new page and copy over contents of the page that
has the uprobed vaddr
b. modify the copy and insert the breakpoint at the required
address
c. switch the original page with the copy containing the
breakpoint
d. flush page tables.
replace_page() is being replicated here because of some minor
changes in the type of pages and also because Hugh Dickins had
plans to improve replace_page() for KSM specific work.
Instruction analysis on x86 is based on instruction decoder and
determines if an instruction can be probed and determines the
necessary fixups after singlestep. Instruction analysis is done
at probe insertion time so that we avoid having to repeat the
same analysis every time a probe is hit.
A lot of code here is due to the improvement/suggestions/inputs
from Peter Zijlstra.
Changelog:
(v10):
- Add code to clear REX.B prefix as suggested by Denys Vlasenko
and Masami Hiramatsu.
(v9):
- Use insn_offset_modrm as suggested by Masami Hiramatsu.
(v7):
Handle comments from Peter Zijlstra:
- Dont take reference to inode. (expect inode to uprobe_register to be sane).
- Use PTR_ERR to set the return value.
- No need to take reference to inode.
- use PTR_ERR to return error value.
- register and uprobe_unregister share code.
(v5):
- Modified del_consumer as per comments from Peter.
- Drop reference to inode before dropping reference to uprobe.
- Use i_size_read(inode) instead of inode->i_size.
- Ensure uprobe->consumers is NULL, before __uprobe_unregister() is called.
- Includes errno.h as recommended by Stephen Rothwell to fix a build issue
on sparc defconfig
- Remove restrictions while unregistering.
- Earlier code leaked inode references under some conditions while
registering/unregistering.
- Continue the vma-rmap walk even if the intermediate vma doesnt
meet the requirements.
- Validate the vma found by find_vma before inserting/removing the
breakpoint
- Call del_consumer under mutex_lock.
- Use hash locks.
- Handle mremap.
- Introduce find_least_offset_node() instead of close match logic in
find_uprobe
- Uprobes no more depends on MM_OWNER; No reference to task_structs
while inserting/removing a probe.
- Uses read_mapping_page instead of grab_cache_page so that the pages
have valid content.
- pass NULL to get_user_pages for the task parameter.
- call SetPageUptodate on the new page allocated in write_opcode.
- fix leaking a reference to the new page under certain conditions.
- Include Instruction Decoder if Uprobes gets defined.
- Remove const attributes for instruction prefix arrays.
- Uses mm_context to know if the application is 32 bit.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Also-written-by: Jim Keniston <jkenisto@us.ibm.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Roland McGrath <roland@hack.frob.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Anton Arapov <anton@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Denys Vlasenko <vda.linux@googlemail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linux-mm <linux-mm@kvack.org>
Link: http://lkml.kernel.org/r/20120209092642.GE16600@linux.vnet.ibm.com
[ Made various small edits to the commit log ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-02-09 17:26:42 +08:00
|
|
|
{
|
|
|
|
struct mm_struct *mm = vma->vm_mm;
|
2017-02-25 06:58:07 +08:00
|
|
|
struct page_vma_mapped_walk pvmw = {
|
|
|
|
.page = old_page,
|
|
|
|
.vma = vma,
|
|
|
|
.address = addr,
|
|
|
|
};
|
2012-07-30 02:22:20 +08:00
|
|
|
int err;
|
2012-10-09 07:33:35 +08:00
|
|
|
/* For mmu_notifiers */
|
|
|
|
const unsigned long mmun_start = addr;
|
|
|
|
const unsigned long mmun_end = addr + PAGE_SIZE;
|
mm: memcontrol: rewrite charge API
These patches rework memcg charge lifetime to integrate more naturally
with the lifetime of user pages. This drastically simplifies the code and
reduces charging and uncharging overhead. The most expensive part of
charging and uncharging is the page_cgroup bit spinlock, which is removed
entirely after this series.
Here are the top-10 profile entries of a stress test that reads a 128G
sparse file on a freshly booted box, without even a dedicated cgroup (i.e.
executing in the root memcg). Before:
15.36% cat [kernel.kallsyms] [k] copy_user_generic_string
13.31% cat [kernel.kallsyms] [k] memset
11.48% cat [kernel.kallsyms] [k] do_mpage_readpage
4.23% cat [kernel.kallsyms] [k] get_page_from_freelist
2.38% cat [kernel.kallsyms] [k] put_page
2.32% cat [kernel.kallsyms] [k] __mem_cgroup_commit_charge
2.18% kswapd0 [kernel.kallsyms] [k] __mem_cgroup_uncharge_common
1.92% kswapd0 [kernel.kallsyms] [k] shrink_page_list
1.86% cat [kernel.kallsyms] [k] __radix_tree_lookup
1.62% cat [kernel.kallsyms] [k] __pagevec_lru_add_fn
After:
15.67% cat [kernel.kallsyms] [k] copy_user_generic_string
13.48% cat [kernel.kallsyms] [k] memset
11.42% cat [kernel.kallsyms] [k] do_mpage_readpage
3.98% cat [kernel.kallsyms] [k] get_page_from_freelist
2.46% cat [kernel.kallsyms] [k] put_page
2.13% kswapd0 [kernel.kallsyms] [k] shrink_page_list
1.88% cat [kernel.kallsyms] [k] __radix_tree_lookup
1.67% cat [kernel.kallsyms] [k] __pagevec_lru_add_fn
1.39% kswapd0 [kernel.kallsyms] [k] free_pcppages_bulk
1.30% cat [kernel.kallsyms] [k] kfree
As you can see, the memcg footprint has shrunk quite a bit.
text data bss dec hex filename
37970 9892 400 48262 bc86 mm/memcontrol.o.old
35239 9892 400 45531 b1db mm/memcontrol.o
This patch (of 4):
The memcg charge API charges pages before they are rmapped - i.e. have an
actual "type" - and so every callsite needs its own set of charge and
uncharge functions to know what type is being operated on. Worse,
uncharge has to happen from a context that is still type-specific, rather
than at the end of the page's lifetime with exclusive access, and so
requires a lot of synchronization.
Rewrite the charge API to provide a generic set of try_charge(),
commit_charge() and cancel_charge() transaction operations, much like
what's currently done for swap-in:
mem_cgroup_try_charge() attempts to reserve a charge, reclaiming
pages from the memcg if necessary.
mem_cgroup_commit_charge() commits the page to the charge once it
has a valid page->mapping and PageAnon() reliably tells the type.
mem_cgroup_cancel_charge() aborts the transaction.
This reduces the charge API and enables subsequent patches to
drastically simplify uncharging.
As pages need to be committed after rmap is established but before they
are added to the LRU, page_add_new_anon_rmap() must stop doing LRU
additions again. Revive lru_cache_add_active_or_unevictable().
[hughd@google.com: fix shmem_unuse]
[hughd@google.com: Add comments on the private use of -EAGAIN]
Signed-off-by: Johannes Weiner <hannes@cmpxchg.org>
Acked-by: Michal Hocko <mhocko@suse.cz>
Cc: Tejun Heo <tj@kernel.org>
Cc: Vladimir Davydov <vdavydov@parallels.com>
Signed-off-by: Hugh Dickins <hughd@google.com>
Cc: Naoya Horiguchi <n-horiguchi@ah.jp.nec.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2014-08-09 05:19:20 +08:00
|
|
|
struct mem_cgroup *memcg;
|
|
|
|
|
2017-02-25 06:58:07 +08:00
|
|
|
VM_BUG_ON_PAGE(PageTransHuge(old_page), old_page);
|
|
|
|
|
2016-08-17 23:37:04 +08:00
|
|
|
err = mem_cgroup_try_charge(new_page, vma->vm_mm, GFP_KERNEL, &memcg,
|
2016-01-16 08:52:20 +08:00
|
|
|
false);
|
mm: memcontrol: rewrite charge API
These patches rework memcg charge lifetime to integrate more naturally
with the lifetime of user pages. This drastically simplifies the code and
reduces charging and uncharging overhead. The most expensive part of
charging and uncharging is the page_cgroup bit spinlock, which is removed
entirely after this series.
Here are the top-10 profile entries of a stress test that reads a 128G
sparse file on a freshly booted box, without even a dedicated cgroup (i.e.
executing in the root memcg). Before:
15.36% cat [kernel.kallsyms] [k] copy_user_generic_string
13.31% cat [kernel.kallsyms] [k] memset
11.48% cat [kernel.kallsyms] [k] do_mpage_readpage
4.23% cat [kernel.kallsyms] [k] get_page_from_freelist
2.38% cat [kernel.kallsyms] [k] put_page
2.32% cat [kernel.kallsyms] [k] __mem_cgroup_commit_charge
2.18% kswapd0 [kernel.kallsyms] [k] __mem_cgroup_uncharge_common
1.92% kswapd0 [kernel.kallsyms] [k] shrink_page_list
1.86% cat [kernel.kallsyms] [k] __radix_tree_lookup
1.62% cat [kernel.kallsyms] [k] __pagevec_lru_add_fn
After:
15.67% cat [kernel.kallsyms] [k] copy_user_generic_string
13.48% cat [kernel.kallsyms] [k] memset
11.42% cat [kernel.kallsyms] [k] do_mpage_readpage
3.98% cat [kernel.kallsyms] [k] get_page_from_freelist
2.46% cat [kernel.kallsyms] [k] put_page
2.13% kswapd0 [kernel.kallsyms] [k] shrink_page_list
1.88% cat [kernel.kallsyms] [k] __radix_tree_lookup
1.67% cat [kernel.kallsyms] [k] __pagevec_lru_add_fn
1.39% kswapd0 [kernel.kallsyms] [k] free_pcppages_bulk
1.30% cat [kernel.kallsyms] [k] kfree
As you can see, the memcg footprint has shrunk quite a bit.
text data bss dec hex filename
37970 9892 400 48262 bc86 mm/memcontrol.o.old
35239 9892 400 45531 b1db mm/memcontrol.o
This patch (of 4):
The memcg charge API charges pages before they are rmapped - i.e. have an
actual "type" - and so every callsite needs its own set of charge and
uncharge functions to know what type is being operated on. Worse,
uncharge has to happen from a context that is still type-specific, rather
than at the end of the page's lifetime with exclusive access, and so
requires a lot of synchronization.
Rewrite the charge API to provide a generic set of try_charge(),
commit_charge() and cancel_charge() transaction operations, much like
what's currently done for swap-in:
mem_cgroup_try_charge() attempts to reserve a charge, reclaiming
pages from the memcg if necessary.
mem_cgroup_commit_charge() commits the page to the charge once it
has a valid page->mapping and PageAnon() reliably tells the type.
mem_cgroup_cancel_charge() aborts the transaction.
This reduces the charge API and enables subsequent patches to
drastically simplify uncharging.
As pages need to be committed after rmap is established but before they
are added to the LRU, page_add_new_anon_rmap() must stop doing LRU
additions again. Revive lru_cache_add_active_or_unevictable().
[hughd@google.com: fix shmem_unuse]
[hughd@google.com: Add comments on the private use of -EAGAIN]
Signed-off-by: Johannes Weiner <hannes@cmpxchg.org>
Acked-by: Michal Hocko <mhocko@suse.cz>
Cc: Tejun Heo <tj@kernel.org>
Cc: Vladimir Davydov <vdavydov@parallels.com>
Signed-off-by: Hugh Dickins <hughd@google.com>
Cc: Naoya Horiguchi <n-horiguchi@ah.jp.nec.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2014-08-09 05:19:20 +08:00
|
|
|
if (err)
|
|
|
|
return err;
|
uprobes, mm, x86: Add the ability to install and remove uprobes breakpoints
Add uprobes support to the core kernel, with x86 support.
This commit adds the kernel facilities, the actual uprobes
user-space ABI and perf probe support comes in later commits.
General design:
Uprobes are maintained in an rb-tree indexed by inode and offset
(the offset here is from the start of the mapping). For a unique
(inode, offset) tuple, there can be at most one uprobe in the
rb-tree.
Since the (inode, offset) tuple identifies a unique uprobe, more
than one user may be interested in the same uprobe. This provides
the ability to connect multiple 'consumers' to the same uprobe.
Each consumer defines a handler and a filter (optional). The
'handler' is run every time the uprobe is hit, if it matches the
'filter' criteria.
The first consumer of a uprobe causes the breakpoint to be
inserted at the specified address and subsequent consumers are
appended to this list. On subsequent probes, the consumer gets
appended to the existing list of consumers. The breakpoint is
removed when the last consumer unregisters. For all other
unregisterations, the consumer is removed from the list of
consumers.
Given a inode, we get a list of the mms that have mapped the
inode. Do the actual registration if mm maps the page where a
probe needs to be inserted/removed.
We use a temporary list to walk through the vmas that map the
inode.
- The number of maps that map the inode, is not known before we
walk the rmap and keeps changing.
- extending vm_area_struct wasn't recommended, it's a
size-critical data structure.
- There can be more than one maps of the inode in the same mm.
We add callbacks to the mmap methods to keep an eye on text vmas
that are of interest to uprobes. When a vma of interest is mapped,
we insert the breakpoint at the right address.
Uprobe works by replacing the instruction at the address defined
by (inode, offset) with the arch specific breakpoint
instruction. We save a copy of the original instruction at the
uprobed address.
This is needed for:
a. executing the instruction out-of-line (xol).
b. instruction analysis for any subsequent fixups.
c. restoring the instruction back when the uprobe is unregistered.
We insert or delete a breakpoint instruction, and this
breakpoint instruction is assumed to be the smallest instruction
available on the platform. For fixed size instruction platforms
this is trivially true, for variable size instruction platforms
the breakpoint instruction is typically the smallest (often a
single byte).
Writing the instruction is done by COWing the page and changing
the instruction during the copy, this even though most platforms
allow atomic writes of the breakpoint instruction. This also
mirrors the behaviour of a ptrace() memory write to a PRIVATE
file map.
The core worker is derived from KSM's replace_page() logic.
In essence, similar to KSM:
a. allocate a new page and copy over contents of the page that
has the uprobed vaddr
b. modify the copy and insert the breakpoint at the required
address
c. switch the original page with the copy containing the
breakpoint
d. flush page tables.
replace_page() is being replicated here because of some minor
changes in the type of pages and also because Hugh Dickins had
plans to improve replace_page() for KSM specific work.
Instruction analysis on x86 is based on instruction decoder and
determines if an instruction can be probed and determines the
necessary fixups after singlestep. Instruction analysis is done
at probe insertion time so that we avoid having to repeat the
same analysis every time a probe is hit.
A lot of code here is due to the improvement/suggestions/inputs
from Peter Zijlstra.
Changelog:
(v10):
- Add code to clear REX.B prefix as suggested by Denys Vlasenko
and Masami Hiramatsu.
(v9):
- Use insn_offset_modrm as suggested by Masami Hiramatsu.
(v7):
Handle comments from Peter Zijlstra:
- Dont take reference to inode. (expect inode to uprobe_register to be sane).
- Use PTR_ERR to set the return value.
- No need to take reference to inode.
- use PTR_ERR to return error value.
- register and uprobe_unregister share code.
(v5):
- Modified del_consumer as per comments from Peter.
- Drop reference to inode before dropping reference to uprobe.
- Use i_size_read(inode) instead of inode->i_size.
- Ensure uprobe->consumers is NULL, before __uprobe_unregister() is called.
- Includes errno.h as recommended by Stephen Rothwell to fix a build issue
on sparc defconfig
- Remove restrictions while unregistering.
- Earlier code leaked inode references under some conditions while
registering/unregistering.
- Continue the vma-rmap walk even if the intermediate vma doesnt
meet the requirements.
- Validate the vma found by find_vma before inserting/removing the
breakpoint
- Call del_consumer under mutex_lock.
- Use hash locks.
- Handle mremap.
- Introduce find_least_offset_node() instead of close match logic in
find_uprobe
- Uprobes no more depends on MM_OWNER; No reference to task_structs
while inserting/removing a probe.
- Uses read_mapping_page instead of grab_cache_page so that the pages
have valid content.
- pass NULL to get_user_pages for the task parameter.
- call SetPageUptodate on the new page allocated in write_opcode.
- fix leaking a reference to the new page under certain conditions.
- Include Instruction Decoder if Uprobes gets defined.
- Remove const attributes for instruction prefix arrays.
- Uses mm_context to know if the application is 32 bit.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Also-written-by: Jim Keniston <jkenisto@us.ibm.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Roland McGrath <roland@hack.frob.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Anton Arapov <anton@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Denys Vlasenko <vda.linux@googlemail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linux-mm <linux-mm@kvack.org>
Link: http://lkml.kernel.org/r/20120209092642.GE16600@linux.vnet.ibm.com
[ Made various small edits to the commit log ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-02-09 17:26:42 +08:00
|
|
|
|
2012-07-30 02:22:49 +08:00
|
|
|
/* For try_to_free_swap() and munlock_vma_page() below */
|
2016-08-17 23:37:04 +08:00
|
|
|
lock_page(old_page);
|
2012-07-30 02:22:20 +08:00
|
|
|
|
2012-10-09 07:33:35 +08:00
|
|
|
mmu_notifier_invalidate_range_start(mm, mmun_start, mmun_end);
|
2012-07-30 02:22:20 +08:00
|
|
|
err = -EAGAIN;
|
2017-02-25 06:58:07 +08:00
|
|
|
if (!page_vma_mapped_walk(&pvmw)) {
|
2016-08-17 23:37:04 +08:00
|
|
|
mem_cgroup_cancel_charge(new_page, memcg, false);
|
2012-07-30 02:22:20 +08:00
|
|
|
goto unlock;
|
2016-08-17 23:36:29 +08:00
|
|
|
}
|
2017-02-25 06:58:07 +08:00
|
|
|
VM_BUG_ON_PAGE(addr != pvmw.address, old_page);
|
uprobes, mm, x86: Add the ability to install and remove uprobes breakpoints
Add uprobes support to the core kernel, with x86 support.
This commit adds the kernel facilities, the actual uprobes
user-space ABI and perf probe support comes in later commits.
General design:
Uprobes are maintained in an rb-tree indexed by inode and offset
(the offset here is from the start of the mapping). For a unique
(inode, offset) tuple, there can be at most one uprobe in the
rb-tree.
Since the (inode, offset) tuple identifies a unique uprobe, more
than one user may be interested in the same uprobe. This provides
the ability to connect multiple 'consumers' to the same uprobe.
Each consumer defines a handler and a filter (optional). The
'handler' is run every time the uprobe is hit, if it matches the
'filter' criteria.
The first consumer of a uprobe causes the breakpoint to be
inserted at the specified address and subsequent consumers are
appended to this list. On subsequent probes, the consumer gets
appended to the existing list of consumers. The breakpoint is
removed when the last consumer unregisters. For all other
unregisterations, the consumer is removed from the list of
consumers.
Given a inode, we get a list of the mms that have mapped the
inode. Do the actual registration if mm maps the page where a
probe needs to be inserted/removed.
We use a temporary list to walk through the vmas that map the
inode.
- The number of maps that map the inode, is not known before we
walk the rmap and keeps changing.
- extending vm_area_struct wasn't recommended, it's a
size-critical data structure.
- There can be more than one maps of the inode in the same mm.
We add callbacks to the mmap methods to keep an eye on text vmas
that are of interest to uprobes. When a vma of interest is mapped,
we insert the breakpoint at the right address.
Uprobe works by replacing the instruction at the address defined
by (inode, offset) with the arch specific breakpoint
instruction. We save a copy of the original instruction at the
uprobed address.
This is needed for:
a. executing the instruction out-of-line (xol).
b. instruction analysis for any subsequent fixups.
c. restoring the instruction back when the uprobe is unregistered.
We insert or delete a breakpoint instruction, and this
breakpoint instruction is assumed to be the smallest instruction
available on the platform. For fixed size instruction platforms
this is trivially true, for variable size instruction platforms
the breakpoint instruction is typically the smallest (often a
single byte).
Writing the instruction is done by COWing the page and changing
the instruction during the copy, this even though most platforms
allow atomic writes of the breakpoint instruction. This also
mirrors the behaviour of a ptrace() memory write to a PRIVATE
file map.
The core worker is derived from KSM's replace_page() logic.
In essence, similar to KSM:
a. allocate a new page and copy over contents of the page that
has the uprobed vaddr
b. modify the copy and insert the breakpoint at the required
address
c. switch the original page with the copy containing the
breakpoint
d. flush page tables.
replace_page() is being replicated here because of some minor
changes in the type of pages and also because Hugh Dickins had
plans to improve replace_page() for KSM specific work.
Instruction analysis on x86 is based on instruction decoder and
determines if an instruction can be probed and determines the
necessary fixups after singlestep. Instruction analysis is done
at probe insertion time so that we avoid having to repeat the
same analysis every time a probe is hit.
A lot of code here is due to the improvement/suggestions/inputs
from Peter Zijlstra.
Changelog:
(v10):
- Add code to clear REX.B prefix as suggested by Denys Vlasenko
and Masami Hiramatsu.
(v9):
- Use insn_offset_modrm as suggested by Masami Hiramatsu.
(v7):
Handle comments from Peter Zijlstra:
- Dont take reference to inode. (expect inode to uprobe_register to be sane).
- Use PTR_ERR to set the return value.
- No need to take reference to inode.
- use PTR_ERR to return error value.
- register and uprobe_unregister share code.
(v5):
- Modified del_consumer as per comments from Peter.
- Drop reference to inode before dropping reference to uprobe.
- Use i_size_read(inode) instead of inode->i_size.
- Ensure uprobe->consumers is NULL, before __uprobe_unregister() is called.
- Includes errno.h as recommended by Stephen Rothwell to fix a build issue
on sparc defconfig
- Remove restrictions while unregistering.
- Earlier code leaked inode references under some conditions while
registering/unregistering.
- Continue the vma-rmap walk even if the intermediate vma doesnt
meet the requirements.
- Validate the vma found by find_vma before inserting/removing the
breakpoint
- Call del_consumer under mutex_lock.
- Use hash locks.
- Handle mremap.
- Introduce find_least_offset_node() instead of close match logic in
find_uprobe
- Uprobes no more depends on MM_OWNER; No reference to task_structs
while inserting/removing a probe.
- Uses read_mapping_page instead of grab_cache_page so that the pages
have valid content.
- pass NULL to get_user_pages for the task parameter.
- call SetPageUptodate on the new page allocated in write_opcode.
- fix leaking a reference to the new page under certain conditions.
- Include Instruction Decoder if Uprobes gets defined.
- Remove const attributes for instruction prefix arrays.
- Uses mm_context to know if the application is 32 bit.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Also-written-by: Jim Keniston <jkenisto@us.ibm.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Roland McGrath <roland@hack.frob.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Anton Arapov <anton@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Denys Vlasenko <vda.linux@googlemail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linux-mm <linux-mm@kvack.org>
Link: http://lkml.kernel.org/r/20120209092642.GE16600@linux.vnet.ibm.com
[ Made various small edits to the commit log ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-02-09 17:26:42 +08:00
|
|
|
|
2016-08-17 23:37:04 +08:00
|
|
|
get_page(new_page);
|
|
|
|
page_add_new_anon_rmap(new_page, vma, addr, false);
|
|
|
|
mem_cgroup_commit_charge(new_page, memcg, false, false);
|
|
|
|
lru_cache_add_active_or_unevictable(new_page, vma);
|
uprobes, mm, x86: Add the ability to install and remove uprobes breakpoints
Add uprobes support to the core kernel, with x86 support.
This commit adds the kernel facilities, the actual uprobes
user-space ABI and perf probe support comes in later commits.
General design:
Uprobes are maintained in an rb-tree indexed by inode and offset
(the offset here is from the start of the mapping). For a unique
(inode, offset) tuple, there can be at most one uprobe in the
rb-tree.
Since the (inode, offset) tuple identifies a unique uprobe, more
than one user may be interested in the same uprobe. This provides
the ability to connect multiple 'consumers' to the same uprobe.
Each consumer defines a handler and a filter (optional). The
'handler' is run every time the uprobe is hit, if it matches the
'filter' criteria.
The first consumer of a uprobe causes the breakpoint to be
inserted at the specified address and subsequent consumers are
appended to this list. On subsequent probes, the consumer gets
appended to the existing list of consumers. The breakpoint is
removed when the last consumer unregisters. For all other
unregisterations, the consumer is removed from the list of
consumers.
Given a inode, we get a list of the mms that have mapped the
inode. Do the actual registration if mm maps the page where a
probe needs to be inserted/removed.
We use a temporary list to walk through the vmas that map the
inode.
- The number of maps that map the inode, is not known before we
walk the rmap and keeps changing.
- extending vm_area_struct wasn't recommended, it's a
size-critical data structure.
- There can be more than one maps of the inode in the same mm.
We add callbacks to the mmap methods to keep an eye on text vmas
that are of interest to uprobes. When a vma of interest is mapped,
we insert the breakpoint at the right address.
Uprobe works by replacing the instruction at the address defined
by (inode, offset) with the arch specific breakpoint
instruction. We save a copy of the original instruction at the
uprobed address.
This is needed for:
a. executing the instruction out-of-line (xol).
b. instruction analysis for any subsequent fixups.
c. restoring the instruction back when the uprobe is unregistered.
We insert or delete a breakpoint instruction, and this
breakpoint instruction is assumed to be the smallest instruction
available on the platform. For fixed size instruction platforms
this is trivially true, for variable size instruction platforms
the breakpoint instruction is typically the smallest (often a
single byte).
Writing the instruction is done by COWing the page and changing
the instruction during the copy, this even though most platforms
allow atomic writes of the breakpoint instruction. This also
mirrors the behaviour of a ptrace() memory write to a PRIVATE
file map.
The core worker is derived from KSM's replace_page() logic.
In essence, similar to KSM:
a. allocate a new page and copy over contents of the page that
has the uprobed vaddr
b. modify the copy and insert the breakpoint at the required
address
c. switch the original page with the copy containing the
breakpoint
d. flush page tables.
replace_page() is being replicated here because of some minor
changes in the type of pages and also because Hugh Dickins had
plans to improve replace_page() for KSM specific work.
Instruction analysis on x86 is based on instruction decoder and
determines if an instruction can be probed and determines the
necessary fixups after singlestep. Instruction analysis is done
at probe insertion time so that we avoid having to repeat the
same analysis every time a probe is hit.
A lot of code here is due to the improvement/suggestions/inputs
from Peter Zijlstra.
Changelog:
(v10):
- Add code to clear REX.B prefix as suggested by Denys Vlasenko
and Masami Hiramatsu.
(v9):
- Use insn_offset_modrm as suggested by Masami Hiramatsu.
(v7):
Handle comments from Peter Zijlstra:
- Dont take reference to inode. (expect inode to uprobe_register to be sane).
- Use PTR_ERR to set the return value.
- No need to take reference to inode.
- use PTR_ERR to return error value.
- register and uprobe_unregister share code.
(v5):
- Modified del_consumer as per comments from Peter.
- Drop reference to inode before dropping reference to uprobe.
- Use i_size_read(inode) instead of inode->i_size.
- Ensure uprobe->consumers is NULL, before __uprobe_unregister() is called.
- Includes errno.h as recommended by Stephen Rothwell to fix a build issue
on sparc defconfig
- Remove restrictions while unregistering.
- Earlier code leaked inode references under some conditions while
registering/unregistering.
- Continue the vma-rmap walk even if the intermediate vma doesnt
meet the requirements.
- Validate the vma found by find_vma before inserting/removing the
breakpoint
- Call del_consumer under mutex_lock.
- Use hash locks.
- Handle mremap.
- Introduce find_least_offset_node() instead of close match logic in
find_uprobe
- Uprobes no more depends on MM_OWNER; No reference to task_structs
while inserting/removing a probe.
- Uses read_mapping_page instead of grab_cache_page so that the pages
have valid content.
- pass NULL to get_user_pages for the task parameter.
- call SetPageUptodate on the new page allocated in write_opcode.
- fix leaking a reference to the new page under certain conditions.
- Include Instruction Decoder if Uprobes gets defined.
- Remove const attributes for instruction prefix arrays.
- Uses mm_context to know if the application is 32 bit.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Also-written-by: Jim Keniston <jkenisto@us.ibm.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Roland McGrath <roland@hack.frob.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Anton Arapov <anton@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Denys Vlasenko <vda.linux@googlemail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linux-mm <linux-mm@kvack.org>
Link: http://lkml.kernel.org/r/20120209092642.GE16600@linux.vnet.ibm.com
[ Made various small edits to the commit log ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-02-09 17:26:42 +08:00
|
|
|
|
2016-08-17 23:37:04 +08:00
|
|
|
if (!PageAnon(old_page)) {
|
|
|
|
dec_mm_counter(mm, mm_counter_file(old_page));
|
2012-04-11 18:35:16 +08:00
|
|
|
inc_mm_counter(mm, MM_ANONPAGES);
|
|
|
|
}
|
|
|
|
|
2017-02-25 06:58:07 +08:00
|
|
|
flush_cache_page(vma, addr, pte_pfn(*pvmw.pte));
|
|
|
|
ptep_clear_flush_notify(vma, addr, pvmw.pte);
|
|
|
|
set_pte_at_notify(mm, addr, pvmw.pte,
|
|
|
|
mk_pte(new_page, vma->vm_page_prot));
|
uprobes, mm, x86: Add the ability to install and remove uprobes breakpoints
Add uprobes support to the core kernel, with x86 support.
This commit adds the kernel facilities, the actual uprobes
user-space ABI and perf probe support comes in later commits.
General design:
Uprobes are maintained in an rb-tree indexed by inode and offset
(the offset here is from the start of the mapping). For a unique
(inode, offset) tuple, there can be at most one uprobe in the
rb-tree.
Since the (inode, offset) tuple identifies a unique uprobe, more
than one user may be interested in the same uprobe. This provides
the ability to connect multiple 'consumers' to the same uprobe.
Each consumer defines a handler and a filter (optional). The
'handler' is run every time the uprobe is hit, if it matches the
'filter' criteria.
The first consumer of a uprobe causes the breakpoint to be
inserted at the specified address and subsequent consumers are
appended to this list. On subsequent probes, the consumer gets
appended to the existing list of consumers. The breakpoint is
removed when the last consumer unregisters. For all other
unregisterations, the consumer is removed from the list of
consumers.
Given a inode, we get a list of the mms that have mapped the
inode. Do the actual registration if mm maps the page where a
probe needs to be inserted/removed.
We use a temporary list to walk through the vmas that map the
inode.
- The number of maps that map the inode, is not known before we
walk the rmap and keeps changing.
- extending vm_area_struct wasn't recommended, it's a
size-critical data structure.
- There can be more than one maps of the inode in the same mm.
We add callbacks to the mmap methods to keep an eye on text vmas
that are of interest to uprobes. When a vma of interest is mapped,
we insert the breakpoint at the right address.
Uprobe works by replacing the instruction at the address defined
by (inode, offset) with the arch specific breakpoint
instruction. We save a copy of the original instruction at the
uprobed address.
This is needed for:
a. executing the instruction out-of-line (xol).
b. instruction analysis for any subsequent fixups.
c. restoring the instruction back when the uprobe is unregistered.
We insert or delete a breakpoint instruction, and this
breakpoint instruction is assumed to be the smallest instruction
available on the platform. For fixed size instruction platforms
this is trivially true, for variable size instruction platforms
the breakpoint instruction is typically the smallest (often a
single byte).
Writing the instruction is done by COWing the page and changing
the instruction during the copy, this even though most platforms
allow atomic writes of the breakpoint instruction. This also
mirrors the behaviour of a ptrace() memory write to a PRIVATE
file map.
The core worker is derived from KSM's replace_page() logic.
In essence, similar to KSM:
a. allocate a new page and copy over contents of the page that
has the uprobed vaddr
b. modify the copy and insert the breakpoint at the required
address
c. switch the original page with the copy containing the
breakpoint
d. flush page tables.
replace_page() is being replicated here because of some minor
changes in the type of pages and also because Hugh Dickins had
plans to improve replace_page() for KSM specific work.
Instruction analysis on x86 is based on instruction decoder and
determines if an instruction can be probed and determines the
necessary fixups after singlestep. Instruction analysis is done
at probe insertion time so that we avoid having to repeat the
same analysis every time a probe is hit.
A lot of code here is due to the improvement/suggestions/inputs
from Peter Zijlstra.
Changelog:
(v10):
- Add code to clear REX.B prefix as suggested by Denys Vlasenko
and Masami Hiramatsu.
(v9):
- Use insn_offset_modrm as suggested by Masami Hiramatsu.
(v7):
Handle comments from Peter Zijlstra:
- Dont take reference to inode. (expect inode to uprobe_register to be sane).
- Use PTR_ERR to set the return value.
- No need to take reference to inode.
- use PTR_ERR to return error value.
- register and uprobe_unregister share code.
(v5):
- Modified del_consumer as per comments from Peter.
- Drop reference to inode before dropping reference to uprobe.
- Use i_size_read(inode) instead of inode->i_size.
- Ensure uprobe->consumers is NULL, before __uprobe_unregister() is called.
- Includes errno.h as recommended by Stephen Rothwell to fix a build issue
on sparc defconfig
- Remove restrictions while unregistering.
- Earlier code leaked inode references under some conditions while
registering/unregistering.
- Continue the vma-rmap walk even if the intermediate vma doesnt
meet the requirements.
- Validate the vma found by find_vma before inserting/removing the
breakpoint
- Call del_consumer under mutex_lock.
- Use hash locks.
- Handle mremap.
- Introduce find_least_offset_node() instead of close match logic in
find_uprobe
- Uprobes no more depends on MM_OWNER; No reference to task_structs
while inserting/removing a probe.
- Uses read_mapping_page instead of grab_cache_page so that the pages
have valid content.
- pass NULL to get_user_pages for the task parameter.
- call SetPageUptodate on the new page allocated in write_opcode.
- fix leaking a reference to the new page under certain conditions.
- Include Instruction Decoder if Uprobes gets defined.
- Remove const attributes for instruction prefix arrays.
- Uses mm_context to know if the application is 32 bit.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Also-written-by: Jim Keniston <jkenisto@us.ibm.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Roland McGrath <roland@hack.frob.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Anton Arapov <anton@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Denys Vlasenko <vda.linux@googlemail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linux-mm <linux-mm@kvack.org>
Link: http://lkml.kernel.org/r/20120209092642.GE16600@linux.vnet.ibm.com
[ Made various small edits to the commit log ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-02-09 17:26:42 +08:00
|
|
|
|
2016-08-17 23:37:04 +08:00
|
|
|
page_remove_rmap(old_page, false);
|
|
|
|
if (!page_mapped(old_page))
|
|
|
|
try_to_free_swap(old_page);
|
2017-02-25 06:58:07 +08:00
|
|
|
page_vma_mapped_walk_done(&pvmw);
|
uprobes, mm, x86: Add the ability to install and remove uprobes breakpoints
Add uprobes support to the core kernel, with x86 support.
This commit adds the kernel facilities, the actual uprobes
user-space ABI and perf probe support comes in later commits.
General design:
Uprobes are maintained in an rb-tree indexed by inode and offset
(the offset here is from the start of the mapping). For a unique
(inode, offset) tuple, there can be at most one uprobe in the
rb-tree.
Since the (inode, offset) tuple identifies a unique uprobe, more
than one user may be interested in the same uprobe. This provides
the ability to connect multiple 'consumers' to the same uprobe.
Each consumer defines a handler and a filter (optional). The
'handler' is run every time the uprobe is hit, if it matches the
'filter' criteria.
The first consumer of a uprobe causes the breakpoint to be
inserted at the specified address and subsequent consumers are
appended to this list. On subsequent probes, the consumer gets
appended to the existing list of consumers. The breakpoint is
removed when the last consumer unregisters. For all other
unregisterations, the consumer is removed from the list of
consumers.
Given a inode, we get a list of the mms that have mapped the
inode. Do the actual registration if mm maps the page where a
probe needs to be inserted/removed.
We use a temporary list to walk through the vmas that map the
inode.
- The number of maps that map the inode, is not known before we
walk the rmap and keeps changing.
- extending vm_area_struct wasn't recommended, it's a
size-critical data structure.
- There can be more than one maps of the inode in the same mm.
We add callbacks to the mmap methods to keep an eye on text vmas
that are of interest to uprobes. When a vma of interest is mapped,
we insert the breakpoint at the right address.
Uprobe works by replacing the instruction at the address defined
by (inode, offset) with the arch specific breakpoint
instruction. We save a copy of the original instruction at the
uprobed address.
This is needed for:
a. executing the instruction out-of-line (xol).
b. instruction analysis for any subsequent fixups.
c. restoring the instruction back when the uprobe is unregistered.
We insert or delete a breakpoint instruction, and this
breakpoint instruction is assumed to be the smallest instruction
available on the platform. For fixed size instruction platforms
this is trivially true, for variable size instruction platforms
the breakpoint instruction is typically the smallest (often a
single byte).
Writing the instruction is done by COWing the page and changing
the instruction during the copy, this even though most platforms
allow atomic writes of the breakpoint instruction. This also
mirrors the behaviour of a ptrace() memory write to a PRIVATE
file map.
The core worker is derived from KSM's replace_page() logic.
In essence, similar to KSM:
a. allocate a new page and copy over contents of the page that
has the uprobed vaddr
b. modify the copy and insert the breakpoint at the required
address
c. switch the original page with the copy containing the
breakpoint
d. flush page tables.
replace_page() is being replicated here because of some minor
changes in the type of pages and also because Hugh Dickins had
plans to improve replace_page() for KSM specific work.
Instruction analysis on x86 is based on instruction decoder and
determines if an instruction can be probed and determines the
necessary fixups after singlestep. Instruction analysis is done
at probe insertion time so that we avoid having to repeat the
same analysis every time a probe is hit.
A lot of code here is due to the improvement/suggestions/inputs
from Peter Zijlstra.
Changelog:
(v10):
- Add code to clear REX.B prefix as suggested by Denys Vlasenko
and Masami Hiramatsu.
(v9):
- Use insn_offset_modrm as suggested by Masami Hiramatsu.
(v7):
Handle comments from Peter Zijlstra:
- Dont take reference to inode. (expect inode to uprobe_register to be sane).
- Use PTR_ERR to set the return value.
- No need to take reference to inode.
- use PTR_ERR to return error value.
- register and uprobe_unregister share code.
(v5):
- Modified del_consumer as per comments from Peter.
- Drop reference to inode before dropping reference to uprobe.
- Use i_size_read(inode) instead of inode->i_size.
- Ensure uprobe->consumers is NULL, before __uprobe_unregister() is called.
- Includes errno.h as recommended by Stephen Rothwell to fix a build issue
on sparc defconfig
- Remove restrictions while unregistering.
- Earlier code leaked inode references under some conditions while
registering/unregistering.
- Continue the vma-rmap walk even if the intermediate vma doesnt
meet the requirements.
- Validate the vma found by find_vma before inserting/removing the
breakpoint
- Call del_consumer under mutex_lock.
- Use hash locks.
- Handle mremap.
- Introduce find_least_offset_node() instead of close match logic in
find_uprobe
- Uprobes no more depends on MM_OWNER; No reference to task_structs
while inserting/removing a probe.
- Uses read_mapping_page instead of grab_cache_page so that the pages
have valid content.
- pass NULL to get_user_pages for the task parameter.
- call SetPageUptodate on the new page allocated in write_opcode.
- fix leaking a reference to the new page under certain conditions.
- Include Instruction Decoder if Uprobes gets defined.
- Remove const attributes for instruction prefix arrays.
- Uses mm_context to know if the application is 32 bit.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Also-written-by: Jim Keniston <jkenisto@us.ibm.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Roland McGrath <roland@hack.frob.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Anton Arapov <anton@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Denys Vlasenko <vda.linux@googlemail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linux-mm <linux-mm@kvack.org>
Link: http://lkml.kernel.org/r/20120209092642.GE16600@linux.vnet.ibm.com
[ Made various small edits to the commit log ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-02-09 17:26:42 +08:00
|
|
|
|
2012-07-30 02:22:49 +08:00
|
|
|
if (vma->vm_flags & VM_LOCKED)
|
2016-08-17 23:37:04 +08:00
|
|
|
munlock_vma_page(old_page);
|
|
|
|
put_page(old_page);
|
2012-07-30 02:22:49 +08:00
|
|
|
|
2012-07-30 02:22:20 +08:00
|
|
|
err = 0;
|
|
|
|
unlock:
|
2012-10-09 07:33:35 +08:00
|
|
|
mmu_notifier_invalidate_range_end(mm, mmun_start, mmun_end);
|
2016-08-17 23:37:04 +08:00
|
|
|
unlock_page(old_page);
|
2012-07-30 02:22:20 +08:00
|
|
|
return err;
|
uprobes, mm, x86: Add the ability to install and remove uprobes breakpoints
Add uprobes support to the core kernel, with x86 support.
This commit adds the kernel facilities, the actual uprobes
user-space ABI and perf probe support comes in later commits.
General design:
Uprobes are maintained in an rb-tree indexed by inode and offset
(the offset here is from the start of the mapping). For a unique
(inode, offset) tuple, there can be at most one uprobe in the
rb-tree.
Since the (inode, offset) tuple identifies a unique uprobe, more
than one user may be interested in the same uprobe. This provides
the ability to connect multiple 'consumers' to the same uprobe.
Each consumer defines a handler and a filter (optional). The
'handler' is run every time the uprobe is hit, if it matches the
'filter' criteria.
The first consumer of a uprobe causes the breakpoint to be
inserted at the specified address and subsequent consumers are
appended to this list. On subsequent probes, the consumer gets
appended to the existing list of consumers. The breakpoint is
removed when the last consumer unregisters. For all other
unregisterations, the consumer is removed from the list of
consumers.
Given a inode, we get a list of the mms that have mapped the
inode. Do the actual registration if mm maps the page where a
probe needs to be inserted/removed.
We use a temporary list to walk through the vmas that map the
inode.
- The number of maps that map the inode, is not known before we
walk the rmap and keeps changing.
- extending vm_area_struct wasn't recommended, it's a
size-critical data structure.
- There can be more than one maps of the inode in the same mm.
We add callbacks to the mmap methods to keep an eye on text vmas
that are of interest to uprobes. When a vma of interest is mapped,
we insert the breakpoint at the right address.
Uprobe works by replacing the instruction at the address defined
by (inode, offset) with the arch specific breakpoint
instruction. We save a copy of the original instruction at the
uprobed address.
This is needed for:
a. executing the instruction out-of-line (xol).
b. instruction analysis for any subsequent fixups.
c. restoring the instruction back when the uprobe is unregistered.
We insert or delete a breakpoint instruction, and this
breakpoint instruction is assumed to be the smallest instruction
available on the platform. For fixed size instruction platforms
this is trivially true, for variable size instruction platforms
the breakpoint instruction is typically the smallest (often a
single byte).
Writing the instruction is done by COWing the page and changing
the instruction during the copy, this even though most platforms
allow atomic writes of the breakpoint instruction. This also
mirrors the behaviour of a ptrace() memory write to a PRIVATE
file map.
The core worker is derived from KSM's replace_page() logic.
In essence, similar to KSM:
a. allocate a new page and copy over contents of the page that
has the uprobed vaddr
b. modify the copy and insert the breakpoint at the required
address
c. switch the original page with the copy containing the
breakpoint
d. flush page tables.
replace_page() is being replicated here because of some minor
changes in the type of pages and also because Hugh Dickins had
plans to improve replace_page() for KSM specific work.
Instruction analysis on x86 is based on instruction decoder and
determines if an instruction can be probed and determines the
necessary fixups after singlestep. Instruction analysis is done
at probe insertion time so that we avoid having to repeat the
same analysis every time a probe is hit.
A lot of code here is due to the improvement/suggestions/inputs
from Peter Zijlstra.
Changelog:
(v10):
- Add code to clear REX.B prefix as suggested by Denys Vlasenko
and Masami Hiramatsu.
(v9):
- Use insn_offset_modrm as suggested by Masami Hiramatsu.
(v7):
Handle comments from Peter Zijlstra:
- Dont take reference to inode. (expect inode to uprobe_register to be sane).
- Use PTR_ERR to set the return value.
- No need to take reference to inode.
- use PTR_ERR to return error value.
- register and uprobe_unregister share code.
(v5):
- Modified del_consumer as per comments from Peter.
- Drop reference to inode before dropping reference to uprobe.
- Use i_size_read(inode) instead of inode->i_size.
- Ensure uprobe->consumers is NULL, before __uprobe_unregister() is called.
- Includes errno.h as recommended by Stephen Rothwell to fix a build issue
on sparc defconfig
- Remove restrictions while unregistering.
- Earlier code leaked inode references under some conditions while
registering/unregistering.
- Continue the vma-rmap walk even if the intermediate vma doesnt
meet the requirements.
- Validate the vma found by find_vma before inserting/removing the
breakpoint
- Call del_consumer under mutex_lock.
- Use hash locks.
- Handle mremap.
- Introduce find_least_offset_node() instead of close match logic in
find_uprobe
- Uprobes no more depends on MM_OWNER; No reference to task_structs
while inserting/removing a probe.
- Uses read_mapping_page instead of grab_cache_page so that the pages
have valid content.
- pass NULL to get_user_pages for the task parameter.
- call SetPageUptodate on the new page allocated in write_opcode.
- fix leaking a reference to the new page under certain conditions.
- Include Instruction Decoder if Uprobes gets defined.
- Remove const attributes for instruction prefix arrays.
- Uses mm_context to know if the application is 32 bit.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Also-written-by: Jim Keniston <jkenisto@us.ibm.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Roland McGrath <roland@hack.frob.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Anton Arapov <anton@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Denys Vlasenko <vda.linux@googlemail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linux-mm <linux-mm@kvack.org>
Link: http://lkml.kernel.org/r/20120209092642.GE16600@linux.vnet.ibm.com
[ Made various small edits to the commit log ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-02-09 17:26:42 +08:00
|
|
|
}
|
|
|
|
|
|
|
|
/**
|
2012-03-12 17:25:45 +08:00
|
|
|
* is_swbp_insn - check if instruction is breakpoint instruction.
|
uprobes, mm, x86: Add the ability to install and remove uprobes breakpoints
Add uprobes support to the core kernel, with x86 support.
This commit adds the kernel facilities, the actual uprobes
user-space ABI and perf probe support comes in later commits.
General design:
Uprobes are maintained in an rb-tree indexed by inode and offset
(the offset here is from the start of the mapping). For a unique
(inode, offset) tuple, there can be at most one uprobe in the
rb-tree.
Since the (inode, offset) tuple identifies a unique uprobe, more
than one user may be interested in the same uprobe. This provides
the ability to connect multiple 'consumers' to the same uprobe.
Each consumer defines a handler and a filter (optional). The
'handler' is run every time the uprobe is hit, if it matches the
'filter' criteria.
The first consumer of a uprobe causes the breakpoint to be
inserted at the specified address and subsequent consumers are
appended to this list. On subsequent probes, the consumer gets
appended to the existing list of consumers. The breakpoint is
removed when the last consumer unregisters. For all other
unregisterations, the consumer is removed from the list of
consumers.
Given a inode, we get a list of the mms that have mapped the
inode. Do the actual registration if mm maps the page where a
probe needs to be inserted/removed.
We use a temporary list to walk through the vmas that map the
inode.
- The number of maps that map the inode, is not known before we
walk the rmap and keeps changing.
- extending vm_area_struct wasn't recommended, it's a
size-critical data structure.
- There can be more than one maps of the inode in the same mm.
We add callbacks to the mmap methods to keep an eye on text vmas
that are of interest to uprobes. When a vma of interest is mapped,
we insert the breakpoint at the right address.
Uprobe works by replacing the instruction at the address defined
by (inode, offset) with the arch specific breakpoint
instruction. We save a copy of the original instruction at the
uprobed address.
This is needed for:
a. executing the instruction out-of-line (xol).
b. instruction analysis for any subsequent fixups.
c. restoring the instruction back when the uprobe is unregistered.
We insert or delete a breakpoint instruction, and this
breakpoint instruction is assumed to be the smallest instruction
available on the platform. For fixed size instruction platforms
this is trivially true, for variable size instruction platforms
the breakpoint instruction is typically the smallest (often a
single byte).
Writing the instruction is done by COWing the page and changing
the instruction during the copy, this even though most platforms
allow atomic writes of the breakpoint instruction. This also
mirrors the behaviour of a ptrace() memory write to a PRIVATE
file map.
The core worker is derived from KSM's replace_page() logic.
In essence, similar to KSM:
a. allocate a new page and copy over contents of the page that
has the uprobed vaddr
b. modify the copy and insert the breakpoint at the required
address
c. switch the original page with the copy containing the
breakpoint
d. flush page tables.
replace_page() is being replicated here because of some minor
changes in the type of pages and also because Hugh Dickins had
plans to improve replace_page() for KSM specific work.
Instruction analysis on x86 is based on instruction decoder and
determines if an instruction can be probed and determines the
necessary fixups after singlestep. Instruction analysis is done
at probe insertion time so that we avoid having to repeat the
same analysis every time a probe is hit.
A lot of code here is due to the improvement/suggestions/inputs
from Peter Zijlstra.
Changelog:
(v10):
- Add code to clear REX.B prefix as suggested by Denys Vlasenko
and Masami Hiramatsu.
(v9):
- Use insn_offset_modrm as suggested by Masami Hiramatsu.
(v7):
Handle comments from Peter Zijlstra:
- Dont take reference to inode. (expect inode to uprobe_register to be sane).
- Use PTR_ERR to set the return value.
- No need to take reference to inode.
- use PTR_ERR to return error value.
- register and uprobe_unregister share code.
(v5):
- Modified del_consumer as per comments from Peter.
- Drop reference to inode before dropping reference to uprobe.
- Use i_size_read(inode) instead of inode->i_size.
- Ensure uprobe->consumers is NULL, before __uprobe_unregister() is called.
- Includes errno.h as recommended by Stephen Rothwell to fix a build issue
on sparc defconfig
- Remove restrictions while unregistering.
- Earlier code leaked inode references under some conditions while
registering/unregistering.
- Continue the vma-rmap walk even if the intermediate vma doesnt
meet the requirements.
- Validate the vma found by find_vma before inserting/removing the
breakpoint
- Call del_consumer under mutex_lock.
- Use hash locks.
- Handle mremap.
- Introduce find_least_offset_node() instead of close match logic in
find_uprobe
- Uprobes no more depends on MM_OWNER; No reference to task_structs
while inserting/removing a probe.
- Uses read_mapping_page instead of grab_cache_page so that the pages
have valid content.
- pass NULL to get_user_pages for the task parameter.
- call SetPageUptodate on the new page allocated in write_opcode.
- fix leaking a reference to the new page under certain conditions.
- Include Instruction Decoder if Uprobes gets defined.
- Remove const attributes for instruction prefix arrays.
- Uses mm_context to know if the application is 32 bit.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Also-written-by: Jim Keniston <jkenisto@us.ibm.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Roland McGrath <roland@hack.frob.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Anton Arapov <anton@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Denys Vlasenko <vda.linux@googlemail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linux-mm <linux-mm@kvack.org>
Link: http://lkml.kernel.org/r/20120209092642.GE16600@linux.vnet.ibm.com
[ Made various small edits to the commit log ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-02-09 17:26:42 +08:00
|
|
|
* @insn: instruction to be checked.
|
2012-03-12 17:25:45 +08:00
|
|
|
* Default implementation of is_swbp_insn
|
uprobes, mm, x86: Add the ability to install and remove uprobes breakpoints
Add uprobes support to the core kernel, with x86 support.
This commit adds the kernel facilities, the actual uprobes
user-space ABI and perf probe support comes in later commits.
General design:
Uprobes are maintained in an rb-tree indexed by inode and offset
(the offset here is from the start of the mapping). For a unique
(inode, offset) tuple, there can be at most one uprobe in the
rb-tree.
Since the (inode, offset) tuple identifies a unique uprobe, more
than one user may be interested in the same uprobe. This provides
the ability to connect multiple 'consumers' to the same uprobe.
Each consumer defines a handler and a filter (optional). The
'handler' is run every time the uprobe is hit, if it matches the
'filter' criteria.
The first consumer of a uprobe causes the breakpoint to be
inserted at the specified address and subsequent consumers are
appended to this list. On subsequent probes, the consumer gets
appended to the existing list of consumers. The breakpoint is
removed when the last consumer unregisters. For all other
unregisterations, the consumer is removed from the list of
consumers.
Given a inode, we get a list of the mms that have mapped the
inode. Do the actual registration if mm maps the page where a
probe needs to be inserted/removed.
We use a temporary list to walk through the vmas that map the
inode.
- The number of maps that map the inode, is not known before we
walk the rmap and keeps changing.
- extending vm_area_struct wasn't recommended, it's a
size-critical data structure.
- There can be more than one maps of the inode in the same mm.
We add callbacks to the mmap methods to keep an eye on text vmas
that are of interest to uprobes. When a vma of interest is mapped,
we insert the breakpoint at the right address.
Uprobe works by replacing the instruction at the address defined
by (inode, offset) with the arch specific breakpoint
instruction. We save a copy of the original instruction at the
uprobed address.
This is needed for:
a. executing the instruction out-of-line (xol).
b. instruction analysis for any subsequent fixups.
c. restoring the instruction back when the uprobe is unregistered.
We insert or delete a breakpoint instruction, and this
breakpoint instruction is assumed to be the smallest instruction
available on the platform. For fixed size instruction platforms
this is trivially true, for variable size instruction platforms
the breakpoint instruction is typically the smallest (often a
single byte).
Writing the instruction is done by COWing the page and changing
the instruction during the copy, this even though most platforms
allow atomic writes of the breakpoint instruction. This also
mirrors the behaviour of a ptrace() memory write to a PRIVATE
file map.
The core worker is derived from KSM's replace_page() logic.
In essence, similar to KSM:
a. allocate a new page and copy over contents of the page that
has the uprobed vaddr
b. modify the copy and insert the breakpoint at the required
address
c. switch the original page with the copy containing the
breakpoint
d. flush page tables.
replace_page() is being replicated here because of some minor
changes in the type of pages and also because Hugh Dickins had
plans to improve replace_page() for KSM specific work.
Instruction analysis on x86 is based on instruction decoder and
determines if an instruction can be probed and determines the
necessary fixups after singlestep. Instruction analysis is done
at probe insertion time so that we avoid having to repeat the
same analysis every time a probe is hit.
A lot of code here is due to the improvement/suggestions/inputs
from Peter Zijlstra.
Changelog:
(v10):
- Add code to clear REX.B prefix as suggested by Denys Vlasenko
and Masami Hiramatsu.
(v9):
- Use insn_offset_modrm as suggested by Masami Hiramatsu.
(v7):
Handle comments from Peter Zijlstra:
- Dont take reference to inode. (expect inode to uprobe_register to be sane).
- Use PTR_ERR to set the return value.
- No need to take reference to inode.
- use PTR_ERR to return error value.
- register and uprobe_unregister share code.
(v5):
- Modified del_consumer as per comments from Peter.
- Drop reference to inode before dropping reference to uprobe.
- Use i_size_read(inode) instead of inode->i_size.
- Ensure uprobe->consumers is NULL, before __uprobe_unregister() is called.
- Includes errno.h as recommended by Stephen Rothwell to fix a build issue
on sparc defconfig
- Remove restrictions while unregistering.
- Earlier code leaked inode references under some conditions while
registering/unregistering.
- Continue the vma-rmap walk even if the intermediate vma doesnt
meet the requirements.
- Validate the vma found by find_vma before inserting/removing the
breakpoint
- Call del_consumer under mutex_lock.
- Use hash locks.
- Handle mremap.
- Introduce find_least_offset_node() instead of close match logic in
find_uprobe
- Uprobes no more depends on MM_OWNER; No reference to task_structs
while inserting/removing a probe.
- Uses read_mapping_page instead of grab_cache_page so that the pages
have valid content.
- pass NULL to get_user_pages for the task parameter.
- call SetPageUptodate on the new page allocated in write_opcode.
- fix leaking a reference to the new page under certain conditions.
- Include Instruction Decoder if Uprobes gets defined.
- Remove const attributes for instruction prefix arrays.
- Uses mm_context to know if the application is 32 bit.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Also-written-by: Jim Keniston <jkenisto@us.ibm.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Roland McGrath <roland@hack.frob.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Anton Arapov <anton@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Denys Vlasenko <vda.linux@googlemail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linux-mm <linux-mm@kvack.org>
Link: http://lkml.kernel.org/r/20120209092642.GE16600@linux.vnet.ibm.com
[ Made various small edits to the commit log ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-02-09 17:26:42 +08:00
|
|
|
* Returns true if @insn is a breakpoint instruction.
|
|
|
|
*/
|
2012-03-12 17:25:45 +08:00
|
|
|
bool __weak is_swbp_insn(uprobe_opcode_t *insn)
|
uprobes, mm, x86: Add the ability to install and remove uprobes breakpoints
Add uprobes support to the core kernel, with x86 support.
This commit adds the kernel facilities, the actual uprobes
user-space ABI and perf probe support comes in later commits.
General design:
Uprobes are maintained in an rb-tree indexed by inode and offset
(the offset here is from the start of the mapping). For a unique
(inode, offset) tuple, there can be at most one uprobe in the
rb-tree.
Since the (inode, offset) tuple identifies a unique uprobe, more
than one user may be interested in the same uprobe. This provides
the ability to connect multiple 'consumers' to the same uprobe.
Each consumer defines a handler and a filter (optional). The
'handler' is run every time the uprobe is hit, if it matches the
'filter' criteria.
The first consumer of a uprobe causes the breakpoint to be
inserted at the specified address and subsequent consumers are
appended to this list. On subsequent probes, the consumer gets
appended to the existing list of consumers. The breakpoint is
removed when the last consumer unregisters. For all other
unregisterations, the consumer is removed from the list of
consumers.
Given a inode, we get a list of the mms that have mapped the
inode. Do the actual registration if mm maps the page where a
probe needs to be inserted/removed.
We use a temporary list to walk through the vmas that map the
inode.
- The number of maps that map the inode, is not known before we
walk the rmap and keeps changing.
- extending vm_area_struct wasn't recommended, it's a
size-critical data structure.
- There can be more than one maps of the inode in the same mm.
We add callbacks to the mmap methods to keep an eye on text vmas
that are of interest to uprobes. When a vma of interest is mapped,
we insert the breakpoint at the right address.
Uprobe works by replacing the instruction at the address defined
by (inode, offset) with the arch specific breakpoint
instruction. We save a copy of the original instruction at the
uprobed address.
This is needed for:
a. executing the instruction out-of-line (xol).
b. instruction analysis for any subsequent fixups.
c. restoring the instruction back when the uprobe is unregistered.
We insert or delete a breakpoint instruction, and this
breakpoint instruction is assumed to be the smallest instruction
available on the platform. For fixed size instruction platforms
this is trivially true, for variable size instruction platforms
the breakpoint instruction is typically the smallest (often a
single byte).
Writing the instruction is done by COWing the page and changing
the instruction during the copy, this even though most platforms
allow atomic writes of the breakpoint instruction. This also
mirrors the behaviour of a ptrace() memory write to a PRIVATE
file map.
The core worker is derived from KSM's replace_page() logic.
In essence, similar to KSM:
a. allocate a new page and copy over contents of the page that
has the uprobed vaddr
b. modify the copy and insert the breakpoint at the required
address
c. switch the original page with the copy containing the
breakpoint
d. flush page tables.
replace_page() is being replicated here because of some minor
changes in the type of pages and also because Hugh Dickins had
plans to improve replace_page() for KSM specific work.
Instruction analysis on x86 is based on instruction decoder and
determines if an instruction can be probed and determines the
necessary fixups after singlestep. Instruction analysis is done
at probe insertion time so that we avoid having to repeat the
same analysis every time a probe is hit.
A lot of code here is due to the improvement/suggestions/inputs
from Peter Zijlstra.
Changelog:
(v10):
- Add code to clear REX.B prefix as suggested by Denys Vlasenko
and Masami Hiramatsu.
(v9):
- Use insn_offset_modrm as suggested by Masami Hiramatsu.
(v7):
Handle comments from Peter Zijlstra:
- Dont take reference to inode. (expect inode to uprobe_register to be sane).
- Use PTR_ERR to set the return value.
- No need to take reference to inode.
- use PTR_ERR to return error value.
- register and uprobe_unregister share code.
(v5):
- Modified del_consumer as per comments from Peter.
- Drop reference to inode before dropping reference to uprobe.
- Use i_size_read(inode) instead of inode->i_size.
- Ensure uprobe->consumers is NULL, before __uprobe_unregister() is called.
- Includes errno.h as recommended by Stephen Rothwell to fix a build issue
on sparc defconfig
- Remove restrictions while unregistering.
- Earlier code leaked inode references under some conditions while
registering/unregistering.
- Continue the vma-rmap walk even if the intermediate vma doesnt
meet the requirements.
- Validate the vma found by find_vma before inserting/removing the
breakpoint
- Call del_consumer under mutex_lock.
- Use hash locks.
- Handle mremap.
- Introduce find_least_offset_node() instead of close match logic in
find_uprobe
- Uprobes no more depends on MM_OWNER; No reference to task_structs
while inserting/removing a probe.
- Uses read_mapping_page instead of grab_cache_page so that the pages
have valid content.
- pass NULL to get_user_pages for the task parameter.
- call SetPageUptodate on the new page allocated in write_opcode.
- fix leaking a reference to the new page under certain conditions.
- Include Instruction Decoder if Uprobes gets defined.
- Remove const attributes for instruction prefix arrays.
- Uses mm_context to know if the application is 32 bit.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Also-written-by: Jim Keniston <jkenisto@us.ibm.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Roland McGrath <roland@hack.frob.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Anton Arapov <anton@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Denys Vlasenko <vda.linux@googlemail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linux-mm <linux-mm@kvack.org>
Link: http://lkml.kernel.org/r/20120209092642.GE16600@linux.vnet.ibm.com
[ Made various small edits to the commit log ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-02-09 17:26:42 +08:00
|
|
|
{
|
2012-03-12 17:25:45 +08:00
|
|
|
return *insn == UPROBE_SWBP_INSN;
|
uprobes, mm, x86: Add the ability to install and remove uprobes breakpoints
Add uprobes support to the core kernel, with x86 support.
This commit adds the kernel facilities, the actual uprobes
user-space ABI and perf probe support comes in later commits.
General design:
Uprobes are maintained in an rb-tree indexed by inode and offset
(the offset here is from the start of the mapping). For a unique
(inode, offset) tuple, there can be at most one uprobe in the
rb-tree.
Since the (inode, offset) tuple identifies a unique uprobe, more
than one user may be interested in the same uprobe. This provides
the ability to connect multiple 'consumers' to the same uprobe.
Each consumer defines a handler and a filter (optional). The
'handler' is run every time the uprobe is hit, if it matches the
'filter' criteria.
The first consumer of a uprobe causes the breakpoint to be
inserted at the specified address and subsequent consumers are
appended to this list. On subsequent probes, the consumer gets
appended to the existing list of consumers. The breakpoint is
removed when the last consumer unregisters. For all other
unregisterations, the consumer is removed from the list of
consumers.
Given a inode, we get a list of the mms that have mapped the
inode. Do the actual registration if mm maps the page where a
probe needs to be inserted/removed.
We use a temporary list to walk through the vmas that map the
inode.
- The number of maps that map the inode, is not known before we
walk the rmap and keeps changing.
- extending vm_area_struct wasn't recommended, it's a
size-critical data structure.
- There can be more than one maps of the inode in the same mm.
We add callbacks to the mmap methods to keep an eye on text vmas
that are of interest to uprobes. When a vma of interest is mapped,
we insert the breakpoint at the right address.
Uprobe works by replacing the instruction at the address defined
by (inode, offset) with the arch specific breakpoint
instruction. We save a copy of the original instruction at the
uprobed address.
This is needed for:
a. executing the instruction out-of-line (xol).
b. instruction analysis for any subsequent fixups.
c. restoring the instruction back when the uprobe is unregistered.
We insert or delete a breakpoint instruction, and this
breakpoint instruction is assumed to be the smallest instruction
available on the platform. For fixed size instruction platforms
this is trivially true, for variable size instruction platforms
the breakpoint instruction is typically the smallest (often a
single byte).
Writing the instruction is done by COWing the page and changing
the instruction during the copy, this even though most platforms
allow atomic writes of the breakpoint instruction. This also
mirrors the behaviour of a ptrace() memory write to a PRIVATE
file map.
The core worker is derived from KSM's replace_page() logic.
In essence, similar to KSM:
a. allocate a new page and copy over contents of the page that
has the uprobed vaddr
b. modify the copy and insert the breakpoint at the required
address
c. switch the original page with the copy containing the
breakpoint
d. flush page tables.
replace_page() is being replicated here because of some minor
changes in the type of pages and also because Hugh Dickins had
plans to improve replace_page() for KSM specific work.
Instruction analysis on x86 is based on instruction decoder and
determines if an instruction can be probed and determines the
necessary fixups after singlestep. Instruction analysis is done
at probe insertion time so that we avoid having to repeat the
same analysis every time a probe is hit.
A lot of code here is due to the improvement/suggestions/inputs
from Peter Zijlstra.
Changelog:
(v10):
- Add code to clear REX.B prefix as suggested by Denys Vlasenko
and Masami Hiramatsu.
(v9):
- Use insn_offset_modrm as suggested by Masami Hiramatsu.
(v7):
Handle comments from Peter Zijlstra:
- Dont take reference to inode. (expect inode to uprobe_register to be sane).
- Use PTR_ERR to set the return value.
- No need to take reference to inode.
- use PTR_ERR to return error value.
- register and uprobe_unregister share code.
(v5):
- Modified del_consumer as per comments from Peter.
- Drop reference to inode before dropping reference to uprobe.
- Use i_size_read(inode) instead of inode->i_size.
- Ensure uprobe->consumers is NULL, before __uprobe_unregister() is called.
- Includes errno.h as recommended by Stephen Rothwell to fix a build issue
on sparc defconfig
- Remove restrictions while unregistering.
- Earlier code leaked inode references under some conditions while
registering/unregistering.
- Continue the vma-rmap walk even if the intermediate vma doesnt
meet the requirements.
- Validate the vma found by find_vma before inserting/removing the
breakpoint
- Call del_consumer under mutex_lock.
- Use hash locks.
- Handle mremap.
- Introduce find_least_offset_node() instead of close match logic in
find_uprobe
- Uprobes no more depends on MM_OWNER; No reference to task_structs
while inserting/removing a probe.
- Uses read_mapping_page instead of grab_cache_page so that the pages
have valid content.
- pass NULL to get_user_pages for the task parameter.
- call SetPageUptodate on the new page allocated in write_opcode.
- fix leaking a reference to the new page under certain conditions.
- Include Instruction Decoder if Uprobes gets defined.
- Remove const attributes for instruction prefix arrays.
- Uses mm_context to know if the application is 32 bit.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Also-written-by: Jim Keniston <jkenisto@us.ibm.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Roland McGrath <roland@hack.frob.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Anton Arapov <anton@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Denys Vlasenko <vda.linux@googlemail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linux-mm <linux-mm@kvack.org>
Link: http://lkml.kernel.org/r/20120209092642.GE16600@linux.vnet.ibm.com
[ Made various small edits to the commit log ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-02-09 17:26:42 +08:00
|
|
|
}
|
|
|
|
|
2013-03-22 23:16:27 +08:00
|
|
|
/**
|
|
|
|
* is_trap_insn - check if instruction is breakpoint instruction.
|
|
|
|
* @insn: instruction to be checked.
|
|
|
|
* Default implementation of is_trap_insn
|
|
|
|
* Returns true if @insn is a breakpoint instruction.
|
|
|
|
*
|
|
|
|
* This function is needed for the case where an architecture has multiple
|
|
|
|
* trap instructions (like powerpc).
|
|
|
|
*/
|
|
|
|
bool __weak is_trap_insn(uprobe_opcode_t *insn)
|
|
|
|
{
|
|
|
|
return is_swbp_insn(insn);
|
|
|
|
}
|
|
|
|
|
2013-03-25 01:24:37 +08:00
|
|
|
static void copy_from_page(struct page *page, unsigned long vaddr, void *dst, int len)
|
2012-09-24 03:10:18 +08:00
|
|
|
{
|
|
|
|
void *kaddr = kmap_atomic(page);
|
2013-03-25 01:24:37 +08:00
|
|
|
memcpy(dst, kaddr + (vaddr & ~PAGE_MASK), len);
|
2012-09-24 03:10:18 +08:00
|
|
|
kunmap_atomic(kaddr);
|
|
|
|
}
|
|
|
|
|
2013-03-25 01:58:04 +08:00
|
|
|
static void copy_to_page(struct page *page, unsigned long vaddr, const void *src, int len)
|
|
|
|
{
|
|
|
|
void *kaddr = kmap_atomic(page);
|
|
|
|
memcpy(kaddr + (vaddr & ~PAGE_MASK), src, len);
|
|
|
|
kunmap_atomic(kaddr);
|
|
|
|
}
|
|
|
|
|
2012-09-24 03:30:44 +08:00
|
|
|
static int verify_opcode(struct page *page, unsigned long vaddr, uprobe_opcode_t *new_opcode)
|
|
|
|
{
|
|
|
|
uprobe_opcode_t old_opcode;
|
|
|
|
bool is_swbp;
|
|
|
|
|
2013-03-22 23:16:27 +08:00
|
|
|
/*
|
|
|
|
* Note: We only check if the old_opcode is UPROBE_SWBP_INSN here.
|
|
|
|
* We do not check if it is any other 'trap variant' which could
|
|
|
|
* be conditional trap instruction such as the one powerpc supports.
|
|
|
|
*
|
|
|
|
* The logic is that we do not care if the underlying instruction
|
|
|
|
* is a trap variant; uprobes always wins over any other (gdb)
|
|
|
|
* breakpoint.
|
|
|
|
*/
|
2013-03-25 01:24:37 +08:00
|
|
|
copy_from_page(page, vaddr, &old_opcode, UPROBE_SWBP_INSN_SIZE);
|
2012-09-24 03:30:44 +08:00
|
|
|
is_swbp = is_swbp_insn(&old_opcode);
|
|
|
|
|
|
|
|
if (is_swbp_insn(new_opcode)) {
|
|
|
|
if (is_swbp) /* register: already installed? */
|
|
|
|
return 0;
|
|
|
|
} else {
|
|
|
|
if (!is_swbp) /* unregister: was it changed by us? */
|
2012-10-01 00:54:53 +08:00
|
|
|
return 0;
|
2012-09-24 03:30:44 +08:00
|
|
|
}
|
|
|
|
|
|
|
|
return 1;
|
|
|
|
}
|
|
|
|
|
uprobes: Support SDT markers having reference count (semaphore)
Userspace Statically Defined Tracepoints[1] are dtrace style markers
inside userspace applications. Applications like PostgreSQL, MySQL,
Pthread, Perl, Python, Java, Ruby, Node.js, libvirt, QEMU, glib etc
have these markers embedded in them. These markers are added by developer
at important places in the code. Each marker source expands to a single
nop instruction in the compiled code but there may be additional
overhead for computing the marker arguments which expands to couple of
instructions. In case the overhead is more, execution of it can be
omitted by runtime if() condition when no one is tracing on the marker:
if (reference_counter > 0) {
Execute marker instructions;
}
Default value of reference counter is 0. Tracer has to increment the
reference counter before tracing on a marker and decrement it when
done with the tracing.
Implement the reference counter logic in core uprobe. User will be
able to use it from trace_uprobe as well as from kernel module. New
trace_uprobe definition with reference counter will now be:
<path>:<offset>[(ref_ctr_offset)]
where ref_ctr_offset is an optional field. For kernel module, new
variant of uprobe_register() has been introduced:
uprobe_register_refctr(inode, offset, ref_ctr_offset, consumer)
No new variant for uprobe_unregister() because it's assumed to have
only one reference counter for one uprobe.
[1] https://sourceware.org/systemtap/wiki/UserSpaceProbeImplementation
Note: 'reference counter' is called as 'semaphore' in original Dtrace
(or Systemtap, bcc and even in ELF) documentation and code. But the
term 'semaphore' is misleading in this context. This is just a counter
used to hold number of tracers tracing on a marker. This is not really
used for any synchronization. So we are calling it a 'reference counter'
in kernel / perf code.
Link: http://lkml.kernel.org/r/20180820044250.11659-2-ravi.bangoria@linux.ibm.com
Reviewed-by: Masami Hiramatsu <mhiramat@kernel.org>
[Only trace_uprobe.c]
Reviewed-by: Oleg Nesterov <oleg@redhat.com>
Reviewed-by: Song Liu <songliubraving@fb.com>
Tested-by: Song Liu <songliubraving@fb.com>
Signed-off-by: Ravi Bangoria <ravi.bangoria@linux.ibm.com>
Signed-off-by: Steven Rostedt (VMware) <rostedt@goodmis.org>
2018-08-20 12:42:47 +08:00
|
|
|
static struct delayed_uprobe *
|
|
|
|
delayed_uprobe_check(struct uprobe *uprobe, struct mm_struct *mm)
|
|
|
|
{
|
|
|
|
struct delayed_uprobe *du;
|
|
|
|
|
|
|
|
list_for_each_entry(du, &delayed_uprobe_list, list)
|
|
|
|
if (du->uprobe == uprobe && du->mm == mm)
|
|
|
|
return du;
|
|
|
|
return NULL;
|
|
|
|
}
|
|
|
|
|
|
|
|
static int delayed_uprobe_add(struct uprobe *uprobe, struct mm_struct *mm)
|
|
|
|
{
|
|
|
|
struct delayed_uprobe *du;
|
|
|
|
|
|
|
|
if (delayed_uprobe_check(uprobe, mm))
|
|
|
|
return 0;
|
|
|
|
|
|
|
|
du = kzalloc(sizeof(*du), GFP_KERNEL);
|
|
|
|
if (!du)
|
|
|
|
return -ENOMEM;
|
|
|
|
|
|
|
|
du->uprobe = uprobe;
|
|
|
|
du->mm = mm;
|
|
|
|
list_add(&du->list, &delayed_uprobe_list);
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
static void delayed_uprobe_delete(struct delayed_uprobe *du)
|
|
|
|
{
|
|
|
|
if (WARN_ON(!du))
|
|
|
|
return;
|
|
|
|
list_del(&du->list);
|
|
|
|
kfree(du);
|
|
|
|
}
|
|
|
|
|
|
|
|
static void delayed_uprobe_remove(struct uprobe *uprobe, struct mm_struct *mm)
|
|
|
|
{
|
|
|
|
struct list_head *pos, *q;
|
|
|
|
struct delayed_uprobe *du;
|
|
|
|
|
|
|
|
if (!uprobe && !mm)
|
|
|
|
return;
|
|
|
|
|
|
|
|
list_for_each_safe(pos, q, &delayed_uprobe_list) {
|
|
|
|
du = list_entry(pos, struct delayed_uprobe, list);
|
|
|
|
|
|
|
|
if (uprobe && du->uprobe != uprobe)
|
|
|
|
continue;
|
|
|
|
if (mm && du->mm != mm)
|
|
|
|
continue;
|
|
|
|
|
|
|
|
delayed_uprobe_delete(du);
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
static bool valid_ref_ctr_vma(struct uprobe *uprobe,
|
|
|
|
struct vm_area_struct *vma)
|
|
|
|
{
|
|
|
|
unsigned long vaddr = offset_to_vaddr(vma, uprobe->ref_ctr_offset);
|
|
|
|
|
|
|
|
return uprobe->ref_ctr_offset &&
|
|
|
|
vma->vm_file &&
|
|
|
|
file_inode(vma->vm_file) == uprobe->inode &&
|
|
|
|
(vma->vm_flags & (VM_WRITE|VM_SHARED)) == VM_WRITE &&
|
|
|
|
vma->vm_start <= vaddr &&
|
|
|
|
vma->vm_end > vaddr;
|
|
|
|
}
|
|
|
|
|
|
|
|
static struct vm_area_struct *
|
|
|
|
find_ref_ctr_vma(struct uprobe *uprobe, struct mm_struct *mm)
|
|
|
|
{
|
|
|
|
struct vm_area_struct *tmp;
|
|
|
|
|
|
|
|
for (tmp = mm->mmap; tmp; tmp = tmp->vm_next)
|
|
|
|
if (valid_ref_ctr_vma(uprobe, tmp))
|
|
|
|
return tmp;
|
|
|
|
|
|
|
|
return NULL;
|
|
|
|
}
|
|
|
|
|
|
|
|
static int
|
|
|
|
__update_ref_ctr(struct mm_struct *mm, unsigned long vaddr, short d)
|
|
|
|
{
|
|
|
|
void *kaddr;
|
|
|
|
struct page *page;
|
|
|
|
struct vm_area_struct *vma;
|
|
|
|
int ret;
|
|
|
|
short *ptr;
|
|
|
|
|
|
|
|
if (!vaddr || !d)
|
|
|
|
return -EINVAL;
|
|
|
|
|
|
|
|
ret = get_user_pages_remote(NULL, mm, vaddr, 1,
|
|
|
|
FOLL_WRITE, &page, &vma, NULL);
|
|
|
|
if (unlikely(ret <= 0)) {
|
|
|
|
/*
|
|
|
|
* We are asking for 1 page. If get_user_pages_remote() fails,
|
|
|
|
* it may return 0, in that case we have to return error.
|
|
|
|
*/
|
|
|
|
return ret == 0 ? -EBUSY : ret;
|
|
|
|
}
|
|
|
|
|
|
|
|
kaddr = kmap_atomic(page);
|
|
|
|
ptr = kaddr + (vaddr & ~PAGE_MASK);
|
|
|
|
|
|
|
|
if (unlikely(*ptr + d < 0)) {
|
|
|
|
pr_warn("ref_ctr going negative. vaddr: 0x%lx, "
|
|
|
|
"curr val: %d, delta: %d\n", vaddr, *ptr, d);
|
|
|
|
ret = -EINVAL;
|
|
|
|
goto out;
|
|
|
|
}
|
|
|
|
|
|
|
|
*ptr += d;
|
|
|
|
ret = 0;
|
|
|
|
out:
|
|
|
|
kunmap_atomic(kaddr);
|
|
|
|
put_page(page);
|
|
|
|
return ret;
|
|
|
|
}
|
|
|
|
|
|
|
|
static void update_ref_ctr_warn(struct uprobe *uprobe,
|
|
|
|
struct mm_struct *mm, short d)
|
|
|
|
{
|
|
|
|
pr_warn("ref_ctr %s failed for inode: 0x%lx offset: "
|
|
|
|
"0x%llx ref_ctr_offset: 0x%llx of mm: 0x%pK\n",
|
|
|
|
d > 0 ? "increment" : "decrement", uprobe->inode->i_ino,
|
|
|
|
(unsigned long long) uprobe->offset,
|
|
|
|
(unsigned long long) uprobe->ref_ctr_offset, mm);
|
|
|
|
}
|
|
|
|
|
|
|
|
static int update_ref_ctr(struct uprobe *uprobe, struct mm_struct *mm,
|
|
|
|
short d)
|
|
|
|
{
|
|
|
|
struct vm_area_struct *rc_vma;
|
|
|
|
unsigned long rc_vaddr;
|
|
|
|
int ret = 0;
|
|
|
|
|
|
|
|
rc_vma = find_ref_ctr_vma(uprobe, mm);
|
|
|
|
|
|
|
|
if (rc_vma) {
|
|
|
|
rc_vaddr = offset_to_vaddr(rc_vma, uprobe->ref_ctr_offset);
|
|
|
|
ret = __update_ref_ctr(mm, rc_vaddr, d);
|
|
|
|
if (ret)
|
|
|
|
update_ref_ctr_warn(uprobe, mm, d);
|
|
|
|
|
|
|
|
if (d > 0)
|
|
|
|
return ret;
|
|
|
|
}
|
|
|
|
|
|
|
|
mutex_lock(&delayed_uprobe_lock);
|
|
|
|
if (d > 0)
|
|
|
|
ret = delayed_uprobe_add(uprobe, mm);
|
|
|
|
else
|
|
|
|
delayed_uprobe_remove(uprobe, mm);
|
|
|
|
mutex_unlock(&delayed_uprobe_lock);
|
|
|
|
|
|
|
|
return ret;
|
|
|
|
}
|
|
|
|
|
uprobes, mm, x86: Add the ability to install and remove uprobes breakpoints
Add uprobes support to the core kernel, with x86 support.
This commit adds the kernel facilities, the actual uprobes
user-space ABI and perf probe support comes in later commits.
General design:
Uprobes are maintained in an rb-tree indexed by inode and offset
(the offset here is from the start of the mapping). For a unique
(inode, offset) tuple, there can be at most one uprobe in the
rb-tree.
Since the (inode, offset) tuple identifies a unique uprobe, more
than one user may be interested in the same uprobe. This provides
the ability to connect multiple 'consumers' to the same uprobe.
Each consumer defines a handler and a filter (optional). The
'handler' is run every time the uprobe is hit, if it matches the
'filter' criteria.
The first consumer of a uprobe causes the breakpoint to be
inserted at the specified address and subsequent consumers are
appended to this list. On subsequent probes, the consumer gets
appended to the existing list of consumers. The breakpoint is
removed when the last consumer unregisters. For all other
unregisterations, the consumer is removed from the list of
consumers.
Given a inode, we get a list of the mms that have mapped the
inode. Do the actual registration if mm maps the page where a
probe needs to be inserted/removed.
We use a temporary list to walk through the vmas that map the
inode.
- The number of maps that map the inode, is not known before we
walk the rmap and keeps changing.
- extending vm_area_struct wasn't recommended, it's a
size-critical data structure.
- There can be more than one maps of the inode in the same mm.
We add callbacks to the mmap methods to keep an eye on text vmas
that are of interest to uprobes. When a vma of interest is mapped,
we insert the breakpoint at the right address.
Uprobe works by replacing the instruction at the address defined
by (inode, offset) with the arch specific breakpoint
instruction. We save a copy of the original instruction at the
uprobed address.
This is needed for:
a. executing the instruction out-of-line (xol).
b. instruction analysis for any subsequent fixups.
c. restoring the instruction back when the uprobe is unregistered.
We insert or delete a breakpoint instruction, and this
breakpoint instruction is assumed to be the smallest instruction
available on the platform. For fixed size instruction platforms
this is trivially true, for variable size instruction platforms
the breakpoint instruction is typically the smallest (often a
single byte).
Writing the instruction is done by COWing the page and changing
the instruction during the copy, this even though most platforms
allow atomic writes of the breakpoint instruction. This also
mirrors the behaviour of a ptrace() memory write to a PRIVATE
file map.
The core worker is derived from KSM's replace_page() logic.
In essence, similar to KSM:
a. allocate a new page and copy over contents of the page that
has the uprobed vaddr
b. modify the copy and insert the breakpoint at the required
address
c. switch the original page with the copy containing the
breakpoint
d. flush page tables.
replace_page() is being replicated here because of some minor
changes in the type of pages and also because Hugh Dickins had
plans to improve replace_page() for KSM specific work.
Instruction analysis on x86 is based on instruction decoder and
determines if an instruction can be probed and determines the
necessary fixups after singlestep. Instruction analysis is done
at probe insertion time so that we avoid having to repeat the
same analysis every time a probe is hit.
A lot of code here is due to the improvement/suggestions/inputs
from Peter Zijlstra.
Changelog:
(v10):
- Add code to clear REX.B prefix as suggested by Denys Vlasenko
and Masami Hiramatsu.
(v9):
- Use insn_offset_modrm as suggested by Masami Hiramatsu.
(v7):
Handle comments from Peter Zijlstra:
- Dont take reference to inode. (expect inode to uprobe_register to be sane).
- Use PTR_ERR to set the return value.
- No need to take reference to inode.
- use PTR_ERR to return error value.
- register and uprobe_unregister share code.
(v5):
- Modified del_consumer as per comments from Peter.
- Drop reference to inode before dropping reference to uprobe.
- Use i_size_read(inode) instead of inode->i_size.
- Ensure uprobe->consumers is NULL, before __uprobe_unregister() is called.
- Includes errno.h as recommended by Stephen Rothwell to fix a build issue
on sparc defconfig
- Remove restrictions while unregistering.
- Earlier code leaked inode references under some conditions while
registering/unregistering.
- Continue the vma-rmap walk even if the intermediate vma doesnt
meet the requirements.
- Validate the vma found by find_vma before inserting/removing the
breakpoint
- Call del_consumer under mutex_lock.
- Use hash locks.
- Handle mremap.
- Introduce find_least_offset_node() instead of close match logic in
find_uprobe
- Uprobes no more depends on MM_OWNER; No reference to task_structs
while inserting/removing a probe.
- Uses read_mapping_page instead of grab_cache_page so that the pages
have valid content.
- pass NULL to get_user_pages for the task parameter.
- call SetPageUptodate on the new page allocated in write_opcode.
- fix leaking a reference to the new page under certain conditions.
- Include Instruction Decoder if Uprobes gets defined.
- Remove const attributes for instruction prefix arrays.
- Uses mm_context to know if the application is 32 bit.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Also-written-by: Jim Keniston <jkenisto@us.ibm.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Roland McGrath <roland@hack.frob.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Anton Arapov <anton@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Denys Vlasenko <vda.linux@googlemail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linux-mm <linux-mm@kvack.org>
Link: http://lkml.kernel.org/r/20120209092642.GE16600@linux.vnet.ibm.com
[ Made various small edits to the commit log ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-02-09 17:26:42 +08:00
|
|
|
/*
|
|
|
|
* NOTE:
|
|
|
|
* Expect the breakpoint instruction to be the smallest size instruction for
|
|
|
|
* the architecture. If an arch has variable length instruction and the
|
|
|
|
* breakpoint instruction is not of the smallest length instruction
|
2013-03-22 23:16:27 +08:00
|
|
|
* supported by that architecture then we need to modify is_trap_at_addr and
|
2013-11-06 02:50:39 +08:00
|
|
|
* uprobe_write_opcode accordingly. This would never be a problem for archs
|
|
|
|
* that have fixed length instructions.
|
2014-05-05 22:38:18 +08:00
|
|
|
*
|
2013-11-06 02:50:39 +08:00
|
|
|
* uprobe_write_opcode - write the opcode at a given virtual address.
|
uprobes, mm, x86: Add the ability to install and remove uprobes breakpoints
Add uprobes support to the core kernel, with x86 support.
This commit adds the kernel facilities, the actual uprobes
user-space ABI and perf probe support comes in later commits.
General design:
Uprobes are maintained in an rb-tree indexed by inode and offset
(the offset here is from the start of the mapping). For a unique
(inode, offset) tuple, there can be at most one uprobe in the
rb-tree.
Since the (inode, offset) tuple identifies a unique uprobe, more
than one user may be interested in the same uprobe. This provides
the ability to connect multiple 'consumers' to the same uprobe.
Each consumer defines a handler and a filter (optional). The
'handler' is run every time the uprobe is hit, if it matches the
'filter' criteria.
The first consumer of a uprobe causes the breakpoint to be
inserted at the specified address and subsequent consumers are
appended to this list. On subsequent probes, the consumer gets
appended to the existing list of consumers. The breakpoint is
removed when the last consumer unregisters. For all other
unregisterations, the consumer is removed from the list of
consumers.
Given a inode, we get a list of the mms that have mapped the
inode. Do the actual registration if mm maps the page where a
probe needs to be inserted/removed.
We use a temporary list to walk through the vmas that map the
inode.
- The number of maps that map the inode, is not known before we
walk the rmap and keeps changing.
- extending vm_area_struct wasn't recommended, it's a
size-critical data structure.
- There can be more than one maps of the inode in the same mm.
We add callbacks to the mmap methods to keep an eye on text vmas
that are of interest to uprobes. When a vma of interest is mapped,
we insert the breakpoint at the right address.
Uprobe works by replacing the instruction at the address defined
by (inode, offset) with the arch specific breakpoint
instruction. We save a copy of the original instruction at the
uprobed address.
This is needed for:
a. executing the instruction out-of-line (xol).
b. instruction analysis for any subsequent fixups.
c. restoring the instruction back when the uprobe is unregistered.
We insert or delete a breakpoint instruction, and this
breakpoint instruction is assumed to be the smallest instruction
available on the platform. For fixed size instruction platforms
this is trivially true, for variable size instruction platforms
the breakpoint instruction is typically the smallest (often a
single byte).
Writing the instruction is done by COWing the page and changing
the instruction during the copy, this even though most platforms
allow atomic writes of the breakpoint instruction. This also
mirrors the behaviour of a ptrace() memory write to a PRIVATE
file map.
The core worker is derived from KSM's replace_page() logic.
In essence, similar to KSM:
a. allocate a new page and copy over contents of the page that
has the uprobed vaddr
b. modify the copy and insert the breakpoint at the required
address
c. switch the original page with the copy containing the
breakpoint
d. flush page tables.
replace_page() is being replicated here because of some minor
changes in the type of pages and also because Hugh Dickins had
plans to improve replace_page() for KSM specific work.
Instruction analysis on x86 is based on instruction decoder and
determines if an instruction can be probed and determines the
necessary fixups after singlestep. Instruction analysis is done
at probe insertion time so that we avoid having to repeat the
same analysis every time a probe is hit.
A lot of code here is due to the improvement/suggestions/inputs
from Peter Zijlstra.
Changelog:
(v10):
- Add code to clear REX.B prefix as suggested by Denys Vlasenko
and Masami Hiramatsu.
(v9):
- Use insn_offset_modrm as suggested by Masami Hiramatsu.
(v7):
Handle comments from Peter Zijlstra:
- Dont take reference to inode. (expect inode to uprobe_register to be sane).
- Use PTR_ERR to set the return value.
- No need to take reference to inode.
- use PTR_ERR to return error value.
- register and uprobe_unregister share code.
(v5):
- Modified del_consumer as per comments from Peter.
- Drop reference to inode before dropping reference to uprobe.
- Use i_size_read(inode) instead of inode->i_size.
- Ensure uprobe->consumers is NULL, before __uprobe_unregister() is called.
- Includes errno.h as recommended by Stephen Rothwell to fix a build issue
on sparc defconfig
- Remove restrictions while unregistering.
- Earlier code leaked inode references under some conditions while
registering/unregistering.
- Continue the vma-rmap walk even if the intermediate vma doesnt
meet the requirements.
- Validate the vma found by find_vma before inserting/removing the
breakpoint
- Call del_consumer under mutex_lock.
- Use hash locks.
- Handle mremap.
- Introduce find_least_offset_node() instead of close match logic in
find_uprobe
- Uprobes no more depends on MM_OWNER; No reference to task_structs
while inserting/removing a probe.
- Uses read_mapping_page instead of grab_cache_page so that the pages
have valid content.
- pass NULL to get_user_pages for the task parameter.
- call SetPageUptodate on the new page allocated in write_opcode.
- fix leaking a reference to the new page under certain conditions.
- Include Instruction Decoder if Uprobes gets defined.
- Remove const attributes for instruction prefix arrays.
- Uses mm_context to know if the application is 32 bit.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Also-written-by: Jim Keniston <jkenisto@us.ibm.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Roland McGrath <roland@hack.frob.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Anton Arapov <anton@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Denys Vlasenko <vda.linux@googlemail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linux-mm <linux-mm@kvack.org>
Link: http://lkml.kernel.org/r/20120209092642.GE16600@linux.vnet.ibm.com
[ Made various small edits to the commit log ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-02-09 17:26:42 +08:00
|
|
|
* @mm: the probed process address space.
|
|
|
|
* @vaddr: the virtual address to store the opcode.
|
|
|
|
* @opcode: opcode to be written at @vaddr.
|
|
|
|
*
|
2014-05-05 22:38:18 +08:00
|
|
|
* Called with mm->mmap_sem held for write.
|
uprobes, mm, x86: Add the ability to install and remove uprobes breakpoints
Add uprobes support to the core kernel, with x86 support.
This commit adds the kernel facilities, the actual uprobes
user-space ABI and perf probe support comes in later commits.
General design:
Uprobes are maintained in an rb-tree indexed by inode and offset
(the offset here is from the start of the mapping). For a unique
(inode, offset) tuple, there can be at most one uprobe in the
rb-tree.
Since the (inode, offset) tuple identifies a unique uprobe, more
than one user may be interested in the same uprobe. This provides
the ability to connect multiple 'consumers' to the same uprobe.
Each consumer defines a handler and a filter (optional). The
'handler' is run every time the uprobe is hit, if it matches the
'filter' criteria.
The first consumer of a uprobe causes the breakpoint to be
inserted at the specified address and subsequent consumers are
appended to this list. On subsequent probes, the consumer gets
appended to the existing list of consumers. The breakpoint is
removed when the last consumer unregisters. For all other
unregisterations, the consumer is removed from the list of
consumers.
Given a inode, we get a list of the mms that have mapped the
inode. Do the actual registration if mm maps the page where a
probe needs to be inserted/removed.
We use a temporary list to walk through the vmas that map the
inode.
- The number of maps that map the inode, is not known before we
walk the rmap and keeps changing.
- extending vm_area_struct wasn't recommended, it's a
size-critical data structure.
- There can be more than one maps of the inode in the same mm.
We add callbacks to the mmap methods to keep an eye on text vmas
that are of interest to uprobes. When a vma of interest is mapped,
we insert the breakpoint at the right address.
Uprobe works by replacing the instruction at the address defined
by (inode, offset) with the arch specific breakpoint
instruction. We save a copy of the original instruction at the
uprobed address.
This is needed for:
a. executing the instruction out-of-line (xol).
b. instruction analysis for any subsequent fixups.
c. restoring the instruction back when the uprobe is unregistered.
We insert or delete a breakpoint instruction, and this
breakpoint instruction is assumed to be the smallest instruction
available on the platform. For fixed size instruction platforms
this is trivially true, for variable size instruction platforms
the breakpoint instruction is typically the smallest (often a
single byte).
Writing the instruction is done by COWing the page and changing
the instruction during the copy, this even though most platforms
allow atomic writes of the breakpoint instruction. This also
mirrors the behaviour of a ptrace() memory write to a PRIVATE
file map.
The core worker is derived from KSM's replace_page() logic.
In essence, similar to KSM:
a. allocate a new page and copy over contents of the page that
has the uprobed vaddr
b. modify the copy and insert the breakpoint at the required
address
c. switch the original page with the copy containing the
breakpoint
d. flush page tables.
replace_page() is being replicated here because of some minor
changes in the type of pages and also because Hugh Dickins had
plans to improve replace_page() for KSM specific work.
Instruction analysis on x86 is based on instruction decoder and
determines if an instruction can be probed and determines the
necessary fixups after singlestep. Instruction analysis is done
at probe insertion time so that we avoid having to repeat the
same analysis every time a probe is hit.
A lot of code here is due to the improvement/suggestions/inputs
from Peter Zijlstra.
Changelog:
(v10):
- Add code to clear REX.B prefix as suggested by Denys Vlasenko
and Masami Hiramatsu.
(v9):
- Use insn_offset_modrm as suggested by Masami Hiramatsu.
(v7):
Handle comments from Peter Zijlstra:
- Dont take reference to inode. (expect inode to uprobe_register to be sane).
- Use PTR_ERR to set the return value.
- No need to take reference to inode.
- use PTR_ERR to return error value.
- register and uprobe_unregister share code.
(v5):
- Modified del_consumer as per comments from Peter.
- Drop reference to inode before dropping reference to uprobe.
- Use i_size_read(inode) instead of inode->i_size.
- Ensure uprobe->consumers is NULL, before __uprobe_unregister() is called.
- Includes errno.h as recommended by Stephen Rothwell to fix a build issue
on sparc defconfig
- Remove restrictions while unregistering.
- Earlier code leaked inode references under some conditions while
registering/unregistering.
- Continue the vma-rmap walk even if the intermediate vma doesnt
meet the requirements.
- Validate the vma found by find_vma before inserting/removing the
breakpoint
- Call del_consumer under mutex_lock.
- Use hash locks.
- Handle mremap.
- Introduce find_least_offset_node() instead of close match logic in
find_uprobe
- Uprobes no more depends on MM_OWNER; No reference to task_structs
while inserting/removing a probe.
- Uses read_mapping_page instead of grab_cache_page so that the pages
have valid content.
- pass NULL to get_user_pages for the task parameter.
- call SetPageUptodate on the new page allocated in write_opcode.
- fix leaking a reference to the new page under certain conditions.
- Include Instruction Decoder if Uprobes gets defined.
- Remove const attributes for instruction prefix arrays.
- Uses mm_context to know if the application is 32 bit.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Also-written-by: Jim Keniston <jkenisto@us.ibm.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Roland McGrath <roland@hack.frob.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Anton Arapov <anton@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Denys Vlasenko <vda.linux@googlemail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linux-mm <linux-mm@kvack.org>
Link: http://lkml.kernel.org/r/20120209092642.GE16600@linux.vnet.ibm.com
[ Made various small edits to the commit log ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-02-09 17:26:42 +08:00
|
|
|
* Return 0 (success) or a negative errno.
|
|
|
|
*/
|
2018-08-09 12:18:52 +08:00
|
|
|
int uprobe_write_opcode(struct arch_uprobe *auprobe, struct mm_struct *mm,
|
|
|
|
unsigned long vaddr, uprobe_opcode_t opcode)
|
uprobes, mm, x86: Add the ability to install and remove uprobes breakpoints
Add uprobes support to the core kernel, with x86 support.
This commit adds the kernel facilities, the actual uprobes
user-space ABI and perf probe support comes in later commits.
General design:
Uprobes are maintained in an rb-tree indexed by inode and offset
(the offset here is from the start of the mapping). For a unique
(inode, offset) tuple, there can be at most one uprobe in the
rb-tree.
Since the (inode, offset) tuple identifies a unique uprobe, more
than one user may be interested in the same uprobe. This provides
the ability to connect multiple 'consumers' to the same uprobe.
Each consumer defines a handler and a filter (optional). The
'handler' is run every time the uprobe is hit, if it matches the
'filter' criteria.
The first consumer of a uprobe causes the breakpoint to be
inserted at the specified address and subsequent consumers are
appended to this list. On subsequent probes, the consumer gets
appended to the existing list of consumers. The breakpoint is
removed when the last consumer unregisters. For all other
unregisterations, the consumer is removed from the list of
consumers.
Given a inode, we get a list of the mms that have mapped the
inode. Do the actual registration if mm maps the page where a
probe needs to be inserted/removed.
We use a temporary list to walk through the vmas that map the
inode.
- The number of maps that map the inode, is not known before we
walk the rmap and keeps changing.
- extending vm_area_struct wasn't recommended, it's a
size-critical data structure.
- There can be more than one maps of the inode in the same mm.
We add callbacks to the mmap methods to keep an eye on text vmas
that are of interest to uprobes. When a vma of interest is mapped,
we insert the breakpoint at the right address.
Uprobe works by replacing the instruction at the address defined
by (inode, offset) with the arch specific breakpoint
instruction. We save a copy of the original instruction at the
uprobed address.
This is needed for:
a. executing the instruction out-of-line (xol).
b. instruction analysis for any subsequent fixups.
c. restoring the instruction back when the uprobe is unregistered.
We insert or delete a breakpoint instruction, and this
breakpoint instruction is assumed to be the smallest instruction
available on the platform. For fixed size instruction platforms
this is trivially true, for variable size instruction platforms
the breakpoint instruction is typically the smallest (often a
single byte).
Writing the instruction is done by COWing the page and changing
the instruction during the copy, this even though most platforms
allow atomic writes of the breakpoint instruction. This also
mirrors the behaviour of a ptrace() memory write to a PRIVATE
file map.
The core worker is derived from KSM's replace_page() logic.
In essence, similar to KSM:
a. allocate a new page and copy over contents of the page that
has the uprobed vaddr
b. modify the copy and insert the breakpoint at the required
address
c. switch the original page with the copy containing the
breakpoint
d. flush page tables.
replace_page() is being replicated here because of some minor
changes in the type of pages and also because Hugh Dickins had
plans to improve replace_page() for KSM specific work.
Instruction analysis on x86 is based on instruction decoder and
determines if an instruction can be probed and determines the
necessary fixups after singlestep. Instruction analysis is done
at probe insertion time so that we avoid having to repeat the
same analysis every time a probe is hit.
A lot of code here is due to the improvement/suggestions/inputs
from Peter Zijlstra.
Changelog:
(v10):
- Add code to clear REX.B prefix as suggested by Denys Vlasenko
and Masami Hiramatsu.
(v9):
- Use insn_offset_modrm as suggested by Masami Hiramatsu.
(v7):
Handle comments from Peter Zijlstra:
- Dont take reference to inode. (expect inode to uprobe_register to be sane).
- Use PTR_ERR to set the return value.
- No need to take reference to inode.
- use PTR_ERR to return error value.
- register and uprobe_unregister share code.
(v5):
- Modified del_consumer as per comments from Peter.
- Drop reference to inode before dropping reference to uprobe.
- Use i_size_read(inode) instead of inode->i_size.
- Ensure uprobe->consumers is NULL, before __uprobe_unregister() is called.
- Includes errno.h as recommended by Stephen Rothwell to fix a build issue
on sparc defconfig
- Remove restrictions while unregistering.
- Earlier code leaked inode references under some conditions while
registering/unregistering.
- Continue the vma-rmap walk even if the intermediate vma doesnt
meet the requirements.
- Validate the vma found by find_vma before inserting/removing the
breakpoint
- Call del_consumer under mutex_lock.
- Use hash locks.
- Handle mremap.
- Introduce find_least_offset_node() instead of close match logic in
find_uprobe
- Uprobes no more depends on MM_OWNER; No reference to task_structs
while inserting/removing a probe.
- Uses read_mapping_page instead of grab_cache_page so that the pages
have valid content.
- pass NULL to get_user_pages for the task parameter.
- call SetPageUptodate on the new page allocated in write_opcode.
- fix leaking a reference to the new page under certain conditions.
- Include Instruction Decoder if Uprobes gets defined.
- Remove const attributes for instruction prefix arrays.
- Uses mm_context to know if the application is 32 bit.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Also-written-by: Jim Keniston <jkenisto@us.ibm.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Roland McGrath <roland@hack.frob.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Anton Arapov <anton@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Denys Vlasenko <vda.linux@googlemail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linux-mm <linux-mm@kvack.org>
Link: http://lkml.kernel.org/r/20120209092642.GE16600@linux.vnet.ibm.com
[ Made various small edits to the commit log ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-02-09 17:26:42 +08:00
|
|
|
{
|
uprobes: Support SDT markers having reference count (semaphore)
Userspace Statically Defined Tracepoints[1] are dtrace style markers
inside userspace applications. Applications like PostgreSQL, MySQL,
Pthread, Perl, Python, Java, Ruby, Node.js, libvirt, QEMU, glib etc
have these markers embedded in them. These markers are added by developer
at important places in the code. Each marker source expands to a single
nop instruction in the compiled code but there may be additional
overhead for computing the marker arguments which expands to couple of
instructions. In case the overhead is more, execution of it can be
omitted by runtime if() condition when no one is tracing on the marker:
if (reference_counter > 0) {
Execute marker instructions;
}
Default value of reference counter is 0. Tracer has to increment the
reference counter before tracing on a marker and decrement it when
done with the tracing.
Implement the reference counter logic in core uprobe. User will be
able to use it from trace_uprobe as well as from kernel module. New
trace_uprobe definition with reference counter will now be:
<path>:<offset>[(ref_ctr_offset)]
where ref_ctr_offset is an optional field. For kernel module, new
variant of uprobe_register() has been introduced:
uprobe_register_refctr(inode, offset, ref_ctr_offset, consumer)
No new variant for uprobe_unregister() because it's assumed to have
only one reference counter for one uprobe.
[1] https://sourceware.org/systemtap/wiki/UserSpaceProbeImplementation
Note: 'reference counter' is called as 'semaphore' in original Dtrace
(or Systemtap, bcc and even in ELF) documentation and code. But the
term 'semaphore' is misleading in this context. This is just a counter
used to hold number of tracers tracing on a marker. This is not really
used for any synchronization. So we are calling it a 'reference counter'
in kernel / perf code.
Link: http://lkml.kernel.org/r/20180820044250.11659-2-ravi.bangoria@linux.ibm.com
Reviewed-by: Masami Hiramatsu <mhiramat@kernel.org>
[Only trace_uprobe.c]
Reviewed-by: Oleg Nesterov <oleg@redhat.com>
Reviewed-by: Song Liu <songliubraving@fb.com>
Tested-by: Song Liu <songliubraving@fb.com>
Signed-off-by: Ravi Bangoria <ravi.bangoria@linux.ibm.com>
Signed-off-by: Steven Rostedt (VMware) <rostedt@goodmis.org>
2018-08-20 12:42:47 +08:00
|
|
|
struct uprobe *uprobe;
|
uprobes, mm, x86: Add the ability to install and remove uprobes breakpoints
Add uprobes support to the core kernel, with x86 support.
This commit adds the kernel facilities, the actual uprobes
user-space ABI and perf probe support comes in later commits.
General design:
Uprobes are maintained in an rb-tree indexed by inode and offset
(the offset here is from the start of the mapping). For a unique
(inode, offset) tuple, there can be at most one uprobe in the
rb-tree.
Since the (inode, offset) tuple identifies a unique uprobe, more
than one user may be interested in the same uprobe. This provides
the ability to connect multiple 'consumers' to the same uprobe.
Each consumer defines a handler and a filter (optional). The
'handler' is run every time the uprobe is hit, if it matches the
'filter' criteria.
The first consumer of a uprobe causes the breakpoint to be
inserted at the specified address and subsequent consumers are
appended to this list. On subsequent probes, the consumer gets
appended to the existing list of consumers. The breakpoint is
removed when the last consumer unregisters. For all other
unregisterations, the consumer is removed from the list of
consumers.
Given a inode, we get a list of the mms that have mapped the
inode. Do the actual registration if mm maps the page where a
probe needs to be inserted/removed.
We use a temporary list to walk through the vmas that map the
inode.
- The number of maps that map the inode, is not known before we
walk the rmap and keeps changing.
- extending vm_area_struct wasn't recommended, it's a
size-critical data structure.
- There can be more than one maps of the inode in the same mm.
We add callbacks to the mmap methods to keep an eye on text vmas
that are of interest to uprobes. When a vma of interest is mapped,
we insert the breakpoint at the right address.
Uprobe works by replacing the instruction at the address defined
by (inode, offset) with the arch specific breakpoint
instruction. We save a copy of the original instruction at the
uprobed address.
This is needed for:
a. executing the instruction out-of-line (xol).
b. instruction analysis for any subsequent fixups.
c. restoring the instruction back when the uprobe is unregistered.
We insert or delete a breakpoint instruction, and this
breakpoint instruction is assumed to be the smallest instruction
available on the platform. For fixed size instruction platforms
this is trivially true, for variable size instruction platforms
the breakpoint instruction is typically the smallest (often a
single byte).
Writing the instruction is done by COWing the page and changing
the instruction during the copy, this even though most platforms
allow atomic writes of the breakpoint instruction. This also
mirrors the behaviour of a ptrace() memory write to a PRIVATE
file map.
The core worker is derived from KSM's replace_page() logic.
In essence, similar to KSM:
a. allocate a new page and copy over contents of the page that
has the uprobed vaddr
b. modify the copy and insert the breakpoint at the required
address
c. switch the original page with the copy containing the
breakpoint
d. flush page tables.
replace_page() is being replicated here because of some minor
changes in the type of pages and also because Hugh Dickins had
plans to improve replace_page() for KSM specific work.
Instruction analysis on x86 is based on instruction decoder and
determines if an instruction can be probed and determines the
necessary fixups after singlestep. Instruction analysis is done
at probe insertion time so that we avoid having to repeat the
same analysis every time a probe is hit.
A lot of code here is due to the improvement/suggestions/inputs
from Peter Zijlstra.
Changelog:
(v10):
- Add code to clear REX.B prefix as suggested by Denys Vlasenko
and Masami Hiramatsu.
(v9):
- Use insn_offset_modrm as suggested by Masami Hiramatsu.
(v7):
Handle comments from Peter Zijlstra:
- Dont take reference to inode. (expect inode to uprobe_register to be sane).
- Use PTR_ERR to set the return value.
- No need to take reference to inode.
- use PTR_ERR to return error value.
- register and uprobe_unregister share code.
(v5):
- Modified del_consumer as per comments from Peter.
- Drop reference to inode before dropping reference to uprobe.
- Use i_size_read(inode) instead of inode->i_size.
- Ensure uprobe->consumers is NULL, before __uprobe_unregister() is called.
- Includes errno.h as recommended by Stephen Rothwell to fix a build issue
on sparc defconfig
- Remove restrictions while unregistering.
- Earlier code leaked inode references under some conditions while
registering/unregistering.
- Continue the vma-rmap walk even if the intermediate vma doesnt
meet the requirements.
- Validate the vma found by find_vma before inserting/removing the
breakpoint
- Call del_consumer under mutex_lock.
- Use hash locks.
- Handle mremap.
- Introduce find_least_offset_node() instead of close match logic in
find_uprobe
- Uprobes no more depends on MM_OWNER; No reference to task_structs
while inserting/removing a probe.
- Uses read_mapping_page instead of grab_cache_page so that the pages
have valid content.
- pass NULL to get_user_pages for the task parameter.
- call SetPageUptodate on the new page allocated in write_opcode.
- fix leaking a reference to the new page under certain conditions.
- Include Instruction Decoder if Uprobes gets defined.
- Remove const attributes for instruction prefix arrays.
- Uses mm_context to know if the application is 32 bit.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Also-written-by: Jim Keniston <jkenisto@us.ibm.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Roland McGrath <roland@hack.frob.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Anton Arapov <anton@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Denys Vlasenko <vda.linux@googlemail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linux-mm <linux-mm@kvack.org>
Link: http://lkml.kernel.org/r/20120209092642.GE16600@linux.vnet.ibm.com
[ Made various small edits to the commit log ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-02-09 17:26:42 +08:00
|
|
|
struct page *old_page, *new_page;
|
|
|
|
struct vm_area_struct *vma;
|
uprobes: Support SDT markers having reference count (semaphore)
Userspace Statically Defined Tracepoints[1] are dtrace style markers
inside userspace applications. Applications like PostgreSQL, MySQL,
Pthread, Perl, Python, Java, Ruby, Node.js, libvirt, QEMU, glib etc
have these markers embedded in them. These markers are added by developer
at important places in the code. Each marker source expands to a single
nop instruction in the compiled code but there may be additional
overhead for computing the marker arguments which expands to couple of
instructions. In case the overhead is more, execution of it can be
omitted by runtime if() condition when no one is tracing on the marker:
if (reference_counter > 0) {
Execute marker instructions;
}
Default value of reference counter is 0. Tracer has to increment the
reference counter before tracing on a marker and decrement it when
done with the tracing.
Implement the reference counter logic in core uprobe. User will be
able to use it from trace_uprobe as well as from kernel module. New
trace_uprobe definition with reference counter will now be:
<path>:<offset>[(ref_ctr_offset)]
where ref_ctr_offset is an optional field. For kernel module, new
variant of uprobe_register() has been introduced:
uprobe_register_refctr(inode, offset, ref_ctr_offset, consumer)
No new variant for uprobe_unregister() because it's assumed to have
only one reference counter for one uprobe.
[1] https://sourceware.org/systemtap/wiki/UserSpaceProbeImplementation
Note: 'reference counter' is called as 'semaphore' in original Dtrace
(or Systemtap, bcc and even in ELF) documentation and code. But the
term 'semaphore' is misleading in this context. This is just a counter
used to hold number of tracers tracing on a marker. This is not really
used for any synchronization. So we are calling it a 'reference counter'
in kernel / perf code.
Link: http://lkml.kernel.org/r/20180820044250.11659-2-ravi.bangoria@linux.ibm.com
Reviewed-by: Masami Hiramatsu <mhiramat@kernel.org>
[Only trace_uprobe.c]
Reviewed-by: Oleg Nesterov <oleg@redhat.com>
Reviewed-by: Song Liu <songliubraving@fb.com>
Tested-by: Song Liu <songliubraving@fb.com>
Signed-off-by: Ravi Bangoria <ravi.bangoria@linux.ibm.com>
Signed-off-by: Steven Rostedt (VMware) <rostedt@goodmis.org>
2018-08-20 12:42:47 +08:00
|
|
|
int ret, is_register, ref_ctr_updated = 0;
|
|
|
|
|
|
|
|
is_register = is_swbp_insn(&opcode);
|
|
|
|
uprobe = container_of(auprobe, struct uprobe, arch);
|
2012-07-30 02:22:12 +08:00
|
|
|
|
2012-06-15 23:43:28 +08:00
|
|
|
retry:
|
uprobes, mm, x86: Add the ability to install and remove uprobes breakpoints
Add uprobes support to the core kernel, with x86 support.
This commit adds the kernel facilities, the actual uprobes
user-space ABI and perf probe support comes in later commits.
General design:
Uprobes are maintained in an rb-tree indexed by inode and offset
(the offset here is from the start of the mapping). For a unique
(inode, offset) tuple, there can be at most one uprobe in the
rb-tree.
Since the (inode, offset) tuple identifies a unique uprobe, more
than one user may be interested in the same uprobe. This provides
the ability to connect multiple 'consumers' to the same uprobe.
Each consumer defines a handler and a filter (optional). The
'handler' is run every time the uprobe is hit, if it matches the
'filter' criteria.
The first consumer of a uprobe causes the breakpoint to be
inserted at the specified address and subsequent consumers are
appended to this list. On subsequent probes, the consumer gets
appended to the existing list of consumers. The breakpoint is
removed when the last consumer unregisters. For all other
unregisterations, the consumer is removed from the list of
consumers.
Given a inode, we get a list of the mms that have mapped the
inode. Do the actual registration if mm maps the page where a
probe needs to be inserted/removed.
We use a temporary list to walk through the vmas that map the
inode.
- The number of maps that map the inode, is not known before we
walk the rmap and keeps changing.
- extending vm_area_struct wasn't recommended, it's a
size-critical data structure.
- There can be more than one maps of the inode in the same mm.
We add callbacks to the mmap methods to keep an eye on text vmas
that are of interest to uprobes. When a vma of interest is mapped,
we insert the breakpoint at the right address.
Uprobe works by replacing the instruction at the address defined
by (inode, offset) with the arch specific breakpoint
instruction. We save a copy of the original instruction at the
uprobed address.
This is needed for:
a. executing the instruction out-of-line (xol).
b. instruction analysis for any subsequent fixups.
c. restoring the instruction back when the uprobe is unregistered.
We insert or delete a breakpoint instruction, and this
breakpoint instruction is assumed to be the smallest instruction
available on the platform. For fixed size instruction platforms
this is trivially true, for variable size instruction platforms
the breakpoint instruction is typically the smallest (often a
single byte).
Writing the instruction is done by COWing the page and changing
the instruction during the copy, this even though most platforms
allow atomic writes of the breakpoint instruction. This also
mirrors the behaviour of a ptrace() memory write to a PRIVATE
file map.
The core worker is derived from KSM's replace_page() logic.
In essence, similar to KSM:
a. allocate a new page and copy over contents of the page that
has the uprobed vaddr
b. modify the copy and insert the breakpoint at the required
address
c. switch the original page with the copy containing the
breakpoint
d. flush page tables.
replace_page() is being replicated here because of some minor
changes in the type of pages and also because Hugh Dickins had
plans to improve replace_page() for KSM specific work.
Instruction analysis on x86 is based on instruction decoder and
determines if an instruction can be probed and determines the
necessary fixups after singlestep. Instruction analysis is done
at probe insertion time so that we avoid having to repeat the
same analysis every time a probe is hit.
A lot of code here is due to the improvement/suggestions/inputs
from Peter Zijlstra.
Changelog:
(v10):
- Add code to clear REX.B prefix as suggested by Denys Vlasenko
and Masami Hiramatsu.
(v9):
- Use insn_offset_modrm as suggested by Masami Hiramatsu.
(v7):
Handle comments from Peter Zijlstra:
- Dont take reference to inode. (expect inode to uprobe_register to be sane).
- Use PTR_ERR to set the return value.
- No need to take reference to inode.
- use PTR_ERR to return error value.
- register and uprobe_unregister share code.
(v5):
- Modified del_consumer as per comments from Peter.
- Drop reference to inode before dropping reference to uprobe.
- Use i_size_read(inode) instead of inode->i_size.
- Ensure uprobe->consumers is NULL, before __uprobe_unregister() is called.
- Includes errno.h as recommended by Stephen Rothwell to fix a build issue
on sparc defconfig
- Remove restrictions while unregistering.
- Earlier code leaked inode references under some conditions while
registering/unregistering.
- Continue the vma-rmap walk even if the intermediate vma doesnt
meet the requirements.
- Validate the vma found by find_vma before inserting/removing the
breakpoint
- Call del_consumer under mutex_lock.
- Use hash locks.
- Handle mremap.
- Introduce find_least_offset_node() instead of close match logic in
find_uprobe
- Uprobes no more depends on MM_OWNER; No reference to task_structs
while inserting/removing a probe.
- Uses read_mapping_page instead of grab_cache_page so that the pages
have valid content.
- pass NULL to get_user_pages for the task parameter.
- call SetPageUptodate on the new page allocated in write_opcode.
- fix leaking a reference to the new page under certain conditions.
- Include Instruction Decoder if Uprobes gets defined.
- Remove const attributes for instruction prefix arrays.
- Uses mm_context to know if the application is 32 bit.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Also-written-by: Jim Keniston <jkenisto@us.ibm.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Roland McGrath <roland@hack.frob.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Anton Arapov <anton@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Denys Vlasenko <vda.linux@googlemail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linux-mm <linux-mm@kvack.org>
Link: http://lkml.kernel.org/r/20120209092642.GE16600@linux.vnet.ibm.com
[ Made various small edits to the commit log ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-02-09 17:26:42 +08:00
|
|
|
/* Read the page with vaddr into memory */
|
2017-02-25 06:57:42 +08:00
|
|
|
ret = get_user_pages_remote(NULL, mm, vaddr, 1,
|
|
|
|
FOLL_FORCE | FOLL_SPLIT, &old_page, &vma, NULL);
|
uprobes, mm, x86: Add the ability to install and remove uprobes breakpoints
Add uprobes support to the core kernel, with x86 support.
This commit adds the kernel facilities, the actual uprobes
user-space ABI and perf probe support comes in later commits.
General design:
Uprobes are maintained in an rb-tree indexed by inode and offset
(the offset here is from the start of the mapping). For a unique
(inode, offset) tuple, there can be at most one uprobe in the
rb-tree.
Since the (inode, offset) tuple identifies a unique uprobe, more
than one user may be interested in the same uprobe. This provides
the ability to connect multiple 'consumers' to the same uprobe.
Each consumer defines a handler and a filter (optional). The
'handler' is run every time the uprobe is hit, if it matches the
'filter' criteria.
The first consumer of a uprobe causes the breakpoint to be
inserted at the specified address and subsequent consumers are
appended to this list. On subsequent probes, the consumer gets
appended to the existing list of consumers. The breakpoint is
removed when the last consumer unregisters. For all other
unregisterations, the consumer is removed from the list of
consumers.
Given a inode, we get a list of the mms that have mapped the
inode. Do the actual registration if mm maps the page where a
probe needs to be inserted/removed.
We use a temporary list to walk through the vmas that map the
inode.
- The number of maps that map the inode, is not known before we
walk the rmap and keeps changing.
- extending vm_area_struct wasn't recommended, it's a
size-critical data structure.
- There can be more than one maps of the inode in the same mm.
We add callbacks to the mmap methods to keep an eye on text vmas
that are of interest to uprobes. When a vma of interest is mapped,
we insert the breakpoint at the right address.
Uprobe works by replacing the instruction at the address defined
by (inode, offset) with the arch specific breakpoint
instruction. We save a copy of the original instruction at the
uprobed address.
This is needed for:
a. executing the instruction out-of-line (xol).
b. instruction analysis for any subsequent fixups.
c. restoring the instruction back when the uprobe is unregistered.
We insert or delete a breakpoint instruction, and this
breakpoint instruction is assumed to be the smallest instruction
available on the platform. For fixed size instruction platforms
this is trivially true, for variable size instruction platforms
the breakpoint instruction is typically the smallest (often a
single byte).
Writing the instruction is done by COWing the page and changing
the instruction during the copy, this even though most platforms
allow atomic writes of the breakpoint instruction. This also
mirrors the behaviour of a ptrace() memory write to a PRIVATE
file map.
The core worker is derived from KSM's replace_page() logic.
In essence, similar to KSM:
a. allocate a new page and copy over contents of the page that
has the uprobed vaddr
b. modify the copy and insert the breakpoint at the required
address
c. switch the original page with the copy containing the
breakpoint
d. flush page tables.
replace_page() is being replicated here because of some minor
changes in the type of pages and also because Hugh Dickins had
plans to improve replace_page() for KSM specific work.
Instruction analysis on x86 is based on instruction decoder and
determines if an instruction can be probed and determines the
necessary fixups after singlestep. Instruction analysis is done
at probe insertion time so that we avoid having to repeat the
same analysis every time a probe is hit.
A lot of code here is due to the improvement/suggestions/inputs
from Peter Zijlstra.
Changelog:
(v10):
- Add code to clear REX.B prefix as suggested by Denys Vlasenko
and Masami Hiramatsu.
(v9):
- Use insn_offset_modrm as suggested by Masami Hiramatsu.
(v7):
Handle comments from Peter Zijlstra:
- Dont take reference to inode. (expect inode to uprobe_register to be sane).
- Use PTR_ERR to set the return value.
- No need to take reference to inode.
- use PTR_ERR to return error value.
- register and uprobe_unregister share code.
(v5):
- Modified del_consumer as per comments from Peter.
- Drop reference to inode before dropping reference to uprobe.
- Use i_size_read(inode) instead of inode->i_size.
- Ensure uprobe->consumers is NULL, before __uprobe_unregister() is called.
- Includes errno.h as recommended by Stephen Rothwell to fix a build issue
on sparc defconfig
- Remove restrictions while unregistering.
- Earlier code leaked inode references under some conditions while
registering/unregistering.
- Continue the vma-rmap walk even if the intermediate vma doesnt
meet the requirements.
- Validate the vma found by find_vma before inserting/removing the
breakpoint
- Call del_consumer under mutex_lock.
- Use hash locks.
- Handle mremap.
- Introduce find_least_offset_node() instead of close match logic in
find_uprobe
- Uprobes no more depends on MM_OWNER; No reference to task_structs
while inserting/removing a probe.
- Uses read_mapping_page instead of grab_cache_page so that the pages
have valid content.
- pass NULL to get_user_pages for the task parameter.
- call SetPageUptodate on the new page allocated in write_opcode.
- fix leaking a reference to the new page under certain conditions.
- Include Instruction Decoder if Uprobes gets defined.
- Remove const attributes for instruction prefix arrays.
- Uses mm_context to know if the application is 32 bit.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Also-written-by: Jim Keniston <jkenisto@us.ibm.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Roland McGrath <roland@hack.frob.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Anton Arapov <anton@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Denys Vlasenko <vda.linux@googlemail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linux-mm <linux-mm@kvack.org>
Link: http://lkml.kernel.org/r/20120209092642.GE16600@linux.vnet.ibm.com
[ Made various small edits to the commit log ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-02-09 17:26:42 +08:00
|
|
|
if (ret <= 0)
|
|
|
|
return ret;
|
2012-02-17 16:27:41 +08:00
|
|
|
|
2012-09-24 03:30:44 +08:00
|
|
|
ret = verify_opcode(old_page, vaddr, &opcode);
|
|
|
|
if (ret <= 0)
|
|
|
|
goto put_old;
|
|
|
|
|
uprobes: Support SDT markers having reference count (semaphore)
Userspace Statically Defined Tracepoints[1] are dtrace style markers
inside userspace applications. Applications like PostgreSQL, MySQL,
Pthread, Perl, Python, Java, Ruby, Node.js, libvirt, QEMU, glib etc
have these markers embedded in them. These markers are added by developer
at important places in the code. Each marker source expands to a single
nop instruction in the compiled code but there may be additional
overhead for computing the marker arguments which expands to couple of
instructions. In case the overhead is more, execution of it can be
omitted by runtime if() condition when no one is tracing on the marker:
if (reference_counter > 0) {
Execute marker instructions;
}
Default value of reference counter is 0. Tracer has to increment the
reference counter before tracing on a marker and decrement it when
done with the tracing.
Implement the reference counter logic in core uprobe. User will be
able to use it from trace_uprobe as well as from kernel module. New
trace_uprobe definition with reference counter will now be:
<path>:<offset>[(ref_ctr_offset)]
where ref_ctr_offset is an optional field. For kernel module, new
variant of uprobe_register() has been introduced:
uprobe_register_refctr(inode, offset, ref_ctr_offset, consumer)
No new variant for uprobe_unregister() because it's assumed to have
only one reference counter for one uprobe.
[1] https://sourceware.org/systemtap/wiki/UserSpaceProbeImplementation
Note: 'reference counter' is called as 'semaphore' in original Dtrace
(or Systemtap, bcc and even in ELF) documentation and code. But the
term 'semaphore' is misleading in this context. This is just a counter
used to hold number of tracers tracing on a marker. This is not really
used for any synchronization. So we are calling it a 'reference counter'
in kernel / perf code.
Link: http://lkml.kernel.org/r/20180820044250.11659-2-ravi.bangoria@linux.ibm.com
Reviewed-by: Masami Hiramatsu <mhiramat@kernel.org>
[Only trace_uprobe.c]
Reviewed-by: Oleg Nesterov <oleg@redhat.com>
Reviewed-by: Song Liu <songliubraving@fb.com>
Tested-by: Song Liu <songliubraving@fb.com>
Signed-off-by: Ravi Bangoria <ravi.bangoria@linux.ibm.com>
Signed-off-by: Steven Rostedt (VMware) <rostedt@goodmis.org>
2018-08-20 12:42:47 +08:00
|
|
|
/* We are going to replace instruction, update ref_ctr. */
|
|
|
|
if (!ref_ctr_updated && uprobe->ref_ctr_offset) {
|
|
|
|
ret = update_ref_ctr(uprobe, mm, is_register ? 1 : -1);
|
|
|
|
if (ret)
|
|
|
|
goto put_old;
|
|
|
|
|
|
|
|
ref_ctr_updated = 1;
|
|
|
|
}
|
|
|
|
|
2014-05-05 22:38:18 +08:00
|
|
|
ret = anon_vma_prepare(vma);
|
|
|
|
if (ret)
|
|
|
|
goto put_old;
|
|
|
|
|
uprobes, mm, x86: Add the ability to install and remove uprobes breakpoints
Add uprobes support to the core kernel, with x86 support.
This commit adds the kernel facilities, the actual uprobes
user-space ABI and perf probe support comes in later commits.
General design:
Uprobes are maintained in an rb-tree indexed by inode and offset
(the offset here is from the start of the mapping). For a unique
(inode, offset) tuple, there can be at most one uprobe in the
rb-tree.
Since the (inode, offset) tuple identifies a unique uprobe, more
than one user may be interested in the same uprobe. This provides
the ability to connect multiple 'consumers' to the same uprobe.
Each consumer defines a handler and a filter (optional). The
'handler' is run every time the uprobe is hit, if it matches the
'filter' criteria.
The first consumer of a uprobe causes the breakpoint to be
inserted at the specified address and subsequent consumers are
appended to this list. On subsequent probes, the consumer gets
appended to the existing list of consumers. The breakpoint is
removed when the last consumer unregisters. For all other
unregisterations, the consumer is removed from the list of
consumers.
Given a inode, we get a list of the mms that have mapped the
inode. Do the actual registration if mm maps the page where a
probe needs to be inserted/removed.
We use a temporary list to walk through the vmas that map the
inode.
- The number of maps that map the inode, is not known before we
walk the rmap and keeps changing.
- extending vm_area_struct wasn't recommended, it's a
size-critical data structure.
- There can be more than one maps of the inode in the same mm.
We add callbacks to the mmap methods to keep an eye on text vmas
that are of interest to uprobes. When a vma of interest is mapped,
we insert the breakpoint at the right address.
Uprobe works by replacing the instruction at the address defined
by (inode, offset) with the arch specific breakpoint
instruction. We save a copy of the original instruction at the
uprobed address.
This is needed for:
a. executing the instruction out-of-line (xol).
b. instruction analysis for any subsequent fixups.
c. restoring the instruction back when the uprobe is unregistered.
We insert or delete a breakpoint instruction, and this
breakpoint instruction is assumed to be the smallest instruction
available on the platform. For fixed size instruction platforms
this is trivially true, for variable size instruction platforms
the breakpoint instruction is typically the smallest (often a
single byte).
Writing the instruction is done by COWing the page and changing
the instruction during the copy, this even though most platforms
allow atomic writes of the breakpoint instruction. This also
mirrors the behaviour of a ptrace() memory write to a PRIVATE
file map.
The core worker is derived from KSM's replace_page() logic.
In essence, similar to KSM:
a. allocate a new page and copy over contents of the page that
has the uprobed vaddr
b. modify the copy and insert the breakpoint at the required
address
c. switch the original page with the copy containing the
breakpoint
d. flush page tables.
replace_page() is being replicated here because of some minor
changes in the type of pages and also because Hugh Dickins had
plans to improve replace_page() for KSM specific work.
Instruction analysis on x86 is based on instruction decoder and
determines if an instruction can be probed and determines the
necessary fixups after singlestep. Instruction analysis is done
at probe insertion time so that we avoid having to repeat the
same analysis every time a probe is hit.
A lot of code here is due to the improvement/suggestions/inputs
from Peter Zijlstra.
Changelog:
(v10):
- Add code to clear REX.B prefix as suggested by Denys Vlasenko
and Masami Hiramatsu.
(v9):
- Use insn_offset_modrm as suggested by Masami Hiramatsu.
(v7):
Handle comments from Peter Zijlstra:
- Dont take reference to inode. (expect inode to uprobe_register to be sane).
- Use PTR_ERR to set the return value.
- No need to take reference to inode.
- use PTR_ERR to return error value.
- register and uprobe_unregister share code.
(v5):
- Modified del_consumer as per comments from Peter.
- Drop reference to inode before dropping reference to uprobe.
- Use i_size_read(inode) instead of inode->i_size.
- Ensure uprobe->consumers is NULL, before __uprobe_unregister() is called.
- Includes errno.h as recommended by Stephen Rothwell to fix a build issue
on sparc defconfig
- Remove restrictions while unregistering.
- Earlier code leaked inode references under some conditions while
registering/unregistering.
- Continue the vma-rmap walk even if the intermediate vma doesnt
meet the requirements.
- Validate the vma found by find_vma before inserting/removing the
breakpoint
- Call del_consumer under mutex_lock.
- Use hash locks.
- Handle mremap.
- Introduce find_least_offset_node() instead of close match logic in
find_uprobe
- Uprobes no more depends on MM_OWNER; No reference to task_structs
while inserting/removing a probe.
- Uses read_mapping_page instead of grab_cache_page so that the pages
have valid content.
- pass NULL to get_user_pages for the task parameter.
- call SetPageUptodate on the new page allocated in write_opcode.
- fix leaking a reference to the new page under certain conditions.
- Include Instruction Decoder if Uprobes gets defined.
- Remove const attributes for instruction prefix arrays.
- Uses mm_context to know if the application is 32 bit.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Also-written-by: Jim Keniston <jkenisto@us.ibm.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Roland McGrath <roland@hack.frob.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Anton Arapov <anton@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Denys Vlasenko <vda.linux@googlemail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linux-mm <linux-mm@kvack.org>
Link: http://lkml.kernel.org/r/20120209092642.GE16600@linux.vnet.ibm.com
[ Made various small edits to the commit log ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-02-09 17:26:42 +08:00
|
|
|
ret = -ENOMEM;
|
|
|
|
new_page = alloc_page_vma(GFP_HIGHUSER_MOVABLE, vma, vaddr);
|
|
|
|
if (!new_page)
|
2012-07-30 02:22:20 +08:00
|
|
|
goto put_old;
|
uprobes, mm, x86: Add the ability to install and remove uprobes breakpoints
Add uprobes support to the core kernel, with x86 support.
This commit adds the kernel facilities, the actual uprobes
user-space ABI and perf probe support comes in later commits.
General design:
Uprobes are maintained in an rb-tree indexed by inode and offset
(the offset here is from the start of the mapping). For a unique
(inode, offset) tuple, there can be at most one uprobe in the
rb-tree.
Since the (inode, offset) tuple identifies a unique uprobe, more
than one user may be interested in the same uprobe. This provides
the ability to connect multiple 'consumers' to the same uprobe.
Each consumer defines a handler and a filter (optional). The
'handler' is run every time the uprobe is hit, if it matches the
'filter' criteria.
The first consumer of a uprobe causes the breakpoint to be
inserted at the specified address and subsequent consumers are
appended to this list. On subsequent probes, the consumer gets
appended to the existing list of consumers. The breakpoint is
removed when the last consumer unregisters. For all other
unregisterations, the consumer is removed from the list of
consumers.
Given a inode, we get a list of the mms that have mapped the
inode. Do the actual registration if mm maps the page where a
probe needs to be inserted/removed.
We use a temporary list to walk through the vmas that map the
inode.
- The number of maps that map the inode, is not known before we
walk the rmap and keeps changing.
- extending vm_area_struct wasn't recommended, it's a
size-critical data structure.
- There can be more than one maps of the inode in the same mm.
We add callbacks to the mmap methods to keep an eye on text vmas
that are of interest to uprobes. When a vma of interest is mapped,
we insert the breakpoint at the right address.
Uprobe works by replacing the instruction at the address defined
by (inode, offset) with the arch specific breakpoint
instruction. We save a copy of the original instruction at the
uprobed address.
This is needed for:
a. executing the instruction out-of-line (xol).
b. instruction analysis for any subsequent fixups.
c. restoring the instruction back when the uprobe is unregistered.
We insert or delete a breakpoint instruction, and this
breakpoint instruction is assumed to be the smallest instruction
available on the platform. For fixed size instruction platforms
this is trivially true, for variable size instruction platforms
the breakpoint instruction is typically the smallest (often a
single byte).
Writing the instruction is done by COWing the page and changing
the instruction during the copy, this even though most platforms
allow atomic writes of the breakpoint instruction. This also
mirrors the behaviour of a ptrace() memory write to a PRIVATE
file map.
The core worker is derived from KSM's replace_page() logic.
In essence, similar to KSM:
a. allocate a new page and copy over contents of the page that
has the uprobed vaddr
b. modify the copy and insert the breakpoint at the required
address
c. switch the original page with the copy containing the
breakpoint
d. flush page tables.
replace_page() is being replicated here because of some minor
changes in the type of pages and also because Hugh Dickins had
plans to improve replace_page() for KSM specific work.
Instruction analysis on x86 is based on instruction decoder and
determines if an instruction can be probed and determines the
necessary fixups after singlestep. Instruction analysis is done
at probe insertion time so that we avoid having to repeat the
same analysis every time a probe is hit.
A lot of code here is due to the improvement/suggestions/inputs
from Peter Zijlstra.
Changelog:
(v10):
- Add code to clear REX.B prefix as suggested by Denys Vlasenko
and Masami Hiramatsu.
(v9):
- Use insn_offset_modrm as suggested by Masami Hiramatsu.
(v7):
Handle comments from Peter Zijlstra:
- Dont take reference to inode. (expect inode to uprobe_register to be sane).
- Use PTR_ERR to set the return value.
- No need to take reference to inode.
- use PTR_ERR to return error value.
- register and uprobe_unregister share code.
(v5):
- Modified del_consumer as per comments from Peter.
- Drop reference to inode before dropping reference to uprobe.
- Use i_size_read(inode) instead of inode->i_size.
- Ensure uprobe->consumers is NULL, before __uprobe_unregister() is called.
- Includes errno.h as recommended by Stephen Rothwell to fix a build issue
on sparc defconfig
- Remove restrictions while unregistering.
- Earlier code leaked inode references under some conditions while
registering/unregistering.
- Continue the vma-rmap walk even if the intermediate vma doesnt
meet the requirements.
- Validate the vma found by find_vma before inserting/removing the
breakpoint
- Call del_consumer under mutex_lock.
- Use hash locks.
- Handle mremap.
- Introduce find_least_offset_node() instead of close match logic in
find_uprobe
- Uprobes no more depends on MM_OWNER; No reference to task_structs
while inserting/removing a probe.
- Uses read_mapping_page instead of grab_cache_page so that the pages
have valid content.
- pass NULL to get_user_pages for the task parameter.
- call SetPageUptodate on the new page allocated in write_opcode.
- fix leaking a reference to the new page under certain conditions.
- Include Instruction Decoder if Uprobes gets defined.
- Remove const attributes for instruction prefix arrays.
- Uses mm_context to know if the application is 32 bit.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Also-written-by: Jim Keniston <jkenisto@us.ibm.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Roland McGrath <roland@hack.frob.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Anton Arapov <anton@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Denys Vlasenko <vda.linux@googlemail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linux-mm <linux-mm@kvack.org>
Link: http://lkml.kernel.org/r/20120209092642.GE16600@linux.vnet.ibm.com
[ Made various small edits to the commit log ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-02-09 17:26:42 +08:00
|
|
|
|
2014-05-05 22:38:18 +08:00
|
|
|
__SetPageUptodate(new_page);
|
2013-03-25 02:04:36 +08:00
|
|
|
copy_highpage(new_page, old_page);
|
|
|
|
copy_to_page(new_page, vaddr, &opcode, UPROBE_SWBP_INSN_SIZE);
|
uprobes, mm, x86: Add the ability to install and remove uprobes breakpoints
Add uprobes support to the core kernel, with x86 support.
This commit adds the kernel facilities, the actual uprobes
user-space ABI and perf probe support comes in later commits.
General design:
Uprobes are maintained in an rb-tree indexed by inode and offset
(the offset here is from the start of the mapping). For a unique
(inode, offset) tuple, there can be at most one uprobe in the
rb-tree.
Since the (inode, offset) tuple identifies a unique uprobe, more
than one user may be interested in the same uprobe. This provides
the ability to connect multiple 'consumers' to the same uprobe.
Each consumer defines a handler and a filter (optional). The
'handler' is run every time the uprobe is hit, if it matches the
'filter' criteria.
The first consumer of a uprobe causes the breakpoint to be
inserted at the specified address and subsequent consumers are
appended to this list. On subsequent probes, the consumer gets
appended to the existing list of consumers. The breakpoint is
removed when the last consumer unregisters. For all other
unregisterations, the consumer is removed from the list of
consumers.
Given a inode, we get a list of the mms that have mapped the
inode. Do the actual registration if mm maps the page where a
probe needs to be inserted/removed.
We use a temporary list to walk through the vmas that map the
inode.
- The number of maps that map the inode, is not known before we
walk the rmap and keeps changing.
- extending vm_area_struct wasn't recommended, it's a
size-critical data structure.
- There can be more than one maps of the inode in the same mm.
We add callbacks to the mmap methods to keep an eye on text vmas
that are of interest to uprobes. When a vma of interest is mapped,
we insert the breakpoint at the right address.
Uprobe works by replacing the instruction at the address defined
by (inode, offset) with the arch specific breakpoint
instruction. We save a copy of the original instruction at the
uprobed address.
This is needed for:
a. executing the instruction out-of-line (xol).
b. instruction analysis for any subsequent fixups.
c. restoring the instruction back when the uprobe is unregistered.
We insert or delete a breakpoint instruction, and this
breakpoint instruction is assumed to be the smallest instruction
available on the platform. For fixed size instruction platforms
this is trivially true, for variable size instruction platforms
the breakpoint instruction is typically the smallest (often a
single byte).
Writing the instruction is done by COWing the page and changing
the instruction during the copy, this even though most platforms
allow atomic writes of the breakpoint instruction. This also
mirrors the behaviour of a ptrace() memory write to a PRIVATE
file map.
The core worker is derived from KSM's replace_page() logic.
In essence, similar to KSM:
a. allocate a new page and copy over contents of the page that
has the uprobed vaddr
b. modify the copy and insert the breakpoint at the required
address
c. switch the original page with the copy containing the
breakpoint
d. flush page tables.
replace_page() is being replicated here because of some minor
changes in the type of pages and also because Hugh Dickins had
plans to improve replace_page() for KSM specific work.
Instruction analysis on x86 is based on instruction decoder and
determines if an instruction can be probed and determines the
necessary fixups after singlestep. Instruction analysis is done
at probe insertion time so that we avoid having to repeat the
same analysis every time a probe is hit.
A lot of code here is due to the improvement/suggestions/inputs
from Peter Zijlstra.
Changelog:
(v10):
- Add code to clear REX.B prefix as suggested by Denys Vlasenko
and Masami Hiramatsu.
(v9):
- Use insn_offset_modrm as suggested by Masami Hiramatsu.
(v7):
Handle comments from Peter Zijlstra:
- Dont take reference to inode. (expect inode to uprobe_register to be sane).
- Use PTR_ERR to set the return value.
- No need to take reference to inode.
- use PTR_ERR to return error value.
- register and uprobe_unregister share code.
(v5):
- Modified del_consumer as per comments from Peter.
- Drop reference to inode before dropping reference to uprobe.
- Use i_size_read(inode) instead of inode->i_size.
- Ensure uprobe->consumers is NULL, before __uprobe_unregister() is called.
- Includes errno.h as recommended by Stephen Rothwell to fix a build issue
on sparc defconfig
- Remove restrictions while unregistering.
- Earlier code leaked inode references under some conditions while
registering/unregistering.
- Continue the vma-rmap walk even if the intermediate vma doesnt
meet the requirements.
- Validate the vma found by find_vma before inserting/removing the
breakpoint
- Call del_consumer under mutex_lock.
- Use hash locks.
- Handle mremap.
- Introduce find_least_offset_node() instead of close match logic in
find_uprobe
- Uprobes no more depends on MM_OWNER; No reference to task_structs
while inserting/removing a probe.
- Uses read_mapping_page instead of grab_cache_page so that the pages
have valid content.
- pass NULL to get_user_pages for the task parameter.
- call SetPageUptodate on the new page allocated in write_opcode.
- fix leaking a reference to the new page under certain conditions.
- Include Instruction Decoder if Uprobes gets defined.
- Remove const attributes for instruction prefix arrays.
- Uses mm_context to know if the application is 32 bit.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Also-written-by: Jim Keniston <jkenisto@us.ibm.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Roland McGrath <roland@hack.frob.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Anton Arapov <anton@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Denys Vlasenko <vda.linux@googlemail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linux-mm <linux-mm@kvack.org>
Link: http://lkml.kernel.org/r/20120209092642.GE16600@linux.vnet.ibm.com
[ Made various small edits to the commit log ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-02-09 17:26:42 +08:00
|
|
|
|
2012-07-30 02:22:16 +08:00
|
|
|
ret = __replace_page(vma, vaddr, old_page, new_page);
|
mm, fs: get rid of PAGE_CACHE_* and page_cache_{get,release} macros
PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} macros were introduced *long* time
ago with promise that one day it will be possible to implement page
cache with bigger chunks than PAGE_SIZE.
This promise never materialized. And unlikely will.
We have many places where PAGE_CACHE_SIZE assumed to be equal to
PAGE_SIZE. And it's constant source of confusion on whether
PAGE_CACHE_* or PAGE_* constant should be used in a particular case,
especially on the border between fs and mm.
Global switching to PAGE_CACHE_SIZE != PAGE_SIZE would cause to much
breakage to be doable.
Let's stop pretending that pages in page cache are special. They are
not.
The changes are pretty straight-forward:
- <foo> << (PAGE_CACHE_SHIFT - PAGE_SHIFT) -> <foo>;
- <foo> >> (PAGE_CACHE_SHIFT - PAGE_SHIFT) -> <foo>;
- PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} -> PAGE_{SIZE,SHIFT,MASK,ALIGN};
- page_cache_get() -> get_page();
- page_cache_release() -> put_page();
This patch contains automated changes generated with coccinelle using
script below. For some reason, coccinelle doesn't patch header files.
I've called spatch for them manually.
The only adjustment after coccinelle is revert of changes to
PAGE_CAHCE_ALIGN definition: we are going to drop it later.
There are few places in the code where coccinelle didn't reach. I'll
fix them manually in a separate patch. Comments and documentation also
will be addressed with the separate patch.
virtual patch
@@
expression E;
@@
- E << (PAGE_CACHE_SHIFT - PAGE_SHIFT)
+ E
@@
expression E;
@@
- E >> (PAGE_CACHE_SHIFT - PAGE_SHIFT)
+ E
@@
@@
- PAGE_CACHE_SHIFT
+ PAGE_SHIFT
@@
@@
- PAGE_CACHE_SIZE
+ PAGE_SIZE
@@
@@
- PAGE_CACHE_MASK
+ PAGE_MASK
@@
expression E;
@@
- PAGE_CACHE_ALIGN(E)
+ PAGE_ALIGN(E)
@@
expression E;
@@
- page_cache_get(E)
+ get_page(E)
@@
expression E;
@@
- page_cache_release(E)
+ put_page(E)
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Acked-by: Michal Hocko <mhocko@suse.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-04-01 20:29:47 +08:00
|
|
|
put_page(new_page);
|
2012-07-30 02:22:20 +08:00
|
|
|
put_old:
|
2012-02-17 16:27:41 +08:00
|
|
|
put_page(old_page);
|
|
|
|
|
2012-06-15 23:43:28 +08:00
|
|
|
if (unlikely(ret == -EAGAIN))
|
|
|
|
goto retry;
|
uprobes: Support SDT markers having reference count (semaphore)
Userspace Statically Defined Tracepoints[1] are dtrace style markers
inside userspace applications. Applications like PostgreSQL, MySQL,
Pthread, Perl, Python, Java, Ruby, Node.js, libvirt, QEMU, glib etc
have these markers embedded in them. These markers are added by developer
at important places in the code. Each marker source expands to a single
nop instruction in the compiled code but there may be additional
overhead for computing the marker arguments which expands to couple of
instructions. In case the overhead is more, execution of it can be
omitted by runtime if() condition when no one is tracing on the marker:
if (reference_counter > 0) {
Execute marker instructions;
}
Default value of reference counter is 0. Tracer has to increment the
reference counter before tracing on a marker and decrement it when
done with the tracing.
Implement the reference counter logic in core uprobe. User will be
able to use it from trace_uprobe as well as from kernel module. New
trace_uprobe definition with reference counter will now be:
<path>:<offset>[(ref_ctr_offset)]
where ref_ctr_offset is an optional field. For kernel module, new
variant of uprobe_register() has been introduced:
uprobe_register_refctr(inode, offset, ref_ctr_offset, consumer)
No new variant for uprobe_unregister() because it's assumed to have
only one reference counter for one uprobe.
[1] https://sourceware.org/systemtap/wiki/UserSpaceProbeImplementation
Note: 'reference counter' is called as 'semaphore' in original Dtrace
(or Systemtap, bcc and even in ELF) documentation and code. But the
term 'semaphore' is misleading in this context. This is just a counter
used to hold number of tracers tracing on a marker. This is not really
used for any synchronization. So we are calling it a 'reference counter'
in kernel / perf code.
Link: http://lkml.kernel.org/r/20180820044250.11659-2-ravi.bangoria@linux.ibm.com
Reviewed-by: Masami Hiramatsu <mhiramat@kernel.org>
[Only trace_uprobe.c]
Reviewed-by: Oleg Nesterov <oleg@redhat.com>
Reviewed-by: Song Liu <songliubraving@fb.com>
Tested-by: Song Liu <songliubraving@fb.com>
Signed-off-by: Ravi Bangoria <ravi.bangoria@linux.ibm.com>
Signed-off-by: Steven Rostedt (VMware) <rostedt@goodmis.org>
2018-08-20 12:42:47 +08:00
|
|
|
|
|
|
|
/* Revert back reference counter if instruction update failed. */
|
|
|
|
if (ret && is_register && ref_ctr_updated)
|
|
|
|
update_ref_ctr(uprobe, mm, -1);
|
|
|
|
|
uprobes, mm, x86: Add the ability to install and remove uprobes breakpoints
Add uprobes support to the core kernel, with x86 support.
This commit adds the kernel facilities, the actual uprobes
user-space ABI and perf probe support comes in later commits.
General design:
Uprobes are maintained in an rb-tree indexed by inode and offset
(the offset here is from the start of the mapping). For a unique
(inode, offset) tuple, there can be at most one uprobe in the
rb-tree.
Since the (inode, offset) tuple identifies a unique uprobe, more
than one user may be interested in the same uprobe. This provides
the ability to connect multiple 'consumers' to the same uprobe.
Each consumer defines a handler and a filter (optional). The
'handler' is run every time the uprobe is hit, if it matches the
'filter' criteria.
The first consumer of a uprobe causes the breakpoint to be
inserted at the specified address and subsequent consumers are
appended to this list. On subsequent probes, the consumer gets
appended to the existing list of consumers. The breakpoint is
removed when the last consumer unregisters. For all other
unregisterations, the consumer is removed from the list of
consumers.
Given a inode, we get a list of the mms that have mapped the
inode. Do the actual registration if mm maps the page where a
probe needs to be inserted/removed.
We use a temporary list to walk through the vmas that map the
inode.
- The number of maps that map the inode, is not known before we
walk the rmap and keeps changing.
- extending vm_area_struct wasn't recommended, it's a
size-critical data structure.
- There can be more than one maps of the inode in the same mm.
We add callbacks to the mmap methods to keep an eye on text vmas
that are of interest to uprobes. When a vma of interest is mapped,
we insert the breakpoint at the right address.
Uprobe works by replacing the instruction at the address defined
by (inode, offset) with the arch specific breakpoint
instruction. We save a copy of the original instruction at the
uprobed address.
This is needed for:
a. executing the instruction out-of-line (xol).
b. instruction analysis for any subsequent fixups.
c. restoring the instruction back when the uprobe is unregistered.
We insert or delete a breakpoint instruction, and this
breakpoint instruction is assumed to be the smallest instruction
available on the platform. For fixed size instruction platforms
this is trivially true, for variable size instruction platforms
the breakpoint instruction is typically the smallest (often a
single byte).
Writing the instruction is done by COWing the page and changing
the instruction during the copy, this even though most platforms
allow atomic writes of the breakpoint instruction. This also
mirrors the behaviour of a ptrace() memory write to a PRIVATE
file map.
The core worker is derived from KSM's replace_page() logic.
In essence, similar to KSM:
a. allocate a new page and copy over contents of the page that
has the uprobed vaddr
b. modify the copy and insert the breakpoint at the required
address
c. switch the original page with the copy containing the
breakpoint
d. flush page tables.
replace_page() is being replicated here because of some minor
changes in the type of pages and also because Hugh Dickins had
plans to improve replace_page() for KSM specific work.
Instruction analysis on x86 is based on instruction decoder and
determines if an instruction can be probed and determines the
necessary fixups after singlestep. Instruction analysis is done
at probe insertion time so that we avoid having to repeat the
same analysis every time a probe is hit.
A lot of code here is due to the improvement/suggestions/inputs
from Peter Zijlstra.
Changelog:
(v10):
- Add code to clear REX.B prefix as suggested by Denys Vlasenko
and Masami Hiramatsu.
(v9):
- Use insn_offset_modrm as suggested by Masami Hiramatsu.
(v7):
Handle comments from Peter Zijlstra:
- Dont take reference to inode. (expect inode to uprobe_register to be sane).
- Use PTR_ERR to set the return value.
- No need to take reference to inode.
- use PTR_ERR to return error value.
- register and uprobe_unregister share code.
(v5):
- Modified del_consumer as per comments from Peter.
- Drop reference to inode before dropping reference to uprobe.
- Use i_size_read(inode) instead of inode->i_size.
- Ensure uprobe->consumers is NULL, before __uprobe_unregister() is called.
- Includes errno.h as recommended by Stephen Rothwell to fix a build issue
on sparc defconfig
- Remove restrictions while unregistering.
- Earlier code leaked inode references under some conditions while
registering/unregistering.
- Continue the vma-rmap walk even if the intermediate vma doesnt
meet the requirements.
- Validate the vma found by find_vma before inserting/removing the
breakpoint
- Call del_consumer under mutex_lock.
- Use hash locks.
- Handle mremap.
- Introduce find_least_offset_node() instead of close match logic in
find_uprobe
- Uprobes no more depends on MM_OWNER; No reference to task_structs
while inserting/removing a probe.
- Uses read_mapping_page instead of grab_cache_page so that the pages
have valid content.
- pass NULL to get_user_pages for the task parameter.
- call SetPageUptodate on the new page allocated in write_opcode.
- fix leaking a reference to the new page under certain conditions.
- Include Instruction Decoder if Uprobes gets defined.
- Remove const attributes for instruction prefix arrays.
- Uses mm_context to know if the application is 32 bit.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Also-written-by: Jim Keniston <jkenisto@us.ibm.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Roland McGrath <roland@hack.frob.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Anton Arapov <anton@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Denys Vlasenko <vda.linux@googlemail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linux-mm <linux-mm@kvack.org>
Link: http://lkml.kernel.org/r/20120209092642.GE16600@linux.vnet.ibm.com
[ Made various small edits to the commit log ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-02-09 17:26:42 +08:00
|
|
|
return ret;
|
|
|
|
}
|
|
|
|
|
|
|
|
/**
|
2012-03-12 17:25:45 +08:00
|
|
|
* set_swbp - store breakpoint at a given address.
|
2012-03-12 17:25:30 +08:00
|
|
|
* @auprobe: arch specific probepoint information.
|
uprobes, mm, x86: Add the ability to install and remove uprobes breakpoints
Add uprobes support to the core kernel, with x86 support.
This commit adds the kernel facilities, the actual uprobes
user-space ABI and perf probe support comes in later commits.
General design:
Uprobes are maintained in an rb-tree indexed by inode and offset
(the offset here is from the start of the mapping). For a unique
(inode, offset) tuple, there can be at most one uprobe in the
rb-tree.
Since the (inode, offset) tuple identifies a unique uprobe, more
than one user may be interested in the same uprobe. This provides
the ability to connect multiple 'consumers' to the same uprobe.
Each consumer defines a handler and a filter (optional). The
'handler' is run every time the uprobe is hit, if it matches the
'filter' criteria.
The first consumer of a uprobe causes the breakpoint to be
inserted at the specified address and subsequent consumers are
appended to this list. On subsequent probes, the consumer gets
appended to the existing list of consumers. The breakpoint is
removed when the last consumer unregisters. For all other
unregisterations, the consumer is removed from the list of
consumers.
Given a inode, we get a list of the mms that have mapped the
inode. Do the actual registration if mm maps the page where a
probe needs to be inserted/removed.
We use a temporary list to walk through the vmas that map the
inode.
- The number of maps that map the inode, is not known before we
walk the rmap and keeps changing.
- extending vm_area_struct wasn't recommended, it's a
size-critical data structure.
- There can be more than one maps of the inode in the same mm.
We add callbacks to the mmap methods to keep an eye on text vmas
that are of interest to uprobes. When a vma of interest is mapped,
we insert the breakpoint at the right address.
Uprobe works by replacing the instruction at the address defined
by (inode, offset) with the arch specific breakpoint
instruction. We save a copy of the original instruction at the
uprobed address.
This is needed for:
a. executing the instruction out-of-line (xol).
b. instruction analysis for any subsequent fixups.
c. restoring the instruction back when the uprobe is unregistered.
We insert or delete a breakpoint instruction, and this
breakpoint instruction is assumed to be the smallest instruction
available on the platform. For fixed size instruction platforms
this is trivially true, for variable size instruction platforms
the breakpoint instruction is typically the smallest (often a
single byte).
Writing the instruction is done by COWing the page and changing
the instruction during the copy, this even though most platforms
allow atomic writes of the breakpoint instruction. This also
mirrors the behaviour of a ptrace() memory write to a PRIVATE
file map.
The core worker is derived from KSM's replace_page() logic.
In essence, similar to KSM:
a. allocate a new page and copy over contents of the page that
has the uprobed vaddr
b. modify the copy and insert the breakpoint at the required
address
c. switch the original page with the copy containing the
breakpoint
d. flush page tables.
replace_page() is being replicated here because of some minor
changes in the type of pages and also because Hugh Dickins had
plans to improve replace_page() for KSM specific work.
Instruction analysis on x86 is based on instruction decoder and
determines if an instruction can be probed and determines the
necessary fixups after singlestep. Instruction analysis is done
at probe insertion time so that we avoid having to repeat the
same analysis every time a probe is hit.
A lot of code here is due to the improvement/suggestions/inputs
from Peter Zijlstra.
Changelog:
(v10):
- Add code to clear REX.B prefix as suggested by Denys Vlasenko
and Masami Hiramatsu.
(v9):
- Use insn_offset_modrm as suggested by Masami Hiramatsu.
(v7):
Handle comments from Peter Zijlstra:
- Dont take reference to inode. (expect inode to uprobe_register to be sane).
- Use PTR_ERR to set the return value.
- No need to take reference to inode.
- use PTR_ERR to return error value.
- register and uprobe_unregister share code.
(v5):
- Modified del_consumer as per comments from Peter.
- Drop reference to inode before dropping reference to uprobe.
- Use i_size_read(inode) instead of inode->i_size.
- Ensure uprobe->consumers is NULL, before __uprobe_unregister() is called.
- Includes errno.h as recommended by Stephen Rothwell to fix a build issue
on sparc defconfig
- Remove restrictions while unregistering.
- Earlier code leaked inode references under some conditions while
registering/unregistering.
- Continue the vma-rmap walk even if the intermediate vma doesnt
meet the requirements.
- Validate the vma found by find_vma before inserting/removing the
breakpoint
- Call del_consumer under mutex_lock.
- Use hash locks.
- Handle mremap.
- Introduce find_least_offset_node() instead of close match logic in
find_uprobe
- Uprobes no more depends on MM_OWNER; No reference to task_structs
while inserting/removing a probe.
- Uses read_mapping_page instead of grab_cache_page so that the pages
have valid content.
- pass NULL to get_user_pages for the task parameter.
- call SetPageUptodate on the new page allocated in write_opcode.
- fix leaking a reference to the new page under certain conditions.
- Include Instruction Decoder if Uprobes gets defined.
- Remove const attributes for instruction prefix arrays.
- Uses mm_context to know if the application is 32 bit.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Also-written-by: Jim Keniston <jkenisto@us.ibm.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Roland McGrath <roland@hack.frob.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Anton Arapov <anton@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Denys Vlasenko <vda.linux@googlemail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linux-mm <linux-mm@kvack.org>
Link: http://lkml.kernel.org/r/20120209092642.GE16600@linux.vnet.ibm.com
[ Made various small edits to the commit log ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-02-09 17:26:42 +08:00
|
|
|
* @mm: the probed process address space.
|
|
|
|
* @vaddr: the virtual address to insert the opcode.
|
|
|
|
*
|
|
|
|
* For mm @mm, store the breakpoint instruction at @vaddr.
|
|
|
|
* Return 0 (success) or a negative errno.
|
|
|
|
*/
|
2012-03-12 17:25:45 +08:00
|
|
|
int __weak set_swbp(struct arch_uprobe *auprobe, struct mm_struct *mm, unsigned long vaddr)
|
uprobes, mm, x86: Add the ability to install and remove uprobes breakpoints
Add uprobes support to the core kernel, with x86 support.
This commit adds the kernel facilities, the actual uprobes
user-space ABI and perf probe support comes in later commits.
General design:
Uprobes are maintained in an rb-tree indexed by inode and offset
(the offset here is from the start of the mapping). For a unique
(inode, offset) tuple, there can be at most one uprobe in the
rb-tree.
Since the (inode, offset) tuple identifies a unique uprobe, more
than one user may be interested in the same uprobe. This provides
the ability to connect multiple 'consumers' to the same uprobe.
Each consumer defines a handler and a filter (optional). The
'handler' is run every time the uprobe is hit, if it matches the
'filter' criteria.
The first consumer of a uprobe causes the breakpoint to be
inserted at the specified address and subsequent consumers are
appended to this list. On subsequent probes, the consumer gets
appended to the existing list of consumers. The breakpoint is
removed when the last consumer unregisters. For all other
unregisterations, the consumer is removed from the list of
consumers.
Given a inode, we get a list of the mms that have mapped the
inode. Do the actual registration if mm maps the page where a
probe needs to be inserted/removed.
We use a temporary list to walk through the vmas that map the
inode.
- The number of maps that map the inode, is not known before we
walk the rmap and keeps changing.
- extending vm_area_struct wasn't recommended, it's a
size-critical data structure.
- There can be more than one maps of the inode in the same mm.
We add callbacks to the mmap methods to keep an eye on text vmas
that are of interest to uprobes. When a vma of interest is mapped,
we insert the breakpoint at the right address.
Uprobe works by replacing the instruction at the address defined
by (inode, offset) with the arch specific breakpoint
instruction. We save a copy of the original instruction at the
uprobed address.
This is needed for:
a. executing the instruction out-of-line (xol).
b. instruction analysis for any subsequent fixups.
c. restoring the instruction back when the uprobe is unregistered.
We insert or delete a breakpoint instruction, and this
breakpoint instruction is assumed to be the smallest instruction
available on the platform. For fixed size instruction platforms
this is trivially true, for variable size instruction platforms
the breakpoint instruction is typically the smallest (often a
single byte).
Writing the instruction is done by COWing the page and changing
the instruction during the copy, this even though most platforms
allow atomic writes of the breakpoint instruction. This also
mirrors the behaviour of a ptrace() memory write to a PRIVATE
file map.
The core worker is derived from KSM's replace_page() logic.
In essence, similar to KSM:
a. allocate a new page and copy over contents of the page that
has the uprobed vaddr
b. modify the copy and insert the breakpoint at the required
address
c. switch the original page with the copy containing the
breakpoint
d. flush page tables.
replace_page() is being replicated here because of some minor
changes in the type of pages and also because Hugh Dickins had
plans to improve replace_page() for KSM specific work.
Instruction analysis on x86 is based on instruction decoder and
determines if an instruction can be probed and determines the
necessary fixups after singlestep. Instruction analysis is done
at probe insertion time so that we avoid having to repeat the
same analysis every time a probe is hit.
A lot of code here is due to the improvement/suggestions/inputs
from Peter Zijlstra.
Changelog:
(v10):
- Add code to clear REX.B prefix as suggested by Denys Vlasenko
and Masami Hiramatsu.
(v9):
- Use insn_offset_modrm as suggested by Masami Hiramatsu.
(v7):
Handle comments from Peter Zijlstra:
- Dont take reference to inode. (expect inode to uprobe_register to be sane).
- Use PTR_ERR to set the return value.
- No need to take reference to inode.
- use PTR_ERR to return error value.
- register and uprobe_unregister share code.
(v5):
- Modified del_consumer as per comments from Peter.
- Drop reference to inode before dropping reference to uprobe.
- Use i_size_read(inode) instead of inode->i_size.
- Ensure uprobe->consumers is NULL, before __uprobe_unregister() is called.
- Includes errno.h as recommended by Stephen Rothwell to fix a build issue
on sparc defconfig
- Remove restrictions while unregistering.
- Earlier code leaked inode references under some conditions while
registering/unregistering.
- Continue the vma-rmap walk even if the intermediate vma doesnt
meet the requirements.
- Validate the vma found by find_vma before inserting/removing the
breakpoint
- Call del_consumer under mutex_lock.
- Use hash locks.
- Handle mremap.
- Introduce find_least_offset_node() instead of close match logic in
find_uprobe
- Uprobes no more depends on MM_OWNER; No reference to task_structs
while inserting/removing a probe.
- Uses read_mapping_page instead of grab_cache_page so that the pages
have valid content.
- pass NULL to get_user_pages for the task parameter.
- call SetPageUptodate on the new page allocated in write_opcode.
- fix leaking a reference to the new page under certain conditions.
- Include Instruction Decoder if Uprobes gets defined.
- Remove const attributes for instruction prefix arrays.
- Uses mm_context to know if the application is 32 bit.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Also-written-by: Jim Keniston <jkenisto@us.ibm.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Roland McGrath <roland@hack.frob.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Anton Arapov <anton@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Denys Vlasenko <vda.linux@googlemail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linux-mm <linux-mm@kvack.org>
Link: http://lkml.kernel.org/r/20120209092642.GE16600@linux.vnet.ibm.com
[ Made various small edits to the commit log ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-02-09 17:26:42 +08:00
|
|
|
{
|
2018-08-09 12:18:52 +08:00
|
|
|
return uprobe_write_opcode(auprobe, mm, vaddr, UPROBE_SWBP_INSN);
|
uprobes, mm, x86: Add the ability to install and remove uprobes breakpoints
Add uprobes support to the core kernel, with x86 support.
This commit adds the kernel facilities, the actual uprobes
user-space ABI and perf probe support comes in later commits.
General design:
Uprobes are maintained in an rb-tree indexed by inode and offset
(the offset here is from the start of the mapping). For a unique
(inode, offset) tuple, there can be at most one uprobe in the
rb-tree.
Since the (inode, offset) tuple identifies a unique uprobe, more
than one user may be interested in the same uprobe. This provides
the ability to connect multiple 'consumers' to the same uprobe.
Each consumer defines a handler and a filter (optional). The
'handler' is run every time the uprobe is hit, if it matches the
'filter' criteria.
The first consumer of a uprobe causes the breakpoint to be
inserted at the specified address and subsequent consumers are
appended to this list. On subsequent probes, the consumer gets
appended to the existing list of consumers. The breakpoint is
removed when the last consumer unregisters. For all other
unregisterations, the consumer is removed from the list of
consumers.
Given a inode, we get a list of the mms that have mapped the
inode. Do the actual registration if mm maps the page where a
probe needs to be inserted/removed.
We use a temporary list to walk through the vmas that map the
inode.
- The number of maps that map the inode, is not known before we
walk the rmap and keeps changing.
- extending vm_area_struct wasn't recommended, it's a
size-critical data structure.
- There can be more than one maps of the inode in the same mm.
We add callbacks to the mmap methods to keep an eye on text vmas
that are of interest to uprobes. When a vma of interest is mapped,
we insert the breakpoint at the right address.
Uprobe works by replacing the instruction at the address defined
by (inode, offset) with the arch specific breakpoint
instruction. We save a copy of the original instruction at the
uprobed address.
This is needed for:
a. executing the instruction out-of-line (xol).
b. instruction analysis for any subsequent fixups.
c. restoring the instruction back when the uprobe is unregistered.
We insert or delete a breakpoint instruction, and this
breakpoint instruction is assumed to be the smallest instruction
available on the platform. For fixed size instruction platforms
this is trivially true, for variable size instruction platforms
the breakpoint instruction is typically the smallest (often a
single byte).
Writing the instruction is done by COWing the page and changing
the instruction during the copy, this even though most platforms
allow atomic writes of the breakpoint instruction. This also
mirrors the behaviour of a ptrace() memory write to a PRIVATE
file map.
The core worker is derived from KSM's replace_page() logic.
In essence, similar to KSM:
a. allocate a new page and copy over contents of the page that
has the uprobed vaddr
b. modify the copy and insert the breakpoint at the required
address
c. switch the original page with the copy containing the
breakpoint
d. flush page tables.
replace_page() is being replicated here because of some minor
changes in the type of pages and also because Hugh Dickins had
plans to improve replace_page() for KSM specific work.
Instruction analysis on x86 is based on instruction decoder and
determines if an instruction can be probed and determines the
necessary fixups after singlestep. Instruction analysis is done
at probe insertion time so that we avoid having to repeat the
same analysis every time a probe is hit.
A lot of code here is due to the improvement/suggestions/inputs
from Peter Zijlstra.
Changelog:
(v10):
- Add code to clear REX.B prefix as suggested by Denys Vlasenko
and Masami Hiramatsu.
(v9):
- Use insn_offset_modrm as suggested by Masami Hiramatsu.
(v7):
Handle comments from Peter Zijlstra:
- Dont take reference to inode. (expect inode to uprobe_register to be sane).
- Use PTR_ERR to set the return value.
- No need to take reference to inode.
- use PTR_ERR to return error value.
- register and uprobe_unregister share code.
(v5):
- Modified del_consumer as per comments from Peter.
- Drop reference to inode before dropping reference to uprobe.
- Use i_size_read(inode) instead of inode->i_size.
- Ensure uprobe->consumers is NULL, before __uprobe_unregister() is called.
- Includes errno.h as recommended by Stephen Rothwell to fix a build issue
on sparc defconfig
- Remove restrictions while unregistering.
- Earlier code leaked inode references under some conditions while
registering/unregistering.
- Continue the vma-rmap walk even if the intermediate vma doesnt
meet the requirements.
- Validate the vma found by find_vma before inserting/removing the
breakpoint
- Call del_consumer under mutex_lock.
- Use hash locks.
- Handle mremap.
- Introduce find_least_offset_node() instead of close match logic in
find_uprobe
- Uprobes no more depends on MM_OWNER; No reference to task_structs
while inserting/removing a probe.
- Uses read_mapping_page instead of grab_cache_page so that the pages
have valid content.
- pass NULL to get_user_pages for the task parameter.
- call SetPageUptodate on the new page allocated in write_opcode.
- fix leaking a reference to the new page under certain conditions.
- Include Instruction Decoder if Uprobes gets defined.
- Remove const attributes for instruction prefix arrays.
- Uses mm_context to know if the application is 32 bit.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Also-written-by: Jim Keniston <jkenisto@us.ibm.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Roland McGrath <roland@hack.frob.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Anton Arapov <anton@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Denys Vlasenko <vda.linux@googlemail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linux-mm <linux-mm@kvack.org>
Link: http://lkml.kernel.org/r/20120209092642.GE16600@linux.vnet.ibm.com
[ Made various small edits to the commit log ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-02-09 17:26:42 +08:00
|
|
|
}
|
|
|
|
|
|
|
|
/**
|
|
|
|
* set_orig_insn - Restore the original instruction.
|
|
|
|
* @mm: the probed process address space.
|
2012-03-12 17:25:30 +08:00
|
|
|
* @auprobe: arch specific probepoint information.
|
uprobes, mm, x86: Add the ability to install and remove uprobes breakpoints
Add uprobes support to the core kernel, with x86 support.
This commit adds the kernel facilities, the actual uprobes
user-space ABI and perf probe support comes in later commits.
General design:
Uprobes are maintained in an rb-tree indexed by inode and offset
(the offset here is from the start of the mapping). For a unique
(inode, offset) tuple, there can be at most one uprobe in the
rb-tree.
Since the (inode, offset) tuple identifies a unique uprobe, more
than one user may be interested in the same uprobe. This provides
the ability to connect multiple 'consumers' to the same uprobe.
Each consumer defines a handler and a filter (optional). The
'handler' is run every time the uprobe is hit, if it matches the
'filter' criteria.
The first consumer of a uprobe causes the breakpoint to be
inserted at the specified address and subsequent consumers are
appended to this list. On subsequent probes, the consumer gets
appended to the existing list of consumers. The breakpoint is
removed when the last consumer unregisters. For all other
unregisterations, the consumer is removed from the list of
consumers.
Given a inode, we get a list of the mms that have mapped the
inode. Do the actual registration if mm maps the page where a
probe needs to be inserted/removed.
We use a temporary list to walk through the vmas that map the
inode.
- The number of maps that map the inode, is not known before we
walk the rmap and keeps changing.
- extending vm_area_struct wasn't recommended, it's a
size-critical data structure.
- There can be more than one maps of the inode in the same mm.
We add callbacks to the mmap methods to keep an eye on text vmas
that are of interest to uprobes. When a vma of interest is mapped,
we insert the breakpoint at the right address.
Uprobe works by replacing the instruction at the address defined
by (inode, offset) with the arch specific breakpoint
instruction. We save a copy of the original instruction at the
uprobed address.
This is needed for:
a. executing the instruction out-of-line (xol).
b. instruction analysis for any subsequent fixups.
c. restoring the instruction back when the uprobe is unregistered.
We insert or delete a breakpoint instruction, and this
breakpoint instruction is assumed to be the smallest instruction
available on the platform. For fixed size instruction platforms
this is trivially true, for variable size instruction platforms
the breakpoint instruction is typically the smallest (often a
single byte).
Writing the instruction is done by COWing the page and changing
the instruction during the copy, this even though most platforms
allow atomic writes of the breakpoint instruction. This also
mirrors the behaviour of a ptrace() memory write to a PRIVATE
file map.
The core worker is derived from KSM's replace_page() logic.
In essence, similar to KSM:
a. allocate a new page and copy over contents of the page that
has the uprobed vaddr
b. modify the copy and insert the breakpoint at the required
address
c. switch the original page with the copy containing the
breakpoint
d. flush page tables.
replace_page() is being replicated here because of some minor
changes in the type of pages and also because Hugh Dickins had
plans to improve replace_page() for KSM specific work.
Instruction analysis on x86 is based on instruction decoder and
determines if an instruction can be probed and determines the
necessary fixups after singlestep. Instruction analysis is done
at probe insertion time so that we avoid having to repeat the
same analysis every time a probe is hit.
A lot of code here is due to the improvement/suggestions/inputs
from Peter Zijlstra.
Changelog:
(v10):
- Add code to clear REX.B prefix as suggested by Denys Vlasenko
and Masami Hiramatsu.
(v9):
- Use insn_offset_modrm as suggested by Masami Hiramatsu.
(v7):
Handle comments from Peter Zijlstra:
- Dont take reference to inode. (expect inode to uprobe_register to be sane).
- Use PTR_ERR to set the return value.
- No need to take reference to inode.
- use PTR_ERR to return error value.
- register and uprobe_unregister share code.
(v5):
- Modified del_consumer as per comments from Peter.
- Drop reference to inode before dropping reference to uprobe.
- Use i_size_read(inode) instead of inode->i_size.
- Ensure uprobe->consumers is NULL, before __uprobe_unregister() is called.
- Includes errno.h as recommended by Stephen Rothwell to fix a build issue
on sparc defconfig
- Remove restrictions while unregistering.
- Earlier code leaked inode references under some conditions while
registering/unregistering.
- Continue the vma-rmap walk even if the intermediate vma doesnt
meet the requirements.
- Validate the vma found by find_vma before inserting/removing the
breakpoint
- Call del_consumer under mutex_lock.
- Use hash locks.
- Handle mremap.
- Introduce find_least_offset_node() instead of close match logic in
find_uprobe
- Uprobes no more depends on MM_OWNER; No reference to task_structs
while inserting/removing a probe.
- Uses read_mapping_page instead of grab_cache_page so that the pages
have valid content.
- pass NULL to get_user_pages for the task parameter.
- call SetPageUptodate on the new page allocated in write_opcode.
- fix leaking a reference to the new page under certain conditions.
- Include Instruction Decoder if Uprobes gets defined.
- Remove const attributes for instruction prefix arrays.
- Uses mm_context to know if the application is 32 bit.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Also-written-by: Jim Keniston <jkenisto@us.ibm.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Roland McGrath <roland@hack.frob.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Anton Arapov <anton@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Denys Vlasenko <vda.linux@googlemail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linux-mm <linux-mm@kvack.org>
Link: http://lkml.kernel.org/r/20120209092642.GE16600@linux.vnet.ibm.com
[ Made various small edits to the commit log ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-02-09 17:26:42 +08:00
|
|
|
* @vaddr: the virtual address to insert the opcode.
|
|
|
|
*
|
|
|
|
* For mm @mm, restore the original opcode (opcode) at @vaddr.
|
|
|
|
* Return 0 (success) or a negative errno.
|
|
|
|
*/
|
2012-02-17 16:27:41 +08:00
|
|
|
int __weak
|
2012-08-09 00:07:03 +08:00
|
|
|
set_orig_insn(struct arch_uprobe *auprobe, struct mm_struct *mm, unsigned long vaddr)
|
uprobes, mm, x86: Add the ability to install and remove uprobes breakpoints
Add uprobes support to the core kernel, with x86 support.
This commit adds the kernel facilities, the actual uprobes
user-space ABI and perf probe support comes in later commits.
General design:
Uprobes are maintained in an rb-tree indexed by inode and offset
(the offset here is from the start of the mapping). For a unique
(inode, offset) tuple, there can be at most one uprobe in the
rb-tree.
Since the (inode, offset) tuple identifies a unique uprobe, more
than one user may be interested in the same uprobe. This provides
the ability to connect multiple 'consumers' to the same uprobe.
Each consumer defines a handler and a filter (optional). The
'handler' is run every time the uprobe is hit, if it matches the
'filter' criteria.
The first consumer of a uprobe causes the breakpoint to be
inserted at the specified address and subsequent consumers are
appended to this list. On subsequent probes, the consumer gets
appended to the existing list of consumers. The breakpoint is
removed when the last consumer unregisters. For all other
unregisterations, the consumer is removed from the list of
consumers.
Given a inode, we get a list of the mms that have mapped the
inode. Do the actual registration if mm maps the page where a
probe needs to be inserted/removed.
We use a temporary list to walk through the vmas that map the
inode.
- The number of maps that map the inode, is not known before we
walk the rmap and keeps changing.
- extending vm_area_struct wasn't recommended, it's a
size-critical data structure.
- There can be more than one maps of the inode in the same mm.
We add callbacks to the mmap methods to keep an eye on text vmas
that are of interest to uprobes. When a vma of interest is mapped,
we insert the breakpoint at the right address.
Uprobe works by replacing the instruction at the address defined
by (inode, offset) with the arch specific breakpoint
instruction. We save a copy of the original instruction at the
uprobed address.
This is needed for:
a. executing the instruction out-of-line (xol).
b. instruction analysis for any subsequent fixups.
c. restoring the instruction back when the uprobe is unregistered.
We insert or delete a breakpoint instruction, and this
breakpoint instruction is assumed to be the smallest instruction
available on the platform. For fixed size instruction platforms
this is trivially true, for variable size instruction platforms
the breakpoint instruction is typically the smallest (often a
single byte).
Writing the instruction is done by COWing the page and changing
the instruction during the copy, this even though most platforms
allow atomic writes of the breakpoint instruction. This also
mirrors the behaviour of a ptrace() memory write to a PRIVATE
file map.
The core worker is derived from KSM's replace_page() logic.
In essence, similar to KSM:
a. allocate a new page and copy over contents of the page that
has the uprobed vaddr
b. modify the copy and insert the breakpoint at the required
address
c. switch the original page with the copy containing the
breakpoint
d. flush page tables.
replace_page() is being replicated here because of some minor
changes in the type of pages and also because Hugh Dickins had
plans to improve replace_page() for KSM specific work.
Instruction analysis on x86 is based on instruction decoder and
determines if an instruction can be probed and determines the
necessary fixups after singlestep. Instruction analysis is done
at probe insertion time so that we avoid having to repeat the
same analysis every time a probe is hit.
A lot of code here is due to the improvement/suggestions/inputs
from Peter Zijlstra.
Changelog:
(v10):
- Add code to clear REX.B prefix as suggested by Denys Vlasenko
and Masami Hiramatsu.
(v9):
- Use insn_offset_modrm as suggested by Masami Hiramatsu.
(v7):
Handle comments from Peter Zijlstra:
- Dont take reference to inode. (expect inode to uprobe_register to be sane).
- Use PTR_ERR to set the return value.
- No need to take reference to inode.
- use PTR_ERR to return error value.
- register and uprobe_unregister share code.
(v5):
- Modified del_consumer as per comments from Peter.
- Drop reference to inode before dropping reference to uprobe.
- Use i_size_read(inode) instead of inode->i_size.
- Ensure uprobe->consumers is NULL, before __uprobe_unregister() is called.
- Includes errno.h as recommended by Stephen Rothwell to fix a build issue
on sparc defconfig
- Remove restrictions while unregistering.
- Earlier code leaked inode references under some conditions while
registering/unregistering.
- Continue the vma-rmap walk even if the intermediate vma doesnt
meet the requirements.
- Validate the vma found by find_vma before inserting/removing the
breakpoint
- Call del_consumer under mutex_lock.
- Use hash locks.
- Handle mremap.
- Introduce find_least_offset_node() instead of close match logic in
find_uprobe
- Uprobes no more depends on MM_OWNER; No reference to task_structs
while inserting/removing a probe.
- Uses read_mapping_page instead of grab_cache_page so that the pages
have valid content.
- pass NULL to get_user_pages for the task parameter.
- call SetPageUptodate on the new page allocated in write_opcode.
- fix leaking a reference to the new page under certain conditions.
- Include Instruction Decoder if Uprobes gets defined.
- Remove const attributes for instruction prefix arrays.
- Uses mm_context to know if the application is 32 bit.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Also-written-by: Jim Keniston <jkenisto@us.ibm.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Roland McGrath <roland@hack.frob.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Anton Arapov <anton@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Denys Vlasenko <vda.linux@googlemail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linux-mm <linux-mm@kvack.org>
Link: http://lkml.kernel.org/r/20120209092642.GE16600@linux.vnet.ibm.com
[ Made various small edits to the commit log ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-02-09 17:26:42 +08:00
|
|
|
{
|
2018-08-09 12:18:52 +08:00
|
|
|
return uprobe_write_opcode(auprobe, mm, vaddr,
|
|
|
|
*(uprobe_opcode_t *)&auprobe->insn);
|
uprobes, mm, x86: Add the ability to install and remove uprobes breakpoints
Add uprobes support to the core kernel, with x86 support.
This commit adds the kernel facilities, the actual uprobes
user-space ABI and perf probe support comes in later commits.
General design:
Uprobes are maintained in an rb-tree indexed by inode and offset
(the offset here is from the start of the mapping). For a unique
(inode, offset) tuple, there can be at most one uprobe in the
rb-tree.
Since the (inode, offset) tuple identifies a unique uprobe, more
than one user may be interested in the same uprobe. This provides
the ability to connect multiple 'consumers' to the same uprobe.
Each consumer defines a handler and a filter (optional). The
'handler' is run every time the uprobe is hit, if it matches the
'filter' criteria.
The first consumer of a uprobe causes the breakpoint to be
inserted at the specified address and subsequent consumers are
appended to this list. On subsequent probes, the consumer gets
appended to the existing list of consumers. The breakpoint is
removed when the last consumer unregisters. For all other
unregisterations, the consumer is removed from the list of
consumers.
Given a inode, we get a list of the mms that have mapped the
inode. Do the actual registration if mm maps the page where a
probe needs to be inserted/removed.
We use a temporary list to walk through the vmas that map the
inode.
- The number of maps that map the inode, is not known before we
walk the rmap and keeps changing.
- extending vm_area_struct wasn't recommended, it's a
size-critical data structure.
- There can be more than one maps of the inode in the same mm.
We add callbacks to the mmap methods to keep an eye on text vmas
that are of interest to uprobes. When a vma of interest is mapped,
we insert the breakpoint at the right address.
Uprobe works by replacing the instruction at the address defined
by (inode, offset) with the arch specific breakpoint
instruction. We save a copy of the original instruction at the
uprobed address.
This is needed for:
a. executing the instruction out-of-line (xol).
b. instruction analysis for any subsequent fixups.
c. restoring the instruction back when the uprobe is unregistered.
We insert or delete a breakpoint instruction, and this
breakpoint instruction is assumed to be the smallest instruction
available on the platform. For fixed size instruction platforms
this is trivially true, for variable size instruction platforms
the breakpoint instruction is typically the smallest (often a
single byte).
Writing the instruction is done by COWing the page and changing
the instruction during the copy, this even though most platforms
allow atomic writes of the breakpoint instruction. This also
mirrors the behaviour of a ptrace() memory write to a PRIVATE
file map.
The core worker is derived from KSM's replace_page() logic.
In essence, similar to KSM:
a. allocate a new page and copy over contents of the page that
has the uprobed vaddr
b. modify the copy and insert the breakpoint at the required
address
c. switch the original page with the copy containing the
breakpoint
d. flush page tables.
replace_page() is being replicated here because of some minor
changes in the type of pages and also because Hugh Dickins had
plans to improve replace_page() for KSM specific work.
Instruction analysis on x86 is based on instruction decoder and
determines if an instruction can be probed and determines the
necessary fixups after singlestep. Instruction analysis is done
at probe insertion time so that we avoid having to repeat the
same analysis every time a probe is hit.
A lot of code here is due to the improvement/suggestions/inputs
from Peter Zijlstra.
Changelog:
(v10):
- Add code to clear REX.B prefix as suggested by Denys Vlasenko
and Masami Hiramatsu.
(v9):
- Use insn_offset_modrm as suggested by Masami Hiramatsu.
(v7):
Handle comments from Peter Zijlstra:
- Dont take reference to inode. (expect inode to uprobe_register to be sane).
- Use PTR_ERR to set the return value.
- No need to take reference to inode.
- use PTR_ERR to return error value.
- register and uprobe_unregister share code.
(v5):
- Modified del_consumer as per comments from Peter.
- Drop reference to inode before dropping reference to uprobe.
- Use i_size_read(inode) instead of inode->i_size.
- Ensure uprobe->consumers is NULL, before __uprobe_unregister() is called.
- Includes errno.h as recommended by Stephen Rothwell to fix a build issue
on sparc defconfig
- Remove restrictions while unregistering.
- Earlier code leaked inode references under some conditions while
registering/unregistering.
- Continue the vma-rmap walk even if the intermediate vma doesnt
meet the requirements.
- Validate the vma found by find_vma before inserting/removing the
breakpoint
- Call del_consumer under mutex_lock.
- Use hash locks.
- Handle mremap.
- Introduce find_least_offset_node() instead of close match logic in
find_uprobe
- Uprobes no more depends on MM_OWNER; No reference to task_structs
while inserting/removing a probe.
- Uses read_mapping_page instead of grab_cache_page so that the pages
have valid content.
- pass NULL to get_user_pages for the task parameter.
- call SetPageUptodate on the new page allocated in write_opcode.
- fix leaking a reference to the new page under certain conditions.
- Include Instruction Decoder if Uprobes gets defined.
- Remove const attributes for instruction prefix arrays.
- Uses mm_context to know if the application is 32 bit.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Also-written-by: Jim Keniston <jkenisto@us.ibm.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Roland McGrath <roland@hack.frob.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Anton Arapov <anton@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Denys Vlasenko <vda.linux@googlemail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linux-mm <linux-mm@kvack.org>
Link: http://lkml.kernel.org/r/20120209092642.GE16600@linux.vnet.ibm.com
[ Made various small edits to the commit log ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-02-09 17:26:42 +08:00
|
|
|
}
|
|
|
|
|
2015-07-21 21:40:03 +08:00
|
|
|
static struct uprobe *get_uprobe(struct uprobe *uprobe)
|
|
|
|
{
|
|
|
|
atomic_inc(&uprobe->ref);
|
|
|
|
return uprobe;
|
|
|
|
}
|
|
|
|
|
|
|
|
static void put_uprobe(struct uprobe *uprobe)
|
|
|
|
{
|
uprobes: Support SDT markers having reference count (semaphore)
Userspace Statically Defined Tracepoints[1] are dtrace style markers
inside userspace applications. Applications like PostgreSQL, MySQL,
Pthread, Perl, Python, Java, Ruby, Node.js, libvirt, QEMU, glib etc
have these markers embedded in them. These markers are added by developer
at important places in the code. Each marker source expands to a single
nop instruction in the compiled code but there may be additional
overhead for computing the marker arguments which expands to couple of
instructions. In case the overhead is more, execution of it can be
omitted by runtime if() condition when no one is tracing on the marker:
if (reference_counter > 0) {
Execute marker instructions;
}
Default value of reference counter is 0. Tracer has to increment the
reference counter before tracing on a marker and decrement it when
done with the tracing.
Implement the reference counter logic in core uprobe. User will be
able to use it from trace_uprobe as well as from kernel module. New
trace_uprobe definition with reference counter will now be:
<path>:<offset>[(ref_ctr_offset)]
where ref_ctr_offset is an optional field. For kernel module, new
variant of uprobe_register() has been introduced:
uprobe_register_refctr(inode, offset, ref_ctr_offset, consumer)
No new variant for uprobe_unregister() because it's assumed to have
only one reference counter for one uprobe.
[1] https://sourceware.org/systemtap/wiki/UserSpaceProbeImplementation
Note: 'reference counter' is called as 'semaphore' in original Dtrace
(or Systemtap, bcc and even in ELF) documentation and code. But the
term 'semaphore' is misleading in this context. This is just a counter
used to hold number of tracers tracing on a marker. This is not really
used for any synchronization. So we are calling it a 'reference counter'
in kernel / perf code.
Link: http://lkml.kernel.org/r/20180820044250.11659-2-ravi.bangoria@linux.ibm.com
Reviewed-by: Masami Hiramatsu <mhiramat@kernel.org>
[Only trace_uprobe.c]
Reviewed-by: Oleg Nesterov <oleg@redhat.com>
Reviewed-by: Song Liu <songliubraving@fb.com>
Tested-by: Song Liu <songliubraving@fb.com>
Signed-off-by: Ravi Bangoria <ravi.bangoria@linux.ibm.com>
Signed-off-by: Steven Rostedt (VMware) <rostedt@goodmis.org>
2018-08-20 12:42:47 +08:00
|
|
|
if (atomic_dec_and_test(&uprobe->ref)) {
|
|
|
|
/*
|
|
|
|
* If application munmap(exec_vma) before uprobe_unregister()
|
|
|
|
* gets called, we don't get a chance to remove uprobe from
|
|
|
|
* delayed_uprobe_list from remove_breakpoint(). Do it here.
|
|
|
|
*/
|
2018-12-05 11:34:23 +08:00
|
|
|
mutex_lock(&delayed_uprobe_lock);
|
uprobes: Support SDT markers having reference count (semaphore)
Userspace Statically Defined Tracepoints[1] are dtrace style markers
inside userspace applications. Applications like PostgreSQL, MySQL,
Pthread, Perl, Python, Java, Ruby, Node.js, libvirt, QEMU, glib etc
have these markers embedded in them. These markers are added by developer
at important places in the code. Each marker source expands to a single
nop instruction in the compiled code but there may be additional
overhead for computing the marker arguments which expands to couple of
instructions. In case the overhead is more, execution of it can be
omitted by runtime if() condition when no one is tracing on the marker:
if (reference_counter > 0) {
Execute marker instructions;
}
Default value of reference counter is 0. Tracer has to increment the
reference counter before tracing on a marker and decrement it when
done with the tracing.
Implement the reference counter logic in core uprobe. User will be
able to use it from trace_uprobe as well as from kernel module. New
trace_uprobe definition with reference counter will now be:
<path>:<offset>[(ref_ctr_offset)]
where ref_ctr_offset is an optional field. For kernel module, new
variant of uprobe_register() has been introduced:
uprobe_register_refctr(inode, offset, ref_ctr_offset, consumer)
No new variant for uprobe_unregister() because it's assumed to have
only one reference counter for one uprobe.
[1] https://sourceware.org/systemtap/wiki/UserSpaceProbeImplementation
Note: 'reference counter' is called as 'semaphore' in original Dtrace
(or Systemtap, bcc and even in ELF) documentation and code. But the
term 'semaphore' is misleading in this context. This is just a counter
used to hold number of tracers tracing on a marker. This is not really
used for any synchronization. So we are calling it a 'reference counter'
in kernel / perf code.
Link: http://lkml.kernel.org/r/20180820044250.11659-2-ravi.bangoria@linux.ibm.com
Reviewed-by: Masami Hiramatsu <mhiramat@kernel.org>
[Only trace_uprobe.c]
Reviewed-by: Oleg Nesterov <oleg@redhat.com>
Reviewed-by: Song Liu <songliubraving@fb.com>
Tested-by: Song Liu <songliubraving@fb.com>
Signed-off-by: Ravi Bangoria <ravi.bangoria@linux.ibm.com>
Signed-off-by: Steven Rostedt (VMware) <rostedt@goodmis.org>
2018-08-20 12:42:47 +08:00
|
|
|
delayed_uprobe_remove(uprobe, NULL);
|
2018-12-05 11:34:23 +08:00
|
|
|
mutex_unlock(&delayed_uprobe_lock);
|
2015-07-21 21:40:03 +08:00
|
|
|
kfree(uprobe);
|
uprobes: Support SDT markers having reference count (semaphore)
Userspace Statically Defined Tracepoints[1] are dtrace style markers
inside userspace applications. Applications like PostgreSQL, MySQL,
Pthread, Perl, Python, Java, Ruby, Node.js, libvirt, QEMU, glib etc
have these markers embedded in them. These markers are added by developer
at important places in the code. Each marker source expands to a single
nop instruction in the compiled code but there may be additional
overhead for computing the marker arguments which expands to couple of
instructions. In case the overhead is more, execution of it can be
omitted by runtime if() condition when no one is tracing on the marker:
if (reference_counter > 0) {
Execute marker instructions;
}
Default value of reference counter is 0. Tracer has to increment the
reference counter before tracing on a marker and decrement it when
done with the tracing.
Implement the reference counter logic in core uprobe. User will be
able to use it from trace_uprobe as well as from kernel module. New
trace_uprobe definition with reference counter will now be:
<path>:<offset>[(ref_ctr_offset)]
where ref_ctr_offset is an optional field. For kernel module, new
variant of uprobe_register() has been introduced:
uprobe_register_refctr(inode, offset, ref_ctr_offset, consumer)
No new variant for uprobe_unregister() because it's assumed to have
only one reference counter for one uprobe.
[1] https://sourceware.org/systemtap/wiki/UserSpaceProbeImplementation
Note: 'reference counter' is called as 'semaphore' in original Dtrace
(or Systemtap, bcc and even in ELF) documentation and code. But the
term 'semaphore' is misleading in this context. This is just a counter
used to hold number of tracers tracing on a marker. This is not really
used for any synchronization. So we are calling it a 'reference counter'
in kernel / perf code.
Link: http://lkml.kernel.org/r/20180820044250.11659-2-ravi.bangoria@linux.ibm.com
Reviewed-by: Masami Hiramatsu <mhiramat@kernel.org>
[Only trace_uprobe.c]
Reviewed-by: Oleg Nesterov <oleg@redhat.com>
Reviewed-by: Song Liu <songliubraving@fb.com>
Tested-by: Song Liu <songliubraving@fb.com>
Signed-off-by: Ravi Bangoria <ravi.bangoria@linux.ibm.com>
Signed-off-by: Steven Rostedt (VMware) <rostedt@goodmis.org>
2018-08-20 12:42:47 +08:00
|
|
|
}
|
2015-07-21 21:40:03 +08:00
|
|
|
}
|
|
|
|
|
uprobes, mm, x86: Add the ability to install and remove uprobes breakpoints
Add uprobes support to the core kernel, with x86 support.
This commit adds the kernel facilities, the actual uprobes
user-space ABI and perf probe support comes in later commits.
General design:
Uprobes are maintained in an rb-tree indexed by inode and offset
(the offset here is from the start of the mapping). For a unique
(inode, offset) tuple, there can be at most one uprobe in the
rb-tree.
Since the (inode, offset) tuple identifies a unique uprobe, more
than one user may be interested in the same uprobe. This provides
the ability to connect multiple 'consumers' to the same uprobe.
Each consumer defines a handler and a filter (optional). The
'handler' is run every time the uprobe is hit, if it matches the
'filter' criteria.
The first consumer of a uprobe causes the breakpoint to be
inserted at the specified address and subsequent consumers are
appended to this list. On subsequent probes, the consumer gets
appended to the existing list of consumers. The breakpoint is
removed when the last consumer unregisters. For all other
unregisterations, the consumer is removed from the list of
consumers.
Given a inode, we get a list of the mms that have mapped the
inode. Do the actual registration if mm maps the page where a
probe needs to be inserted/removed.
We use a temporary list to walk through the vmas that map the
inode.
- The number of maps that map the inode, is not known before we
walk the rmap and keeps changing.
- extending vm_area_struct wasn't recommended, it's a
size-critical data structure.
- There can be more than one maps of the inode in the same mm.
We add callbacks to the mmap methods to keep an eye on text vmas
that are of interest to uprobes. When a vma of interest is mapped,
we insert the breakpoint at the right address.
Uprobe works by replacing the instruction at the address defined
by (inode, offset) with the arch specific breakpoint
instruction. We save a copy of the original instruction at the
uprobed address.
This is needed for:
a. executing the instruction out-of-line (xol).
b. instruction analysis for any subsequent fixups.
c. restoring the instruction back when the uprobe is unregistered.
We insert or delete a breakpoint instruction, and this
breakpoint instruction is assumed to be the smallest instruction
available on the platform. For fixed size instruction platforms
this is trivially true, for variable size instruction platforms
the breakpoint instruction is typically the smallest (often a
single byte).
Writing the instruction is done by COWing the page and changing
the instruction during the copy, this even though most platforms
allow atomic writes of the breakpoint instruction. This also
mirrors the behaviour of a ptrace() memory write to a PRIVATE
file map.
The core worker is derived from KSM's replace_page() logic.
In essence, similar to KSM:
a. allocate a new page and copy over contents of the page that
has the uprobed vaddr
b. modify the copy and insert the breakpoint at the required
address
c. switch the original page with the copy containing the
breakpoint
d. flush page tables.
replace_page() is being replicated here because of some minor
changes in the type of pages and also because Hugh Dickins had
plans to improve replace_page() for KSM specific work.
Instruction analysis on x86 is based on instruction decoder and
determines if an instruction can be probed and determines the
necessary fixups after singlestep. Instruction analysis is done
at probe insertion time so that we avoid having to repeat the
same analysis every time a probe is hit.
A lot of code here is due to the improvement/suggestions/inputs
from Peter Zijlstra.
Changelog:
(v10):
- Add code to clear REX.B prefix as suggested by Denys Vlasenko
and Masami Hiramatsu.
(v9):
- Use insn_offset_modrm as suggested by Masami Hiramatsu.
(v7):
Handle comments from Peter Zijlstra:
- Dont take reference to inode. (expect inode to uprobe_register to be sane).
- Use PTR_ERR to set the return value.
- No need to take reference to inode.
- use PTR_ERR to return error value.
- register and uprobe_unregister share code.
(v5):
- Modified del_consumer as per comments from Peter.
- Drop reference to inode before dropping reference to uprobe.
- Use i_size_read(inode) instead of inode->i_size.
- Ensure uprobe->consumers is NULL, before __uprobe_unregister() is called.
- Includes errno.h as recommended by Stephen Rothwell to fix a build issue
on sparc defconfig
- Remove restrictions while unregistering.
- Earlier code leaked inode references under some conditions while
registering/unregistering.
- Continue the vma-rmap walk even if the intermediate vma doesnt
meet the requirements.
- Validate the vma found by find_vma before inserting/removing the
breakpoint
- Call del_consumer under mutex_lock.
- Use hash locks.
- Handle mremap.
- Introduce find_least_offset_node() instead of close match logic in
find_uprobe
- Uprobes no more depends on MM_OWNER; No reference to task_structs
while inserting/removing a probe.
- Uses read_mapping_page instead of grab_cache_page so that the pages
have valid content.
- pass NULL to get_user_pages for the task parameter.
- call SetPageUptodate on the new page allocated in write_opcode.
- fix leaking a reference to the new page under certain conditions.
- Include Instruction Decoder if Uprobes gets defined.
- Remove const attributes for instruction prefix arrays.
- Uses mm_context to know if the application is 32 bit.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Also-written-by: Jim Keniston <jkenisto@us.ibm.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Roland McGrath <roland@hack.frob.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Anton Arapov <anton@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Denys Vlasenko <vda.linux@googlemail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linux-mm <linux-mm@kvack.org>
Link: http://lkml.kernel.org/r/20120209092642.GE16600@linux.vnet.ibm.com
[ Made various small edits to the commit log ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-02-09 17:26:42 +08:00
|
|
|
static int match_uprobe(struct uprobe *l, struct uprobe *r)
|
|
|
|
{
|
|
|
|
if (l->inode < r->inode)
|
|
|
|
return -1;
|
2012-02-17 16:27:41 +08:00
|
|
|
|
uprobes, mm, x86: Add the ability to install and remove uprobes breakpoints
Add uprobes support to the core kernel, with x86 support.
This commit adds the kernel facilities, the actual uprobes
user-space ABI and perf probe support comes in later commits.
General design:
Uprobes are maintained in an rb-tree indexed by inode and offset
(the offset here is from the start of the mapping). For a unique
(inode, offset) tuple, there can be at most one uprobe in the
rb-tree.
Since the (inode, offset) tuple identifies a unique uprobe, more
than one user may be interested in the same uprobe. This provides
the ability to connect multiple 'consumers' to the same uprobe.
Each consumer defines a handler and a filter (optional). The
'handler' is run every time the uprobe is hit, if it matches the
'filter' criteria.
The first consumer of a uprobe causes the breakpoint to be
inserted at the specified address and subsequent consumers are
appended to this list. On subsequent probes, the consumer gets
appended to the existing list of consumers. The breakpoint is
removed when the last consumer unregisters. For all other
unregisterations, the consumer is removed from the list of
consumers.
Given a inode, we get a list of the mms that have mapped the
inode. Do the actual registration if mm maps the page where a
probe needs to be inserted/removed.
We use a temporary list to walk through the vmas that map the
inode.
- The number of maps that map the inode, is not known before we
walk the rmap and keeps changing.
- extending vm_area_struct wasn't recommended, it's a
size-critical data structure.
- There can be more than one maps of the inode in the same mm.
We add callbacks to the mmap methods to keep an eye on text vmas
that are of interest to uprobes. When a vma of interest is mapped,
we insert the breakpoint at the right address.
Uprobe works by replacing the instruction at the address defined
by (inode, offset) with the arch specific breakpoint
instruction. We save a copy of the original instruction at the
uprobed address.
This is needed for:
a. executing the instruction out-of-line (xol).
b. instruction analysis for any subsequent fixups.
c. restoring the instruction back when the uprobe is unregistered.
We insert or delete a breakpoint instruction, and this
breakpoint instruction is assumed to be the smallest instruction
available on the platform. For fixed size instruction platforms
this is trivially true, for variable size instruction platforms
the breakpoint instruction is typically the smallest (often a
single byte).
Writing the instruction is done by COWing the page and changing
the instruction during the copy, this even though most platforms
allow atomic writes of the breakpoint instruction. This also
mirrors the behaviour of a ptrace() memory write to a PRIVATE
file map.
The core worker is derived from KSM's replace_page() logic.
In essence, similar to KSM:
a. allocate a new page and copy over contents of the page that
has the uprobed vaddr
b. modify the copy and insert the breakpoint at the required
address
c. switch the original page with the copy containing the
breakpoint
d. flush page tables.
replace_page() is being replicated here because of some minor
changes in the type of pages and also because Hugh Dickins had
plans to improve replace_page() for KSM specific work.
Instruction analysis on x86 is based on instruction decoder and
determines if an instruction can be probed and determines the
necessary fixups after singlestep. Instruction analysis is done
at probe insertion time so that we avoid having to repeat the
same analysis every time a probe is hit.
A lot of code here is due to the improvement/suggestions/inputs
from Peter Zijlstra.
Changelog:
(v10):
- Add code to clear REX.B prefix as suggested by Denys Vlasenko
and Masami Hiramatsu.
(v9):
- Use insn_offset_modrm as suggested by Masami Hiramatsu.
(v7):
Handle comments from Peter Zijlstra:
- Dont take reference to inode. (expect inode to uprobe_register to be sane).
- Use PTR_ERR to set the return value.
- No need to take reference to inode.
- use PTR_ERR to return error value.
- register and uprobe_unregister share code.
(v5):
- Modified del_consumer as per comments from Peter.
- Drop reference to inode before dropping reference to uprobe.
- Use i_size_read(inode) instead of inode->i_size.
- Ensure uprobe->consumers is NULL, before __uprobe_unregister() is called.
- Includes errno.h as recommended by Stephen Rothwell to fix a build issue
on sparc defconfig
- Remove restrictions while unregistering.
- Earlier code leaked inode references under some conditions while
registering/unregistering.
- Continue the vma-rmap walk even if the intermediate vma doesnt
meet the requirements.
- Validate the vma found by find_vma before inserting/removing the
breakpoint
- Call del_consumer under mutex_lock.
- Use hash locks.
- Handle mremap.
- Introduce find_least_offset_node() instead of close match logic in
find_uprobe
- Uprobes no more depends on MM_OWNER; No reference to task_structs
while inserting/removing a probe.
- Uses read_mapping_page instead of grab_cache_page so that the pages
have valid content.
- pass NULL to get_user_pages for the task parameter.
- call SetPageUptodate on the new page allocated in write_opcode.
- fix leaking a reference to the new page under certain conditions.
- Include Instruction Decoder if Uprobes gets defined.
- Remove const attributes for instruction prefix arrays.
- Uses mm_context to know if the application is 32 bit.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Also-written-by: Jim Keniston <jkenisto@us.ibm.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Roland McGrath <roland@hack.frob.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Anton Arapov <anton@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Denys Vlasenko <vda.linux@googlemail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linux-mm <linux-mm@kvack.org>
Link: http://lkml.kernel.org/r/20120209092642.GE16600@linux.vnet.ibm.com
[ Made various small edits to the commit log ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-02-09 17:26:42 +08:00
|
|
|
if (l->inode > r->inode)
|
|
|
|
return 1;
|
|
|
|
|
2012-02-17 16:27:41 +08:00
|
|
|
if (l->offset < r->offset)
|
|
|
|
return -1;
|
|
|
|
|
|
|
|
if (l->offset > r->offset)
|
|
|
|
return 1;
|
uprobes, mm, x86: Add the ability to install and remove uprobes breakpoints
Add uprobes support to the core kernel, with x86 support.
This commit adds the kernel facilities, the actual uprobes
user-space ABI and perf probe support comes in later commits.
General design:
Uprobes are maintained in an rb-tree indexed by inode and offset
(the offset here is from the start of the mapping). For a unique
(inode, offset) tuple, there can be at most one uprobe in the
rb-tree.
Since the (inode, offset) tuple identifies a unique uprobe, more
than one user may be interested in the same uprobe. This provides
the ability to connect multiple 'consumers' to the same uprobe.
Each consumer defines a handler and a filter (optional). The
'handler' is run every time the uprobe is hit, if it matches the
'filter' criteria.
The first consumer of a uprobe causes the breakpoint to be
inserted at the specified address and subsequent consumers are
appended to this list. On subsequent probes, the consumer gets
appended to the existing list of consumers. The breakpoint is
removed when the last consumer unregisters. For all other
unregisterations, the consumer is removed from the list of
consumers.
Given a inode, we get a list of the mms that have mapped the
inode. Do the actual registration if mm maps the page where a
probe needs to be inserted/removed.
We use a temporary list to walk through the vmas that map the
inode.
- The number of maps that map the inode, is not known before we
walk the rmap and keeps changing.
- extending vm_area_struct wasn't recommended, it's a
size-critical data structure.
- There can be more than one maps of the inode in the same mm.
We add callbacks to the mmap methods to keep an eye on text vmas
that are of interest to uprobes. When a vma of interest is mapped,
we insert the breakpoint at the right address.
Uprobe works by replacing the instruction at the address defined
by (inode, offset) with the arch specific breakpoint
instruction. We save a copy of the original instruction at the
uprobed address.
This is needed for:
a. executing the instruction out-of-line (xol).
b. instruction analysis for any subsequent fixups.
c. restoring the instruction back when the uprobe is unregistered.
We insert or delete a breakpoint instruction, and this
breakpoint instruction is assumed to be the smallest instruction
available on the platform. For fixed size instruction platforms
this is trivially true, for variable size instruction platforms
the breakpoint instruction is typically the smallest (often a
single byte).
Writing the instruction is done by COWing the page and changing
the instruction during the copy, this even though most platforms
allow atomic writes of the breakpoint instruction. This also
mirrors the behaviour of a ptrace() memory write to a PRIVATE
file map.
The core worker is derived from KSM's replace_page() logic.
In essence, similar to KSM:
a. allocate a new page and copy over contents of the page that
has the uprobed vaddr
b. modify the copy and insert the breakpoint at the required
address
c. switch the original page with the copy containing the
breakpoint
d. flush page tables.
replace_page() is being replicated here because of some minor
changes in the type of pages and also because Hugh Dickins had
plans to improve replace_page() for KSM specific work.
Instruction analysis on x86 is based on instruction decoder and
determines if an instruction can be probed and determines the
necessary fixups after singlestep. Instruction analysis is done
at probe insertion time so that we avoid having to repeat the
same analysis every time a probe is hit.
A lot of code here is due to the improvement/suggestions/inputs
from Peter Zijlstra.
Changelog:
(v10):
- Add code to clear REX.B prefix as suggested by Denys Vlasenko
and Masami Hiramatsu.
(v9):
- Use insn_offset_modrm as suggested by Masami Hiramatsu.
(v7):
Handle comments from Peter Zijlstra:
- Dont take reference to inode. (expect inode to uprobe_register to be sane).
- Use PTR_ERR to set the return value.
- No need to take reference to inode.
- use PTR_ERR to return error value.
- register and uprobe_unregister share code.
(v5):
- Modified del_consumer as per comments from Peter.
- Drop reference to inode before dropping reference to uprobe.
- Use i_size_read(inode) instead of inode->i_size.
- Ensure uprobe->consumers is NULL, before __uprobe_unregister() is called.
- Includes errno.h as recommended by Stephen Rothwell to fix a build issue
on sparc defconfig
- Remove restrictions while unregistering.
- Earlier code leaked inode references under some conditions while
registering/unregistering.
- Continue the vma-rmap walk even if the intermediate vma doesnt
meet the requirements.
- Validate the vma found by find_vma before inserting/removing the
breakpoint
- Call del_consumer under mutex_lock.
- Use hash locks.
- Handle mremap.
- Introduce find_least_offset_node() instead of close match logic in
find_uprobe
- Uprobes no more depends on MM_OWNER; No reference to task_structs
while inserting/removing a probe.
- Uses read_mapping_page instead of grab_cache_page so that the pages
have valid content.
- pass NULL to get_user_pages for the task parameter.
- call SetPageUptodate on the new page allocated in write_opcode.
- fix leaking a reference to the new page under certain conditions.
- Include Instruction Decoder if Uprobes gets defined.
- Remove const attributes for instruction prefix arrays.
- Uses mm_context to know if the application is 32 bit.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Also-written-by: Jim Keniston <jkenisto@us.ibm.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Roland McGrath <roland@hack.frob.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Anton Arapov <anton@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Denys Vlasenko <vda.linux@googlemail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linux-mm <linux-mm@kvack.org>
Link: http://lkml.kernel.org/r/20120209092642.GE16600@linux.vnet.ibm.com
[ Made various small edits to the commit log ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-02-09 17:26:42 +08:00
|
|
|
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
static struct uprobe *__find_uprobe(struct inode *inode, loff_t offset)
|
|
|
|
{
|
|
|
|
struct uprobe u = { .inode = inode, .offset = offset };
|
|
|
|
struct rb_node *n = uprobes_tree.rb_node;
|
|
|
|
struct uprobe *uprobe;
|
|
|
|
int match;
|
|
|
|
|
|
|
|
while (n) {
|
|
|
|
uprobe = rb_entry(n, struct uprobe, rb_node);
|
|
|
|
match = match_uprobe(&u, uprobe);
|
2015-07-21 21:40:03 +08:00
|
|
|
if (!match)
|
|
|
|
return get_uprobe(uprobe);
|
2012-02-17 16:27:41 +08:00
|
|
|
|
uprobes, mm, x86: Add the ability to install and remove uprobes breakpoints
Add uprobes support to the core kernel, with x86 support.
This commit adds the kernel facilities, the actual uprobes
user-space ABI and perf probe support comes in later commits.
General design:
Uprobes are maintained in an rb-tree indexed by inode and offset
(the offset here is from the start of the mapping). For a unique
(inode, offset) tuple, there can be at most one uprobe in the
rb-tree.
Since the (inode, offset) tuple identifies a unique uprobe, more
than one user may be interested in the same uprobe. This provides
the ability to connect multiple 'consumers' to the same uprobe.
Each consumer defines a handler and a filter (optional). The
'handler' is run every time the uprobe is hit, if it matches the
'filter' criteria.
The first consumer of a uprobe causes the breakpoint to be
inserted at the specified address and subsequent consumers are
appended to this list. On subsequent probes, the consumer gets
appended to the existing list of consumers. The breakpoint is
removed when the last consumer unregisters. For all other
unregisterations, the consumer is removed from the list of
consumers.
Given a inode, we get a list of the mms that have mapped the
inode. Do the actual registration if mm maps the page where a
probe needs to be inserted/removed.
We use a temporary list to walk through the vmas that map the
inode.
- The number of maps that map the inode, is not known before we
walk the rmap and keeps changing.
- extending vm_area_struct wasn't recommended, it's a
size-critical data structure.
- There can be more than one maps of the inode in the same mm.
We add callbacks to the mmap methods to keep an eye on text vmas
that are of interest to uprobes. When a vma of interest is mapped,
we insert the breakpoint at the right address.
Uprobe works by replacing the instruction at the address defined
by (inode, offset) with the arch specific breakpoint
instruction. We save a copy of the original instruction at the
uprobed address.
This is needed for:
a. executing the instruction out-of-line (xol).
b. instruction analysis for any subsequent fixups.
c. restoring the instruction back when the uprobe is unregistered.
We insert or delete a breakpoint instruction, and this
breakpoint instruction is assumed to be the smallest instruction
available on the platform. For fixed size instruction platforms
this is trivially true, for variable size instruction platforms
the breakpoint instruction is typically the smallest (often a
single byte).
Writing the instruction is done by COWing the page and changing
the instruction during the copy, this even though most platforms
allow atomic writes of the breakpoint instruction. This also
mirrors the behaviour of a ptrace() memory write to a PRIVATE
file map.
The core worker is derived from KSM's replace_page() logic.
In essence, similar to KSM:
a. allocate a new page and copy over contents of the page that
has the uprobed vaddr
b. modify the copy and insert the breakpoint at the required
address
c. switch the original page with the copy containing the
breakpoint
d. flush page tables.
replace_page() is being replicated here because of some minor
changes in the type of pages and also because Hugh Dickins had
plans to improve replace_page() for KSM specific work.
Instruction analysis on x86 is based on instruction decoder and
determines if an instruction can be probed and determines the
necessary fixups after singlestep. Instruction analysis is done
at probe insertion time so that we avoid having to repeat the
same analysis every time a probe is hit.
A lot of code here is due to the improvement/suggestions/inputs
from Peter Zijlstra.
Changelog:
(v10):
- Add code to clear REX.B prefix as suggested by Denys Vlasenko
and Masami Hiramatsu.
(v9):
- Use insn_offset_modrm as suggested by Masami Hiramatsu.
(v7):
Handle comments from Peter Zijlstra:
- Dont take reference to inode. (expect inode to uprobe_register to be sane).
- Use PTR_ERR to set the return value.
- No need to take reference to inode.
- use PTR_ERR to return error value.
- register and uprobe_unregister share code.
(v5):
- Modified del_consumer as per comments from Peter.
- Drop reference to inode before dropping reference to uprobe.
- Use i_size_read(inode) instead of inode->i_size.
- Ensure uprobe->consumers is NULL, before __uprobe_unregister() is called.
- Includes errno.h as recommended by Stephen Rothwell to fix a build issue
on sparc defconfig
- Remove restrictions while unregistering.
- Earlier code leaked inode references under some conditions while
registering/unregistering.
- Continue the vma-rmap walk even if the intermediate vma doesnt
meet the requirements.
- Validate the vma found by find_vma before inserting/removing the
breakpoint
- Call del_consumer under mutex_lock.
- Use hash locks.
- Handle mremap.
- Introduce find_least_offset_node() instead of close match logic in
find_uprobe
- Uprobes no more depends on MM_OWNER; No reference to task_structs
while inserting/removing a probe.
- Uses read_mapping_page instead of grab_cache_page so that the pages
have valid content.
- pass NULL to get_user_pages for the task parameter.
- call SetPageUptodate on the new page allocated in write_opcode.
- fix leaking a reference to the new page under certain conditions.
- Include Instruction Decoder if Uprobes gets defined.
- Remove const attributes for instruction prefix arrays.
- Uses mm_context to know if the application is 32 bit.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Also-written-by: Jim Keniston <jkenisto@us.ibm.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Roland McGrath <roland@hack.frob.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Anton Arapov <anton@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Denys Vlasenko <vda.linux@googlemail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linux-mm <linux-mm@kvack.org>
Link: http://lkml.kernel.org/r/20120209092642.GE16600@linux.vnet.ibm.com
[ Made various small edits to the commit log ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-02-09 17:26:42 +08:00
|
|
|
if (match < 0)
|
|
|
|
n = n->rb_left;
|
|
|
|
else
|
|
|
|
n = n->rb_right;
|
|
|
|
}
|
|
|
|
return NULL;
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Find a uprobe corresponding to a given inode:offset
|
|
|
|
* Acquires uprobes_treelock
|
|
|
|
*/
|
|
|
|
static struct uprobe *find_uprobe(struct inode *inode, loff_t offset)
|
|
|
|
{
|
|
|
|
struct uprobe *uprobe;
|
|
|
|
|
2012-08-18 23:01:57 +08:00
|
|
|
spin_lock(&uprobes_treelock);
|
uprobes, mm, x86: Add the ability to install and remove uprobes breakpoints
Add uprobes support to the core kernel, with x86 support.
This commit adds the kernel facilities, the actual uprobes
user-space ABI and perf probe support comes in later commits.
General design:
Uprobes are maintained in an rb-tree indexed by inode and offset
(the offset here is from the start of the mapping). For a unique
(inode, offset) tuple, there can be at most one uprobe in the
rb-tree.
Since the (inode, offset) tuple identifies a unique uprobe, more
than one user may be interested in the same uprobe. This provides
the ability to connect multiple 'consumers' to the same uprobe.
Each consumer defines a handler and a filter (optional). The
'handler' is run every time the uprobe is hit, if it matches the
'filter' criteria.
The first consumer of a uprobe causes the breakpoint to be
inserted at the specified address and subsequent consumers are
appended to this list. On subsequent probes, the consumer gets
appended to the existing list of consumers. The breakpoint is
removed when the last consumer unregisters. For all other
unregisterations, the consumer is removed from the list of
consumers.
Given a inode, we get a list of the mms that have mapped the
inode. Do the actual registration if mm maps the page where a
probe needs to be inserted/removed.
We use a temporary list to walk through the vmas that map the
inode.
- The number of maps that map the inode, is not known before we
walk the rmap and keeps changing.
- extending vm_area_struct wasn't recommended, it's a
size-critical data structure.
- There can be more than one maps of the inode in the same mm.
We add callbacks to the mmap methods to keep an eye on text vmas
that are of interest to uprobes. When a vma of interest is mapped,
we insert the breakpoint at the right address.
Uprobe works by replacing the instruction at the address defined
by (inode, offset) with the arch specific breakpoint
instruction. We save a copy of the original instruction at the
uprobed address.
This is needed for:
a. executing the instruction out-of-line (xol).
b. instruction analysis for any subsequent fixups.
c. restoring the instruction back when the uprobe is unregistered.
We insert or delete a breakpoint instruction, and this
breakpoint instruction is assumed to be the smallest instruction
available on the platform. For fixed size instruction platforms
this is trivially true, for variable size instruction platforms
the breakpoint instruction is typically the smallest (often a
single byte).
Writing the instruction is done by COWing the page and changing
the instruction during the copy, this even though most platforms
allow atomic writes of the breakpoint instruction. This also
mirrors the behaviour of a ptrace() memory write to a PRIVATE
file map.
The core worker is derived from KSM's replace_page() logic.
In essence, similar to KSM:
a. allocate a new page and copy over contents of the page that
has the uprobed vaddr
b. modify the copy and insert the breakpoint at the required
address
c. switch the original page with the copy containing the
breakpoint
d. flush page tables.
replace_page() is being replicated here because of some minor
changes in the type of pages and also because Hugh Dickins had
plans to improve replace_page() for KSM specific work.
Instruction analysis on x86 is based on instruction decoder and
determines if an instruction can be probed and determines the
necessary fixups after singlestep. Instruction analysis is done
at probe insertion time so that we avoid having to repeat the
same analysis every time a probe is hit.
A lot of code here is due to the improvement/suggestions/inputs
from Peter Zijlstra.
Changelog:
(v10):
- Add code to clear REX.B prefix as suggested by Denys Vlasenko
and Masami Hiramatsu.
(v9):
- Use insn_offset_modrm as suggested by Masami Hiramatsu.
(v7):
Handle comments from Peter Zijlstra:
- Dont take reference to inode. (expect inode to uprobe_register to be sane).
- Use PTR_ERR to set the return value.
- No need to take reference to inode.
- use PTR_ERR to return error value.
- register and uprobe_unregister share code.
(v5):
- Modified del_consumer as per comments from Peter.
- Drop reference to inode before dropping reference to uprobe.
- Use i_size_read(inode) instead of inode->i_size.
- Ensure uprobe->consumers is NULL, before __uprobe_unregister() is called.
- Includes errno.h as recommended by Stephen Rothwell to fix a build issue
on sparc defconfig
- Remove restrictions while unregistering.
- Earlier code leaked inode references under some conditions while
registering/unregistering.
- Continue the vma-rmap walk even if the intermediate vma doesnt
meet the requirements.
- Validate the vma found by find_vma before inserting/removing the
breakpoint
- Call del_consumer under mutex_lock.
- Use hash locks.
- Handle mremap.
- Introduce find_least_offset_node() instead of close match logic in
find_uprobe
- Uprobes no more depends on MM_OWNER; No reference to task_structs
while inserting/removing a probe.
- Uses read_mapping_page instead of grab_cache_page so that the pages
have valid content.
- pass NULL to get_user_pages for the task parameter.
- call SetPageUptodate on the new page allocated in write_opcode.
- fix leaking a reference to the new page under certain conditions.
- Include Instruction Decoder if Uprobes gets defined.
- Remove const attributes for instruction prefix arrays.
- Uses mm_context to know if the application is 32 bit.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Also-written-by: Jim Keniston <jkenisto@us.ibm.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Roland McGrath <roland@hack.frob.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Anton Arapov <anton@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Denys Vlasenko <vda.linux@googlemail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linux-mm <linux-mm@kvack.org>
Link: http://lkml.kernel.org/r/20120209092642.GE16600@linux.vnet.ibm.com
[ Made various small edits to the commit log ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-02-09 17:26:42 +08:00
|
|
|
uprobe = __find_uprobe(inode, offset);
|
2012-08-18 23:01:57 +08:00
|
|
|
spin_unlock(&uprobes_treelock);
|
2012-02-17 16:27:41 +08:00
|
|
|
|
uprobes, mm, x86: Add the ability to install and remove uprobes breakpoints
Add uprobes support to the core kernel, with x86 support.
This commit adds the kernel facilities, the actual uprobes
user-space ABI and perf probe support comes in later commits.
General design:
Uprobes are maintained in an rb-tree indexed by inode and offset
(the offset here is from the start of the mapping). For a unique
(inode, offset) tuple, there can be at most one uprobe in the
rb-tree.
Since the (inode, offset) tuple identifies a unique uprobe, more
than one user may be interested in the same uprobe. This provides
the ability to connect multiple 'consumers' to the same uprobe.
Each consumer defines a handler and a filter (optional). The
'handler' is run every time the uprobe is hit, if it matches the
'filter' criteria.
The first consumer of a uprobe causes the breakpoint to be
inserted at the specified address and subsequent consumers are
appended to this list. On subsequent probes, the consumer gets
appended to the existing list of consumers. The breakpoint is
removed when the last consumer unregisters. For all other
unregisterations, the consumer is removed from the list of
consumers.
Given a inode, we get a list of the mms that have mapped the
inode. Do the actual registration if mm maps the page where a
probe needs to be inserted/removed.
We use a temporary list to walk through the vmas that map the
inode.
- The number of maps that map the inode, is not known before we
walk the rmap and keeps changing.
- extending vm_area_struct wasn't recommended, it's a
size-critical data structure.
- There can be more than one maps of the inode in the same mm.
We add callbacks to the mmap methods to keep an eye on text vmas
that are of interest to uprobes. When a vma of interest is mapped,
we insert the breakpoint at the right address.
Uprobe works by replacing the instruction at the address defined
by (inode, offset) with the arch specific breakpoint
instruction. We save a copy of the original instruction at the
uprobed address.
This is needed for:
a. executing the instruction out-of-line (xol).
b. instruction analysis for any subsequent fixups.
c. restoring the instruction back when the uprobe is unregistered.
We insert or delete a breakpoint instruction, and this
breakpoint instruction is assumed to be the smallest instruction
available on the platform. For fixed size instruction platforms
this is trivially true, for variable size instruction platforms
the breakpoint instruction is typically the smallest (often a
single byte).
Writing the instruction is done by COWing the page and changing
the instruction during the copy, this even though most platforms
allow atomic writes of the breakpoint instruction. This also
mirrors the behaviour of a ptrace() memory write to a PRIVATE
file map.
The core worker is derived from KSM's replace_page() logic.
In essence, similar to KSM:
a. allocate a new page and copy over contents of the page that
has the uprobed vaddr
b. modify the copy and insert the breakpoint at the required
address
c. switch the original page with the copy containing the
breakpoint
d. flush page tables.
replace_page() is being replicated here because of some minor
changes in the type of pages and also because Hugh Dickins had
plans to improve replace_page() for KSM specific work.
Instruction analysis on x86 is based on instruction decoder and
determines if an instruction can be probed and determines the
necessary fixups after singlestep. Instruction analysis is done
at probe insertion time so that we avoid having to repeat the
same analysis every time a probe is hit.
A lot of code here is due to the improvement/suggestions/inputs
from Peter Zijlstra.
Changelog:
(v10):
- Add code to clear REX.B prefix as suggested by Denys Vlasenko
and Masami Hiramatsu.
(v9):
- Use insn_offset_modrm as suggested by Masami Hiramatsu.
(v7):
Handle comments from Peter Zijlstra:
- Dont take reference to inode. (expect inode to uprobe_register to be sane).
- Use PTR_ERR to set the return value.
- No need to take reference to inode.
- use PTR_ERR to return error value.
- register and uprobe_unregister share code.
(v5):
- Modified del_consumer as per comments from Peter.
- Drop reference to inode before dropping reference to uprobe.
- Use i_size_read(inode) instead of inode->i_size.
- Ensure uprobe->consumers is NULL, before __uprobe_unregister() is called.
- Includes errno.h as recommended by Stephen Rothwell to fix a build issue
on sparc defconfig
- Remove restrictions while unregistering.
- Earlier code leaked inode references under some conditions while
registering/unregistering.
- Continue the vma-rmap walk even if the intermediate vma doesnt
meet the requirements.
- Validate the vma found by find_vma before inserting/removing the
breakpoint
- Call del_consumer under mutex_lock.
- Use hash locks.
- Handle mremap.
- Introduce find_least_offset_node() instead of close match logic in
find_uprobe
- Uprobes no more depends on MM_OWNER; No reference to task_structs
while inserting/removing a probe.
- Uses read_mapping_page instead of grab_cache_page so that the pages
have valid content.
- pass NULL to get_user_pages for the task parameter.
- call SetPageUptodate on the new page allocated in write_opcode.
- fix leaking a reference to the new page under certain conditions.
- Include Instruction Decoder if Uprobes gets defined.
- Remove const attributes for instruction prefix arrays.
- Uses mm_context to know if the application is 32 bit.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Also-written-by: Jim Keniston <jkenisto@us.ibm.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Roland McGrath <roland@hack.frob.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Anton Arapov <anton@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Denys Vlasenko <vda.linux@googlemail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linux-mm <linux-mm@kvack.org>
Link: http://lkml.kernel.org/r/20120209092642.GE16600@linux.vnet.ibm.com
[ Made various small edits to the commit log ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-02-09 17:26:42 +08:00
|
|
|
return uprobe;
|
|
|
|
}
|
|
|
|
|
|
|
|
static struct uprobe *__insert_uprobe(struct uprobe *uprobe)
|
|
|
|
{
|
|
|
|
struct rb_node **p = &uprobes_tree.rb_node;
|
|
|
|
struct rb_node *parent = NULL;
|
|
|
|
struct uprobe *u;
|
|
|
|
int match;
|
|
|
|
|
|
|
|
while (*p) {
|
|
|
|
parent = *p;
|
|
|
|
u = rb_entry(parent, struct uprobe, rb_node);
|
|
|
|
match = match_uprobe(uprobe, u);
|
2015-07-21 21:40:03 +08:00
|
|
|
if (!match)
|
|
|
|
return get_uprobe(u);
|
uprobes, mm, x86: Add the ability to install and remove uprobes breakpoints
Add uprobes support to the core kernel, with x86 support.
This commit adds the kernel facilities, the actual uprobes
user-space ABI and perf probe support comes in later commits.
General design:
Uprobes are maintained in an rb-tree indexed by inode and offset
(the offset here is from the start of the mapping). For a unique
(inode, offset) tuple, there can be at most one uprobe in the
rb-tree.
Since the (inode, offset) tuple identifies a unique uprobe, more
than one user may be interested in the same uprobe. This provides
the ability to connect multiple 'consumers' to the same uprobe.
Each consumer defines a handler and a filter (optional). The
'handler' is run every time the uprobe is hit, if it matches the
'filter' criteria.
The first consumer of a uprobe causes the breakpoint to be
inserted at the specified address and subsequent consumers are
appended to this list. On subsequent probes, the consumer gets
appended to the existing list of consumers. The breakpoint is
removed when the last consumer unregisters. For all other
unregisterations, the consumer is removed from the list of
consumers.
Given a inode, we get a list of the mms that have mapped the
inode. Do the actual registration if mm maps the page where a
probe needs to be inserted/removed.
We use a temporary list to walk through the vmas that map the
inode.
- The number of maps that map the inode, is not known before we
walk the rmap and keeps changing.
- extending vm_area_struct wasn't recommended, it's a
size-critical data structure.
- There can be more than one maps of the inode in the same mm.
We add callbacks to the mmap methods to keep an eye on text vmas
that are of interest to uprobes. When a vma of interest is mapped,
we insert the breakpoint at the right address.
Uprobe works by replacing the instruction at the address defined
by (inode, offset) with the arch specific breakpoint
instruction. We save a copy of the original instruction at the
uprobed address.
This is needed for:
a. executing the instruction out-of-line (xol).
b. instruction analysis for any subsequent fixups.
c. restoring the instruction back when the uprobe is unregistered.
We insert or delete a breakpoint instruction, and this
breakpoint instruction is assumed to be the smallest instruction
available on the platform. For fixed size instruction platforms
this is trivially true, for variable size instruction platforms
the breakpoint instruction is typically the smallest (often a
single byte).
Writing the instruction is done by COWing the page and changing
the instruction during the copy, this even though most platforms
allow atomic writes of the breakpoint instruction. This also
mirrors the behaviour of a ptrace() memory write to a PRIVATE
file map.
The core worker is derived from KSM's replace_page() logic.
In essence, similar to KSM:
a. allocate a new page and copy over contents of the page that
has the uprobed vaddr
b. modify the copy and insert the breakpoint at the required
address
c. switch the original page with the copy containing the
breakpoint
d. flush page tables.
replace_page() is being replicated here because of some minor
changes in the type of pages and also because Hugh Dickins had
plans to improve replace_page() for KSM specific work.
Instruction analysis on x86 is based on instruction decoder and
determines if an instruction can be probed and determines the
necessary fixups after singlestep. Instruction analysis is done
at probe insertion time so that we avoid having to repeat the
same analysis every time a probe is hit.
A lot of code here is due to the improvement/suggestions/inputs
from Peter Zijlstra.
Changelog:
(v10):
- Add code to clear REX.B prefix as suggested by Denys Vlasenko
and Masami Hiramatsu.
(v9):
- Use insn_offset_modrm as suggested by Masami Hiramatsu.
(v7):
Handle comments from Peter Zijlstra:
- Dont take reference to inode. (expect inode to uprobe_register to be sane).
- Use PTR_ERR to set the return value.
- No need to take reference to inode.
- use PTR_ERR to return error value.
- register and uprobe_unregister share code.
(v5):
- Modified del_consumer as per comments from Peter.
- Drop reference to inode before dropping reference to uprobe.
- Use i_size_read(inode) instead of inode->i_size.
- Ensure uprobe->consumers is NULL, before __uprobe_unregister() is called.
- Includes errno.h as recommended by Stephen Rothwell to fix a build issue
on sparc defconfig
- Remove restrictions while unregistering.
- Earlier code leaked inode references under some conditions while
registering/unregistering.
- Continue the vma-rmap walk even if the intermediate vma doesnt
meet the requirements.
- Validate the vma found by find_vma before inserting/removing the
breakpoint
- Call del_consumer under mutex_lock.
- Use hash locks.
- Handle mremap.
- Introduce find_least_offset_node() instead of close match logic in
find_uprobe
- Uprobes no more depends on MM_OWNER; No reference to task_structs
while inserting/removing a probe.
- Uses read_mapping_page instead of grab_cache_page so that the pages
have valid content.
- pass NULL to get_user_pages for the task parameter.
- call SetPageUptodate on the new page allocated in write_opcode.
- fix leaking a reference to the new page under certain conditions.
- Include Instruction Decoder if Uprobes gets defined.
- Remove const attributes for instruction prefix arrays.
- Uses mm_context to know if the application is 32 bit.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Also-written-by: Jim Keniston <jkenisto@us.ibm.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Roland McGrath <roland@hack.frob.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Anton Arapov <anton@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Denys Vlasenko <vda.linux@googlemail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linux-mm <linux-mm@kvack.org>
Link: http://lkml.kernel.org/r/20120209092642.GE16600@linux.vnet.ibm.com
[ Made various small edits to the commit log ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-02-09 17:26:42 +08:00
|
|
|
|
|
|
|
if (match < 0)
|
|
|
|
p = &parent->rb_left;
|
|
|
|
else
|
|
|
|
p = &parent->rb_right;
|
|
|
|
|
|
|
|
}
|
2012-02-17 16:27:41 +08:00
|
|
|
|
uprobes, mm, x86: Add the ability to install and remove uprobes breakpoints
Add uprobes support to the core kernel, with x86 support.
This commit adds the kernel facilities, the actual uprobes
user-space ABI and perf probe support comes in later commits.
General design:
Uprobes are maintained in an rb-tree indexed by inode and offset
(the offset here is from the start of the mapping). For a unique
(inode, offset) tuple, there can be at most one uprobe in the
rb-tree.
Since the (inode, offset) tuple identifies a unique uprobe, more
than one user may be interested in the same uprobe. This provides
the ability to connect multiple 'consumers' to the same uprobe.
Each consumer defines a handler and a filter (optional). The
'handler' is run every time the uprobe is hit, if it matches the
'filter' criteria.
The first consumer of a uprobe causes the breakpoint to be
inserted at the specified address and subsequent consumers are
appended to this list. On subsequent probes, the consumer gets
appended to the existing list of consumers. The breakpoint is
removed when the last consumer unregisters. For all other
unregisterations, the consumer is removed from the list of
consumers.
Given a inode, we get a list of the mms that have mapped the
inode. Do the actual registration if mm maps the page where a
probe needs to be inserted/removed.
We use a temporary list to walk through the vmas that map the
inode.
- The number of maps that map the inode, is not known before we
walk the rmap and keeps changing.
- extending vm_area_struct wasn't recommended, it's a
size-critical data structure.
- There can be more than one maps of the inode in the same mm.
We add callbacks to the mmap methods to keep an eye on text vmas
that are of interest to uprobes. When a vma of interest is mapped,
we insert the breakpoint at the right address.
Uprobe works by replacing the instruction at the address defined
by (inode, offset) with the arch specific breakpoint
instruction. We save a copy of the original instruction at the
uprobed address.
This is needed for:
a. executing the instruction out-of-line (xol).
b. instruction analysis for any subsequent fixups.
c. restoring the instruction back when the uprobe is unregistered.
We insert or delete a breakpoint instruction, and this
breakpoint instruction is assumed to be the smallest instruction
available on the platform. For fixed size instruction platforms
this is trivially true, for variable size instruction platforms
the breakpoint instruction is typically the smallest (often a
single byte).
Writing the instruction is done by COWing the page and changing
the instruction during the copy, this even though most platforms
allow atomic writes of the breakpoint instruction. This also
mirrors the behaviour of a ptrace() memory write to a PRIVATE
file map.
The core worker is derived from KSM's replace_page() logic.
In essence, similar to KSM:
a. allocate a new page and copy over contents of the page that
has the uprobed vaddr
b. modify the copy and insert the breakpoint at the required
address
c. switch the original page with the copy containing the
breakpoint
d. flush page tables.
replace_page() is being replicated here because of some minor
changes in the type of pages and also because Hugh Dickins had
plans to improve replace_page() for KSM specific work.
Instruction analysis on x86 is based on instruction decoder and
determines if an instruction can be probed and determines the
necessary fixups after singlestep. Instruction analysis is done
at probe insertion time so that we avoid having to repeat the
same analysis every time a probe is hit.
A lot of code here is due to the improvement/suggestions/inputs
from Peter Zijlstra.
Changelog:
(v10):
- Add code to clear REX.B prefix as suggested by Denys Vlasenko
and Masami Hiramatsu.
(v9):
- Use insn_offset_modrm as suggested by Masami Hiramatsu.
(v7):
Handle comments from Peter Zijlstra:
- Dont take reference to inode. (expect inode to uprobe_register to be sane).
- Use PTR_ERR to set the return value.
- No need to take reference to inode.
- use PTR_ERR to return error value.
- register and uprobe_unregister share code.
(v5):
- Modified del_consumer as per comments from Peter.
- Drop reference to inode before dropping reference to uprobe.
- Use i_size_read(inode) instead of inode->i_size.
- Ensure uprobe->consumers is NULL, before __uprobe_unregister() is called.
- Includes errno.h as recommended by Stephen Rothwell to fix a build issue
on sparc defconfig
- Remove restrictions while unregistering.
- Earlier code leaked inode references under some conditions while
registering/unregistering.
- Continue the vma-rmap walk even if the intermediate vma doesnt
meet the requirements.
- Validate the vma found by find_vma before inserting/removing the
breakpoint
- Call del_consumer under mutex_lock.
- Use hash locks.
- Handle mremap.
- Introduce find_least_offset_node() instead of close match logic in
find_uprobe
- Uprobes no more depends on MM_OWNER; No reference to task_structs
while inserting/removing a probe.
- Uses read_mapping_page instead of grab_cache_page so that the pages
have valid content.
- pass NULL to get_user_pages for the task parameter.
- call SetPageUptodate on the new page allocated in write_opcode.
- fix leaking a reference to the new page under certain conditions.
- Include Instruction Decoder if Uprobes gets defined.
- Remove const attributes for instruction prefix arrays.
- Uses mm_context to know if the application is 32 bit.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Also-written-by: Jim Keniston <jkenisto@us.ibm.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Roland McGrath <roland@hack.frob.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Anton Arapov <anton@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Denys Vlasenko <vda.linux@googlemail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linux-mm <linux-mm@kvack.org>
Link: http://lkml.kernel.org/r/20120209092642.GE16600@linux.vnet.ibm.com
[ Made various small edits to the commit log ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-02-09 17:26:42 +08:00
|
|
|
u = NULL;
|
|
|
|
rb_link_node(&uprobe->rb_node, parent, p);
|
|
|
|
rb_insert_color(&uprobe->rb_node, &uprobes_tree);
|
|
|
|
/* get access + creation ref */
|
|
|
|
atomic_set(&uprobe->ref, 2);
|
2012-02-17 16:27:41 +08:00
|
|
|
|
uprobes, mm, x86: Add the ability to install and remove uprobes breakpoints
Add uprobes support to the core kernel, with x86 support.
This commit adds the kernel facilities, the actual uprobes
user-space ABI and perf probe support comes in later commits.
General design:
Uprobes are maintained in an rb-tree indexed by inode and offset
(the offset here is from the start of the mapping). For a unique
(inode, offset) tuple, there can be at most one uprobe in the
rb-tree.
Since the (inode, offset) tuple identifies a unique uprobe, more
than one user may be interested in the same uprobe. This provides
the ability to connect multiple 'consumers' to the same uprobe.
Each consumer defines a handler and a filter (optional). The
'handler' is run every time the uprobe is hit, if it matches the
'filter' criteria.
The first consumer of a uprobe causes the breakpoint to be
inserted at the specified address and subsequent consumers are
appended to this list. On subsequent probes, the consumer gets
appended to the existing list of consumers. The breakpoint is
removed when the last consumer unregisters. For all other
unregisterations, the consumer is removed from the list of
consumers.
Given a inode, we get a list of the mms that have mapped the
inode. Do the actual registration if mm maps the page where a
probe needs to be inserted/removed.
We use a temporary list to walk through the vmas that map the
inode.
- The number of maps that map the inode, is not known before we
walk the rmap and keeps changing.
- extending vm_area_struct wasn't recommended, it's a
size-critical data structure.
- There can be more than one maps of the inode in the same mm.
We add callbacks to the mmap methods to keep an eye on text vmas
that are of interest to uprobes. When a vma of interest is mapped,
we insert the breakpoint at the right address.
Uprobe works by replacing the instruction at the address defined
by (inode, offset) with the arch specific breakpoint
instruction. We save a copy of the original instruction at the
uprobed address.
This is needed for:
a. executing the instruction out-of-line (xol).
b. instruction analysis for any subsequent fixups.
c. restoring the instruction back when the uprobe is unregistered.
We insert or delete a breakpoint instruction, and this
breakpoint instruction is assumed to be the smallest instruction
available on the platform. For fixed size instruction platforms
this is trivially true, for variable size instruction platforms
the breakpoint instruction is typically the smallest (often a
single byte).
Writing the instruction is done by COWing the page and changing
the instruction during the copy, this even though most platforms
allow atomic writes of the breakpoint instruction. This also
mirrors the behaviour of a ptrace() memory write to a PRIVATE
file map.
The core worker is derived from KSM's replace_page() logic.
In essence, similar to KSM:
a. allocate a new page and copy over contents of the page that
has the uprobed vaddr
b. modify the copy and insert the breakpoint at the required
address
c. switch the original page with the copy containing the
breakpoint
d. flush page tables.
replace_page() is being replicated here because of some minor
changes in the type of pages and also because Hugh Dickins had
plans to improve replace_page() for KSM specific work.
Instruction analysis on x86 is based on instruction decoder and
determines if an instruction can be probed and determines the
necessary fixups after singlestep. Instruction analysis is done
at probe insertion time so that we avoid having to repeat the
same analysis every time a probe is hit.
A lot of code here is due to the improvement/suggestions/inputs
from Peter Zijlstra.
Changelog:
(v10):
- Add code to clear REX.B prefix as suggested by Denys Vlasenko
and Masami Hiramatsu.
(v9):
- Use insn_offset_modrm as suggested by Masami Hiramatsu.
(v7):
Handle comments from Peter Zijlstra:
- Dont take reference to inode. (expect inode to uprobe_register to be sane).
- Use PTR_ERR to set the return value.
- No need to take reference to inode.
- use PTR_ERR to return error value.
- register and uprobe_unregister share code.
(v5):
- Modified del_consumer as per comments from Peter.
- Drop reference to inode before dropping reference to uprobe.
- Use i_size_read(inode) instead of inode->i_size.
- Ensure uprobe->consumers is NULL, before __uprobe_unregister() is called.
- Includes errno.h as recommended by Stephen Rothwell to fix a build issue
on sparc defconfig
- Remove restrictions while unregistering.
- Earlier code leaked inode references under some conditions while
registering/unregistering.
- Continue the vma-rmap walk even if the intermediate vma doesnt
meet the requirements.
- Validate the vma found by find_vma before inserting/removing the
breakpoint
- Call del_consumer under mutex_lock.
- Use hash locks.
- Handle mremap.
- Introduce find_least_offset_node() instead of close match logic in
find_uprobe
- Uprobes no more depends on MM_OWNER; No reference to task_structs
while inserting/removing a probe.
- Uses read_mapping_page instead of grab_cache_page so that the pages
have valid content.
- pass NULL to get_user_pages for the task parameter.
- call SetPageUptodate on the new page allocated in write_opcode.
- fix leaking a reference to the new page under certain conditions.
- Include Instruction Decoder if Uprobes gets defined.
- Remove const attributes for instruction prefix arrays.
- Uses mm_context to know if the application is 32 bit.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Also-written-by: Jim Keniston <jkenisto@us.ibm.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Roland McGrath <roland@hack.frob.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Anton Arapov <anton@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Denys Vlasenko <vda.linux@googlemail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linux-mm <linux-mm@kvack.org>
Link: http://lkml.kernel.org/r/20120209092642.GE16600@linux.vnet.ibm.com
[ Made various small edits to the commit log ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-02-09 17:26:42 +08:00
|
|
|
return u;
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
2012-02-17 16:27:41 +08:00
|
|
|
* Acquire uprobes_treelock.
|
uprobes, mm, x86: Add the ability to install and remove uprobes breakpoints
Add uprobes support to the core kernel, with x86 support.
This commit adds the kernel facilities, the actual uprobes
user-space ABI and perf probe support comes in later commits.
General design:
Uprobes are maintained in an rb-tree indexed by inode and offset
(the offset here is from the start of the mapping). For a unique
(inode, offset) tuple, there can be at most one uprobe in the
rb-tree.
Since the (inode, offset) tuple identifies a unique uprobe, more
than one user may be interested in the same uprobe. This provides
the ability to connect multiple 'consumers' to the same uprobe.
Each consumer defines a handler and a filter (optional). The
'handler' is run every time the uprobe is hit, if it matches the
'filter' criteria.
The first consumer of a uprobe causes the breakpoint to be
inserted at the specified address and subsequent consumers are
appended to this list. On subsequent probes, the consumer gets
appended to the existing list of consumers. The breakpoint is
removed when the last consumer unregisters. For all other
unregisterations, the consumer is removed from the list of
consumers.
Given a inode, we get a list of the mms that have mapped the
inode. Do the actual registration if mm maps the page where a
probe needs to be inserted/removed.
We use a temporary list to walk through the vmas that map the
inode.
- The number of maps that map the inode, is not known before we
walk the rmap and keeps changing.
- extending vm_area_struct wasn't recommended, it's a
size-critical data structure.
- There can be more than one maps of the inode in the same mm.
We add callbacks to the mmap methods to keep an eye on text vmas
that are of interest to uprobes. When a vma of interest is mapped,
we insert the breakpoint at the right address.
Uprobe works by replacing the instruction at the address defined
by (inode, offset) with the arch specific breakpoint
instruction. We save a copy of the original instruction at the
uprobed address.
This is needed for:
a. executing the instruction out-of-line (xol).
b. instruction analysis for any subsequent fixups.
c. restoring the instruction back when the uprobe is unregistered.
We insert or delete a breakpoint instruction, and this
breakpoint instruction is assumed to be the smallest instruction
available on the platform. For fixed size instruction platforms
this is trivially true, for variable size instruction platforms
the breakpoint instruction is typically the smallest (often a
single byte).
Writing the instruction is done by COWing the page and changing
the instruction during the copy, this even though most platforms
allow atomic writes of the breakpoint instruction. This also
mirrors the behaviour of a ptrace() memory write to a PRIVATE
file map.
The core worker is derived from KSM's replace_page() logic.
In essence, similar to KSM:
a. allocate a new page and copy over contents of the page that
has the uprobed vaddr
b. modify the copy and insert the breakpoint at the required
address
c. switch the original page with the copy containing the
breakpoint
d. flush page tables.
replace_page() is being replicated here because of some minor
changes in the type of pages and also because Hugh Dickins had
plans to improve replace_page() for KSM specific work.
Instruction analysis on x86 is based on instruction decoder and
determines if an instruction can be probed and determines the
necessary fixups after singlestep. Instruction analysis is done
at probe insertion time so that we avoid having to repeat the
same analysis every time a probe is hit.
A lot of code here is due to the improvement/suggestions/inputs
from Peter Zijlstra.
Changelog:
(v10):
- Add code to clear REX.B prefix as suggested by Denys Vlasenko
and Masami Hiramatsu.
(v9):
- Use insn_offset_modrm as suggested by Masami Hiramatsu.
(v7):
Handle comments from Peter Zijlstra:
- Dont take reference to inode. (expect inode to uprobe_register to be sane).
- Use PTR_ERR to set the return value.
- No need to take reference to inode.
- use PTR_ERR to return error value.
- register and uprobe_unregister share code.
(v5):
- Modified del_consumer as per comments from Peter.
- Drop reference to inode before dropping reference to uprobe.
- Use i_size_read(inode) instead of inode->i_size.
- Ensure uprobe->consumers is NULL, before __uprobe_unregister() is called.
- Includes errno.h as recommended by Stephen Rothwell to fix a build issue
on sparc defconfig
- Remove restrictions while unregistering.
- Earlier code leaked inode references under some conditions while
registering/unregistering.
- Continue the vma-rmap walk even if the intermediate vma doesnt
meet the requirements.
- Validate the vma found by find_vma before inserting/removing the
breakpoint
- Call del_consumer under mutex_lock.
- Use hash locks.
- Handle mremap.
- Introduce find_least_offset_node() instead of close match logic in
find_uprobe
- Uprobes no more depends on MM_OWNER; No reference to task_structs
while inserting/removing a probe.
- Uses read_mapping_page instead of grab_cache_page so that the pages
have valid content.
- pass NULL to get_user_pages for the task parameter.
- call SetPageUptodate on the new page allocated in write_opcode.
- fix leaking a reference to the new page under certain conditions.
- Include Instruction Decoder if Uprobes gets defined.
- Remove const attributes for instruction prefix arrays.
- Uses mm_context to know if the application is 32 bit.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Also-written-by: Jim Keniston <jkenisto@us.ibm.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Roland McGrath <roland@hack.frob.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Anton Arapov <anton@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Denys Vlasenko <vda.linux@googlemail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linux-mm <linux-mm@kvack.org>
Link: http://lkml.kernel.org/r/20120209092642.GE16600@linux.vnet.ibm.com
[ Made various small edits to the commit log ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-02-09 17:26:42 +08:00
|
|
|
* Matching uprobe already exists in rbtree;
|
|
|
|
* increment (access refcount) and return the matching uprobe.
|
|
|
|
*
|
|
|
|
* No matching uprobe; insert the uprobe in rb_tree;
|
|
|
|
* get a double refcount (access + creation) and return NULL.
|
|
|
|
*/
|
|
|
|
static struct uprobe *insert_uprobe(struct uprobe *uprobe)
|
|
|
|
{
|
|
|
|
struct uprobe *u;
|
|
|
|
|
2012-08-18 23:01:57 +08:00
|
|
|
spin_lock(&uprobes_treelock);
|
uprobes, mm, x86: Add the ability to install and remove uprobes breakpoints
Add uprobes support to the core kernel, with x86 support.
This commit adds the kernel facilities, the actual uprobes
user-space ABI and perf probe support comes in later commits.
General design:
Uprobes are maintained in an rb-tree indexed by inode and offset
(the offset here is from the start of the mapping). For a unique
(inode, offset) tuple, there can be at most one uprobe in the
rb-tree.
Since the (inode, offset) tuple identifies a unique uprobe, more
than one user may be interested in the same uprobe. This provides
the ability to connect multiple 'consumers' to the same uprobe.
Each consumer defines a handler and a filter (optional). The
'handler' is run every time the uprobe is hit, if it matches the
'filter' criteria.
The first consumer of a uprobe causes the breakpoint to be
inserted at the specified address and subsequent consumers are
appended to this list. On subsequent probes, the consumer gets
appended to the existing list of consumers. The breakpoint is
removed when the last consumer unregisters. For all other
unregisterations, the consumer is removed from the list of
consumers.
Given a inode, we get a list of the mms that have mapped the
inode. Do the actual registration if mm maps the page where a
probe needs to be inserted/removed.
We use a temporary list to walk through the vmas that map the
inode.
- The number of maps that map the inode, is not known before we
walk the rmap and keeps changing.
- extending vm_area_struct wasn't recommended, it's a
size-critical data structure.
- There can be more than one maps of the inode in the same mm.
We add callbacks to the mmap methods to keep an eye on text vmas
that are of interest to uprobes. When a vma of interest is mapped,
we insert the breakpoint at the right address.
Uprobe works by replacing the instruction at the address defined
by (inode, offset) with the arch specific breakpoint
instruction. We save a copy of the original instruction at the
uprobed address.
This is needed for:
a. executing the instruction out-of-line (xol).
b. instruction analysis for any subsequent fixups.
c. restoring the instruction back when the uprobe is unregistered.
We insert or delete a breakpoint instruction, and this
breakpoint instruction is assumed to be the smallest instruction
available on the platform. For fixed size instruction platforms
this is trivially true, for variable size instruction platforms
the breakpoint instruction is typically the smallest (often a
single byte).
Writing the instruction is done by COWing the page and changing
the instruction during the copy, this even though most platforms
allow atomic writes of the breakpoint instruction. This also
mirrors the behaviour of a ptrace() memory write to a PRIVATE
file map.
The core worker is derived from KSM's replace_page() logic.
In essence, similar to KSM:
a. allocate a new page and copy over contents of the page that
has the uprobed vaddr
b. modify the copy and insert the breakpoint at the required
address
c. switch the original page with the copy containing the
breakpoint
d. flush page tables.
replace_page() is being replicated here because of some minor
changes in the type of pages and also because Hugh Dickins had
plans to improve replace_page() for KSM specific work.
Instruction analysis on x86 is based on instruction decoder and
determines if an instruction can be probed and determines the
necessary fixups after singlestep. Instruction analysis is done
at probe insertion time so that we avoid having to repeat the
same analysis every time a probe is hit.
A lot of code here is due to the improvement/suggestions/inputs
from Peter Zijlstra.
Changelog:
(v10):
- Add code to clear REX.B prefix as suggested by Denys Vlasenko
and Masami Hiramatsu.
(v9):
- Use insn_offset_modrm as suggested by Masami Hiramatsu.
(v7):
Handle comments from Peter Zijlstra:
- Dont take reference to inode. (expect inode to uprobe_register to be sane).
- Use PTR_ERR to set the return value.
- No need to take reference to inode.
- use PTR_ERR to return error value.
- register and uprobe_unregister share code.
(v5):
- Modified del_consumer as per comments from Peter.
- Drop reference to inode before dropping reference to uprobe.
- Use i_size_read(inode) instead of inode->i_size.
- Ensure uprobe->consumers is NULL, before __uprobe_unregister() is called.
- Includes errno.h as recommended by Stephen Rothwell to fix a build issue
on sparc defconfig
- Remove restrictions while unregistering.
- Earlier code leaked inode references under some conditions while
registering/unregistering.
- Continue the vma-rmap walk even if the intermediate vma doesnt
meet the requirements.
- Validate the vma found by find_vma before inserting/removing the
breakpoint
- Call del_consumer under mutex_lock.
- Use hash locks.
- Handle mremap.
- Introduce find_least_offset_node() instead of close match logic in
find_uprobe
- Uprobes no more depends on MM_OWNER; No reference to task_structs
while inserting/removing a probe.
- Uses read_mapping_page instead of grab_cache_page so that the pages
have valid content.
- pass NULL to get_user_pages for the task parameter.
- call SetPageUptodate on the new page allocated in write_opcode.
- fix leaking a reference to the new page under certain conditions.
- Include Instruction Decoder if Uprobes gets defined.
- Remove const attributes for instruction prefix arrays.
- Uses mm_context to know if the application is 32 bit.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Also-written-by: Jim Keniston <jkenisto@us.ibm.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Roland McGrath <roland@hack.frob.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Anton Arapov <anton@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Denys Vlasenko <vda.linux@googlemail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linux-mm <linux-mm@kvack.org>
Link: http://lkml.kernel.org/r/20120209092642.GE16600@linux.vnet.ibm.com
[ Made various small edits to the commit log ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-02-09 17:26:42 +08:00
|
|
|
u = __insert_uprobe(uprobe);
|
2012-08-18 23:01:57 +08:00
|
|
|
spin_unlock(&uprobes_treelock);
|
2012-02-17 16:27:41 +08:00
|
|
|
|
uprobes, mm, x86: Add the ability to install and remove uprobes breakpoints
Add uprobes support to the core kernel, with x86 support.
This commit adds the kernel facilities, the actual uprobes
user-space ABI and perf probe support comes in later commits.
General design:
Uprobes are maintained in an rb-tree indexed by inode and offset
(the offset here is from the start of the mapping). For a unique
(inode, offset) tuple, there can be at most one uprobe in the
rb-tree.
Since the (inode, offset) tuple identifies a unique uprobe, more
than one user may be interested in the same uprobe. This provides
the ability to connect multiple 'consumers' to the same uprobe.
Each consumer defines a handler and a filter (optional). The
'handler' is run every time the uprobe is hit, if it matches the
'filter' criteria.
The first consumer of a uprobe causes the breakpoint to be
inserted at the specified address and subsequent consumers are
appended to this list. On subsequent probes, the consumer gets
appended to the existing list of consumers. The breakpoint is
removed when the last consumer unregisters. For all other
unregisterations, the consumer is removed from the list of
consumers.
Given a inode, we get a list of the mms that have mapped the
inode. Do the actual registration if mm maps the page where a
probe needs to be inserted/removed.
We use a temporary list to walk through the vmas that map the
inode.
- The number of maps that map the inode, is not known before we
walk the rmap and keeps changing.
- extending vm_area_struct wasn't recommended, it's a
size-critical data structure.
- There can be more than one maps of the inode in the same mm.
We add callbacks to the mmap methods to keep an eye on text vmas
that are of interest to uprobes. When a vma of interest is mapped,
we insert the breakpoint at the right address.
Uprobe works by replacing the instruction at the address defined
by (inode, offset) with the arch specific breakpoint
instruction. We save a copy of the original instruction at the
uprobed address.
This is needed for:
a. executing the instruction out-of-line (xol).
b. instruction analysis for any subsequent fixups.
c. restoring the instruction back when the uprobe is unregistered.
We insert or delete a breakpoint instruction, and this
breakpoint instruction is assumed to be the smallest instruction
available on the platform. For fixed size instruction platforms
this is trivially true, for variable size instruction platforms
the breakpoint instruction is typically the smallest (often a
single byte).
Writing the instruction is done by COWing the page and changing
the instruction during the copy, this even though most platforms
allow atomic writes of the breakpoint instruction. This also
mirrors the behaviour of a ptrace() memory write to a PRIVATE
file map.
The core worker is derived from KSM's replace_page() logic.
In essence, similar to KSM:
a. allocate a new page and copy over contents of the page that
has the uprobed vaddr
b. modify the copy and insert the breakpoint at the required
address
c. switch the original page with the copy containing the
breakpoint
d. flush page tables.
replace_page() is being replicated here because of some minor
changes in the type of pages and also because Hugh Dickins had
plans to improve replace_page() for KSM specific work.
Instruction analysis on x86 is based on instruction decoder and
determines if an instruction can be probed and determines the
necessary fixups after singlestep. Instruction analysis is done
at probe insertion time so that we avoid having to repeat the
same analysis every time a probe is hit.
A lot of code here is due to the improvement/suggestions/inputs
from Peter Zijlstra.
Changelog:
(v10):
- Add code to clear REX.B prefix as suggested by Denys Vlasenko
and Masami Hiramatsu.
(v9):
- Use insn_offset_modrm as suggested by Masami Hiramatsu.
(v7):
Handle comments from Peter Zijlstra:
- Dont take reference to inode. (expect inode to uprobe_register to be sane).
- Use PTR_ERR to set the return value.
- No need to take reference to inode.
- use PTR_ERR to return error value.
- register and uprobe_unregister share code.
(v5):
- Modified del_consumer as per comments from Peter.
- Drop reference to inode before dropping reference to uprobe.
- Use i_size_read(inode) instead of inode->i_size.
- Ensure uprobe->consumers is NULL, before __uprobe_unregister() is called.
- Includes errno.h as recommended by Stephen Rothwell to fix a build issue
on sparc defconfig
- Remove restrictions while unregistering.
- Earlier code leaked inode references under some conditions while
registering/unregistering.
- Continue the vma-rmap walk even if the intermediate vma doesnt
meet the requirements.
- Validate the vma found by find_vma before inserting/removing the
breakpoint
- Call del_consumer under mutex_lock.
- Use hash locks.
- Handle mremap.
- Introduce find_least_offset_node() instead of close match logic in
find_uprobe
- Uprobes no more depends on MM_OWNER; No reference to task_structs
while inserting/removing a probe.
- Uses read_mapping_page instead of grab_cache_page so that the pages
have valid content.
- pass NULL to get_user_pages for the task parameter.
- call SetPageUptodate on the new page allocated in write_opcode.
- fix leaking a reference to the new page under certain conditions.
- Include Instruction Decoder if Uprobes gets defined.
- Remove const attributes for instruction prefix arrays.
- Uses mm_context to know if the application is 32 bit.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Also-written-by: Jim Keniston <jkenisto@us.ibm.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Roland McGrath <roland@hack.frob.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Anton Arapov <anton@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Denys Vlasenko <vda.linux@googlemail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linux-mm <linux-mm@kvack.org>
Link: http://lkml.kernel.org/r/20120209092642.GE16600@linux.vnet.ibm.com
[ Made various small edits to the commit log ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-02-09 17:26:42 +08:00
|
|
|
return u;
|
|
|
|
}
|
|
|
|
|
2018-08-20 12:42:48 +08:00
|
|
|
static void
|
|
|
|
ref_ctr_mismatch_warn(struct uprobe *cur_uprobe, struct uprobe *uprobe)
|
|
|
|
{
|
|
|
|
pr_warn("ref_ctr_offset mismatch. inode: 0x%lx offset: 0x%llx "
|
|
|
|
"ref_ctr_offset(old): 0x%llx ref_ctr_offset(new): 0x%llx\n",
|
|
|
|
uprobe->inode->i_ino, (unsigned long long) uprobe->offset,
|
|
|
|
(unsigned long long) cur_uprobe->ref_ctr_offset,
|
|
|
|
(unsigned long long) uprobe->ref_ctr_offset);
|
|
|
|
}
|
|
|
|
|
uprobes: Support SDT markers having reference count (semaphore)
Userspace Statically Defined Tracepoints[1] are dtrace style markers
inside userspace applications. Applications like PostgreSQL, MySQL,
Pthread, Perl, Python, Java, Ruby, Node.js, libvirt, QEMU, glib etc
have these markers embedded in them. These markers are added by developer
at important places in the code. Each marker source expands to a single
nop instruction in the compiled code but there may be additional
overhead for computing the marker arguments which expands to couple of
instructions. In case the overhead is more, execution of it can be
omitted by runtime if() condition when no one is tracing on the marker:
if (reference_counter > 0) {
Execute marker instructions;
}
Default value of reference counter is 0. Tracer has to increment the
reference counter before tracing on a marker and decrement it when
done with the tracing.
Implement the reference counter logic in core uprobe. User will be
able to use it from trace_uprobe as well as from kernel module. New
trace_uprobe definition with reference counter will now be:
<path>:<offset>[(ref_ctr_offset)]
where ref_ctr_offset is an optional field. For kernel module, new
variant of uprobe_register() has been introduced:
uprobe_register_refctr(inode, offset, ref_ctr_offset, consumer)
No new variant for uprobe_unregister() because it's assumed to have
only one reference counter for one uprobe.
[1] https://sourceware.org/systemtap/wiki/UserSpaceProbeImplementation
Note: 'reference counter' is called as 'semaphore' in original Dtrace
(or Systemtap, bcc and even in ELF) documentation and code. But the
term 'semaphore' is misleading in this context. This is just a counter
used to hold number of tracers tracing on a marker. This is not really
used for any synchronization. So we are calling it a 'reference counter'
in kernel / perf code.
Link: http://lkml.kernel.org/r/20180820044250.11659-2-ravi.bangoria@linux.ibm.com
Reviewed-by: Masami Hiramatsu <mhiramat@kernel.org>
[Only trace_uprobe.c]
Reviewed-by: Oleg Nesterov <oleg@redhat.com>
Reviewed-by: Song Liu <songliubraving@fb.com>
Tested-by: Song Liu <songliubraving@fb.com>
Signed-off-by: Ravi Bangoria <ravi.bangoria@linux.ibm.com>
Signed-off-by: Steven Rostedt (VMware) <rostedt@goodmis.org>
2018-08-20 12:42:47 +08:00
|
|
|
static struct uprobe *alloc_uprobe(struct inode *inode, loff_t offset,
|
|
|
|
loff_t ref_ctr_offset)
|
uprobes, mm, x86: Add the ability to install and remove uprobes breakpoints
Add uprobes support to the core kernel, with x86 support.
This commit adds the kernel facilities, the actual uprobes
user-space ABI and perf probe support comes in later commits.
General design:
Uprobes are maintained in an rb-tree indexed by inode and offset
(the offset here is from the start of the mapping). For a unique
(inode, offset) tuple, there can be at most one uprobe in the
rb-tree.
Since the (inode, offset) tuple identifies a unique uprobe, more
than one user may be interested in the same uprobe. This provides
the ability to connect multiple 'consumers' to the same uprobe.
Each consumer defines a handler and a filter (optional). The
'handler' is run every time the uprobe is hit, if it matches the
'filter' criteria.
The first consumer of a uprobe causes the breakpoint to be
inserted at the specified address and subsequent consumers are
appended to this list. On subsequent probes, the consumer gets
appended to the existing list of consumers. The breakpoint is
removed when the last consumer unregisters. For all other
unregisterations, the consumer is removed from the list of
consumers.
Given a inode, we get a list of the mms that have mapped the
inode. Do the actual registration if mm maps the page where a
probe needs to be inserted/removed.
We use a temporary list to walk through the vmas that map the
inode.
- The number of maps that map the inode, is not known before we
walk the rmap and keeps changing.
- extending vm_area_struct wasn't recommended, it's a
size-critical data structure.
- There can be more than one maps of the inode in the same mm.
We add callbacks to the mmap methods to keep an eye on text vmas
that are of interest to uprobes. When a vma of interest is mapped,
we insert the breakpoint at the right address.
Uprobe works by replacing the instruction at the address defined
by (inode, offset) with the arch specific breakpoint
instruction. We save a copy of the original instruction at the
uprobed address.
This is needed for:
a. executing the instruction out-of-line (xol).
b. instruction analysis for any subsequent fixups.
c. restoring the instruction back when the uprobe is unregistered.
We insert or delete a breakpoint instruction, and this
breakpoint instruction is assumed to be the smallest instruction
available on the platform. For fixed size instruction platforms
this is trivially true, for variable size instruction platforms
the breakpoint instruction is typically the smallest (often a
single byte).
Writing the instruction is done by COWing the page and changing
the instruction during the copy, this even though most platforms
allow atomic writes of the breakpoint instruction. This also
mirrors the behaviour of a ptrace() memory write to a PRIVATE
file map.
The core worker is derived from KSM's replace_page() logic.
In essence, similar to KSM:
a. allocate a new page and copy over contents of the page that
has the uprobed vaddr
b. modify the copy and insert the breakpoint at the required
address
c. switch the original page with the copy containing the
breakpoint
d. flush page tables.
replace_page() is being replicated here because of some minor
changes in the type of pages and also because Hugh Dickins had
plans to improve replace_page() for KSM specific work.
Instruction analysis on x86 is based on instruction decoder and
determines if an instruction can be probed and determines the
necessary fixups after singlestep. Instruction analysis is done
at probe insertion time so that we avoid having to repeat the
same analysis every time a probe is hit.
A lot of code here is due to the improvement/suggestions/inputs
from Peter Zijlstra.
Changelog:
(v10):
- Add code to clear REX.B prefix as suggested by Denys Vlasenko
and Masami Hiramatsu.
(v9):
- Use insn_offset_modrm as suggested by Masami Hiramatsu.
(v7):
Handle comments from Peter Zijlstra:
- Dont take reference to inode. (expect inode to uprobe_register to be sane).
- Use PTR_ERR to set the return value.
- No need to take reference to inode.
- use PTR_ERR to return error value.
- register and uprobe_unregister share code.
(v5):
- Modified del_consumer as per comments from Peter.
- Drop reference to inode before dropping reference to uprobe.
- Use i_size_read(inode) instead of inode->i_size.
- Ensure uprobe->consumers is NULL, before __uprobe_unregister() is called.
- Includes errno.h as recommended by Stephen Rothwell to fix a build issue
on sparc defconfig
- Remove restrictions while unregistering.
- Earlier code leaked inode references under some conditions while
registering/unregistering.
- Continue the vma-rmap walk even if the intermediate vma doesnt
meet the requirements.
- Validate the vma found by find_vma before inserting/removing the
breakpoint
- Call del_consumer under mutex_lock.
- Use hash locks.
- Handle mremap.
- Introduce find_least_offset_node() instead of close match logic in
find_uprobe
- Uprobes no more depends on MM_OWNER; No reference to task_structs
while inserting/removing a probe.
- Uses read_mapping_page instead of grab_cache_page so that the pages
have valid content.
- pass NULL to get_user_pages for the task parameter.
- call SetPageUptodate on the new page allocated in write_opcode.
- fix leaking a reference to the new page under certain conditions.
- Include Instruction Decoder if Uprobes gets defined.
- Remove const attributes for instruction prefix arrays.
- Uses mm_context to know if the application is 32 bit.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Also-written-by: Jim Keniston <jkenisto@us.ibm.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Roland McGrath <roland@hack.frob.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Anton Arapov <anton@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Denys Vlasenko <vda.linux@googlemail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linux-mm <linux-mm@kvack.org>
Link: http://lkml.kernel.org/r/20120209092642.GE16600@linux.vnet.ibm.com
[ Made various small edits to the commit log ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-02-09 17:26:42 +08:00
|
|
|
{
|
|
|
|
struct uprobe *uprobe, *cur_uprobe;
|
|
|
|
|
|
|
|
uprobe = kzalloc(sizeof(struct uprobe), GFP_KERNEL);
|
|
|
|
if (!uprobe)
|
|
|
|
return NULL;
|
|
|
|
|
2018-04-24 01:21:35 +08:00
|
|
|
uprobe->inode = inode;
|
uprobes, mm, x86: Add the ability to install and remove uprobes breakpoints
Add uprobes support to the core kernel, with x86 support.
This commit adds the kernel facilities, the actual uprobes
user-space ABI and perf probe support comes in later commits.
General design:
Uprobes are maintained in an rb-tree indexed by inode and offset
(the offset here is from the start of the mapping). For a unique
(inode, offset) tuple, there can be at most one uprobe in the
rb-tree.
Since the (inode, offset) tuple identifies a unique uprobe, more
than one user may be interested in the same uprobe. This provides
the ability to connect multiple 'consumers' to the same uprobe.
Each consumer defines a handler and a filter (optional). The
'handler' is run every time the uprobe is hit, if it matches the
'filter' criteria.
The first consumer of a uprobe causes the breakpoint to be
inserted at the specified address and subsequent consumers are
appended to this list. On subsequent probes, the consumer gets
appended to the existing list of consumers. The breakpoint is
removed when the last consumer unregisters. For all other
unregisterations, the consumer is removed from the list of
consumers.
Given a inode, we get a list of the mms that have mapped the
inode. Do the actual registration if mm maps the page where a
probe needs to be inserted/removed.
We use a temporary list to walk through the vmas that map the
inode.
- The number of maps that map the inode, is not known before we
walk the rmap and keeps changing.
- extending vm_area_struct wasn't recommended, it's a
size-critical data structure.
- There can be more than one maps of the inode in the same mm.
We add callbacks to the mmap methods to keep an eye on text vmas
that are of interest to uprobes. When a vma of interest is mapped,
we insert the breakpoint at the right address.
Uprobe works by replacing the instruction at the address defined
by (inode, offset) with the arch specific breakpoint
instruction. We save a copy of the original instruction at the
uprobed address.
This is needed for:
a. executing the instruction out-of-line (xol).
b. instruction analysis for any subsequent fixups.
c. restoring the instruction back when the uprobe is unregistered.
We insert or delete a breakpoint instruction, and this
breakpoint instruction is assumed to be the smallest instruction
available on the platform. For fixed size instruction platforms
this is trivially true, for variable size instruction platforms
the breakpoint instruction is typically the smallest (often a
single byte).
Writing the instruction is done by COWing the page and changing
the instruction during the copy, this even though most platforms
allow atomic writes of the breakpoint instruction. This also
mirrors the behaviour of a ptrace() memory write to a PRIVATE
file map.
The core worker is derived from KSM's replace_page() logic.
In essence, similar to KSM:
a. allocate a new page and copy over contents of the page that
has the uprobed vaddr
b. modify the copy and insert the breakpoint at the required
address
c. switch the original page with the copy containing the
breakpoint
d. flush page tables.
replace_page() is being replicated here because of some minor
changes in the type of pages and also because Hugh Dickins had
plans to improve replace_page() for KSM specific work.
Instruction analysis on x86 is based on instruction decoder and
determines if an instruction can be probed and determines the
necessary fixups after singlestep. Instruction analysis is done
at probe insertion time so that we avoid having to repeat the
same analysis every time a probe is hit.
A lot of code here is due to the improvement/suggestions/inputs
from Peter Zijlstra.
Changelog:
(v10):
- Add code to clear REX.B prefix as suggested by Denys Vlasenko
and Masami Hiramatsu.
(v9):
- Use insn_offset_modrm as suggested by Masami Hiramatsu.
(v7):
Handle comments from Peter Zijlstra:
- Dont take reference to inode. (expect inode to uprobe_register to be sane).
- Use PTR_ERR to set the return value.
- No need to take reference to inode.
- use PTR_ERR to return error value.
- register and uprobe_unregister share code.
(v5):
- Modified del_consumer as per comments from Peter.
- Drop reference to inode before dropping reference to uprobe.
- Use i_size_read(inode) instead of inode->i_size.
- Ensure uprobe->consumers is NULL, before __uprobe_unregister() is called.
- Includes errno.h as recommended by Stephen Rothwell to fix a build issue
on sparc defconfig
- Remove restrictions while unregistering.
- Earlier code leaked inode references under some conditions while
registering/unregistering.
- Continue the vma-rmap walk even if the intermediate vma doesnt
meet the requirements.
- Validate the vma found by find_vma before inserting/removing the
breakpoint
- Call del_consumer under mutex_lock.
- Use hash locks.
- Handle mremap.
- Introduce find_least_offset_node() instead of close match logic in
find_uprobe
- Uprobes no more depends on MM_OWNER; No reference to task_structs
while inserting/removing a probe.
- Uses read_mapping_page instead of grab_cache_page so that the pages
have valid content.
- pass NULL to get_user_pages for the task parameter.
- call SetPageUptodate on the new page allocated in write_opcode.
- fix leaking a reference to the new page under certain conditions.
- Include Instruction Decoder if Uprobes gets defined.
- Remove const attributes for instruction prefix arrays.
- Uses mm_context to know if the application is 32 bit.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Also-written-by: Jim Keniston <jkenisto@us.ibm.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Roland McGrath <roland@hack.frob.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Anton Arapov <anton@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Denys Vlasenko <vda.linux@googlemail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linux-mm <linux-mm@kvack.org>
Link: http://lkml.kernel.org/r/20120209092642.GE16600@linux.vnet.ibm.com
[ Made various small edits to the commit log ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-02-09 17:26:42 +08:00
|
|
|
uprobe->offset = offset;
|
uprobes: Support SDT markers having reference count (semaphore)
Userspace Statically Defined Tracepoints[1] are dtrace style markers
inside userspace applications. Applications like PostgreSQL, MySQL,
Pthread, Perl, Python, Java, Ruby, Node.js, libvirt, QEMU, glib etc
have these markers embedded in them. These markers are added by developer
at important places in the code. Each marker source expands to a single
nop instruction in the compiled code but there may be additional
overhead for computing the marker arguments which expands to couple of
instructions. In case the overhead is more, execution of it can be
omitted by runtime if() condition when no one is tracing on the marker:
if (reference_counter > 0) {
Execute marker instructions;
}
Default value of reference counter is 0. Tracer has to increment the
reference counter before tracing on a marker and decrement it when
done with the tracing.
Implement the reference counter logic in core uprobe. User will be
able to use it from trace_uprobe as well as from kernel module. New
trace_uprobe definition with reference counter will now be:
<path>:<offset>[(ref_ctr_offset)]
where ref_ctr_offset is an optional field. For kernel module, new
variant of uprobe_register() has been introduced:
uprobe_register_refctr(inode, offset, ref_ctr_offset, consumer)
No new variant for uprobe_unregister() because it's assumed to have
only one reference counter for one uprobe.
[1] https://sourceware.org/systemtap/wiki/UserSpaceProbeImplementation
Note: 'reference counter' is called as 'semaphore' in original Dtrace
(or Systemtap, bcc and even in ELF) documentation and code. But the
term 'semaphore' is misleading in this context. This is just a counter
used to hold number of tracers tracing on a marker. This is not really
used for any synchronization. So we are calling it a 'reference counter'
in kernel / perf code.
Link: http://lkml.kernel.org/r/20180820044250.11659-2-ravi.bangoria@linux.ibm.com
Reviewed-by: Masami Hiramatsu <mhiramat@kernel.org>
[Only trace_uprobe.c]
Reviewed-by: Oleg Nesterov <oleg@redhat.com>
Reviewed-by: Song Liu <songliubraving@fb.com>
Tested-by: Song Liu <songliubraving@fb.com>
Signed-off-by: Ravi Bangoria <ravi.bangoria@linux.ibm.com>
Signed-off-by: Steven Rostedt (VMware) <rostedt@goodmis.org>
2018-08-20 12:42:47 +08:00
|
|
|
uprobe->ref_ctr_offset = ref_ctr_offset;
|
2012-11-25 00:29:40 +08:00
|
|
|
init_rwsem(&uprobe->register_rwsem);
|
uprobes, mm, x86: Add the ability to install and remove uprobes breakpoints
Add uprobes support to the core kernel, with x86 support.
This commit adds the kernel facilities, the actual uprobes
user-space ABI and perf probe support comes in later commits.
General design:
Uprobes are maintained in an rb-tree indexed by inode and offset
(the offset here is from the start of the mapping). For a unique
(inode, offset) tuple, there can be at most one uprobe in the
rb-tree.
Since the (inode, offset) tuple identifies a unique uprobe, more
than one user may be interested in the same uprobe. This provides
the ability to connect multiple 'consumers' to the same uprobe.
Each consumer defines a handler and a filter (optional). The
'handler' is run every time the uprobe is hit, if it matches the
'filter' criteria.
The first consumer of a uprobe causes the breakpoint to be
inserted at the specified address and subsequent consumers are
appended to this list. On subsequent probes, the consumer gets
appended to the existing list of consumers. The breakpoint is
removed when the last consumer unregisters. For all other
unregisterations, the consumer is removed from the list of
consumers.
Given a inode, we get a list of the mms that have mapped the
inode. Do the actual registration if mm maps the page where a
probe needs to be inserted/removed.
We use a temporary list to walk through the vmas that map the
inode.
- The number of maps that map the inode, is not known before we
walk the rmap and keeps changing.
- extending vm_area_struct wasn't recommended, it's a
size-critical data structure.
- There can be more than one maps of the inode in the same mm.
We add callbacks to the mmap methods to keep an eye on text vmas
that are of interest to uprobes. When a vma of interest is mapped,
we insert the breakpoint at the right address.
Uprobe works by replacing the instruction at the address defined
by (inode, offset) with the arch specific breakpoint
instruction. We save a copy of the original instruction at the
uprobed address.
This is needed for:
a. executing the instruction out-of-line (xol).
b. instruction analysis for any subsequent fixups.
c. restoring the instruction back when the uprobe is unregistered.
We insert or delete a breakpoint instruction, and this
breakpoint instruction is assumed to be the smallest instruction
available on the platform. For fixed size instruction platforms
this is trivially true, for variable size instruction platforms
the breakpoint instruction is typically the smallest (often a
single byte).
Writing the instruction is done by COWing the page and changing
the instruction during the copy, this even though most platforms
allow atomic writes of the breakpoint instruction. This also
mirrors the behaviour of a ptrace() memory write to a PRIVATE
file map.
The core worker is derived from KSM's replace_page() logic.
In essence, similar to KSM:
a. allocate a new page and copy over contents of the page that
has the uprobed vaddr
b. modify the copy and insert the breakpoint at the required
address
c. switch the original page with the copy containing the
breakpoint
d. flush page tables.
replace_page() is being replicated here because of some minor
changes in the type of pages and also because Hugh Dickins had
plans to improve replace_page() for KSM specific work.
Instruction analysis on x86 is based on instruction decoder and
determines if an instruction can be probed and determines the
necessary fixups after singlestep. Instruction analysis is done
at probe insertion time so that we avoid having to repeat the
same analysis every time a probe is hit.
A lot of code here is due to the improvement/suggestions/inputs
from Peter Zijlstra.
Changelog:
(v10):
- Add code to clear REX.B prefix as suggested by Denys Vlasenko
and Masami Hiramatsu.
(v9):
- Use insn_offset_modrm as suggested by Masami Hiramatsu.
(v7):
Handle comments from Peter Zijlstra:
- Dont take reference to inode. (expect inode to uprobe_register to be sane).
- Use PTR_ERR to set the return value.
- No need to take reference to inode.
- use PTR_ERR to return error value.
- register and uprobe_unregister share code.
(v5):
- Modified del_consumer as per comments from Peter.
- Drop reference to inode before dropping reference to uprobe.
- Use i_size_read(inode) instead of inode->i_size.
- Ensure uprobe->consumers is NULL, before __uprobe_unregister() is called.
- Includes errno.h as recommended by Stephen Rothwell to fix a build issue
on sparc defconfig
- Remove restrictions while unregistering.
- Earlier code leaked inode references under some conditions while
registering/unregistering.
- Continue the vma-rmap walk even if the intermediate vma doesnt
meet the requirements.
- Validate the vma found by find_vma before inserting/removing the
breakpoint
- Call del_consumer under mutex_lock.
- Use hash locks.
- Handle mremap.
- Introduce find_least_offset_node() instead of close match logic in
find_uprobe
- Uprobes no more depends on MM_OWNER; No reference to task_structs
while inserting/removing a probe.
- Uses read_mapping_page instead of grab_cache_page so that the pages
have valid content.
- pass NULL to get_user_pages for the task parameter.
- call SetPageUptodate on the new page allocated in write_opcode.
- fix leaking a reference to the new page under certain conditions.
- Include Instruction Decoder if Uprobes gets defined.
- Remove const attributes for instruction prefix arrays.
- Uses mm_context to know if the application is 32 bit.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Also-written-by: Jim Keniston <jkenisto@us.ibm.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Roland McGrath <roland@hack.frob.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Anton Arapov <anton@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Denys Vlasenko <vda.linux@googlemail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linux-mm <linux-mm@kvack.org>
Link: http://lkml.kernel.org/r/20120209092642.GE16600@linux.vnet.ibm.com
[ Made various small edits to the commit log ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-02-09 17:26:42 +08:00
|
|
|
init_rwsem(&uprobe->consumer_rwsem);
|
|
|
|
|
|
|
|
/* add to uprobes_tree, sorted on inode:offset */
|
|
|
|
cur_uprobe = insert_uprobe(uprobe);
|
|
|
|
/* a uprobe exists for this inode:offset combination */
|
|
|
|
if (cur_uprobe) {
|
2018-08-20 12:42:48 +08:00
|
|
|
if (cur_uprobe->ref_ctr_offset != uprobe->ref_ctr_offset) {
|
|
|
|
ref_ctr_mismatch_warn(cur_uprobe, uprobe);
|
|
|
|
put_uprobe(cur_uprobe);
|
|
|
|
kfree(uprobe);
|
|
|
|
return ERR_PTR(-EINVAL);
|
|
|
|
}
|
uprobes, mm, x86: Add the ability to install and remove uprobes breakpoints
Add uprobes support to the core kernel, with x86 support.
This commit adds the kernel facilities, the actual uprobes
user-space ABI and perf probe support comes in later commits.
General design:
Uprobes are maintained in an rb-tree indexed by inode and offset
(the offset here is from the start of the mapping). For a unique
(inode, offset) tuple, there can be at most one uprobe in the
rb-tree.
Since the (inode, offset) tuple identifies a unique uprobe, more
than one user may be interested in the same uprobe. This provides
the ability to connect multiple 'consumers' to the same uprobe.
Each consumer defines a handler and a filter (optional). The
'handler' is run every time the uprobe is hit, if it matches the
'filter' criteria.
The first consumer of a uprobe causes the breakpoint to be
inserted at the specified address and subsequent consumers are
appended to this list. On subsequent probes, the consumer gets
appended to the existing list of consumers. The breakpoint is
removed when the last consumer unregisters. For all other
unregisterations, the consumer is removed from the list of
consumers.
Given a inode, we get a list of the mms that have mapped the
inode. Do the actual registration if mm maps the page where a
probe needs to be inserted/removed.
We use a temporary list to walk through the vmas that map the
inode.
- The number of maps that map the inode, is not known before we
walk the rmap and keeps changing.
- extending vm_area_struct wasn't recommended, it's a
size-critical data structure.
- There can be more than one maps of the inode in the same mm.
We add callbacks to the mmap methods to keep an eye on text vmas
that are of interest to uprobes. When a vma of interest is mapped,
we insert the breakpoint at the right address.
Uprobe works by replacing the instruction at the address defined
by (inode, offset) with the arch specific breakpoint
instruction. We save a copy of the original instruction at the
uprobed address.
This is needed for:
a. executing the instruction out-of-line (xol).
b. instruction analysis for any subsequent fixups.
c. restoring the instruction back when the uprobe is unregistered.
We insert or delete a breakpoint instruction, and this
breakpoint instruction is assumed to be the smallest instruction
available on the platform. For fixed size instruction platforms
this is trivially true, for variable size instruction platforms
the breakpoint instruction is typically the smallest (often a
single byte).
Writing the instruction is done by COWing the page and changing
the instruction during the copy, this even though most platforms
allow atomic writes of the breakpoint instruction. This also
mirrors the behaviour of a ptrace() memory write to a PRIVATE
file map.
The core worker is derived from KSM's replace_page() logic.
In essence, similar to KSM:
a. allocate a new page and copy over contents of the page that
has the uprobed vaddr
b. modify the copy and insert the breakpoint at the required
address
c. switch the original page with the copy containing the
breakpoint
d. flush page tables.
replace_page() is being replicated here because of some minor
changes in the type of pages and also because Hugh Dickins had
plans to improve replace_page() for KSM specific work.
Instruction analysis on x86 is based on instruction decoder and
determines if an instruction can be probed and determines the
necessary fixups after singlestep. Instruction analysis is done
at probe insertion time so that we avoid having to repeat the
same analysis every time a probe is hit.
A lot of code here is due to the improvement/suggestions/inputs
from Peter Zijlstra.
Changelog:
(v10):
- Add code to clear REX.B prefix as suggested by Denys Vlasenko
and Masami Hiramatsu.
(v9):
- Use insn_offset_modrm as suggested by Masami Hiramatsu.
(v7):
Handle comments from Peter Zijlstra:
- Dont take reference to inode. (expect inode to uprobe_register to be sane).
- Use PTR_ERR to set the return value.
- No need to take reference to inode.
- use PTR_ERR to return error value.
- register and uprobe_unregister share code.
(v5):
- Modified del_consumer as per comments from Peter.
- Drop reference to inode before dropping reference to uprobe.
- Use i_size_read(inode) instead of inode->i_size.
- Ensure uprobe->consumers is NULL, before __uprobe_unregister() is called.
- Includes errno.h as recommended by Stephen Rothwell to fix a build issue
on sparc defconfig
- Remove restrictions while unregistering.
- Earlier code leaked inode references under some conditions while
registering/unregistering.
- Continue the vma-rmap walk even if the intermediate vma doesnt
meet the requirements.
- Validate the vma found by find_vma before inserting/removing the
breakpoint
- Call del_consumer under mutex_lock.
- Use hash locks.
- Handle mremap.
- Introduce find_least_offset_node() instead of close match logic in
find_uprobe
- Uprobes no more depends on MM_OWNER; No reference to task_structs
while inserting/removing a probe.
- Uses read_mapping_page instead of grab_cache_page so that the pages
have valid content.
- pass NULL to get_user_pages for the task parameter.
- call SetPageUptodate on the new page allocated in write_opcode.
- fix leaking a reference to the new page under certain conditions.
- Include Instruction Decoder if Uprobes gets defined.
- Remove const attributes for instruction prefix arrays.
- Uses mm_context to know if the application is 32 bit.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Also-written-by: Jim Keniston <jkenisto@us.ibm.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Roland McGrath <roland@hack.frob.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Anton Arapov <anton@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Denys Vlasenko <vda.linux@googlemail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linux-mm <linux-mm@kvack.org>
Link: http://lkml.kernel.org/r/20120209092642.GE16600@linux.vnet.ibm.com
[ Made various small edits to the commit log ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-02-09 17:26:42 +08:00
|
|
|
kfree(uprobe);
|
|
|
|
uprobe = cur_uprobe;
|
2012-02-17 16:27:41 +08:00
|
|
|
}
|
|
|
|
|
uprobes, mm, x86: Add the ability to install and remove uprobes breakpoints
Add uprobes support to the core kernel, with x86 support.
This commit adds the kernel facilities, the actual uprobes
user-space ABI and perf probe support comes in later commits.
General design:
Uprobes are maintained in an rb-tree indexed by inode and offset
(the offset here is from the start of the mapping). For a unique
(inode, offset) tuple, there can be at most one uprobe in the
rb-tree.
Since the (inode, offset) tuple identifies a unique uprobe, more
than one user may be interested in the same uprobe. This provides
the ability to connect multiple 'consumers' to the same uprobe.
Each consumer defines a handler and a filter (optional). The
'handler' is run every time the uprobe is hit, if it matches the
'filter' criteria.
The first consumer of a uprobe causes the breakpoint to be
inserted at the specified address and subsequent consumers are
appended to this list. On subsequent probes, the consumer gets
appended to the existing list of consumers. The breakpoint is
removed when the last consumer unregisters. For all other
unregisterations, the consumer is removed from the list of
consumers.
Given a inode, we get a list of the mms that have mapped the
inode. Do the actual registration if mm maps the page where a
probe needs to be inserted/removed.
We use a temporary list to walk through the vmas that map the
inode.
- The number of maps that map the inode, is not known before we
walk the rmap and keeps changing.
- extending vm_area_struct wasn't recommended, it's a
size-critical data structure.
- There can be more than one maps of the inode in the same mm.
We add callbacks to the mmap methods to keep an eye on text vmas
that are of interest to uprobes. When a vma of interest is mapped,
we insert the breakpoint at the right address.
Uprobe works by replacing the instruction at the address defined
by (inode, offset) with the arch specific breakpoint
instruction. We save a copy of the original instruction at the
uprobed address.
This is needed for:
a. executing the instruction out-of-line (xol).
b. instruction analysis for any subsequent fixups.
c. restoring the instruction back when the uprobe is unregistered.
We insert or delete a breakpoint instruction, and this
breakpoint instruction is assumed to be the smallest instruction
available on the platform. For fixed size instruction platforms
this is trivially true, for variable size instruction platforms
the breakpoint instruction is typically the smallest (often a
single byte).
Writing the instruction is done by COWing the page and changing
the instruction during the copy, this even though most platforms
allow atomic writes of the breakpoint instruction. This also
mirrors the behaviour of a ptrace() memory write to a PRIVATE
file map.
The core worker is derived from KSM's replace_page() logic.
In essence, similar to KSM:
a. allocate a new page and copy over contents of the page that
has the uprobed vaddr
b. modify the copy and insert the breakpoint at the required
address
c. switch the original page with the copy containing the
breakpoint
d. flush page tables.
replace_page() is being replicated here because of some minor
changes in the type of pages and also because Hugh Dickins had
plans to improve replace_page() for KSM specific work.
Instruction analysis on x86 is based on instruction decoder and
determines if an instruction can be probed and determines the
necessary fixups after singlestep. Instruction analysis is done
at probe insertion time so that we avoid having to repeat the
same analysis every time a probe is hit.
A lot of code here is due to the improvement/suggestions/inputs
from Peter Zijlstra.
Changelog:
(v10):
- Add code to clear REX.B prefix as suggested by Denys Vlasenko
and Masami Hiramatsu.
(v9):
- Use insn_offset_modrm as suggested by Masami Hiramatsu.
(v7):
Handle comments from Peter Zijlstra:
- Dont take reference to inode. (expect inode to uprobe_register to be sane).
- Use PTR_ERR to set the return value.
- No need to take reference to inode.
- use PTR_ERR to return error value.
- register and uprobe_unregister share code.
(v5):
- Modified del_consumer as per comments from Peter.
- Drop reference to inode before dropping reference to uprobe.
- Use i_size_read(inode) instead of inode->i_size.
- Ensure uprobe->consumers is NULL, before __uprobe_unregister() is called.
- Includes errno.h as recommended by Stephen Rothwell to fix a build issue
on sparc defconfig
- Remove restrictions while unregistering.
- Earlier code leaked inode references under some conditions while
registering/unregistering.
- Continue the vma-rmap walk even if the intermediate vma doesnt
meet the requirements.
- Validate the vma found by find_vma before inserting/removing the
breakpoint
- Call del_consumer under mutex_lock.
- Use hash locks.
- Handle mremap.
- Introduce find_least_offset_node() instead of close match logic in
find_uprobe
- Uprobes no more depends on MM_OWNER; No reference to task_structs
while inserting/removing a probe.
- Uses read_mapping_page instead of grab_cache_page so that the pages
have valid content.
- pass NULL to get_user_pages for the task parameter.
- call SetPageUptodate on the new page allocated in write_opcode.
- fix leaking a reference to the new page under certain conditions.
- Include Instruction Decoder if Uprobes gets defined.
- Remove const attributes for instruction prefix arrays.
- Uses mm_context to know if the application is 32 bit.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Also-written-by: Jim Keniston <jkenisto@us.ibm.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Roland McGrath <roland@hack.frob.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Anton Arapov <anton@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Denys Vlasenko <vda.linux@googlemail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linux-mm <linux-mm@kvack.org>
Link: http://lkml.kernel.org/r/20120209092642.GE16600@linux.vnet.ibm.com
[ Made various small edits to the commit log ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-02-09 17:26:42 +08:00
|
|
|
return uprobe;
|
|
|
|
}
|
|
|
|
|
2012-11-24 03:15:17 +08:00
|
|
|
static void consumer_add(struct uprobe *uprobe, struct uprobe_consumer *uc)
|
uprobes, mm, x86: Add the ability to install and remove uprobes breakpoints
Add uprobes support to the core kernel, with x86 support.
This commit adds the kernel facilities, the actual uprobes
user-space ABI and perf probe support comes in later commits.
General design:
Uprobes are maintained in an rb-tree indexed by inode and offset
(the offset here is from the start of the mapping). For a unique
(inode, offset) tuple, there can be at most one uprobe in the
rb-tree.
Since the (inode, offset) tuple identifies a unique uprobe, more
than one user may be interested in the same uprobe. This provides
the ability to connect multiple 'consumers' to the same uprobe.
Each consumer defines a handler and a filter (optional). The
'handler' is run every time the uprobe is hit, if it matches the
'filter' criteria.
The first consumer of a uprobe causes the breakpoint to be
inserted at the specified address and subsequent consumers are
appended to this list. On subsequent probes, the consumer gets
appended to the existing list of consumers. The breakpoint is
removed when the last consumer unregisters. For all other
unregisterations, the consumer is removed from the list of
consumers.
Given a inode, we get a list of the mms that have mapped the
inode. Do the actual registration if mm maps the page where a
probe needs to be inserted/removed.
We use a temporary list to walk through the vmas that map the
inode.
- The number of maps that map the inode, is not known before we
walk the rmap and keeps changing.
- extending vm_area_struct wasn't recommended, it's a
size-critical data structure.
- There can be more than one maps of the inode in the same mm.
We add callbacks to the mmap methods to keep an eye on text vmas
that are of interest to uprobes. When a vma of interest is mapped,
we insert the breakpoint at the right address.
Uprobe works by replacing the instruction at the address defined
by (inode, offset) with the arch specific breakpoint
instruction. We save a copy of the original instruction at the
uprobed address.
This is needed for:
a. executing the instruction out-of-line (xol).
b. instruction analysis for any subsequent fixups.
c. restoring the instruction back when the uprobe is unregistered.
We insert or delete a breakpoint instruction, and this
breakpoint instruction is assumed to be the smallest instruction
available on the platform. For fixed size instruction platforms
this is trivially true, for variable size instruction platforms
the breakpoint instruction is typically the smallest (often a
single byte).
Writing the instruction is done by COWing the page and changing
the instruction during the copy, this even though most platforms
allow atomic writes of the breakpoint instruction. This also
mirrors the behaviour of a ptrace() memory write to a PRIVATE
file map.
The core worker is derived from KSM's replace_page() logic.
In essence, similar to KSM:
a. allocate a new page and copy over contents of the page that
has the uprobed vaddr
b. modify the copy and insert the breakpoint at the required
address
c. switch the original page with the copy containing the
breakpoint
d. flush page tables.
replace_page() is being replicated here because of some minor
changes in the type of pages and also because Hugh Dickins had
plans to improve replace_page() for KSM specific work.
Instruction analysis on x86 is based on instruction decoder and
determines if an instruction can be probed and determines the
necessary fixups after singlestep. Instruction analysis is done
at probe insertion time so that we avoid having to repeat the
same analysis every time a probe is hit.
A lot of code here is due to the improvement/suggestions/inputs
from Peter Zijlstra.
Changelog:
(v10):
- Add code to clear REX.B prefix as suggested by Denys Vlasenko
and Masami Hiramatsu.
(v9):
- Use insn_offset_modrm as suggested by Masami Hiramatsu.
(v7):
Handle comments from Peter Zijlstra:
- Dont take reference to inode. (expect inode to uprobe_register to be sane).
- Use PTR_ERR to set the return value.
- No need to take reference to inode.
- use PTR_ERR to return error value.
- register and uprobe_unregister share code.
(v5):
- Modified del_consumer as per comments from Peter.
- Drop reference to inode before dropping reference to uprobe.
- Use i_size_read(inode) instead of inode->i_size.
- Ensure uprobe->consumers is NULL, before __uprobe_unregister() is called.
- Includes errno.h as recommended by Stephen Rothwell to fix a build issue
on sparc defconfig
- Remove restrictions while unregistering.
- Earlier code leaked inode references under some conditions while
registering/unregistering.
- Continue the vma-rmap walk even if the intermediate vma doesnt
meet the requirements.
- Validate the vma found by find_vma before inserting/removing the
breakpoint
- Call del_consumer under mutex_lock.
- Use hash locks.
- Handle mremap.
- Introduce find_least_offset_node() instead of close match logic in
find_uprobe
- Uprobes no more depends on MM_OWNER; No reference to task_structs
while inserting/removing a probe.
- Uses read_mapping_page instead of grab_cache_page so that the pages
have valid content.
- pass NULL to get_user_pages for the task parameter.
- call SetPageUptodate on the new page allocated in write_opcode.
- fix leaking a reference to the new page under certain conditions.
- Include Instruction Decoder if Uprobes gets defined.
- Remove const attributes for instruction prefix arrays.
- Uses mm_context to know if the application is 32 bit.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Also-written-by: Jim Keniston <jkenisto@us.ibm.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Roland McGrath <roland@hack.frob.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Anton Arapov <anton@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Denys Vlasenko <vda.linux@googlemail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linux-mm <linux-mm@kvack.org>
Link: http://lkml.kernel.org/r/20120209092642.GE16600@linux.vnet.ibm.com
[ Made various small edits to the commit log ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-02-09 17:26:42 +08:00
|
|
|
{
|
|
|
|
down_write(&uprobe->consumer_rwsem);
|
2012-03-12 17:25:30 +08:00
|
|
|
uc->next = uprobe->consumers;
|
|
|
|
uprobe->consumers = uc;
|
uprobes, mm, x86: Add the ability to install and remove uprobes breakpoints
Add uprobes support to the core kernel, with x86 support.
This commit adds the kernel facilities, the actual uprobes
user-space ABI and perf probe support comes in later commits.
General design:
Uprobes are maintained in an rb-tree indexed by inode and offset
(the offset here is from the start of the mapping). For a unique
(inode, offset) tuple, there can be at most one uprobe in the
rb-tree.
Since the (inode, offset) tuple identifies a unique uprobe, more
than one user may be interested in the same uprobe. This provides
the ability to connect multiple 'consumers' to the same uprobe.
Each consumer defines a handler and a filter (optional). The
'handler' is run every time the uprobe is hit, if it matches the
'filter' criteria.
The first consumer of a uprobe causes the breakpoint to be
inserted at the specified address and subsequent consumers are
appended to this list. On subsequent probes, the consumer gets
appended to the existing list of consumers. The breakpoint is
removed when the last consumer unregisters. For all other
unregisterations, the consumer is removed from the list of
consumers.
Given a inode, we get a list of the mms that have mapped the
inode. Do the actual registration if mm maps the page where a
probe needs to be inserted/removed.
We use a temporary list to walk through the vmas that map the
inode.
- The number of maps that map the inode, is not known before we
walk the rmap and keeps changing.
- extending vm_area_struct wasn't recommended, it's a
size-critical data structure.
- There can be more than one maps of the inode in the same mm.
We add callbacks to the mmap methods to keep an eye on text vmas
that are of interest to uprobes. When a vma of interest is mapped,
we insert the breakpoint at the right address.
Uprobe works by replacing the instruction at the address defined
by (inode, offset) with the arch specific breakpoint
instruction. We save a copy of the original instruction at the
uprobed address.
This is needed for:
a. executing the instruction out-of-line (xol).
b. instruction analysis for any subsequent fixups.
c. restoring the instruction back when the uprobe is unregistered.
We insert or delete a breakpoint instruction, and this
breakpoint instruction is assumed to be the smallest instruction
available on the platform. For fixed size instruction platforms
this is trivially true, for variable size instruction platforms
the breakpoint instruction is typically the smallest (often a
single byte).
Writing the instruction is done by COWing the page and changing
the instruction during the copy, this even though most platforms
allow atomic writes of the breakpoint instruction. This also
mirrors the behaviour of a ptrace() memory write to a PRIVATE
file map.
The core worker is derived from KSM's replace_page() logic.
In essence, similar to KSM:
a. allocate a new page and copy over contents of the page that
has the uprobed vaddr
b. modify the copy and insert the breakpoint at the required
address
c. switch the original page with the copy containing the
breakpoint
d. flush page tables.
replace_page() is being replicated here because of some minor
changes in the type of pages and also because Hugh Dickins had
plans to improve replace_page() for KSM specific work.
Instruction analysis on x86 is based on instruction decoder and
determines if an instruction can be probed and determines the
necessary fixups after singlestep. Instruction analysis is done
at probe insertion time so that we avoid having to repeat the
same analysis every time a probe is hit.
A lot of code here is due to the improvement/suggestions/inputs
from Peter Zijlstra.
Changelog:
(v10):
- Add code to clear REX.B prefix as suggested by Denys Vlasenko
and Masami Hiramatsu.
(v9):
- Use insn_offset_modrm as suggested by Masami Hiramatsu.
(v7):
Handle comments from Peter Zijlstra:
- Dont take reference to inode. (expect inode to uprobe_register to be sane).
- Use PTR_ERR to set the return value.
- No need to take reference to inode.
- use PTR_ERR to return error value.
- register and uprobe_unregister share code.
(v5):
- Modified del_consumer as per comments from Peter.
- Drop reference to inode before dropping reference to uprobe.
- Use i_size_read(inode) instead of inode->i_size.
- Ensure uprobe->consumers is NULL, before __uprobe_unregister() is called.
- Includes errno.h as recommended by Stephen Rothwell to fix a build issue
on sparc defconfig
- Remove restrictions while unregistering.
- Earlier code leaked inode references under some conditions while
registering/unregistering.
- Continue the vma-rmap walk even if the intermediate vma doesnt
meet the requirements.
- Validate the vma found by find_vma before inserting/removing the
breakpoint
- Call del_consumer under mutex_lock.
- Use hash locks.
- Handle mremap.
- Introduce find_least_offset_node() instead of close match logic in
find_uprobe
- Uprobes no more depends on MM_OWNER; No reference to task_structs
while inserting/removing a probe.
- Uses read_mapping_page instead of grab_cache_page so that the pages
have valid content.
- pass NULL to get_user_pages for the task parameter.
- call SetPageUptodate on the new page allocated in write_opcode.
- fix leaking a reference to the new page under certain conditions.
- Include Instruction Decoder if Uprobes gets defined.
- Remove const attributes for instruction prefix arrays.
- Uses mm_context to know if the application is 32 bit.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Also-written-by: Jim Keniston <jkenisto@us.ibm.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Roland McGrath <roland@hack.frob.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Anton Arapov <anton@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Denys Vlasenko <vda.linux@googlemail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linux-mm <linux-mm@kvack.org>
Link: http://lkml.kernel.org/r/20120209092642.GE16600@linux.vnet.ibm.com
[ Made various small edits to the commit log ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-02-09 17:26:42 +08:00
|
|
|
up_write(&uprobe->consumer_rwsem);
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
2012-03-12 17:25:30 +08:00
|
|
|
* For uprobe @uprobe, delete the consumer @uc.
|
|
|
|
* Return true if the @uc is deleted successfully
|
uprobes, mm, x86: Add the ability to install and remove uprobes breakpoints
Add uprobes support to the core kernel, with x86 support.
This commit adds the kernel facilities, the actual uprobes
user-space ABI and perf probe support comes in later commits.
General design:
Uprobes are maintained in an rb-tree indexed by inode and offset
(the offset here is from the start of the mapping). For a unique
(inode, offset) tuple, there can be at most one uprobe in the
rb-tree.
Since the (inode, offset) tuple identifies a unique uprobe, more
than one user may be interested in the same uprobe. This provides
the ability to connect multiple 'consumers' to the same uprobe.
Each consumer defines a handler and a filter (optional). The
'handler' is run every time the uprobe is hit, if it matches the
'filter' criteria.
The first consumer of a uprobe causes the breakpoint to be
inserted at the specified address and subsequent consumers are
appended to this list. On subsequent probes, the consumer gets
appended to the existing list of consumers. The breakpoint is
removed when the last consumer unregisters. For all other
unregisterations, the consumer is removed from the list of
consumers.
Given a inode, we get a list of the mms that have mapped the
inode. Do the actual registration if mm maps the page where a
probe needs to be inserted/removed.
We use a temporary list to walk through the vmas that map the
inode.
- The number of maps that map the inode, is not known before we
walk the rmap and keeps changing.
- extending vm_area_struct wasn't recommended, it's a
size-critical data structure.
- There can be more than one maps of the inode in the same mm.
We add callbacks to the mmap methods to keep an eye on text vmas
that are of interest to uprobes. When a vma of interest is mapped,
we insert the breakpoint at the right address.
Uprobe works by replacing the instruction at the address defined
by (inode, offset) with the arch specific breakpoint
instruction. We save a copy of the original instruction at the
uprobed address.
This is needed for:
a. executing the instruction out-of-line (xol).
b. instruction analysis for any subsequent fixups.
c. restoring the instruction back when the uprobe is unregistered.
We insert or delete a breakpoint instruction, and this
breakpoint instruction is assumed to be the smallest instruction
available on the platform. For fixed size instruction platforms
this is trivially true, for variable size instruction platforms
the breakpoint instruction is typically the smallest (often a
single byte).
Writing the instruction is done by COWing the page and changing
the instruction during the copy, this even though most platforms
allow atomic writes of the breakpoint instruction. This also
mirrors the behaviour of a ptrace() memory write to a PRIVATE
file map.
The core worker is derived from KSM's replace_page() logic.
In essence, similar to KSM:
a. allocate a new page and copy over contents of the page that
has the uprobed vaddr
b. modify the copy and insert the breakpoint at the required
address
c. switch the original page with the copy containing the
breakpoint
d. flush page tables.
replace_page() is being replicated here because of some minor
changes in the type of pages and also because Hugh Dickins had
plans to improve replace_page() for KSM specific work.
Instruction analysis on x86 is based on instruction decoder and
determines if an instruction can be probed and determines the
necessary fixups after singlestep. Instruction analysis is done
at probe insertion time so that we avoid having to repeat the
same analysis every time a probe is hit.
A lot of code here is due to the improvement/suggestions/inputs
from Peter Zijlstra.
Changelog:
(v10):
- Add code to clear REX.B prefix as suggested by Denys Vlasenko
and Masami Hiramatsu.
(v9):
- Use insn_offset_modrm as suggested by Masami Hiramatsu.
(v7):
Handle comments from Peter Zijlstra:
- Dont take reference to inode. (expect inode to uprobe_register to be sane).
- Use PTR_ERR to set the return value.
- No need to take reference to inode.
- use PTR_ERR to return error value.
- register and uprobe_unregister share code.
(v5):
- Modified del_consumer as per comments from Peter.
- Drop reference to inode before dropping reference to uprobe.
- Use i_size_read(inode) instead of inode->i_size.
- Ensure uprobe->consumers is NULL, before __uprobe_unregister() is called.
- Includes errno.h as recommended by Stephen Rothwell to fix a build issue
on sparc defconfig
- Remove restrictions while unregistering.
- Earlier code leaked inode references under some conditions while
registering/unregistering.
- Continue the vma-rmap walk even if the intermediate vma doesnt
meet the requirements.
- Validate the vma found by find_vma before inserting/removing the
breakpoint
- Call del_consumer under mutex_lock.
- Use hash locks.
- Handle mremap.
- Introduce find_least_offset_node() instead of close match logic in
find_uprobe
- Uprobes no more depends on MM_OWNER; No reference to task_structs
while inserting/removing a probe.
- Uses read_mapping_page instead of grab_cache_page so that the pages
have valid content.
- pass NULL to get_user_pages for the task parameter.
- call SetPageUptodate on the new page allocated in write_opcode.
- fix leaking a reference to the new page under certain conditions.
- Include Instruction Decoder if Uprobes gets defined.
- Remove const attributes for instruction prefix arrays.
- Uses mm_context to know if the application is 32 bit.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Also-written-by: Jim Keniston <jkenisto@us.ibm.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Roland McGrath <roland@hack.frob.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Anton Arapov <anton@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Denys Vlasenko <vda.linux@googlemail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linux-mm <linux-mm@kvack.org>
Link: http://lkml.kernel.org/r/20120209092642.GE16600@linux.vnet.ibm.com
[ Made various small edits to the commit log ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-02-09 17:26:42 +08:00
|
|
|
* or return false.
|
|
|
|
*/
|
2012-03-12 17:25:30 +08:00
|
|
|
static bool consumer_del(struct uprobe *uprobe, struct uprobe_consumer *uc)
|
uprobes, mm, x86: Add the ability to install and remove uprobes breakpoints
Add uprobes support to the core kernel, with x86 support.
This commit adds the kernel facilities, the actual uprobes
user-space ABI and perf probe support comes in later commits.
General design:
Uprobes are maintained in an rb-tree indexed by inode and offset
(the offset here is from the start of the mapping). For a unique
(inode, offset) tuple, there can be at most one uprobe in the
rb-tree.
Since the (inode, offset) tuple identifies a unique uprobe, more
than one user may be interested in the same uprobe. This provides
the ability to connect multiple 'consumers' to the same uprobe.
Each consumer defines a handler and a filter (optional). The
'handler' is run every time the uprobe is hit, if it matches the
'filter' criteria.
The first consumer of a uprobe causes the breakpoint to be
inserted at the specified address and subsequent consumers are
appended to this list. On subsequent probes, the consumer gets
appended to the existing list of consumers. The breakpoint is
removed when the last consumer unregisters. For all other
unregisterations, the consumer is removed from the list of
consumers.
Given a inode, we get a list of the mms that have mapped the
inode. Do the actual registration if mm maps the page where a
probe needs to be inserted/removed.
We use a temporary list to walk through the vmas that map the
inode.
- The number of maps that map the inode, is not known before we
walk the rmap and keeps changing.
- extending vm_area_struct wasn't recommended, it's a
size-critical data structure.
- There can be more than one maps of the inode in the same mm.
We add callbacks to the mmap methods to keep an eye on text vmas
that are of interest to uprobes. When a vma of interest is mapped,
we insert the breakpoint at the right address.
Uprobe works by replacing the instruction at the address defined
by (inode, offset) with the arch specific breakpoint
instruction. We save a copy of the original instruction at the
uprobed address.
This is needed for:
a. executing the instruction out-of-line (xol).
b. instruction analysis for any subsequent fixups.
c. restoring the instruction back when the uprobe is unregistered.
We insert or delete a breakpoint instruction, and this
breakpoint instruction is assumed to be the smallest instruction
available on the platform. For fixed size instruction platforms
this is trivially true, for variable size instruction platforms
the breakpoint instruction is typically the smallest (often a
single byte).
Writing the instruction is done by COWing the page and changing
the instruction during the copy, this even though most platforms
allow atomic writes of the breakpoint instruction. This also
mirrors the behaviour of a ptrace() memory write to a PRIVATE
file map.
The core worker is derived from KSM's replace_page() logic.
In essence, similar to KSM:
a. allocate a new page and copy over contents of the page that
has the uprobed vaddr
b. modify the copy and insert the breakpoint at the required
address
c. switch the original page with the copy containing the
breakpoint
d. flush page tables.
replace_page() is being replicated here because of some minor
changes in the type of pages and also because Hugh Dickins had
plans to improve replace_page() for KSM specific work.
Instruction analysis on x86 is based on instruction decoder and
determines if an instruction can be probed and determines the
necessary fixups after singlestep. Instruction analysis is done
at probe insertion time so that we avoid having to repeat the
same analysis every time a probe is hit.
A lot of code here is due to the improvement/suggestions/inputs
from Peter Zijlstra.
Changelog:
(v10):
- Add code to clear REX.B prefix as suggested by Denys Vlasenko
and Masami Hiramatsu.
(v9):
- Use insn_offset_modrm as suggested by Masami Hiramatsu.
(v7):
Handle comments from Peter Zijlstra:
- Dont take reference to inode. (expect inode to uprobe_register to be sane).
- Use PTR_ERR to set the return value.
- No need to take reference to inode.
- use PTR_ERR to return error value.
- register and uprobe_unregister share code.
(v5):
- Modified del_consumer as per comments from Peter.
- Drop reference to inode before dropping reference to uprobe.
- Use i_size_read(inode) instead of inode->i_size.
- Ensure uprobe->consumers is NULL, before __uprobe_unregister() is called.
- Includes errno.h as recommended by Stephen Rothwell to fix a build issue
on sparc defconfig
- Remove restrictions while unregistering.
- Earlier code leaked inode references under some conditions while
registering/unregistering.
- Continue the vma-rmap walk even if the intermediate vma doesnt
meet the requirements.
- Validate the vma found by find_vma before inserting/removing the
breakpoint
- Call del_consumer under mutex_lock.
- Use hash locks.
- Handle mremap.
- Introduce find_least_offset_node() instead of close match logic in
find_uprobe
- Uprobes no more depends on MM_OWNER; No reference to task_structs
while inserting/removing a probe.
- Uses read_mapping_page instead of grab_cache_page so that the pages
have valid content.
- pass NULL to get_user_pages for the task parameter.
- call SetPageUptodate on the new page allocated in write_opcode.
- fix leaking a reference to the new page under certain conditions.
- Include Instruction Decoder if Uprobes gets defined.
- Remove const attributes for instruction prefix arrays.
- Uses mm_context to know if the application is 32 bit.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Also-written-by: Jim Keniston <jkenisto@us.ibm.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Roland McGrath <roland@hack.frob.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Anton Arapov <anton@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Denys Vlasenko <vda.linux@googlemail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linux-mm <linux-mm@kvack.org>
Link: http://lkml.kernel.org/r/20120209092642.GE16600@linux.vnet.ibm.com
[ Made various small edits to the commit log ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-02-09 17:26:42 +08:00
|
|
|
{
|
|
|
|
struct uprobe_consumer **con;
|
|
|
|
bool ret = false;
|
|
|
|
|
|
|
|
down_write(&uprobe->consumer_rwsem);
|
|
|
|
for (con = &uprobe->consumers; *con; con = &(*con)->next) {
|
2012-03-12 17:25:30 +08:00
|
|
|
if (*con == uc) {
|
|
|
|
*con = uc->next;
|
uprobes, mm, x86: Add the ability to install and remove uprobes breakpoints
Add uprobes support to the core kernel, with x86 support.
This commit adds the kernel facilities, the actual uprobes
user-space ABI and perf probe support comes in later commits.
General design:
Uprobes are maintained in an rb-tree indexed by inode and offset
(the offset here is from the start of the mapping). For a unique
(inode, offset) tuple, there can be at most one uprobe in the
rb-tree.
Since the (inode, offset) tuple identifies a unique uprobe, more
than one user may be interested in the same uprobe. This provides
the ability to connect multiple 'consumers' to the same uprobe.
Each consumer defines a handler and a filter (optional). The
'handler' is run every time the uprobe is hit, if it matches the
'filter' criteria.
The first consumer of a uprobe causes the breakpoint to be
inserted at the specified address and subsequent consumers are
appended to this list. On subsequent probes, the consumer gets
appended to the existing list of consumers. The breakpoint is
removed when the last consumer unregisters. For all other
unregisterations, the consumer is removed from the list of
consumers.
Given a inode, we get a list of the mms that have mapped the
inode. Do the actual registration if mm maps the page where a
probe needs to be inserted/removed.
We use a temporary list to walk through the vmas that map the
inode.
- The number of maps that map the inode, is not known before we
walk the rmap and keeps changing.
- extending vm_area_struct wasn't recommended, it's a
size-critical data structure.
- There can be more than one maps of the inode in the same mm.
We add callbacks to the mmap methods to keep an eye on text vmas
that are of interest to uprobes. When a vma of interest is mapped,
we insert the breakpoint at the right address.
Uprobe works by replacing the instruction at the address defined
by (inode, offset) with the arch specific breakpoint
instruction. We save a copy of the original instruction at the
uprobed address.
This is needed for:
a. executing the instruction out-of-line (xol).
b. instruction analysis for any subsequent fixups.
c. restoring the instruction back when the uprobe is unregistered.
We insert or delete a breakpoint instruction, and this
breakpoint instruction is assumed to be the smallest instruction
available on the platform. For fixed size instruction platforms
this is trivially true, for variable size instruction platforms
the breakpoint instruction is typically the smallest (often a
single byte).
Writing the instruction is done by COWing the page and changing
the instruction during the copy, this even though most platforms
allow atomic writes of the breakpoint instruction. This also
mirrors the behaviour of a ptrace() memory write to a PRIVATE
file map.
The core worker is derived from KSM's replace_page() logic.
In essence, similar to KSM:
a. allocate a new page and copy over contents of the page that
has the uprobed vaddr
b. modify the copy and insert the breakpoint at the required
address
c. switch the original page with the copy containing the
breakpoint
d. flush page tables.
replace_page() is being replicated here because of some minor
changes in the type of pages and also because Hugh Dickins had
plans to improve replace_page() for KSM specific work.
Instruction analysis on x86 is based on instruction decoder and
determines if an instruction can be probed and determines the
necessary fixups after singlestep. Instruction analysis is done
at probe insertion time so that we avoid having to repeat the
same analysis every time a probe is hit.
A lot of code here is due to the improvement/suggestions/inputs
from Peter Zijlstra.
Changelog:
(v10):
- Add code to clear REX.B prefix as suggested by Denys Vlasenko
and Masami Hiramatsu.
(v9):
- Use insn_offset_modrm as suggested by Masami Hiramatsu.
(v7):
Handle comments from Peter Zijlstra:
- Dont take reference to inode. (expect inode to uprobe_register to be sane).
- Use PTR_ERR to set the return value.
- No need to take reference to inode.
- use PTR_ERR to return error value.
- register and uprobe_unregister share code.
(v5):
- Modified del_consumer as per comments from Peter.
- Drop reference to inode before dropping reference to uprobe.
- Use i_size_read(inode) instead of inode->i_size.
- Ensure uprobe->consumers is NULL, before __uprobe_unregister() is called.
- Includes errno.h as recommended by Stephen Rothwell to fix a build issue
on sparc defconfig
- Remove restrictions while unregistering.
- Earlier code leaked inode references under some conditions while
registering/unregistering.
- Continue the vma-rmap walk even if the intermediate vma doesnt
meet the requirements.
- Validate the vma found by find_vma before inserting/removing the
breakpoint
- Call del_consumer under mutex_lock.
- Use hash locks.
- Handle mremap.
- Introduce find_least_offset_node() instead of close match logic in
find_uprobe
- Uprobes no more depends on MM_OWNER; No reference to task_structs
while inserting/removing a probe.
- Uses read_mapping_page instead of grab_cache_page so that the pages
have valid content.
- pass NULL to get_user_pages for the task parameter.
- call SetPageUptodate on the new page allocated in write_opcode.
- fix leaking a reference to the new page under certain conditions.
- Include Instruction Decoder if Uprobes gets defined.
- Remove const attributes for instruction prefix arrays.
- Uses mm_context to know if the application is 32 bit.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Also-written-by: Jim Keniston <jkenisto@us.ibm.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Roland McGrath <roland@hack.frob.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Anton Arapov <anton@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Denys Vlasenko <vda.linux@googlemail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linux-mm <linux-mm@kvack.org>
Link: http://lkml.kernel.org/r/20120209092642.GE16600@linux.vnet.ibm.com
[ Made various small edits to the commit log ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-02-09 17:26:42 +08:00
|
|
|
ret = true;
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
up_write(&uprobe->consumer_rwsem);
|
2012-02-17 16:27:41 +08:00
|
|
|
|
uprobes, mm, x86: Add the ability to install and remove uprobes breakpoints
Add uprobes support to the core kernel, with x86 support.
This commit adds the kernel facilities, the actual uprobes
user-space ABI and perf probe support comes in later commits.
General design:
Uprobes are maintained in an rb-tree indexed by inode and offset
(the offset here is from the start of the mapping). For a unique
(inode, offset) tuple, there can be at most one uprobe in the
rb-tree.
Since the (inode, offset) tuple identifies a unique uprobe, more
than one user may be interested in the same uprobe. This provides
the ability to connect multiple 'consumers' to the same uprobe.
Each consumer defines a handler and a filter (optional). The
'handler' is run every time the uprobe is hit, if it matches the
'filter' criteria.
The first consumer of a uprobe causes the breakpoint to be
inserted at the specified address and subsequent consumers are
appended to this list. On subsequent probes, the consumer gets
appended to the existing list of consumers. The breakpoint is
removed when the last consumer unregisters. For all other
unregisterations, the consumer is removed from the list of
consumers.
Given a inode, we get a list of the mms that have mapped the
inode. Do the actual registration if mm maps the page where a
probe needs to be inserted/removed.
We use a temporary list to walk through the vmas that map the
inode.
- The number of maps that map the inode, is not known before we
walk the rmap and keeps changing.
- extending vm_area_struct wasn't recommended, it's a
size-critical data structure.
- There can be more than one maps of the inode in the same mm.
We add callbacks to the mmap methods to keep an eye on text vmas
that are of interest to uprobes. When a vma of interest is mapped,
we insert the breakpoint at the right address.
Uprobe works by replacing the instruction at the address defined
by (inode, offset) with the arch specific breakpoint
instruction. We save a copy of the original instruction at the
uprobed address.
This is needed for:
a. executing the instruction out-of-line (xol).
b. instruction analysis for any subsequent fixups.
c. restoring the instruction back when the uprobe is unregistered.
We insert or delete a breakpoint instruction, and this
breakpoint instruction is assumed to be the smallest instruction
available on the platform. For fixed size instruction platforms
this is trivially true, for variable size instruction platforms
the breakpoint instruction is typically the smallest (often a
single byte).
Writing the instruction is done by COWing the page and changing
the instruction during the copy, this even though most platforms
allow atomic writes of the breakpoint instruction. This also
mirrors the behaviour of a ptrace() memory write to a PRIVATE
file map.
The core worker is derived from KSM's replace_page() logic.
In essence, similar to KSM:
a. allocate a new page and copy over contents of the page that
has the uprobed vaddr
b. modify the copy and insert the breakpoint at the required
address
c. switch the original page with the copy containing the
breakpoint
d. flush page tables.
replace_page() is being replicated here because of some minor
changes in the type of pages and also because Hugh Dickins had
plans to improve replace_page() for KSM specific work.
Instruction analysis on x86 is based on instruction decoder and
determines if an instruction can be probed and determines the
necessary fixups after singlestep. Instruction analysis is done
at probe insertion time so that we avoid having to repeat the
same analysis every time a probe is hit.
A lot of code here is due to the improvement/suggestions/inputs
from Peter Zijlstra.
Changelog:
(v10):
- Add code to clear REX.B prefix as suggested by Denys Vlasenko
and Masami Hiramatsu.
(v9):
- Use insn_offset_modrm as suggested by Masami Hiramatsu.
(v7):
Handle comments from Peter Zijlstra:
- Dont take reference to inode. (expect inode to uprobe_register to be sane).
- Use PTR_ERR to set the return value.
- No need to take reference to inode.
- use PTR_ERR to return error value.
- register and uprobe_unregister share code.
(v5):
- Modified del_consumer as per comments from Peter.
- Drop reference to inode before dropping reference to uprobe.
- Use i_size_read(inode) instead of inode->i_size.
- Ensure uprobe->consumers is NULL, before __uprobe_unregister() is called.
- Includes errno.h as recommended by Stephen Rothwell to fix a build issue
on sparc defconfig
- Remove restrictions while unregistering.
- Earlier code leaked inode references under some conditions while
registering/unregistering.
- Continue the vma-rmap walk even if the intermediate vma doesnt
meet the requirements.
- Validate the vma found by find_vma before inserting/removing the
breakpoint
- Call del_consumer under mutex_lock.
- Use hash locks.
- Handle mremap.
- Introduce find_least_offset_node() instead of close match logic in
find_uprobe
- Uprobes no more depends on MM_OWNER; No reference to task_structs
while inserting/removing a probe.
- Uses read_mapping_page instead of grab_cache_page so that the pages
have valid content.
- pass NULL to get_user_pages for the task parameter.
- call SetPageUptodate on the new page allocated in write_opcode.
- fix leaking a reference to the new page under certain conditions.
- Include Instruction Decoder if Uprobes gets defined.
- Remove const attributes for instruction prefix arrays.
- Uses mm_context to know if the application is 32 bit.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Also-written-by: Jim Keniston <jkenisto@us.ibm.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Roland McGrath <roland@hack.frob.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Anton Arapov <anton@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Denys Vlasenko <vda.linux@googlemail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linux-mm <linux-mm@kvack.org>
Link: http://lkml.kernel.org/r/20120209092642.GE16600@linux.vnet.ibm.com
[ Made various small edits to the commit log ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-02-09 17:26:42 +08:00
|
|
|
return ret;
|
|
|
|
}
|
|
|
|
|
2013-11-08 02:41:57 +08:00
|
|
|
static int __copy_insn(struct address_space *mapping, struct file *filp,
|
|
|
|
void *insn, int nbytes, loff_t offset)
|
uprobes, mm, x86: Add the ability to install and remove uprobes breakpoints
Add uprobes support to the core kernel, with x86 support.
This commit adds the kernel facilities, the actual uprobes
user-space ABI and perf probe support comes in later commits.
General design:
Uprobes are maintained in an rb-tree indexed by inode and offset
(the offset here is from the start of the mapping). For a unique
(inode, offset) tuple, there can be at most one uprobe in the
rb-tree.
Since the (inode, offset) tuple identifies a unique uprobe, more
than one user may be interested in the same uprobe. This provides
the ability to connect multiple 'consumers' to the same uprobe.
Each consumer defines a handler and a filter (optional). The
'handler' is run every time the uprobe is hit, if it matches the
'filter' criteria.
The first consumer of a uprobe causes the breakpoint to be
inserted at the specified address and subsequent consumers are
appended to this list. On subsequent probes, the consumer gets
appended to the existing list of consumers. The breakpoint is
removed when the last consumer unregisters. For all other
unregisterations, the consumer is removed from the list of
consumers.
Given a inode, we get a list of the mms that have mapped the
inode. Do the actual registration if mm maps the page where a
probe needs to be inserted/removed.
We use a temporary list to walk through the vmas that map the
inode.
- The number of maps that map the inode, is not known before we
walk the rmap and keeps changing.
- extending vm_area_struct wasn't recommended, it's a
size-critical data structure.
- There can be more than one maps of the inode in the same mm.
We add callbacks to the mmap methods to keep an eye on text vmas
that are of interest to uprobes. When a vma of interest is mapped,
we insert the breakpoint at the right address.
Uprobe works by replacing the instruction at the address defined
by (inode, offset) with the arch specific breakpoint
instruction. We save a copy of the original instruction at the
uprobed address.
This is needed for:
a. executing the instruction out-of-line (xol).
b. instruction analysis for any subsequent fixups.
c. restoring the instruction back when the uprobe is unregistered.
We insert or delete a breakpoint instruction, and this
breakpoint instruction is assumed to be the smallest instruction
available on the platform. For fixed size instruction platforms
this is trivially true, for variable size instruction platforms
the breakpoint instruction is typically the smallest (often a
single byte).
Writing the instruction is done by COWing the page and changing
the instruction during the copy, this even though most platforms
allow atomic writes of the breakpoint instruction. This also
mirrors the behaviour of a ptrace() memory write to a PRIVATE
file map.
The core worker is derived from KSM's replace_page() logic.
In essence, similar to KSM:
a. allocate a new page and copy over contents of the page that
has the uprobed vaddr
b. modify the copy and insert the breakpoint at the required
address
c. switch the original page with the copy containing the
breakpoint
d. flush page tables.
replace_page() is being replicated here because of some minor
changes in the type of pages and also because Hugh Dickins had
plans to improve replace_page() for KSM specific work.
Instruction analysis on x86 is based on instruction decoder and
determines if an instruction can be probed and determines the
necessary fixups after singlestep. Instruction analysis is done
at probe insertion time so that we avoid having to repeat the
same analysis every time a probe is hit.
A lot of code here is due to the improvement/suggestions/inputs
from Peter Zijlstra.
Changelog:
(v10):
- Add code to clear REX.B prefix as suggested by Denys Vlasenko
and Masami Hiramatsu.
(v9):
- Use insn_offset_modrm as suggested by Masami Hiramatsu.
(v7):
Handle comments from Peter Zijlstra:
- Dont take reference to inode. (expect inode to uprobe_register to be sane).
- Use PTR_ERR to set the return value.
- No need to take reference to inode.
- use PTR_ERR to return error value.
- register and uprobe_unregister share code.
(v5):
- Modified del_consumer as per comments from Peter.
- Drop reference to inode before dropping reference to uprobe.
- Use i_size_read(inode) instead of inode->i_size.
- Ensure uprobe->consumers is NULL, before __uprobe_unregister() is called.
- Includes errno.h as recommended by Stephen Rothwell to fix a build issue
on sparc defconfig
- Remove restrictions while unregistering.
- Earlier code leaked inode references under some conditions while
registering/unregistering.
- Continue the vma-rmap walk even if the intermediate vma doesnt
meet the requirements.
- Validate the vma found by find_vma before inserting/removing the
breakpoint
- Call del_consumer under mutex_lock.
- Use hash locks.
- Handle mremap.
- Introduce find_least_offset_node() instead of close match logic in
find_uprobe
- Uprobes no more depends on MM_OWNER; No reference to task_structs
while inserting/removing a probe.
- Uses read_mapping_page instead of grab_cache_page so that the pages
have valid content.
- pass NULL to get_user_pages for the task parameter.
- call SetPageUptodate on the new page allocated in write_opcode.
- fix leaking a reference to the new page under certain conditions.
- Include Instruction Decoder if Uprobes gets defined.
- Remove const attributes for instruction prefix arrays.
- Uses mm_context to know if the application is 32 bit.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Also-written-by: Jim Keniston <jkenisto@us.ibm.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Roland McGrath <roland@hack.frob.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Anton Arapov <anton@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Denys Vlasenko <vda.linux@googlemail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linux-mm <linux-mm@kvack.org>
Link: http://lkml.kernel.org/r/20120209092642.GE16600@linux.vnet.ibm.com
[ Made various small edits to the commit log ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-02-09 17:26:42 +08:00
|
|
|
{
|
|
|
|
struct page *page;
|
|
|
|
/*
|
2014-05-20 02:41:36 +08:00
|
|
|
* Ensure that the page that has the original instruction is populated
|
|
|
|
* and in page-cache. If ->readpage == NULL it must be shmem_mapping(),
|
|
|
|
* see uprobe_register().
|
uprobes, mm, x86: Add the ability to install and remove uprobes breakpoints
Add uprobes support to the core kernel, with x86 support.
This commit adds the kernel facilities, the actual uprobes
user-space ABI and perf probe support comes in later commits.
General design:
Uprobes are maintained in an rb-tree indexed by inode and offset
(the offset here is from the start of the mapping). For a unique
(inode, offset) tuple, there can be at most one uprobe in the
rb-tree.
Since the (inode, offset) tuple identifies a unique uprobe, more
than one user may be interested in the same uprobe. This provides
the ability to connect multiple 'consumers' to the same uprobe.
Each consumer defines a handler and a filter (optional). The
'handler' is run every time the uprobe is hit, if it matches the
'filter' criteria.
The first consumer of a uprobe causes the breakpoint to be
inserted at the specified address and subsequent consumers are
appended to this list. On subsequent probes, the consumer gets
appended to the existing list of consumers. The breakpoint is
removed when the last consumer unregisters. For all other
unregisterations, the consumer is removed from the list of
consumers.
Given a inode, we get a list of the mms that have mapped the
inode. Do the actual registration if mm maps the page where a
probe needs to be inserted/removed.
We use a temporary list to walk through the vmas that map the
inode.
- The number of maps that map the inode, is not known before we
walk the rmap and keeps changing.
- extending vm_area_struct wasn't recommended, it's a
size-critical data structure.
- There can be more than one maps of the inode in the same mm.
We add callbacks to the mmap methods to keep an eye on text vmas
that are of interest to uprobes. When a vma of interest is mapped,
we insert the breakpoint at the right address.
Uprobe works by replacing the instruction at the address defined
by (inode, offset) with the arch specific breakpoint
instruction. We save a copy of the original instruction at the
uprobed address.
This is needed for:
a. executing the instruction out-of-line (xol).
b. instruction analysis for any subsequent fixups.
c. restoring the instruction back when the uprobe is unregistered.
We insert or delete a breakpoint instruction, and this
breakpoint instruction is assumed to be the smallest instruction
available on the platform. For fixed size instruction platforms
this is trivially true, for variable size instruction platforms
the breakpoint instruction is typically the smallest (often a
single byte).
Writing the instruction is done by COWing the page and changing
the instruction during the copy, this even though most platforms
allow atomic writes of the breakpoint instruction. This also
mirrors the behaviour of a ptrace() memory write to a PRIVATE
file map.
The core worker is derived from KSM's replace_page() logic.
In essence, similar to KSM:
a. allocate a new page and copy over contents of the page that
has the uprobed vaddr
b. modify the copy and insert the breakpoint at the required
address
c. switch the original page with the copy containing the
breakpoint
d. flush page tables.
replace_page() is being replicated here because of some minor
changes in the type of pages and also because Hugh Dickins had
plans to improve replace_page() for KSM specific work.
Instruction analysis on x86 is based on instruction decoder and
determines if an instruction can be probed and determines the
necessary fixups after singlestep. Instruction analysis is done
at probe insertion time so that we avoid having to repeat the
same analysis every time a probe is hit.
A lot of code here is due to the improvement/suggestions/inputs
from Peter Zijlstra.
Changelog:
(v10):
- Add code to clear REX.B prefix as suggested by Denys Vlasenko
and Masami Hiramatsu.
(v9):
- Use insn_offset_modrm as suggested by Masami Hiramatsu.
(v7):
Handle comments from Peter Zijlstra:
- Dont take reference to inode. (expect inode to uprobe_register to be sane).
- Use PTR_ERR to set the return value.
- No need to take reference to inode.
- use PTR_ERR to return error value.
- register and uprobe_unregister share code.
(v5):
- Modified del_consumer as per comments from Peter.
- Drop reference to inode before dropping reference to uprobe.
- Use i_size_read(inode) instead of inode->i_size.
- Ensure uprobe->consumers is NULL, before __uprobe_unregister() is called.
- Includes errno.h as recommended by Stephen Rothwell to fix a build issue
on sparc defconfig
- Remove restrictions while unregistering.
- Earlier code leaked inode references under some conditions while
registering/unregistering.
- Continue the vma-rmap walk even if the intermediate vma doesnt
meet the requirements.
- Validate the vma found by find_vma before inserting/removing the
breakpoint
- Call del_consumer under mutex_lock.
- Use hash locks.
- Handle mremap.
- Introduce find_least_offset_node() instead of close match logic in
find_uprobe
- Uprobes no more depends on MM_OWNER; No reference to task_structs
while inserting/removing a probe.
- Uses read_mapping_page instead of grab_cache_page so that the pages
have valid content.
- pass NULL to get_user_pages for the task parameter.
- call SetPageUptodate on the new page allocated in write_opcode.
- fix leaking a reference to the new page under certain conditions.
- Include Instruction Decoder if Uprobes gets defined.
- Remove const attributes for instruction prefix arrays.
- Uses mm_context to know if the application is 32 bit.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Also-written-by: Jim Keniston <jkenisto@us.ibm.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Roland McGrath <roland@hack.frob.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Anton Arapov <anton@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Denys Vlasenko <vda.linux@googlemail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linux-mm <linux-mm@kvack.org>
Link: http://lkml.kernel.org/r/20120209092642.GE16600@linux.vnet.ibm.com
[ Made various small edits to the commit log ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-02-09 17:26:42 +08:00
|
|
|
*/
|
2014-05-20 02:41:36 +08:00
|
|
|
if (mapping->a_ops->readpage)
|
mm, fs: get rid of PAGE_CACHE_* and page_cache_{get,release} macros
PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} macros were introduced *long* time
ago with promise that one day it will be possible to implement page
cache with bigger chunks than PAGE_SIZE.
This promise never materialized. And unlikely will.
We have many places where PAGE_CACHE_SIZE assumed to be equal to
PAGE_SIZE. And it's constant source of confusion on whether
PAGE_CACHE_* or PAGE_* constant should be used in a particular case,
especially on the border between fs and mm.
Global switching to PAGE_CACHE_SIZE != PAGE_SIZE would cause to much
breakage to be doable.
Let's stop pretending that pages in page cache are special. They are
not.
The changes are pretty straight-forward:
- <foo> << (PAGE_CACHE_SHIFT - PAGE_SHIFT) -> <foo>;
- <foo> >> (PAGE_CACHE_SHIFT - PAGE_SHIFT) -> <foo>;
- PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} -> PAGE_{SIZE,SHIFT,MASK,ALIGN};
- page_cache_get() -> get_page();
- page_cache_release() -> put_page();
This patch contains automated changes generated with coccinelle using
script below. For some reason, coccinelle doesn't patch header files.
I've called spatch for them manually.
The only adjustment after coccinelle is revert of changes to
PAGE_CAHCE_ALIGN definition: we are going to drop it later.
There are few places in the code where coccinelle didn't reach. I'll
fix them manually in a separate patch. Comments and documentation also
will be addressed with the separate patch.
virtual patch
@@
expression E;
@@
- E << (PAGE_CACHE_SHIFT - PAGE_SHIFT)
+ E
@@
expression E;
@@
- E >> (PAGE_CACHE_SHIFT - PAGE_SHIFT)
+ E
@@
@@
- PAGE_CACHE_SHIFT
+ PAGE_SHIFT
@@
@@
- PAGE_CACHE_SIZE
+ PAGE_SIZE
@@
@@
- PAGE_CACHE_MASK
+ PAGE_MASK
@@
expression E;
@@
- PAGE_CACHE_ALIGN(E)
+ PAGE_ALIGN(E)
@@
expression E;
@@
- page_cache_get(E)
+ get_page(E)
@@
expression E;
@@
- page_cache_release(E)
+ put_page(E)
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Acked-by: Michal Hocko <mhocko@suse.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-04-01 20:29:47 +08:00
|
|
|
page = read_mapping_page(mapping, offset >> PAGE_SHIFT, filp);
|
2014-05-20 02:41:36 +08:00
|
|
|
else
|
mm, fs: get rid of PAGE_CACHE_* and page_cache_{get,release} macros
PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} macros were introduced *long* time
ago with promise that one day it will be possible to implement page
cache with bigger chunks than PAGE_SIZE.
This promise never materialized. And unlikely will.
We have many places where PAGE_CACHE_SIZE assumed to be equal to
PAGE_SIZE. And it's constant source of confusion on whether
PAGE_CACHE_* or PAGE_* constant should be used in a particular case,
especially on the border between fs and mm.
Global switching to PAGE_CACHE_SIZE != PAGE_SIZE would cause to much
breakage to be doable.
Let's stop pretending that pages in page cache are special. They are
not.
The changes are pretty straight-forward:
- <foo> << (PAGE_CACHE_SHIFT - PAGE_SHIFT) -> <foo>;
- <foo> >> (PAGE_CACHE_SHIFT - PAGE_SHIFT) -> <foo>;
- PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} -> PAGE_{SIZE,SHIFT,MASK,ALIGN};
- page_cache_get() -> get_page();
- page_cache_release() -> put_page();
This patch contains automated changes generated with coccinelle using
script below. For some reason, coccinelle doesn't patch header files.
I've called spatch for them manually.
The only adjustment after coccinelle is revert of changes to
PAGE_CAHCE_ALIGN definition: we are going to drop it later.
There are few places in the code where coccinelle didn't reach. I'll
fix them manually in a separate patch. Comments and documentation also
will be addressed with the separate patch.
virtual patch
@@
expression E;
@@
- E << (PAGE_CACHE_SHIFT - PAGE_SHIFT)
+ E
@@
expression E;
@@
- E >> (PAGE_CACHE_SHIFT - PAGE_SHIFT)
+ E
@@
@@
- PAGE_CACHE_SHIFT
+ PAGE_SHIFT
@@
@@
- PAGE_CACHE_SIZE
+ PAGE_SIZE
@@
@@
- PAGE_CACHE_MASK
+ PAGE_MASK
@@
expression E;
@@
- PAGE_CACHE_ALIGN(E)
+ PAGE_ALIGN(E)
@@
expression E;
@@
- page_cache_get(E)
+ get_page(E)
@@
expression E;
@@
- page_cache_release(E)
+ put_page(E)
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Acked-by: Michal Hocko <mhocko@suse.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-04-01 20:29:47 +08:00
|
|
|
page = shmem_read_mapping_page(mapping, offset >> PAGE_SHIFT);
|
uprobes, mm, x86: Add the ability to install and remove uprobes breakpoints
Add uprobes support to the core kernel, with x86 support.
This commit adds the kernel facilities, the actual uprobes
user-space ABI and perf probe support comes in later commits.
General design:
Uprobes are maintained in an rb-tree indexed by inode and offset
(the offset here is from the start of the mapping). For a unique
(inode, offset) tuple, there can be at most one uprobe in the
rb-tree.
Since the (inode, offset) tuple identifies a unique uprobe, more
than one user may be interested in the same uprobe. This provides
the ability to connect multiple 'consumers' to the same uprobe.
Each consumer defines a handler and a filter (optional). The
'handler' is run every time the uprobe is hit, if it matches the
'filter' criteria.
The first consumer of a uprobe causes the breakpoint to be
inserted at the specified address and subsequent consumers are
appended to this list. On subsequent probes, the consumer gets
appended to the existing list of consumers. The breakpoint is
removed when the last consumer unregisters. For all other
unregisterations, the consumer is removed from the list of
consumers.
Given a inode, we get a list of the mms that have mapped the
inode. Do the actual registration if mm maps the page where a
probe needs to be inserted/removed.
We use a temporary list to walk through the vmas that map the
inode.
- The number of maps that map the inode, is not known before we
walk the rmap and keeps changing.
- extending vm_area_struct wasn't recommended, it's a
size-critical data structure.
- There can be more than one maps of the inode in the same mm.
We add callbacks to the mmap methods to keep an eye on text vmas
that are of interest to uprobes. When a vma of interest is mapped,
we insert the breakpoint at the right address.
Uprobe works by replacing the instruction at the address defined
by (inode, offset) with the arch specific breakpoint
instruction. We save a copy of the original instruction at the
uprobed address.
This is needed for:
a. executing the instruction out-of-line (xol).
b. instruction analysis for any subsequent fixups.
c. restoring the instruction back when the uprobe is unregistered.
We insert or delete a breakpoint instruction, and this
breakpoint instruction is assumed to be the smallest instruction
available on the platform. For fixed size instruction platforms
this is trivially true, for variable size instruction platforms
the breakpoint instruction is typically the smallest (often a
single byte).
Writing the instruction is done by COWing the page and changing
the instruction during the copy, this even though most platforms
allow atomic writes of the breakpoint instruction. This also
mirrors the behaviour of a ptrace() memory write to a PRIVATE
file map.
The core worker is derived from KSM's replace_page() logic.
In essence, similar to KSM:
a. allocate a new page and copy over contents of the page that
has the uprobed vaddr
b. modify the copy and insert the breakpoint at the required
address
c. switch the original page with the copy containing the
breakpoint
d. flush page tables.
replace_page() is being replicated here because of some minor
changes in the type of pages and also because Hugh Dickins had
plans to improve replace_page() for KSM specific work.
Instruction analysis on x86 is based on instruction decoder and
determines if an instruction can be probed and determines the
necessary fixups after singlestep. Instruction analysis is done
at probe insertion time so that we avoid having to repeat the
same analysis every time a probe is hit.
A lot of code here is due to the improvement/suggestions/inputs
from Peter Zijlstra.
Changelog:
(v10):
- Add code to clear REX.B prefix as suggested by Denys Vlasenko
and Masami Hiramatsu.
(v9):
- Use insn_offset_modrm as suggested by Masami Hiramatsu.
(v7):
Handle comments from Peter Zijlstra:
- Dont take reference to inode. (expect inode to uprobe_register to be sane).
- Use PTR_ERR to set the return value.
- No need to take reference to inode.
- use PTR_ERR to return error value.
- register and uprobe_unregister share code.
(v5):
- Modified del_consumer as per comments from Peter.
- Drop reference to inode before dropping reference to uprobe.
- Use i_size_read(inode) instead of inode->i_size.
- Ensure uprobe->consumers is NULL, before __uprobe_unregister() is called.
- Includes errno.h as recommended by Stephen Rothwell to fix a build issue
on sparc defconfig
- Remove restrictions while unregistering.
- Earlier code leaked inode references under some conditions while
registering/unregistering.
- Continue the vma-rmap walk even if the intermediate vma doesnt
meet the requirements.
- Validate the vma found by find_vma before inserting/removing the
breakpoint
- Call del_consumer under mutex_lock.
- Use hash locks.
- Handle mremap.
- Introduce find_least_offset_node() instead of close match logic in
find_uprobe
- Uprobes no more depends on MM_OWNER; No reference to task_structs
while inserting/removing a probe.
- Uses read_mapping_page instead of grab_cache_page so that the pages
have valid content.
- pass NULL to get_user_pages for the task parameter.
- call SetPageUptodate on the new page allocated in write_opcode.
- fix leaking a reference to the new page under certain conditions.
- Include Instruction Decoder if Uprobes gets defined.
- Remove const attributes for instruction prefix arrays.
- Uses mm_context to know if the application is 32 bit.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Also-written-by: Jim Keniston <jkenisto@us.ibm.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Roland McGrath <roland@hack.frob.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Anton Arapov <anton@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Denys Vlasenko <vda.linux@googlemail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linux-mm <linux-mm@kvack.org>
Link: http://lkml.kernel.org/r/20120209092642.GE16600@linux.vnet.ibm.com
[ Made various small edits to the commit log ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-02-09 17:26:42 +08:00
|
|
|
if (IS_ERR(page))
|
|
|
|
return PTR_ERR(page);
|
|
|
|
|
2013-03-25 01:37:48 +08:00
|
|
|
copy_from_page(page, offset, insn, nbytes);
|
mm, fs: get rid of PAGE_CACHE_* and page_cache_{get,release} macros
PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} macros were introduced *long* time
ago with promise that one day it will be possible to implement page
cache with bigger chunks than PAGE_SIZE.
This promise never materialized. And unlikely will.
We have many places where PAGE_CACHE_SIZE assumed to be equal to
PAGE_SIZE. And it's constant source of confusion on whether
PAGE_CACHE_* or PAGE_* constant should be used in a particular case,
especially on the border between fs and mm.
Global switching to PAGE_CACHE_SIZE != PAGE_SIZE would cause to much
breakage to be doable.
Let's stop pretending that pages in page cache are special. They are
not.
The changes are pretty straight-forward:
- <foo> << (PAGE_CACHE_SHIFT - PAGE_SHIFT) -> <foo>;
- <foo> >> (PAGE_CACHE_SHIFT - PAGE_SHIFT) -> <foo>;
- PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} -> PAGE_{SIZE,SHIFT,MASK,ALIGN};
- page_cache_get() -> get_page();
- page_cache_release() -> put_page();
This patch contains automated changes generated with coccinelle using
script below. For some reason, coccinelle doesn't patch header files.
I've called spatch for them manually.
The only adjustment after coccinelle is revert of changes to
PAGE_CAHCE_ALIGN definition: we are going to drop it later.
There are few places in the code where coccinelle didn't reach. I'll
fix them manually in a separate patch. Comments and documentation also
will be addressed with the separate patch.
virtual patch
@@
expression E;
@@
- E << (PAGE_CACHE_SHIFT - PAGE_SHIFT)
+ E
@@
expression E;
@@
- E >> (PAGE_CACHE_SHIFT - PAGE_SHIFT)
+ E
@@
@@
- PAGE_CACHE_SHIFT
+ PAGE_SHIFT
@@
@@
- PAGE_CACHE_SIZE
+ PAGE_SIZE
@@
@@
- PAGE_CACHE_MASK
+ PAGE_MASK
@@
expression E;
@@
- PAGE_CACHE_ALIGN(E)
+ PAGE_ALIGN(E)
@@
expression E;
@@
- page_cache_get(E)
+ get_page(E)
@@
expression E;
@@
- page_cache_release(E)
+ put_page(E)
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Acked-by: Michal Hocko <mhocko@suse.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-04-01 20:29:47 +08:00
|
|
|
put_page(page);
|
2012-02-17 16:27:41 +08:00
|
|
|
|
uprobes, mm, x86: Add the ability to install and remove uprobes breakpoints
Add uprobes support to the core kernel, with x86 support.
This commit adds the kernel facilities, the actual uprobes
user-space ABI and perf probe support comes in later commits.
General design:
Uprobes are maintained in an rb-tree indexed by inode and offset
(the offset here is from the start of the mapping). For a unique
(inode, offset) tuple, there can be at most one uprobe in the
rb-tree.
Since the (inode, offset) tuple identifies a unique uprobe, more
than one user may be interested in the same uprobe. This provides
the ability to connect multiple 'consumers' to the same uprobe.
Each consumer defines a handler and a filter (optional). The
'handler' is run every time the uprobe is hit, if it matches the
'filter' criteria.
The first consumer of a uprobe causes the breakpoint to be
inserted at the specified address and subsequent consumers are
appended to this list. On subsequent probes, the consumer gets
appended to the existing list of consumers. The breakpoint is
removed when the last consumer unregisters. For all other
unregisterations, the consumer is removed from the list of
consumers.
Given a inode, we get a list of the mms that have mapped the
inode. Do the actual registration if mm maps the page where a
probe needs to be inserted/removed.
We use a temporary list to walk through the vmas that map the
inode.
- The number of maps that map the inode, is not known before we
walk the rmap and keeps changing.
- extending vm_area_struct wasn't recommended, it's a
size-critical data structure.
- There can be more than one maps of the inode in the same mm.
We add callbacks to the mmap methods to keep an eye on text vmas
that are of interest to uprobes. When a vma of interest is mapped,
we insert the breakpoint at the right address.
Uprobe works by replacing the instruction at the address defined
by (inode, offset) with the arch specific breakpoint
instruction. We save a copy of the original instruction at the
uprobed address.
This is needed for:
a. executing the instruction out-of-line (xol).
b. instruction analysis for any subsequent fixups.
c. restoring the instruction back when the uprobe is unregistered.
We insert or delete a breakpoint instruction, and this
breakpoint instruction is assumed to be the smallest instruction
available on the platform. For fixed size instruction platforms
this is trivially true, for variable size instruction platforms
the breakpoint instruction is typically the smallest (often a
single byte).
Writing the instruction is done by COWing the page and changing
the instruction during the copy, this even though most platforms
allow atomic writes of the breakpoint instruction. This also
mirrors the behaviour of a ptrace() memory write to a PRIVATE
file map.
The core worker is derived from KSM's replace_page() logic.
In essence, similar to KSM:
a. allocate a new page and copy over contents of the page that
has the uprobed vaddr
b. modify the copy and insert the breakpoint at the required
address
c. switch the original page with the copy containing the
breakpoint
d. flush page tables.
replace_page() is being replicated here because of some minor
changes in the type of pages and also because Hugh Dickins had
plans to improve replace_page() for KSM specific work.
Instruction analysis on x86 is based on instruction decoder and
determines if an instruction can be probed and determines the
necessary fixups after singlestep. Instruction analysis is done
at probe insertion time so that we avoid having to repeat the
same analysis every time a probe is hit.
A lot of code here is due to the improvement/suggestions/inputs
from Peter Zijlstra.
Changelog:
(v10):
- Add code to clear REX.B prefix as suggested by Denys Vlasenko
and Masami Hiramatsu.
(v9):
- Use insn_offset_modrm as suggested by Masami Hiramatsu.
(v7):
Handle comments from Peter Zijlstra:
- Dont take reference to inode. (expect inode to uprobe_register to be sane).
- Use PTR_ERR to set the return value.
- No need to take reference to inode.
- use PTR_ERR to return error value.
- register and uprobe_unregister share code.
(v5):
- Modified del_consumer as per comments from Peter.
- Drop reference to inode before dropping reference to uprobe.
- Use i_size_read(inode) instead of inode->i_size.
- Ensure uprobe->consumers is NULL, before __uprobe_unregister() is called.
- Includes errno.h as recommended by Stephen Rothwell to fix a build issue
on sparc defconfig
- Remove restrictions while unregistering.
- Earlier code leaked inode references under some conditions while
registering/unregistering.
- Continue the vma-rmap walk even if the intermediate vma doesnt
meet the requirements.
- Validate the vma found by find_vma before inserting/removing the
breakpoint
- Call del_consumer under mutex_lock.
- Use hash locks.
- Handle mremap.
- Introduce find_least_offset_node() instead of close match logic in
find_uprobe
- Uprobes no more depends on MM_OWNER; No reference to task_structs
while inserting/removing a probe.
- Uses read_mapping_page instead of grab_cache_page so that the pages
have valid content.
- pass NULL to get_user_pages for the task parameter.
- call SetPageUptodate on the new page allocated in write_opcode.
- fix leaking a reference to the new page under certain conditions.
- Include Instruction Decoder if Uprobes gets defined.
- Remove const attributes for instruction prefix arrays.
- Uses mm_context to know if the application is 32 bit.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Also-written-by: Jim Keniston <jkenisto@us.ibm.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Roland McGrath <roland@hack.frob.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Anton Arapov <anton@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Denys Vlasenko <vda.linux@googlemail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linux-mm <linux-mm@kvack.org>
Link: http://lkml.kernel.org/r/20120209092642.GE16600@linux.vnet.ibm.com
[ Made various small edits to the commit log ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-02-09 17:26:42 +08:00
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2012-06-15 23:43:42 +08:00
|
|
|
static int copy_insn(struct uprobe *uprobe, struct file *filp)
|
uprobes, mm, x86: Add the ability to install and remove uprobes breakpoints
Add uprobes support to the core kernel, with x86 support.
This commit adds the kernel facilities, the actual uprobes
user-space ABI and perf probe support comes in later commits.
General design:
Uprobes are maintained in an rb-tree indexed by inode and offset
(the offset here is from the start of the mapping). For a unique
(inode, offset) tuple, there can be at most one uprobe in the
rb-tree.
Since the (inode, offset) tuple identifies a unique uprobe, more
than one user may be interested in the same uprobe. This provides
the ability to connect multiple 'consumers' to the same uprobe.
Each consumer defines a handler and a filter (optional). The
'handler' is run every time the uprobe is hit, if it matches the
'filter' criteria.
The first consumer of a uprobe causes the breakpoint to be
inserted at the specified address and subsequent consumers are
appended to this list. On subsequent probes, the consumer gets
appended to the existing list of consumers. The breakpoint is
removed when the last consumer unregisters. For all other
unregisterations, the consumer is removed from the list of
consumers.
Given a inode, we get a list of the mms that have mapped the
inode. Do the actual registration if mm maps the page where a
probe needs to be inserted/removed.
We use a temporary list to walk through the vmas that map the
inode.
- The number of maps that map the inode, is not known before we
walk the rmap and keeps changing.
- extending vm_area_struct wasn't recommended, it's a
size-critical data structure.
- There can be more than one maps of the inode in the same mm.
We add callbacks to the mmap methods to keep an eye on text vmas
that are of interest to uprobes. When a vma of interest is mapped,
we insert the breakpoint at the right address.
Uprobe works by replacing the instruction at the address defined
by (inode, offset) with the arch specific breakpoint
instruction. We save a copy of the original instruction at the
uprobed address.
This is needed for:
a. executing the instruction out-of-line (xol).
b. instruction analysis for any subsequent fixups.
c. restoring the instruction back when the uprobe is unregistered.
We insert or delete a breakpoint instruction, and this
breakpoint instruction is assumed to be the smallest instruction
available on the platform. For fixed size instruction platforms
this is trivially true, for variable size instruction platforms
the breakpoint instruction is typically the smallest (often a
single byte).
Writing the instruction is done by COWing the page and changing
the instruction during the copy, this even though most platforms
allow atomic writes of the breakpoint instruction. This also
mirrors the behaviour of a ptrace() memory write to a PRIVATE
file map.
The core worker is derived from KSM's replace_page() logic.
In essence, similar to KSM:
a. allocate a new page and copy over contents of the page that
has the uprobed vaddr
b. modify the copy and insert the breakpoint at the required
address
c. switch the original page with the copy containing the
breakpoint
d. flush page tables.
replace_page() is being replicated here because of some minor
changes in the type of pages and also because Hugh Dickins had
plans to improve replace_page() for KSM specific work.
Instruction analysis on x86 is based on instruction decoder and
determines if an instruction can be probed and determines the
necessary fixups after singlestep. Instruction analysis is done
at probe insertion time so that we avoid having to repeat the
same analysis every time a probe is hit.
A lot of code here is due to the improvement/suggestions/inputs
from Peter Zijlstra.
Changelog:
(v10):
- Add code to clear REX.B prefix as suggested by Denys Vlasenko
and Masami Hiramatsu.
(v9):
- Use insn_offset_modrm as suggested by Masami Hiramatsu.
(v7):
Handle comments from Peter Zijlstra:
- Dont take reference to inode. (expect inode to uprobe_register to be sane).
- Use PTR_ERR to set the return value.
- No need to take reference to inode.
- use PTR_ERR to return error value.
- register and uprobe_unregister share code.
(v5):
- Modified del_consumer as per comments from Peter.
- Drop reference to inode before dropping reference to uprobe.
- Use i_size_read(inode) instead of inode->i_size.
- Ensure uprobe->consumers is NULL, before __uprobe_unregister() is called.
- Includes errno.h as recommended by Stephen Rothwell to fix a build issue
on sparc defconfig
- Remove restrictions while unregistering.
- Earlier code leaked inode references under some conditions while
registering/unregistering.
- Continue the vma-rmap walk even if the intermediate vma doesnt
meet the requirements.
- Validate the vma found by find_vma before inserting/removing the
breakpoint
- Call del_consumer under mutex_lock.
- Use hash locks.
- Handle mremap.
- Introduce find_least_offset_node() instead of close match logic in
find_uprobe
- Uprobes no more depends on MM_OWNER; No reference to task_structs
while inserting/removing a probe.
- Uses read_mapping_page instead of grab_cache_page so that the pages
have valid content.
- pass NULL to get_user_pages for the task parameter.
- call SetPageUptodate on the new page allocated in write_opcode.
- fix leaking a reference to the new page under certain conditions.
- Include Instruction Decoder if Uprobes gets defined.
- Remove const attributes for instruction prefix arrays.
- Uses mm_context to know if the application is 32 bit.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Also-written-by: Jim Keniston <jkenisto@us.ibm.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Roland McGrath <roland@hack.frob.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Anton Arapov <anton@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Denys Vlasenko <vda.linux@googlemail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linux-mm <linux-mm@kvack.org>
Link: http://lkml.kernel.org/r/20120209092642.GE16600@linux.vnet.ibm.com
[ Made various small edits to the commit log ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-02-09 17:26:42 +08:00
|
|
|
{
|
2013-11-08 02:41:57 +08:00
|
|
|
struct address_space *mapping = uprobe->inode->i_mapping;
|
|
|
|
loff_t offs = uprobe->offset;
|
2013-11-10 00:58:54 +08:00
|
|
|
void *insn = &uprobe->arch.insn;
|
|
|
|
int size = sizeof(uprobe->arch.insn);
|
2013-11-08 02:41:57 +08:00
|
|
|
int len, err = -EIO;
|
|
|
|
|
|
|
|
/* Copy only available bytes, -EIO if nothing was read */
|
|
|
|
do {
|
|
|
|
if (offs >= i_size_read(uprobe->inode))
|
|
|
|
break;
|
|
|
|
|
|
|
|
len = min_t(int, size, PAGE_SIZE - (offs & ~PAGE_MASK));
|
|
|
|
err = __copy_insn(mapping, filp, insn, len, offs);
|
2012-06-15 23:43:44 +08:00
|
|
|
if (err)
|
2013-11-08 02:41:57 +08:00
|
|
|
break;
|
|
|
|
|
|
|
|
insn += len;
|
|
|
|
offs += len;
|
|
|
|
size -= len;
|
|
|
|
} while (size);
|
|
|
|
|
|
|
|
return err;
|
uprobes, mm, x86: Add the ability to install and remove uprobes breakpoints
Add uprobes support to the core kernel, with x86 support.
This commit adds the kernel facilities, the actual uprobes
user-space ABI and perf probe support comes in later commits.
General design:
Uprobes are maintained in an rb-tree indexed by inode and offset
(the offset here is from the start of the mapping). For a unique
(inode, offset) tuple, there can be at most one uprobe in the
rb-tree.
Since the (inode, offset) tuple identifies a unique uprobe, more
than one user may be interested in the same uprobe. This provides
the ability to connect multiple 'consumers' to the same uprobe.
Each consumer defines a handler and a filter (optional). The
'handler' is run every time the uprobe is hit, if it matches the
'filter' criteria.
The first consumer of a uprobe causes the breakpoint to be
inserted at the specified address and subsequent consumers are
appended to this list. On subsequent probes, the consumer gets
appended to the existing list of consumers. The breakpoint is
removed when the last consumer unregisters. For all other
unregisterations, the consumer is removed from the list of
consumers.
Given a inode, we get a list of the mms that have mapped the
inode. Do the actual registration if mm maps the page where a
probe needs to be inserted/removed.
We use a temporary list to walk through the vmas that map the
inode.
- The number of maps that map the inode, is not known before we
walk the rmap and keeps changing.
- extending vm_area_struct wasn't recommended, it's a
size-critical data structure.
- There can be more than one maps of the inode in the same mm.
We add callbacks to the mmap methods to keep an eye on text vmas
that are of interest to uprobes. When a vma of interest is mapped,
we insert the breakpoint at the right address.
Uprobe works by replacing the instruction at the address defined
by (inode, offset) with the arch specific breakpoint
instruction. We save a copy of the original instruction at the
uprobed address.
This is needed for:
a. executing the instruction out-of-line (xol).
b. instruction analysis for any subsequent fixups.
c. restoring the instruction back when the uprobe is unregistered.
We insert or delete a breakpoint instruction, and this
breakpoint instruction is assumed to be the smallest instruction
available on the platform. For fixed size instruction platforms
this is trivially true, for variable size instruction platforms
the breakpoint instruction is typically the smallest (often a
single byte).
Writing the instruction is done by COWing the page and changing
the instruction during the copy, this even though most platforms
allow atomic writes of the breakpoint instruction. This also
mirrors the behaviour of a ptrace() memory write to a PRIVATE
file map.
The core worker is derived from KSM's replace_page() logic.
In essence, similar to KSM:
a. allocate a new page and copy over contents of the page that
has the uprobed vaddr
b. modify the copy and insert the breakpoint at the required
address
c. switch the original page with the copy containing the
breakpoint
d. flush page tables.
replace_page() is being replicated here because of some minor
changes in the type of pages and also because Hugh Dickins had
plans to improve replace_page() for KSM specific work.
Instruction analysis on x86 is based on instruction decoder and
determines if an instruction can be probed and determines the
necessary fixups after singlestep. Instruction analysis is done
at probe insertion time so that we avoid having to repeat the
same analysis every time a probe is hit.
A lot of code here is due to the improvement/suggestions/inputs
from Peter Zijlstra.
Changelog:
(v10):
- Add code to clear REX.B prefix as suggested by Denys Vlasenko
and Masami Hiramatsu.
(v9):
- Use insn_offset_modrm as suggested by Masami Hiramatsu.
(v7):
Handle comments from Peter Zijlstra:
- Dont take reference to inode. (expect inode to uprobe_register to be sane).
- Use PTR_ERR to set the return value.
- No need to take reference to inode.
- use PTR_ERR to return error value.
- register and uprobe_unregister share code.
(v5):
- Modified del_consumer as per comments from Peter.
- Drop reference to inode before dropping reference to uprobe.
- Use i_size_read(inode) instead of inode->i_size.
- Ensure uprobe->consumers is NULL, before __uprobe_unregister() is called.
- Includes errno.h as recommended by Stephen Rothwell to fix a build issue
on sparc defconfig
- Remove restrictions while unregistering.
- Earlier code leaked inode references under some conditions while
registering/unregistering.
- Continue the vma-rmap walk even if the intermediate vma doesnt
meet the requirements.
- Validate the vma found by find_vma before inserting/removing the
breakpoint
- Call del_consumer under mutex_lock.
- Use hash locks.
- Handle mremap.
- Introduce find_least_offset_node() instead of close match logic in
find_uprobe
- Uprobes no more depends on MM_OWNER; No reference to task_structs
while inserting/removing a probe.
- Uses read_mapping_page instead of grab_cache_page so that the pages
have valid content.
- pass NULL to get_user_pages for the task parameter.
- call SetPageUptodate on the new page allocated in write_opcode.
- fix leaking a reference to the new page under certain conditions.
- Include Instruction Decoder if Uprobes gets defined.
- Remove const attributes for instruction prefix arrays.
- Uses mm_context to know if the application is 32 bit.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Also-written-by: Jim Keniston <jkenisto@us.ibm.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Roland McGrath <roland@hack.frob.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Anton Arapov <anton@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Denys Vlasenko <vda.linux@googlemail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linux-mm <linux-mm@kvack.org>
Link: http://lkml.kernel.org/r/20120209092642.GE16600@linux.vnet.ibm.com
[ Made various small edits to the commit log ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-02-09 17:26:42 +08:00
|
|
|
}
|
|
|
|
|
2012-10-01 02:11:45 +08:00
|
|
|
static int prepare_uprobe(struct uprobe *uprobe, struct file *file,
|
|
|
|
struct mm_struct *mm, unsigned long vaddr)
|
|
|
|
{
|
|
|
|
int ret = 0;
|
|
|
|
|
2012-10-01 03:12:44 +08:00
|
|
|
if (test_bit(UPROBE_COPY_INSN, &uprobe->flags))
|
2012-10-01 02:11:45 +08:00
|
|
|
return ret;
|
|
|
|
|
2012-11-25 01:51:34 +08:00
|
|
|
/* TODO: move this into _register, until then we abuse this sem. */
|
|
|
|
down_write(&uprobe->consumer_rwsem);
|
2012-10-01 03:12:44 +08:00
|
|
|
if (test_bit(UPROBE_COPY_INSN, &uprobe->flags))
|
2012-10-01 02:31:41 +08:00
|
|
|
goto out;
|
|
|
|
|
2012-10-01 02:11:45 +08:00
|
|
|
ret = copy_insn(uprobe, file);
|
|
|
|
if (ret)
|
|
|
|
goto out;
|
|
|
|
|
|
|
|
ret = -ENOTSUPP;
|
2013-11-10 00:58:54 +08:00
|
|
|
if (is_trap_insn((uprobe_opcode_t *)&uprobe->arch.insn))
|
2012-10-01 02:11:45 +08:00
|
|
|
goto out;
|
|
|
|
|
|
|
|
ret = arch_uprobe_analyze_insn(&uprobe->arch, mm, vaddr);
|
|
|
|
if (ret)
|
|
|
|
goto out;
|
|
|
|
|
2013-11-06 02:50:39 +08:00
|
|
|
/* uprobe_write_opcode() assumes we don't cross page boundary */
|
2012-10-01 02:11:45 +08:00
|
|
|
BUG_ON((uprobe->offset & ~PAGE_MASK) +
|
|
|
|
UPROBE_SWBP_INSN_SIZE > PAGE_SIZE);
|
|
|
|
|
2018-11-23 00:10:31 +08:00
|
|
|
smp_wmb(); /* pairs with the smp_rmb() in handle_swbp() */
|
2012-10-01 03:12:44 +08:00
|
|
|
set_bit(UPROBE_COPY_INSN, &uprobe->flags);
|
2012-10-01 02:11:45 +08:00
|
|
|
|
|
|
|
out:
|
2012-11-25 01:51:34 +08:00
|
|
|
up_write(&uprobe->consumer_rwsem);
|
2012-10-01 02:31:41 +08:00
|
|
|
|
2012-10-01 02:11:45 +08:00
|
|
|
return ret;
|
|
|
|
}
|
|
|
|
|
2012-12-29 00:58:38 +08:00
|
|
|
static inline bool consumer_filter(struct uprobe_consumer *uc,
|
|
|
|
enum uprobe_filter_ctx ctx, struct mm_struct *mm)
|
2012-12-28 01:21:11 +08:00
|
|
|
{
|
2012-12-29 00:58:38 +08:00
|
|
|
return !uc->filter || uc->filter(uc, ctx, mm);
|
2012-12-28 01:21:11 +08:00
|
|
|
}
|
|
|
|
|
2012-12-29 00:58:38 +08:00
|
|
|
static bool filter_chain(struct uprobe *uprobe,
|
|
|
|
enum uprobe_filter_ctx ctx, struct mm_struct *mm)
|
2012-11-23 01:30:15 +08:00
|
|
|
{
|
2012-11-25 01:15:46 +08:00
|
|
|
struct uprobe_consumer *uc;
|
|
|
|
bool ret = false;
|
|
|
|
|
|
|
|
down_read(&uprobe->consumer_rwsem);
|
|
|
|
for (uc = uprobe->consumers; uc; uc = uc->next) {
|
2012-12-29 00:58:38 +08:00
|
|
|
ret = consumer_filter(uc, ctx, mm);
|
2012-11-25 01:15:46 +08:00
|
|
|
if (ret)
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
up_read(&uprobe->consumer_rwsem);
|
|
|
|
|
|
|
|
return ret;
|
2012-11-23 01:30:15 +08:00
|
|
|
}
|
|
|
|
|
2012-03-12 17:25:30 +08:00
|
|
|
static int
|
|
|
|
install_breakpoint(struct uprobe *uprobe, struct mm_struct *mm,
|
2012-06-15 23:43:55 +08:00
|
|
|
struct vm_area_struct *vma, unsigned long vaddr)
|
uprobes, mm, x86: Add the ability to install and remove uprobes breakpoints
Add uprobes support to the core kernel, with x86 support.
This commit adds the kernel facilities, the actual uprobes
user-space ABI and perf probe support comes in later commits.
General design:
Uprobes are maintained in an rb-tree indexed by inode and offset
(the offset here is from the start of the mapping). For a unique
(inode, offset) tuple, there can be at most one uprobe in the
rb-tree.
Since the (inode, offset) tuple identifies a unique uprobe, more
than one user may be interested in the same uprobe. This provides
the ability to connect multiple 'consumers' to the same uprobe.
Each consumer defines a handler and a filter (optional). The
'handler' is run every time the uprobe is hit, if it matches the
'filter' criteria.
The first consumer of a uprobe causes the breakpoint to be
inserted at the specified address and subsequent consumers are
appended to this list. On subsequent probes, the consumer gets
appended to the existing list of consumers. The breakpoint is
removed when the last consumer unregisters. For all other
unregisterations, the consumer is removed from the list of
consumers.
Given a inode, we get a list of the mms that have mapped the
inode. Do the actual registration if mm maps the page where a
probe needs to be inserted/removed.
We use a temporary list to walk through the vmas that map the
inode.
- The number of maps that map the inode, is not known before we
walk the rmap and keeps changing.
- extending vm_area_struct wasn't recommended, it's a
size-critical data structure.
- There can be more than one maps of the inode in the same mm.
We add callbacks to the mmap methods to keep an eye on text vmas
that are of interest to uprobes. When a vma of interest is mapped,
we insert the breakpoint at the right address.
Uprobe works by replacing the instruction at the address defined
by (inode, offset) with the arch specific breakpoint
instruction. We save a copy of the original instruction at the
uprobed address.
This is needed for:
a. executing the instruction out-of-line (xol).
b. instruction analysis for any subsequent fixups.
c. restoring the instruction back when the uprobe is unregistered.
We insert or delete a breakpoint instruction, and this
breakpoint instruction is assumed to be the smallest instruction
available on the platform. For fixed size instruction platforms
this is trivially true, for variable size instruction platforms
the breakpoint instruction is typically the smallest (often a
single byte).
Writing the instruction is done by COWing the page and changing
the instruction during the copy, this even though most platforms
allow atomic writes of the breakpoint instruction. This also
mirrors the behaviour of a ptrace() memory write to a PRIVATE
file map.
The core worker is derived from KSM's replace_page() logic.
In essence, similar to KSM:
a. allocate a new page and copy over contents of the page that
has the uprobed vaddr
b. modify the copy and insert the breakpoint at the required
address
c. switch the original page with the copy containing the
breakpoint
d. flush page tables.
replace_page() is being replicated here because of some minor
changes in the type of pages and also because Hugh Dickins had
plans to improve replace_page() for KSM specific work.
Instruction analysis on x86 is based on instruction decoder and
determines if an instruction can be probed and determines the
necessary fixups after singlestep. Instruction analysis is done
at probe insertion time so that we avoid having to repeat the
same analysis every time a probe is hit.
A lot of code here is due to the improvement/suggestions/inputs
from Peter Zijlstra.
Changelog:
(v10):
- Add code to clear REX.B prefix as suggested by Denys Vlasenko
and Masami Hiramatsu.
(v9):
- Use insn_offset_modrm as suggested by Masami Hiramatsu.
(v7):
Handle comments from Peter Zijlstra:
- Dont take reference to inode. (expect inode to uprobe_register to be sane).
- Use PTR_ERR to set the return value.
- No need to take reference to inode.
- use PTR_ERR to return error value.
- register and uprobe_unregister share code.
(v5):
- Modified del_consumer as per comments from Peter.
- Drop reference to inode before dropping reference to uprobe.
- Use i_size_read(inode) instead of inode->i_size.
- Ensure uprobe->consumers is NULL, before __uprobe_unregister() is called.
- Includes errno.h as recommended by Stephen Rothwell to fix a build issue
on sparc defconfig
- Remove restrictions while unregistering.
- Earlier code leaked inode references under some conditions while
registering/unregistering.
- Continue the vma-rmap walk even if the intermediate vma doesnt
meet the requirements.
- Validate the vma found by find_vma before inserting/removing the
breakpoint
- Call del_consumer under mutex_lock.
- Use hash locks.
- Handle mremap.
- Introduce find_least_offset_node() instead of close match logic in
find_uprobe
- Uprobes no more depends on MM_OWNER; No reference to task_structs
while inserting/removing a probe.
- Uses read_mapping_page instead of grab_cache_page so that the pages
have valid content.
- pass NULL to get_user_pages for the task parameter.
- call SetPageUptodate on the new page allocated in write_opcode.
- fix leaking a reference to the new page under certain conditions.
- Include Instruction Decoder if Uprobes gets defined.
- Remove const attributes for instruction prefix arrays.
- Uses mm_context to know if the application is 32 bit.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Also-written-by: Jim Keniston <jkenisto@us.ibm.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Roland McGrath <roland@hack.frob.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Anton Arapov <anton@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Denys Vlasenko <vda.linux@googlemail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linux-mm <linux-mm@kvack.org>
Link: http://lkml.kernel.org/r/20120209092642.GE16600@linux.vnet.ibm.com
[ Made various small edits to the commit log ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-02-09 17:26:42 +08:00
|
|
|
{
|
2012-08-08 23:11:42 +08:00
|
|
|
bool first_uprobe;
|
uprobes, mm, x86: Add the ability to install and remove uprobes breakpoints
Add uprobes support to the core kernel, with x86 support.
This commit adds the kernel facilities, the actual uprobes
user-space ABI and perf probe support comes in later commits.
General design:
Uprobes are maintained in an rb-tree indexed by inode and offset
(the offset here is from the start of the mapping). For a unique
(inode, offset) tuple, there can be at most one uprobe in the
rb-tree.
Since the (inode, offset) tuple identifies a unique uprobe, more
than one user may be interested in the same uprobe. This provides
the ability to connect multiple 'consumers' to the same uprobe.
Each consumer defines a handler and a filter (optional). The
'handler' is run every time the uprobe is hit, if it matches the
'filter' criteria.
The first consumer of a uprobe causes the breakpoint to be
inserted at the specified address and subsequent consumers are
appended to this list. On subsequent probes, the consumer gets
appended to the existing list of consumers. The breakpoint is
removed when the last consumer unregisters. For all other
unregisterations, the consumer is removed from the list of
consumers.
Given a inode, we get a list of the mms that have mapped the
inode. Do the actual registration if mm maps the page where a
probe needs to be inserted/removed.
We use a temporary list to walk through the vmas that map the
inode.
- The number of maps that map the inode, is not known before we
walk the rmap and keeps changing.
- extending vm_area_struct wasn't recommended, it's a
size-critical data structure.
- There can be more than one maps of the inode in the same mm.
We add callbacks to the mmap methods to keep an eye on text vmas
that are of interest to uprobes. When a vma of interest is mapped,
we insert the breakpoint at the right address.
Uprobe works by replacing the instruction at the address defined
by (inode, offset) with the arch specific breakpoint
instruction. We save a copy of the original instruction at the
uprobed address.
This is needed for:
a. executing the instruction out-of-line (xol).
b. instruction analysis for any subsequent fixups.
c. restoring the instruction back when the uprobe is unregistered.
We insert or delete a breakpoint instruction, and this
breakpoint instruction is assumed to be the smallest instruction
available on the platform. For fixed size instruction platforms
this is trivially true, for variable size instruction platforms
the breakpoint instruction is typically the smallest (often a
single byte).
Writing the instruction is done by COWing the page and changing
the instruction during the copy, this even though most platforms
allow atomic writes of the breakpoint instruction. This also
mirrors the behaviour of a ptrace() memory write to a PRIVATE
file map.
The core worker is derived from KSM's replace_page() logic.
In essence, similar to KSM:
a. allocate a new page and copy over contents of the page that
has the uprobed vaddr
b. modify the copy and insert the breakpoint at the required
address
c. switch the original page with the copy containing the
breakpoint
d. flush page tables.
replace_page() is being replicated here because of some minor
changes in the type of pages and also because Hugh Dickins had
plans to improve replace_page() for KSM specific work.
Instruction analysis on x86 is based on instruction decoder and
determines if an instruction can be probed and determines the
necessary fixups after singlestep. Instruction analysis is done
at probe insertion time so that we avoid having to repeat the
same analysis every time a probe is hit.
A lot of code here is due to the improvement/suggestions/inputs
from Peter Zijlstra.
Changelog:
(v10):
- Add code to clear REX.B prefix as suggested by Denys Vlasenko
and Masami Hiramatsu.
(v9):
- Use insn_offset_modrm as suggested by Masami Hiramatsu.
(v7):
Handle comments from Peter Zijlstra:
- Dont take reference to inode. (expect inode to uprobe_register to be sane).
- Use PTR_ERR to set the return value.
- No need to take reference to inode.
- use PTR_ERR to return error value.
- register and uprobe_unregister share code.
(v5):
- Modified del_consumer as per comments from Peter.
- Drop reference to inode before dropping reference to uprobe.
- Use i_size_read(inode) instead of inode->i_size.
- Ensure uprobe->consumers is NULL, before __uprobe_unregister() is called.
- Includes errno.h as recommended by Stephen Rothwell to fix a build issue
on sparc defconfig
- Remove restrictions while unregistering.
- Earlier code leaked inode references under some conditions while
registering/unregistering.
- Continue the vma-rmap walk even if the intermediate vma doesnt
meet the requirements.
- Validate the vma found by find_vma before inserting/removing the
breakpoint
- Call del_consumer under mutex_lock.
- Use hash locks.
- Handle mremap.
- Introduce find_least_offset_node() instead of close match logic in
find_uprobe
- Uprobes no more depends on MM_OWNER; No reference to task_structs
while inserting/removing a probe.
- Uses read_mapping_page instead of grab_cache_page so that the pages
have valid content.
- pass NULL to get_user_pages for the task parameter.
- call SetPageUptodate on the new page allocated in write_opcode.
- fix leaking a reference to the new page under certain conditions.
- Include Instruction Decoder if Uprobes gets defined.
- Remove const attributes for instruction prefix arrays.
- Uses mm_context to know if the application is 32 bit.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Also-written-by: Jim Keniston <jkenisto@us.ibm.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Roland McGrath <roland@hack.frob.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Anton Arapov <anton@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Denys Vlasenko <vda.linux@googlemail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linux-mm <linux-mm@kvack.org>
Link: http://lkml.kernel.org/r/20120209092642.GE16600@linux.vnet.ibm.com
[ Made various small edits to the commit log ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-02-09 17:26:42 +08:00
|
|
|
int ret;
|
|
|
|
|
2012-10-01 02:11:45 +08:00
|
|
|
ret = prepare_uprobe(uprobe, vma->vm_file, mm, vaddr);
|
|
|
|
if (ret)
|
|
|
|
return ret;
|
2012-03-31 02:26:46 +08:00
|
|
|
|
2012-08-08 23:11:42 +08:00
|
|
|
/*
|
|
|
|
* set MMF_HAS_UPROBES in advance for uprobe_pre_sstep_notifier(),
|
|
|
|
* the task can hit this breakpoint right after __replace_page().
|
|
|
|
*/
|
|
|
|
first_uprobe = !test_bit(MMF_HAS_UPROBES, &mm->flags);
|
|
|
|
if (first_uprobe)
|
|
|
|
set_bit(MMF_HAS_UPROBES, &mm->flags);
|
|
|
|
|
2012-06-15 23:43:55 +08:00
|
|
|
ret = set_swbp(&uprobe->arch, mm, vaddr);
|
2012-08-19 22:15:09 +08:00
|
|
|
if (!ret)
|
|
|
|
clear_bit(MMF_RECALC_UPROBES, &mm->flags);
|
|
|
|
else if (first_uprobe)
|
2012-08-08 23:11:42 +08:00
|
|
|
clear_bit(MMF_HAS_UPROBES, &mm->flags);
|
uprobes, mm, x86: Add the ability to install and remove uprobes breakpoints
Add uprobes support to the core kernel, with x86 support.
This commit adds the kernel facilities, the actual uprobes
user-space ABI and perf probe support comes in later commits.
General design:
Uprobes are maintained in an rb-tree indexed by inode and offset
(the offset here is from the start of the mapping). For a unique
(inode, offset) tuple, there can be at most one uprobe in the
rb-tree.
Since the (inode, offset) tuple identifies a unique uprobe, more
than one user may be interested in the same uprobe. This provides
the ability to connect multiple 'consumers' to the same uprobe.
Each consumer defines a handler and a filter (optional). The
'handler' is run every time the uprobe is hit, if it matches the
'filter' criteria.
The first consumer of a uprobe causes the breakpoint to be
inserted at the specified address and subsequent consumers are
appended to this list. On subsequent probes, the consumer gets
appended to the existing list of consumers. The breakpoint is
removed when the last consumer unregisters. For all other
unregisterations, the consumer is removed from the list of
consumers.
Given a inode, we get a list of the mms that have mapped the
inode. Do the actual registration if mm maps the page where a
probe needs to be inserted/removed.
We use a temporary list to walk through the vmas that map the
inode.
- The number of maps that map the inode, is not known before we
walk the rmap and keeps changing.
- extending vm_area_struct wasn't recommended, it's a
size-critical data structure.
- There can be more than one maps of the inode in the same mm.
We add callbacks to the mmap methods to keep an eye on text vmas
that are of interest to uprobes. When a vma of interest is mapped,
we insert the breakpoint at the right address.
Uprobe works by replacing the instruction at the address defined
by (inode, offset) with the arch specific breakpoint
instruction. We save a copy of the original instruction at the
uprobed address.
This is needed for:
a. executing the instruction out-of-line (xol).
b. instruction analysis for any subsequent fixups.
c. restoring the instruction back when the uprobe is unregistered.
We insert or delete a breakpoint instruction, and this
breakpoint instruction is assumed to be the smallest instruction
available on the platform. For fixed size instruction platforms
this is trivially true, for variable size instruction platforms
the breakpoint instruction is typically the smallest (often a
single byte).
Writing the instruction is done by COWing the page and changing
the instruction during the copy, this even though most platforms
allow atomic writes of the breakpoint instruction. This also
mirrors the behaviour of a ptrace() memory write to a PRIVATE
file map.
The core worker is derived from KSM's replace_page() logic.
In essence, similar to KSM:
a. allocate a new page and copy over contents of the page that
has the uprobed vaddr
b. modify the copy and insert the breakpoint at the required
address
c. switch the original page with the copy containing the
breakpoint
d. flush page tables.
replace_page() is being replicated here because of some minor
changes in the type of pages and also because Hugh Dickins had
plans to improve replace_page() for KSM specific work.
Instruction analysis on x86 is based on instruction decoder and
determines if an instruction can be probed and determines the
necessary fixups after singlestep. Instruction analysis is done
at probe insertion time so that we avoid having to repeat the
same analysis every time a probe is hit.
A lot of code here is due to the improvement/suggestions/inputs
from Peter Zijlstra.
Changelog:
(v10):
- Add code to clear REX.B prefix as suggested by Denys Vlasenko
and Masami Hiramatsu.
(v9):
- Use insn_offset_modrm as suggested by Masami Hiramatsu.
(v7):
Handle comments from Peter Zijlstra:
- Dont take reference to inode. (expect inode to uprobe_register to be sane).
- Use PTR_ERR to set the return value.
- No need to take reference to inode.
- use PTR_ERR to return error value.
- register and uprobe_unregister share code.
(v5):
- Modified del_consumer as per comments from Peter.
- Drop reference to inode before dropping reference to uprobe.
- Use i_size_read(inode) instead of inode->i_size.
- Ensure uprobe->consumers is NULL, before __uprobe_unregister() is called.
- Includes errno.h as recommended by Stephen Rothwell to fix a build issue
on sparc defconfig
- Remove restrictions while unregistering.
- Earlier code leaked inode references under some conditions while
registering/unregistering.
- Continue the vma-rmap walk even if the intermediate vma doesnt
meet the requirements.
- Validate the vma found by find_vma before inserting/removing the
breakpoint
- Call del_consumer under mutex_lock.
- Use hash locks.
- Handle mremap.
- Introduce find_least_offset_node() instead of close match logic in
find_uprobe
- Uprobes no more depends on MM_OWNER; No reference to task_structs
while inserting/removing a probe.
- Uses read_mapping_page instead of grab_cache_page so that the pages
have valid content.
- pass NULL to get_user_pages for the task parameter.
- call SetPageUptodate on the new page allocated in write_opcode.
- fix leaking a reference to the new page under certain conditions.
- Include Instruction Decoder if Uprobes gets defined.
- Remove const attributes for instruction prefix arrays.
- Uses mm_context to know if the application is 32 bit.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Also-written-by: Jim Keniston <jkenisto@us.ibm.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Roland McGrath <roland@hack.frob.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Anton Arapov <anton@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Denys Vlasenko <vda.linux@googlemail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linux-mm <linux-mm@kvack.org>
Link: http://lkml.kernel.org/r/20120209092642.GE16600@linux.vnet.ibm.com
[ Made various small edits to the commit log ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-02-09 17:26:42 +08:00
|
|
|
|
|
|
|
return ret;
|
|
|
|
}
|
|
|
|
|
2012-10-01 00:54:53 +08:00
|
|
|
static int
|
2012-06-15 23:43:55 +08:00
|
|
|
remove_breakpoint(struct uprobe *uprobe, struct mm_struct *mm, unsigned long vaddr)
|
uprobes, mm, x86: Add the ability to install and remove uprobes breakpoints
Add uprobes support to the core kernel, with x86 support.
This commit adds the kernel facilities, the actual uprobes
user-space ABI and perf probe support comes in later commits.
General design:
Uprobes are maintained in an rb-tree indexed by inode and offset
(the offset here is from the start of the mapping). For a unique
(inode, offset) tuple, there can be at most one uprobe in the
rb-tree.
Since the (inode, offset) tuple identifies a unique uprobe, more
than one user may be interested in the same uprobe. This provides
the ability to connect multiple 'consumers' to the same uprobe.
Each consumer defines a handler and a filter (optional). The
'handler' is run every time the uprobe is hit, if it matches the
'filter' criteria.
The first consumer of a uprobe causes the breakpoint to be
inserted at the specified address and subsequent consumers are
appended to this list. On subsequent probes, the consumer gets
appended to the existing list of consumers. The breakpoint is
removed when the last consumer unregisters. For all other
unregisterations, the consumer is removed from the list of
consumers.
Given a inode, we get a list of the mms that have mapped the
inode. Do the actual registration if mm maps the page where a
probe needs to be inserted/removed.
We use a temporary list to walk through the vmas that map the
inode.
- The number of maps that map the inode, is not known before we
walk the rmap and keeps changing.
- extending vm_area_struct wasn't recommended, it's a
size-critical data structure.
- There can be more than one maps of the inode in the same mm.
We add callbacks to the mmap methods to keep an eye on text vmas
that are of interest to uprobes. When a vma of interest is mapped,
we insert the breakpoint at the right address.
Uprobe works by replacing the instruction at the address defined
by (inode, offset) with the arch specific breakpoint
instruction. We save a copy of the original instruction at the
uprobed address.
This is needed for:
a. executing the instruction out-of-line (xol).
b. instruction analysis for any subsequent fixups.
c. restoring the instruction back when the uprobe is unregistered.
We insert or delete a breakpoint instruction, and this
breakpoint instruction is assumed to be the smallest instruction
available on the platform. For fixed size instruction platforms
this is trivially true, for variable size instruction platforms
the breakpoint instruction is typically the smallest (often a
single byte).
Writing the instruction is done by COWing the page and changing
the instruction during the copy, this even though most platforms
allow atomic writes of the breakpoint instruction. This also
mirrors the behaviour of a ptrace() memory write to a PRIVATE
file map.
The core worker is derived from KSM's replace_page() logic.
In essence, similar to KSM:
a. allocate a new page and copy over contents of the page that
has the uprobed vaddr
b. modify the copy and insert the breakpoint at the required
address
c. switch the original page with the copy containing the
breakpoint
d. flush page tables.
replace_page() is being replicated here because of some minor
changes in the type of pages and also because Hugh Dickins had
plans to improve replace_page() for KSM specific work.
Instruction analysis on x86 is based on instruction decoder and
determines if an instruction can be probed and determines the
necessary fixups after singlestep. Instruction analysis is done
at probe insertion time so that we avoid having to repeat the
same analysis every time a probe is hit.
A lot of code here is due to the improvement/suggestions/inputs
from Peter Zijlstra.
Changelog:
(v10):
- Add code to clear REX.B prefix as suggested by Denys Vlasenko
and Masami Hiramatsu.
(v9):
- Use insn_offset_modrm as suggested by Masami Hiramatsu.
(v7):
Handle comments from Peter Zijlstra:
- Dont take reference to inode. (expect inode to uprobe_register to be sane).
- Use PTR_ERR to set the return value.
- No need to take reference to inode.
- use PTR_ERR to return error value.
- register and uprobe_unregister share code.
(v5):
- Modified del_consumer as per comments from Peter.
- Drop reference to inode before dropping reference to uprobe.
- Use i_size_read(inode) instead of inode->i_size.
- Ensure uprobe->consumers is NULL, before __uprobe_unregister() is called.
- Includes errno.h as recommended by Stephen Rothwell to fix a build issue
on sparc defconfig
- Remove restrictions while unregistering.
- Earlier code leaked inode references under some conditions while
registering/unregistering.
- Continue the vma-rmap walk even if the intermediate vma doesnt
meet the requirements.
- Validate the vma found by find_vma before inserting/removing the
breakpoint
- Call del_consumer under mutex_lock.
- Use hash locks.
- Handle mremap.
- Introduce find_least_offset_node() instead of close match logic in
find_uprobe
- Uprobes no more depends on MM_OWNER; No reference to task_structs
while inserting/removing a probe.
- Uses read_mapping_page instead of grab_cache_page so that the pages
have valid content.
- pass NULL to get_user_pages for the task parameter.
- call SetPageUptodate on the new page allocated in write_opcode.
- fix leaking a reference to the new page under certain conditions.
- Include Instruction Decoder if Uprobes gets defined.
- Remove const attributes for instruction prefix arrays.
- Uses mm_context to know if the application is 32 bit.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Also-written-by: Jim Keniston <jkenisto@us.ibm.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Roland McGrath <roland@hack.frob.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Anton Arapov <anton@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Denys Vlasenko <vda.linux@googlemail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linux-mm <linux-mm@kvack.org>
Link: http://lkml.kernel.org/r/20120209092642.GE16600@linux.vnet.ibm.com
[ Made various small edits to the commit log ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-02-09 17:26:42 +08:00
|
|
|
{
|
2012-08-19 22:15:09 +08:00
|
|
|
set_bit(MMF_RECALC_UPROBES, &mm->flags);
|
2012-10-01 00:54:53 +08:00
|
|
|
return set_orig_insn(&uprobe->arch, mm, vaddr);
|
uprobes, mm, x86: Add the ability to install and remove uprobes breakpoints
Add uprobes support to the core kernel, with x86 support.
This commit adds the kernel facilities, the actual uprobes
user-space ABI and perf probe support comes in later commits.
General design:
Uprobes are maintained in an rb-tree indexed by inode and offset
(the offset here is from the start of the mapping). For a unique
(inode, offset) tuple, there can be at most one uprobe in the
rb-tree.
Since the (inode, offset) tuple identifies a unique uprobe, more
than one user may be interested in the same uprobe. This provides
the ability to connect multiple 'consumers' to the same uprobe.
Each consumer defines a handler and a filter (optional). The
'handler' is run every time the uprobe is hit, if it matches the
'filter' criteria.
The first consumer of a uprobe causes the breakpoint to be
inserted at the specified address and subsequent consumers are
appended to this list. On subsequent probes, the consumer gets
appended to the existing list of consumers. The breakpoint is
removed when the last consumer unregisters. For all other
unregisterations, the consumer is removed from the list of
consumers.
Given a inode, we get a list of the mms that have mapped the
inode. Do the actual registration if mm maps the page where a
probe needs to be inserted/removed.
We use a temporary list to walk through the vmas that map the
inode.
- The number of maps that map the inode, is not known before we
walk the rmap and keeps changing.
- extending vm_area_struct wasn't recommended, it's a
size-critical data structure.
- There can be more than one maps of the inode in the same mm.
We add callbacks to the mmap methods to keep an eye on text vmas
that are of interest to uprobes. When a vma of interest is mapped,
we insert the breakpoint at the right address.
Uprobe works by replacing the instruction at the address defined
by (inode, offset) with the arch specific breakpoint
instruction. We save a copy of the original instruction at the
uprobed address.
This is needed for:
a. executing the instruction out-of-line (xol).
b. instruction analysis for any subsequent fixups.
c. restoring the instruction back when the uprobe is unregistered.
We insert or delete a breakpoint instruction, and this
breakpoint instruction is assumed to be the smallest instruction
available on the platform. For fixed size instruction platforms
this is trivially true, for variable size instruction platforms
the breakpoint instruction is typically the smallest (often a
single byte).
Writing the instruction is done by COWing the page and changing
the instruction during the copy, this even though most platforms
allow atomic writes of the breakpoint instruction. This also
mirrors the behaviour of a ptrace() memory write to a PRIVATE
file map.
The core worker is derived from KSM's replace_page() logic.
In essence, similar to KSM:
a. allocate a new page and copy over contents of the page that
has the uprobed vaddr
b. modify the copy and insert the breakpoint at the required
address
c. switch the original page with the copy containing the
breakpoint
d. flush page tables.
replace_page() is being replicated here because of some minor
changes in the type of pages and also because Hugh Dickins had
plans to improve replace_page() for KSM specific work.
Instruction analysis on x86 is based on instruction decoder and
determines if an instruction can be probed and determines the
necessary fixups after singlestep. Instruction analysis is done
at probe insertion time so that we avoid having to repeat the
same analysis every time a probe is hit.
A lot of code here is due to the improvement/suggestions/inputs
from Peter Zijlstra.
Changelog:
(v10):
- Add code to clear REX.B prefix as suggested by Denys Vlasenko
and Masami Hiramatsu.
(v9):
- Use insn_offset_modrm as suggested by Masami Hiramatsu.
(v7):
Handle comments from Peter Zijlstra:
- Dont take reference to inode. (expect inode to uprobe_register to be sane).
- Use PTR_ERR to set the return value.
- No need to take reference to inode.
- use PTR_ERR to return error value.
- register and uprobe_unregister share code.
(v5):
- Modified del_consumer as per comments from Peter.
- Drop reference to inode before dropping reference to uprobe.
- Use i_size_read(inode) instead of inode->i_size.
- Ensure uprobe->consumers is NULL, before __uprobe_unregister() is called.
- Includes errno.h as recommended by Stephen Rothwell to fix a build issue
on sparc defconfig
- Remove restrictions while unregistering.
- Earlier code leaked inode references under some conditions while
registering/unregistering.
- Continue the vma-rmap walk even if the intermediate vma doesnt
meet the requirements.
- Validate the vma found by find_vma before inserting/removing the
breakpoint
- Call del_consumer under mutex_lock.
- Use hash locks.
- Handle mremap.
- Introduce find_least_offset_node() instead of close match logic in
find_uprobe
- Uprobes no more depends on MM_OWNER; No reference to task_structs
while inserting/removing a probe.
- Uses read_mapping_page instead of grab_cache_page so that the pages
have valid content.
- pass NULL to get_user_pages for the task parameter.
- call SetPageUptodate on the new page allocated in write_opcode.
- fix leaking a reference to the new page under certain conditions.
- Include Instruction Decoder if Uprobes gets defined.
- Remove const attributes for instruction prefix arrays.
- Uses mm_context to know if the application is 32 bit.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Also-written-by: Jim Keniston <jkenisto@us.ibm.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Roland McGrath <roland@hack.frob.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Anton Arapov <anton@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Denys Vlasenko <vda.linux@googlemail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linux-mm <linux-mm@kvack.org>
Link: http://lkml.kernel.org/r/20120209092642.GE16600@linux.vnet.ibm.com
[ Made various small edits to the commit log ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-02-09 17:26:42 +08:00
|
|
|
}
|
|
|
|
|
2012-11-26 05:01:42 +08:00
|
|
|
static inline bool uprobe_is_active(struct uprobe *uprobe)
|
|
|
|
{
|
|
|
|
return !RB_EMPTY_NODE(&uprobe->rb_node);
|
|
|
|
}
|
uprobes/core: Handle breakpoint and singlestep exceptions
Uprobes uses exception notifiers to get to know if a thread hit
a breakpoint or a singlestep exception.
When a thread hits a uprobe or is singlestepping post a uprobe
hit, the uprobe exception notifier sets its TIF_UPROBE bit,
which will then be checked on its return to userspace path
(do_notify_resume() ->uprobe_notify_resume()), where the
consumers handlers are run (in task context) based on the
defined filters.
Uprobe hits are thread specific and hence we need to maintain
information about if a task hit a uprobe, what uprobe was hit,
the slot where the original instruction was copied for xol so
that it can be singlestepped with appropriate fixups.
In some cases, special care is needed for instructions that are
executed out of line (xol). These are architecture specific
artefacts, such as handling RIP relative instructions on x86_64.
Since the instruction at which the uprobe was inserted is
executed out of line, architecture specific fixups are added so
that the thread continues normal execution in the presence of a
uprobe.
Postpone the signals until we execute the probed insn.
post_xol() path does a recalc_sigpending() before return to
user-mode, this ensures the signal can't be lost.
Uprobes relies on DIE_DEBUG notification to notify if a
singlestep is complete.
Adds x86 specific uprobe exception notifiers and appropriate
hooks needed to determine a uprobe hit and subsequent post
processing.
Add requisite x86 fixups for xol for uprobes. Specific cases
needing fixups include relative jumps (x86_64), calls, etc.
Where possible, we check and skip singlestepping the
breakpointed instructions. For now we skip single byte as well
as few multibyte nop instructions. However this can be extended
to other instructions too.
Credits to Oleg Nesterov for suggestions/patches related to
signal, breakpoint, singlestep handling code.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Jim Keniston <jkenisto@linux.vnet.ibm.com>
Cc: Linux-mm <linux-mm@kvack.org>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/20120313180011.29771.89027.sendpatchset@srdronam.in.ibm.com
[ Performed various cleanliness edits ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-03-14 02:00:11 +08:00
|
|
|
/*
|
2012-05-30 03:30:08 +08:00
|
|
|
* There could be threads that have already hit the breakpoint. They
|
|
|
|
* will recheck the current insn and restart if find_uprobe() fails.
|
|
|
|
* See find_active_uprobe().
|
uprobes/core: Handle breakpoint and singlestep exceptions
Uprobes uses exception notifiers to get to know if a thread hit
a breakpoint or a singlestep exception.
When a thread hits a uprobe or is singlestepping post a uprobe
hit, the uprobe exception notifier sets its TIF_UPROBE bit,
which will then be checked on its return to userspace path
(do_notify_resume() ->uprobe_notify_resume()), where the
consumers handlers are run (in task context) based on the
defined filters.
Uprobe hits are thread specific and hence we need to maintain
information about if a task hit a uprobe, what uprobe was hit,
the slot where the original instruction was copied for xol so
that it can be singlestepped with appropriate fixups.
In some cases, special care is needed for instructions that are
executed out of line (xol). These are architecture specific
artefacts, such as handling RIP relative instructions on x86_64.
Since the instruction at which the uprobe was inserted is
executed out of line, architecture specific fixups are added so
that the thread continues normal execution in the presence of a
uprobe.
Postpone the signals until we execute the probed insn.
post_xol() path does a recalc_sigpending() before return to
user-mode, this ensures the signal can't be lost.
Uprobes relies on DIE_DEBUG notification to notify if a
singlestep is complete.
Adds x86 specific uprobe exception notifiers and appropriate
hooks needed to determine a uprobe hit and subsequent post
processing.
Add requisite x86 fixups for xol for uprobes. Specific cases
needing fixups include relative jumps (x86_64), calls, etc.
Where possible, we check and skip singlestepping the
breakpointed instructions. For now we skip single byte as well
as few multibyte nop instructions. However this can be extended
to other instructions too.
Credits to Oleg Nesterov for suggestions/patches related to
signal, breakpoint, singlestep handling code.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Jim Keniston <jkenisto@linux.vnet.ibm.com>
Cc: Linux-mm <linux-mm@kvack.org>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/20120313180011.29771.89027.sendpatchset@srdronam.in.ibm.com
[ Performed various cleanliness edits ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-03-14 02:00:11 +08:00
|
|
|
*/
|
uprobes, mm, x86: Add the ability to install and remove uprobes breakpoints
Add uprobes support to the core kernel, with x86 support.
This commit adds the kernel facilities, the actual uprobes
user-space ABI and perf probe support comes in later commits.
General design:
Uprobes are maintained in an rb-tree indexed by inode and offset
(the offset here is from the start of the mapping). For a unique
(inode, offset) tuple, there can be at most one uprobe in the
rb-tree.
Since the (inode, offset) tuple identifies a unique uprobe, more
than one user may be interested in the same uprobe. This provides
the ability to connect multiple 'consumers' to the same uprobe.
Each consumer defines a handler and a filter (optional). The
'handler' is run every time the uprobe is hit, if it matches the
'filter' criteria.
The first consumer of a uprobe causes the breakpoint to be
inserted at the specified address and subsequent consumers are
appended to this list. On subsequent probes, the consumer gets
appended to the existing list of consumers. The breakpoint is
removed when the last consumer unregisters. For all other
unregisterations, the consumer is removed from the list of
consumers.
Given a inode, we get a list of the mms that have mapped the
inode. Do the actual registration if mm maps the page where a
probe needs to be inserted/removed.
We use a temporary list to walk through the vmas that map the
inode.
- The number of maps that map the inode, is not known before we
walk the rmap and keeps changing.
- extending vm_area_struct wasn't recommended, it's a
size-critical data structure.
- There can be more than one maps of the inode in the same mm.
We add callbacks to the mmap methods to keep an eye on text vmas
that are of interest to uprobes. When a vma of interest is mapped,
we insert the breakpoint at the right address.
Uprobe works by replacing the instruction at the address defined
by (inode, offset) with the arch specific breakpoint
instruction. We save a copy of the original instruction at the
uprobed address.
This is needed for:
a. executing the instruction out-of-line (xol).
b. instruction analysis for any subsequent fixups.
c. restoring the instruction back when the uprobe is unregistered.
We insert or delete a breakpoint instruction, and this
breakpoint instruction is assumed to be the smallest instruction
available on the platform. For fixed size instruction platforms
this is trivially true, for variable size instruction platforms
the breakpoint instruction is typically the smallest (often a
single byte).
Writing the instruction is done by COWing the page and changing
the instruction during the copy, this even though most platforms
allow atomic writes of the breakpoint instruction. This also
mirrors the behaviour of a ptrace() memory write to a PRIVATE
file map.
The core worker is derived from KSM's replace_page() logic.
In essence, similar to KSM:
a. allocate a new page and copy over contents of the page that
has the uprobed vaddr
b. modify the copy and insert the breakpoint at the required
address
c. switch the original page with the copy containing the
breakpoint
d. flush page tables.
replace_page() is being replicated here because of some minor
changes in the type of pages and also because Hugh Dickins had
plans to improve replace_page() for KSM specific work.
Instruction analysis on x86 is based on instruction decoder and
determines if an instruction can be probed and determines the
necessary fixups after singlestep. Instruction analysis is done
at probe insertion time so that we avoid having to repeat the
same analysis every time a probe is hit.
A lot of code here is due to the improvement/suggestions/inputs
from Peter Zijlstra.
Changelog:
(v10):
- Add code to clear REX.B prefix as suggested by Denys Vlasenko
and Masami Hiramatsu.
(v9):
- Use insn_offset_modrm as suggested by Masami Hiramatsu.
(v7):
Handle comments from Peter Zijlstra:
- Dont take reference to inode. (expect inode to uprobe_register to be sane).
- Use PTR_ERR to set the return value.
- No need to take reference to inode.
- use PTR_ERR to return error value.
- register and uprobe_unregister share code.
(v5):
- Modified del_consumer as per comments from Peter.
- Drop reference to inode before dropping reference to uprobe.
- Use i_size_read(inode) instead of inode->i_size.
- Ensure uprobe->consumers is NULL, before __uprobe_unregister() is called.
- Includes errno.h as recommended by Stephen Rothwell to fix a build issue
on sparc defconfig
- Remove restrictions while unregistering.
- Earlier code leaked inode references under some conditions while
registering/unregistering.
- Continue the vma-rmap walk even if the intermediate vma doesnt
meet the requirements.
- Validate the vma found by find_vma before inserting/removing the
breakpoint
- Call del_consumer under mutex_lock.
- Use hash locks.
- Handle mremap.
- Introduce find_least_offset_node() instead of close match logic in
find_uprobe
- Uprobes no more depends on MM_OWNER; No reference to task_structs
while inserting/removing a probe.
- Uses read_mapping_page instead of grab_cache_page so that the pages
have valid content.
- pass NULL to get_user_pages for the task parameter.
- call SetPageUptodate on the new page allocated in write_opcode.
- fix leaking a reference to the new page under certain conditions.
- Include Instruction Decoder if Uprobes gets defined.
- Remove const attributes for instruction prefix arrays.
- Uses mm_context to know if the application is 32 bit.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Also-written-by: Jim Keniston <jkenisto@us.ibm.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Roland McGrath <roland@hack.frob.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Anton Arapov <anton@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Denys Vlasenko <vda.linux@googlemail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linux-mm <linux-mm@kvack.org>
Link: http://lkml.kernel.org/r/20120209092642.GE16600@linux.vnet.ibm.com
[ Made various small edits to the commit log ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-02-09 17:26:42 +08:00
|
|
|
static void delete_uprobe(struct uprobe *uprobe)
|
|
|
|
{
|
2012-11-26 05:01:42 +08:00
|
|
|
if (WARN_ON(!uprobe_is_active(uprobe)))
|
|
|
|
return;
|
|
|
|
|
2012-08-18 23:01:57 +08:00
|
|
|
spin_lock(&uprobes_treelock);
|
uprobes, mm, x86: Add the ability to install and remove uprobes breakpoints
Add uprobes support to the core kernel, with x86 support.
This commit adds the kernel facilities, the actual uprobes
user-space ABI and perf probe support comes in later commits.
General design:
Uprobes are maintained in an rb-tree indexed by inode and offset
(the offset here is from the start of the mapping). For a unique
(inode, offset) tuple, there can be at most one uprobe in the
rb-tree.
Since the (inode, offset) tuple identifies a unique uprobe, more
than one user may be interested in the same uprobe. This provides
the ability to connect multiple 'consumers' to the same uprobe.
Each consumer defines a handler and a filter (optional). The
'handler' is run every time the uprobe is hit, if it matches the
'filter' criteria.
The first consumer of a uprobe causes the breakpoint to be
inserted at the specified address and subsequent consumers are
appended to this list. On subsequent probes, the consumer gets
appended to the existing list of consumers. The breakpoint is
removed when the last consumer unregisters. For all other
unregisterations, the consumer is removed from the list of
consumers.
Given a inode, we get a list of the mms that have mapped the
inode. Do the actual registration if mm maps the page where a
probe needs to be inserted/removed.
We use a temporary list to walk through the vmas that map the
inode.
- The number of maps that map the inode, is not known before we
walk the rmap and keeps changing.
- extending vm_area_struct wasn't recommended, it's a
size-critical data structure.
- There can be more than one maps of the inode in the same mm.
We add callbacks to the mmap methods to keep an eye on text vmas
that are of interest to uprobes. When a vma of interest is mapped,
we insert the breakpoint at the right address.
Uprobe works by replacing the instruction at the address defined
by (inode, offset) with the arch specific breakpoint
instruction. We save a copy of the original instruction at the
uprobed address.
This is needed for:
a. executing the instruction out-of-line (xol).
b. instruction analysis for any subsequent fixups.
c. restoring the instruction back when the uprobe is unregistered.
We insert or delete a breakpoint instruction, and this
breakpoint instruction is assumed to be the smallest instruction
available on the platform. For fixed size instruction platforms
this is trivially true, for variable size instruction platforms
the breakpoint instruction is typically the smallest (often a
single byte).
Writing the instruction is done by COWing the page and changing
the instruction during the copy, this even though most platforms
allow atomic writes of the breakpoint instruction. This also
mirrors the behaviour of a ptrace() memory write to a PRIVATE
file map.
The core worker is derived from KSM's replace_page() logic.
In essence, similar to KSM:
a. allocate a new page and copy over contents of the page that
has the uprobed vaddr
b. modify the copy and insert the breakpoint at the required
address
c. switch the original page with the copy containing the
breakpoint
d. flush page tables.
replace_page() is being replicated here because of some minor
changes in the type of pages and also because Hugh Dickins had
plans to improve replace_page() for KSM specific work.
Instruction analysis on x86 is based on instruction decoder and
determines if an instruction can be probed and determines the
necessary fixups after singlestep. Instruction analysis is done
at probe insertion time so that we avoid having to repeat the
same analysis every time a probe is hit.
A lot of code here is due to the improvement/suggestions/inputs
from Peter Zijlstra.
Changelog:
(v10):
- Add code to clear REX.B prefix as suggested by Denys Vlasenko
and Masami Hiramatsu.
(v9):
- Use insn_offset_modrm as suggested by Masami Hiramatsu.
(v7):
Handle comments from Peter Zijlstra:
- Dont take reference to inode. (expect inode to uprobe_register to be sane).
- Use PTR_ERR to set the return value.
- No need to take reference to inode.
- use PTR_ERR to return error value.
- register and uprobe_unregister share code.
(v5):
- Modified del_consumer as per comments from Peter.
- Drop reference to inode before dropping reference to uprobe.
- Use i_size_read(inode) instead of inode->i_size.
- Ensure uprobe->consumers is NULL, before __uprobe_unregister() is called.
- Includes errno.h as recommended by Stephen Rothwell to fix a build issue
on sparc defconfig
- Remove restrictions while unregistering.
- Earlier code leaked inode references under some conditions while
registering/unregistering.
- Continue the vma-rmap walk even if the intermediate vma doesnt
meet the requirements.
- Validate the vma found by find_vma before inserting/removing the
breakpoint
- Call del_consumer under mutex_lock.
- Use hash locks.
- Handle mremap.
- Introduce find_least_offset_node() instead of close match logic in
find_uprobe
- Uprobes no more depends on MM_OWNER; No reference to task_structs
while inserting/removing a probe.
- Uses read_mapping_page instead of grab_cache_page so that the pages
have valid content.
- pass NULL to get_user_pages for the task parameter.
- call SetPageUptodate on the new page allocated in write_opcode.
- fix leaking a reference to the new page under certain conditions.
- Include Instruction Decoder if Uprobes gets defined.
- Remove const attributes for instruction prefix arrays.
- Uses mm_context to know if the application is 32 bit.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Also-written-by: Jim Keniston <jkenisto@us.ibm.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Roland McGrath <roland@hack.frob.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Anton Arapov <anton@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Denys Vlasenko <vda.linux@googlemail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linux-mm <linux-mm@kvack.org>
Link: http://lkml.kernel.org/r/20120209092642.GE16600@linux.vnet.ibm.com
[ Made various small edits to the commit log ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-02-09 17:26:42 +08:00
|
|
|
rb_erase(&uprobe->rb_node, &uprobes_tree);
|
2012-08-18 23:01:57 +08:00
|
|
|
spin_unlock(&uprobes_treelock);
|
2012-11-26 05:01:42 +08:00
|
|
|
RB_CLEAR_NODE(&uprobe->rb_node); /* for uprobe_is_active() */
|
uprobes, mm, x86: Add the ability to install and remove uprobes breakpoints
Add uprobes support to the core kernel, with x86 support.
This commit adds the kernel facilities, the actual uprobes
user-space ABI and perf probe support comes in later commits.
General design:
Uprobes are maintained in an rb-tree indexed by inode and offset
(the offset here is from the start of the mapping). For a unique
(inode, offset) tuple, there can be at most one uprobe in the
rb-tree.
Since the (inode, offset) tuple identifies a unique uprobe, more
than one user may be interested in the same uprobe. This provides
the ability to connect multiple 'consumers' to the same uprobe.
Each consumer defines a handler and a filter (optional). The
'handler' is run every time the uprobe is hit, if it matches the
'filter' criteria.
The first consumer of a uprobe causes the breakpoint to be
inserted at the specified address and subsequent consumers are
appended to this list. On subsequent probes, the consumer gets
appended to the existing list of consumers. The breakpoint is
removed when the last consumer unregisters. For all other
unregisterations, the consumer is removed from the list of
consumers.
Given a inode, we get a list of the mms that have mapped the
inode. Do the actual registration if mm maps the page where a
probe needs to be inserted/removed.
We use a temporary list to walk through the vmas that map the
inode.
- The number of maps that map the inode, is not known before we
walk the rmap and keeps changing.
- extending vm_area_struct wasn't recommended, it's a
size-critical data structure.
- There can be more than one maps of the inode in the same mm.
We add callbacks to the mmap methods to keep an eye on text vmas
that are of interest to uprobes. When a vma of interest is mapped,
we insert the breakpoint at the right address.
Uprobe works by replacing the instruction at the address defined
by (inode, offset) with the arch specific breakpoint
instruction. We save a copy of the original instruction at the
uprobed address.
This is needed for:
a. executing the instruction out-of-line (xol).
b. instruction analysis for any subsequent fixups.
c. restoring the instruction back when the uprobe is unregistered.
We insert or delete a breakpoint instruction, and this
breakpoint instruction is assumed to be the smallest instruction
available on the platform. For fixed size instruction platforms
this is trivially true, for variable size instruction platforms
the breakpoint instruction is typically the smallest (often a
single byte).
Writing the instruction is done by COWing the page and changing
the instruction during the copy, this even though most platforms
allow atomic writes of the breakpoint instruction. This also
mirrors the behaviour of a ptrace() memory write to a PRIVATE
file map.
The core worker is derived from KSM's replace_page() logic.
In essence, similar to KSM:
a. allocate a new page and copy over contents of the page that
has the uprobed vaddr
b. modify the copy and insert the breakpoint at the required
address
c. switch the original page with the copy containing the
breakpoint
d. flush page tables.
replace_page() is being replicated here because of some minor
changes in the type of pages and also because Hugh Dickins had
plans to improve replace_page() for KSM specific work.
Instruction analysis on x86 is based on instruction decoder and
determines if an instruction can be probed and determines the
necessary fixups after singlestep. Instruction analysis is done
at probe insertion time so that we avoid having to repeat the
same analysis every time a probe is hit.
A lot of code here is due to the improvement/suggestions/inputs
from Peter Zijlstra.
Changelog:
(v10):
- Add code to clear REX.B prefix as suggested by Denys Vlasenko
and Masami Hiramatsu.
(v9):
- Use insn_offset_modrm as suggested by Masami Hiramatsu.
(v7):
Handle comments from Peter Zijlstra:
- Dont take reference to inode. (expect inode to uprobe_register to be sane).
- Use PTR_ERR to set the return value.
- No need to take reference to inode.
- use PTR_ERR to return error value.
- register and uprobe_unregister share code.
(v5):
- Modified del_consumer as per comments from Peter.
- Drop reference to inode before dropping reference to uprobe.
- Use i_size_read(inode) instead of inode->i_size.
- Ensure uprobe->consumers is NULL, before __uprobe_unregister() is called.
- Includes errno.h as recommended by Stephen Rothwell to fix a build issue
on sparc defconfig
- Remove restrictions while unregistering.
- Earlier code leaked inode references under some conditions while
registering/unregistering.
- Continue the vma-rmap walk even if the intermediate vma doesnt
meet the requirements.
- Validate the vma found by find_vma before inserting/removing the
breakpoint
- Call del_consumer under mutex_lock.
- Use hash locks.
- Handle mremap.
- Introduce find_least_offset_node() instead of close match logic in
find_uprobe
- Uprobes no more depends on MM_OWNER; No reference to task_structs
while inserting/removing a probe.
- Uses read_mapping_page instead of grab_cache_page so that the pages
have valid content.
- pass NULL to get_user_pages for the task parameter.
- call SetPageUptodate on the new page allocated in write_opcode.
- fix leaking a reference to the new page under certain conditions.
- Include Instruction Decoder if Uprobes gets defined.
- Remove const attributes for instruction prefix arrays.
- Uses mm_context to know if the application is 32 bit.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Also-written-by: Jim Keniston <jkenisto@us.ibm.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Roland McGrath <roland@hack.frob.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Anton Arapov <anton@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Denys Vlasenko <vda.linux@googlemail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linux-mm <linux-mm@kvack.org>
Link: http://lkml.kernel.org/r/20120209092642.GE16600@linux.vnet.ibm.com
[ Made various small edits to the commit log ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-02-09 17:26:42 +08:00
|
|
|
put_uprobe(uprobe);
|
|
|
|
}
|
|
|
|
|
2012-06-15 23:43:33 +08:00
|
|
|
struct map_info {
|
|
|
|
struct map_info *next;
|
|
|
|
struct mm_struct *mm;
|
2012-06-15 23:43:55 +08:00
|
|
|
unsigned long vaddr;
|
2012-06-15 23:43:33 +08:00
|
|
|
};
|
|
|
|
|
|
|
|
static inline struct map_info *free_map_info(struct map_info *info)
|
uprobes, mm, x86: Add the ability to install and remove uprobes breakpoints
Add uprobes support to the core kernel, with x86 support.
This commit adds the kernel facilities, the actual uprobes
user-space ABI and perf probe support comes in later commits.
General design:
Uprobes are maintained in an rb-tree indexed by inode and offset
(the offset here is from the start of the mapping). For a unique
(inode, offset) tuple, there can be at most one uprobe in the
rb-tree.
Since the (inode, offset) tuple identifies a unique uprobe, more
than one user may be interested in the same uprobe. This provides
the ability to connect multiple 'consumers' to the same uprobe.
Each consumer defines a handler and a filter (optional). The
'handler' is run every time the uprobe is hit, if it matches the
'filter' criteria.
The first consumer of a uprobe causes the breakpoint to be
inserted at the specified address and subsequent consumers are
appended to this list. On subsequent probes, the consumer gets
appended to the existing list of consumers. The breakpoint is
removed when the last consumer unregisters. For all other
unregisterations, the consumer is removed from the list of
consumers.
Given a inode, we get a list of the mms that have mapped the
inode. Do the actual registration if mm maps the page where a
probe needs to be inserted/removed.
We use a temporary list to walk through the vmas that map the
inode.
- The number of maps that map the inode, is not known before we
walk the rmap and keeps changing.
- extending vm_area_struct wasn't recommended, it's a
size-critical data structure.
- There can be more than one maps of the inode in the same mm.
We add callbacks to the mmap methods to keep an eye on text vmas
that are of interest to uprobes. When a vma of interest is mapped,
we insert the breakpoint at the right address.
Uprobe works by replacing the instruction at the address defined
by (inode, offset) with the arch specific breakpoint
instruction. We save a copy of the original instruction at the
uprobed address.
This is needed for:
a. executing the instruction out-of-line (xol).
b. instruction analysis for any subsequent fixups.
c. restoring the instruction back when the uprobe is unregistered.
We insert or delete a breakpoint instruction, and this
breakpoint instruction is assumed to be the smallest instruction
available on the platform. For fixed size instruction platforms
this is trivially true, for variable size instruction platforms
the breakpoint instruction is typically the smallest (often a
single byte).
Writing the instruction is done by COWing the page and changing
the instruction during the copy, this even though most platforms
allow atomic writes of the breakpoint instruction. This also
mirrors the behaviour of a ptrace() memory write to a PRIVATE
file map.
The core worker is derived from KSM's replace_page() logic.
In essence, similar to KSM:
a. allocate a new page and copy over contents of the page that
has the uprobed vaddr
b. modify the copy and insert the breakpoint at the required
address
c. switch the original page with the copy containing the
breakpoint
d. flush page tables.
replace_page() is being replicated here because of some minor
changes in the type of pages and also because Hugh Dickins had
plans to improve replace_page() for KSM specific work.
Instruction analysis on x86 is based on instruction decoder and
determines if an instruction can be probed and determines the
necessary fixups after singlestep. Instruction analysis is done
at probe insertion time so that we avoid having to repeat the
same analysis every time a probe is hit.
A lot of code here is due to the improvement/suggestions/inputs
from Peter Zijlstra.
Changelog:
(v10):
- Add code to clear REX.B prefix as suggested by Denys Vlasenko
and Masami Hiramatsu.
(v9):
- Use insn_offset_modrm as suggested by Masami Hiramatsu.
(v7):
Handle comments from Peter Zijlstra:
- Dont take reference to inode. (expect inode to uprobe_register to be sane).
- Use PTR_ERR to set the return value.
- No need to take reference to inode.
- use PTR_ERR to return error value.
- register and uprobe_unregister share code.
(v5):
- Modified del_consumer as per comments from Peter.
- Drop reference to inode before dropping reference to uprobe.
- Use i_size_read(inode) instead of inode->i_size.
- Ensure uprobe->consumers is NULL, before __uprobe_unregister() is called.
- Includes errno.h as recommended by Stephen Rothwell to fix a build issue
on sparc defconfig
- Remove restrictions while unregistering.
- Earlier code leaked inode references under some conditions while
registering/unregistering.
- Continue the vma-rmap walk even if the intermediate vma doesnt
meet the requirements.
- Validate the vma found by find_vma before inserting/removing the
breakpoint
- Call del_consumer under mutex_lock.
- Use hash locks.
- Handle mremap.
- Introduce find_least_offset_node() instead of close match logic in
find_uprobe
- Uprobes no more depends on MM_OWNER; No reference to task_structs
while inserting/removing a probe.
- Uses read_mapping_page instead of grab_cache_page so that the pages
have valid content.
- pass NULL to get_user_pages for the task parameter.
- call SetPageUptodate on the new page allocated in write_opcode.
- fix leaking a reference to the new page under certain conditions.
- Include Instruction Decoder if Uprobes gets defined.
- Remove const attributes for instruction prefix arrays.
- Uses mm_context to know if the application is 32 bit.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Also-written-by: Jim Keniston <jkenisto@us.ibm.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Roland McGrath <roland@hack.frob.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Anton Arapov <anton@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Denys Vlasenko <vda.linux@googlemail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linux-mm <linux-mm@kvack.org>
Link: http://lkml.kernel.org/r/20120209092642.GE16600@linux.vnet.ibm.com
[ Made various small edits to the commit log ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-02-09 17:26:42 +08:00
|
|
|
{
|
2012-06-15 23:43:33 +08:00
|
|
|
struct map_info *next = info->next;
|
|
|
|
kfree(info);
|
|
|
|
return next;
|
|
|
|
}
|
|
|
|
|
|
|
|
static struct map_info *
|
|
|
|
build_map_info(struct address_space *mapping, loff_t offset, bool is_register)
|
|
|
|
{
|
|
|
|
unsigned long pgoff = offset >> PAGE_SHIFT;
|
uprobes, mm, x86: Add the ability to install and remove uprobes breakpoints
Add uprobes support to the core kernel, with x86 support.
This commit adds the kernel facilities, the actual uprobes
user-space ABI and perf probe support comes in later commits.
General design:
Uprobes are maintained in an rb-tree indexed by inode and offset
(the offset here is from the start of the mapping). For a unique
(inode, offset) tuple, there can be at most one uprobe in the
rb-tree.
Since the (inode, offset) tuple identifies a unique uprobe, more
than one user may be interested in the same uprobe. This provides
the ability to connect multiple 'consumers' to the same uprobe.
Each consumer defines a handler and a filter (optional). The
'handler' is run every time the uprobe is hit, if it matches the
'filter' criteria.
The first consumer of a uprobe causes the breakpoint to be
inserted at the specified address and subsequent consumers are
appended to this list. On subsequent probes, the consumer gets
appended to the existing list of consumers. The breakpoint is
removed when the last consumer unregisters. For all other
unregisterations, the consumer is removed from the list of
consumers.
Given a inode, we get a list of the mms that have mapped the
inode. Do the actual registration if mm maps the page where a
probe needs to be inserted/removed.
We use a temporary list to walk through the vmas that map the
inode.
- The number of maps that map the inode, is not known before we
walk the rmap and keeps changing.
- extending vm_area_struct wasn't recommended, it's a
size-critical data structure.
- There can be more than one maps of the inode in the same mm.
We add callbacks to the mmap methods to keep an eye on text vmas
that are of interest to uprobes. When a vma of interest is mapped,
we insert the breakpoint at the right address.
Uprobe works by replacing the instruction at the address defined
by (inode, offset) with the arch specific breakpoint
instruction. We save a copy of the original instruction at the
uprobed address.
This is needed for:
a. executing the instruction out-of-line (xol).
b. instruction analysis for any subsequent fixups.
c. restoring the instruction back when the uprobe is unregistered.
We insert or delete a breakpoint instruction, and this
breakpoint instruction is assumed to be the smallest instruction
available on the platform. For fixed size instruction platforms
this is trivially true, for variable size instruction platforms
the breakpoint instruction is typically the smallest (often a
single byte).
Writing the instruction is done by COWing the page and changing
the instruction during the copy, this even though most platforms
allow atomic writes of the breakpoint instruction. This also
mirrors the behaviour of a ptrace() memory write to a PRIVATE
file map.
The core worker is derived from KSM's replace_page() logic.
In essence, similar to KSM:
a. allocate a new page and copy over contents of the page that
has the uprobed vaddr
b. modify the copy and insert the breakpoint at the required
address
c. switch the original page with the copy containing the
breakpoint
d. flush page tables.
replace_page() is being replicated here because of some minor
changes in the type of pages and also because Hugh Dickins had
plans to improve replace_page() for KSM specific work.
Instruction analysis on x86 is based on instruction decoder and
determines if an instruction can be probed and determines the
necessary fixups after singlestep. Instruction analysis is done
at probe insertion time so that we avoid having to repeat the
same analysis every time a probe is hit.
A lot of code here is due to the improvement/suggestions/inputs
from Peter Zijlstra.
Changelog:
(v10):
- Add code to clear REX.B prefix as suggested by Denys Vlasenko
and Masami Hiramatsu.
(v9):
- Use insn_offset_modrm as suggested by Masami Hiramatsu.
(v7):
Handle comments from Peter Zijlstra:
- Dont take reference to inode. (expect inode to uprobe_register to be sane).
- Use PTR_ERR to set the return value.
- No need to take reference to inode.
- use PTR_ERR to return error value.
- register and uprobe_unregister share code.
(v5):
- Modified del_consumer as per comments from Peter.
- Drop reference to inode before dropping reference to uprobe.
- Use i_size_read(inode) instead of inode->i_size.
- Ensure uprobe->consumers is NULL, before __uprobe_unregister() is called.
- Includes errno.h as recommended by Stephen Rothwell to fix a build issue
on sparc defconfig
- Remove restrictions while unregistering.
- Earlier code leaked inode references under some conditions while
registering/unregistering.
- Continue the vma-rmap walk even if the intermediate vma doesnt
meet the requirements.
- Validate the vma found by find_vma before inserting/removing the
breakpoint
- Call del_consumer under mutex_lock.
- Use hash locks.
- Handle mremap.
- Introduce find_least_offset_node() instead of close match logic in
find_uprobe
- Uprobes no more depends on MM_OWNER; No reference to task_structs
while inserting/removing a probe.
- Uses read_mapping_page instead of grab_cache_page so that the pages
have valid content.
- pass NULL to get_user_pages for the task parameter.
- call SetPageUptodate on the new page allocated in write_opcode.
- fix leaking a reference to the new page under certain conditions.
- Include Instruction Decoder if Uprobes gets defined.
- Remove const attributes for instruction prefix arrays.
- Uses mm_context to know if the application is 32 bit.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Also-written-by: Jim Keniston <jkenisto@us.ibm.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Roland McGrath <roland@hack.frob.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Anton Arapov <anton@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Denys Vlasenko <vda.linux@googlemail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linux-mm <linux-mm@kvack.org>
Link: http://lkml.kernel.org/r/20120209092642.GE16600@linux.vnet.ibm.com
[ Made various small edits to the commit log ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-02-09 17:26:42 +08:00
|
|
|
struct vm_area_struct *vma;
|
2012-06-15 23:43:33 +08:00
|
|
|
struct map_info *curr = NULL;
|
|
|
|
struct map_info *prev = NULL;
|
|
|
|
struct map_info *info;
|
|
|
|
int more = 0;
|
uprobes, mm, x86: Add the ability to install and remove uprobes breakpoints
Add uprobes support to the core kernel, with x86 support.
This commit adds the kernel facilities, the actual uprobes
user-space ABI and perf probe support comes in later commits.
General design:
Uprobes are maintained in an rb-tree indexed by inode and offset
(the offset here is from the start of the mapping). For a unique
(inode, offset) tuple, there can be at most one uprobe in the
rb-tree.
Since the (inode, offset) tuple identifies a unique uprobe, more
than one user may be interested in the same uprobe. This provides
the ability to connect multiple 'consumers' to the same uprobe.
Each consumer defines a handler and a filter (optional). The
'handler' is run every time the uprobe is hit, if it matches the
'filter' criteria.
The first consumer of a uprobe causes the breakpoint to be
inserted at the specified address and subsequent consumers are
appended to this list. On subsequent probes, the consumer gets
appended to the existing list of consumers. The breakpoint is
removed when the last consumer unregisters. For all other
unregisterations, the consumer is removed from the list of
consumers.
Given a inode, we get a list of the mms that have mapped the
inode. Do the actual registration if mm maps the page where a
probe needs to be inserted/removed.
We use a temporary list to walk through the vmas that map the
inode.
- The number of maps that map the inode, is not known before we
walk the rmap and keeps changing.
- extending vm_area_struct wasn't recommended, it's a
size-critical data structure.
- There can be more than one maps of the inode in the same mm.
We add callbacks to the mmap methods to keep an eye on text vmas
that are of interest to uprobes. When a vma of interest is mapped,
we insert the breakpoint at the right address.
Uprobe works by replacing the instruction at the address defined
by (inode, offset) with the arch specific breakpoint
instruction. We save a copy of the original instruction at the
uprobed address.
This is needed for:
a. executing the instruction out-of-line (xol).
b. instruction analysis for any subsequent fixups.
c. restoring the instruction back when the uprobe is unregistered.
We insert or delete a breakpoint instruction, and this
breakpoint instruction is assumed to be the smallest instruction
available on the platform. For fixed size instruction platforms
this is trivially true, for variable size instruction platforms
the breakpoint instruction is typically the smallest (often a
single byte).
Writing the instruction is done by COWing the page and changing
the instruction during the copy, this even though most platforms
allow atomic writes of the breakpoint instruction. This also
mirrors the behaviour of a ptrace() memory write to a PRIVATE
file map.
The core worker is derived from KSM's replace_page() logic.
In essence, similar to KSM:
a. allocate a new page and copy over contents of the page that
has the uprobed vaddr
b. modify the copy and insert the breakpoint at the required
address
c. switch the original page with the copy containing the
breakpoint
d. flush page tables.
replace_page() is being replicated here because of some minor
changes in the type of pages and also because Hugh Dickins had
plans to improve replace_page() for KSM specific work.
Instruction analysis on x86 is based on instruction decoder and
determines if an instruction can be probed and determines the
necessary fixups after singlestep. Instruction analysis is done
at probe insertion time so that we avoid having to repeat the
same analysis every time a probe is hit.
A lot of code here is due to the improvement/suggestions/inputs
from Peter Zijlstra.
Changelog:
(v10):
- Add code to clear REX.B prefix as suggested by Denys Vlasenko
and Masami Hiramatsu.
(v9):
- Use insn_offset_modrm as suggested by Masami Hiramatsu.
(v7):
Handle comments from Peter Zijlstra:
- Dont take reference to inode. (expect inode to uprobe_register to be sane).
- Use PTR_ERR to set the return value.
- No need to take reference to inode.
- use PTR_ERR to return error value.
- register and uprobe_unregister share code.
(v5):
- Modified del_consumer as per comments from Peter.
- Drop reference to inode before dropping reference to uprobe.
- Use i_size_read(inode) instead of inode->i_size.
- Ensure uprobe->consumers is NULL, before __uprobe_unregister() is called.
- Includes errno.h as recommended by Stephen Rothwell to fix a build issue
on sparc defconfig
- Remove restrictions while unregistering.
- Earlier code leaked inode references under some conditions while
registering/unregistering.
- Continue the vma-rmap walk even if the intermediate vma doesnt
meet the requirements.
- Validate the vma found by find_vma before inserting/removing the
breakpoint
- Call del_consumer under mutex_lock.
- Use hash locks.
- Handle mremap.
- Introduce find_least_offset_node() instead of close match logic in
find_uprobe
- Uprobes no more depends on MM_OWNER; No reference to task_structs
while inserting/removing a probe.
- Uses read_mapping_page instead of grab_cache_page so that the pages
have valid content.
- pass NULL to get_user_pages for the task parameter.
- call SetPageUptodate on the new page allocated in write_opcode.
- fix leaking a reference to the new page under certain conditions.
- Include Instruction Decoder if Uprobes gets defined.
- Remove const attributes for instruction prefix arrays.
- Uses mm_context to know if the application is 32 bit.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Also-written-by: Jim Keniston <jkenisto@us.ibm.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Roland McGrath <roland@hack.frob.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Anton Arapov <anton@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Denys Vlasenko <vda.linux@googlemail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linux-mm <linux-mm@kvack.org>
Link: http://lkml.kernel.org/r/20120209092642.GE16600@linux.vnet.ibm.com
[ Made various small edits to the commit log ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-02-09 17:26:42 +08:00
|
|
|
|
2012-06-15 23:43:33 +08:00
|
|
|
again:
|
2014-12-13 08:54:30 +08:00
|
|
|
i_mmap_lock_read(mapping);
|
2012-10-09 07:31:25 +08:00
|
|
|
vma_interval_tree_foreach(vma, &mapping->i_mmap, pgoff, pgoff) {
|
uprobes, mm, x86: Add the ability to install and remove uprobes breakpoints
Add uprobes support to the core kernel, with x86 support.
This commit adds the kernel facilities, the actual uprobes
user-space ABI and perf probe support comes in later commits.
General design:
Uprobes are maintained in an rb-tree indexed by inode and offset
(the offset here is from the start of the mapping). For a unique
(inode, offset) tuple, there can be at most one uprobe in the
rb-tree.
Since the (inode, offset) tuple identifies a unique uprobe, more
than one user may be interested in the same uprobe. This provides
the ability to connect multiple 'consumers' to the same uprobe.
Each consumer defines a handler and a filter (optional). The
'handler' is run every time the uprobe is hit, if it matches the
'filter' criteria.
The first consumer of a uprobe causes the breakpoint to be
inserted at the specified address and subsequent consumers are
appended to this list. On subsequent probes, the consumer gets
appended to the existing list of consumers. The breakpoint is
removed when the last consumer unregisters. For all other
unregisterations, the consumer is removed from the list of
consumers.
Given a inode, we get a list of the mms that have mapped the
inode. Do the actual registration if mm maps the page where a
probe needs to be inserted/removed.
We use a temporary list to walk through the vmas that map the
inode.
- The number of maps that map the inode, is not known before we
walk the rmap and keeps changing.
- extending vm_area_struct wasn't recommended, it's a
size-critical data structure.
- There can be more than one maps of the inode in the same mm.
We add callbacks to the mmap methods to keep an eye on text vmas
that are of interest to uprobes. When a vma of interest is mapped,
we insert the breakpoint at the right address.
Uprobe works by replacing the instruction at the address defined
by (inode, offset) with the arch specific breakpoint
instruction. We save a copy of the original instruction at the
uprobed address.
This is needed for:
a. executing the instruction out-of-line (xol).
b. instruction analysis for any subsequent fixups.
c. restoring the instruction back when the uprobe is unregistered.
We insert or delete a breakpoint instruction, and this
breakpoint instruction is assumed to be the smallest instruction
available on the platform. For fixed size instruction platforms
this is trivially true, for variable size instruction platforms
the breakpoint instruction is typically the smallest (often a
single byte).
Writing the instruction is done by COWing the page and changing
the instruction during the copy, this even though most platforms
allow atomic writes of the breakpoint instruction. This also
mirrors the behaviour of a ptrace() memory write to a PRIVATE
file map.
The core worker is derived from KSM's replace_page() logic.
In essence, similar to KSM:
a. allocate a new page and copy over contents of the page that
has the uprobed vaddr
b. modify the copy and insert the breakpoint at the required
address
c. switch the original page with the copy containing the
breakpoint
d. flush page tables.
replace_page() is being replicated here because of some minor
changes in the type of pages and also because Hugh Dickins had
plans to improve replace_page() for KSM specific work.
Instruction analysis on x86 is based on instruction decoder and
determines if an instruction can be probed and determines the
necessary fixups after singlestep. Instruction analysis is done
at probe insertion time so that we avoid having to repeat the
same analysis every time a probe is hit.
A lot of code here is due to the improvement/suggestions/inputs
from Peter Zijlstra.
Changelog:
(v10):
- Add code to clear REX.B prefix as suggested by Denys Vlasenko
and Masami Hiramatsu.
(v9):
- Use insn_offset_modrm as suggested by Masami Hiramatsu.
(v7):
Handle comments from Peter Zijlstra:
- Dont take reference to inode. (expect inode to uprobe_register to be sane).
- Use PTR_ERR to set the return value.
- No need to take reference to inode.
- use PTR_ERR to return error value.
- register and uprobe_unregister share code.
(v5):
- Modified del_consumer as per comments from Peter.
- Drop reference to inode before dropping reference to uprobe.
- Use i_size_read(inode) instead of inode->i_size.
- Ensure uprobe->consumers is NULL, before __uprobe_unregister() is called.
- Includes errno.h as recommended by Stephen Rothwell to fix a build issue
on sparc defconfig
- Remove restrictions while unregistering.
- Earlier code leaked inode references under some conditions while
registering/unregistering.
- Continue the vma-rmap walk even if the intermediate vma doesnt
meet the requirements.
- Validate the vma found by find_vma before inserting/removing the
breakpoint
- Call del_consumer under mutex_lock.
- Use hash locks.
- Handle mremap.
- Introduce find_least_offset_node() instead of close match logic in
find_uprobe
- Uprobes no more depends on MM_OWNER; No reference to task_structs
while inserting/removing a probe.
- Uses read_mapping_page instead of grab_cache_page so that the pages
have valid content.
- pass NULL to get_user_pages for the task parameter.
- call SetPageUptodate on the new page allocated in write_opcode.
- fix leaking a reference to the new page under certain conditions.
- Include Instruction Decoder if Uprobes gets defined.
- Remove const attributes for instruction prefix arrays.
- Uses mm_context to know if the application is 32 bit.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Also-written-by: Jim Keniston <jkenisto@us.ibm.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Roland McGrath <roland@hack.frob.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Anton Arapov <anton@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Denys Vlasenko <vda.linux@googlemail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linux-mm <linux-mm@kvack.org>
Link: http://lkml.kernel.org/r/20120209092642.GE16600@linux.vnet.ibm.com
[ Made various small edits to the commit log ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-02-09 17:26:42 +08:00
|
|
|
if (!valid_vma(vma, is_register))
|
|
|
|
continue;
|
|
|
|
|
2012-06-15 23:43:36 +08:00
|
|
|
if (!prev && !more) {
|
|
|
|
/*
|
2014-12-13 08:54:24 +08:00
|
|
|
* Needs GFP_NOWAIT to avoid i_mmap_rwsem recursion through
|
2012-06-15 23:43:36 +08:00
|
|
|
* reclaim. This is optimistic, no harm done if it fails.
|
|
|
|
*/
|
|
|
|
prev = kmalloc(sizeof(struct map_info),
|
|
|
|
GFP_NOWAIT | __GFP_NOMEMALLOC | __GFP_NOWARN);
|
|
|
|
if (prev)
|
|
|
|
prev->next = NULL;
|
|
|
|
}
|
2012-06-15 23:43:33 +08:00
|
|
|
if (!prev) {
|
|
|
|
more++;
|
|
|
|
continue;
|
uprobes, mm, x86: Add the ability to install and remove uprobes breakpoints
Add uprobes support to the core kernel, with x86 support.
This commit adds the kernel facilities, the actual uprobes
user-space ABI and perf probe support comes in later commits.
General design:
Uprobes are maintained in an rb-tree indexed by inode and offset
(the offset here is from the start of the mapping). For a unique
(inode, offset) tuple, there can be at most one uprobe in the
rb-tree.
Since the (inode, offset) tuple identifies a unique uprobe, more
than one user may be interested in the same uprobe. This provides
the ability to connect multiple 'consumers' to the same uprobe.
Each consumer defines a handler and a filter (optional). The
'handler' is run every time the uprobe is hit, if it matches the
'filter' criteria.
The first consumer of a uprobe causes the breakpoint to be
inserted at the specified address and subsequent consumers are
appended to this list. On subsequent probes, the consumer gets
appended to the existing list of consumers. The breakpoint is
removed when the last consumer unregisters. For all other
unregisterations, the consumer is removed from the list of
consumers.
Given a inode, we get a list of the mms that have mapped the
inode. Do the actual registration if mm maps the page where a
probe needs to be inserted/removed.
We use a temporary list to walk through the vmas that map the
inode.
- The number of maps that map the inode, is not known before we
walk the rmap and keeps changing.
- extending vm_area_struct wasn't recommended, it's a
size-critical data structure.
- There can be more than one maps of the inode in the same mm.
We add callbacks to the mmap methods to keep an eye on text vmas
that are of interest to uprobes. When a vma of interest is mapped,
we insert the breakpoint at the right address.
Uprobe works by replacing the instruction at the address defined
by (inode, offset) with the arch specific breakpoint
instruction. We save a copy of the original instruction at the
uprobed address.
This is needed for:
a. executing the instruction out-of-line (xol).
b. instruction analysis for any subsequent fixups.
c. restoring the instruction back when the uprobe is unregistered.
We insert or delete a breakpoint instruction, and this
breakpoint instruction is assumed to be the smallest instruction
available on the platform. For fixed size instruction platforms
this is trivially true, for variable size instruction platforms
the breakpoint instruction is typically the smallest (often a
single byte).
Writing the instruction is done by COWing the page and changing
the instruction during the copy, this even though most platforms
allow atomic writes of the breakpoint instruction. This also
mirrors the behaviour of a ptrace() memory write to a PRIVATE
file map.
The core worker is derived from KSM's replace_page() logic.
In essence, similar to KSM:
a. allocate a new page and copy over contents of the page that
has the uprobed vaddr
b. modify the copy and insert the breakpoint at the required
address
c. switch the original page with the copy containing the
breakpoint
d. flush page tables.
replace_page() is being replicated here because of some minor
changes in the type of pages and also because Hugh Dickins had
plans to improve replace_page() for KSM specific work.
Instruction analysis on x86 is based on instruction decoder and
determines if an instruction can be probed and determines the
necessary fixups after singlestep. Instruction analysis is done
at probe insertion time so that we avoid having to repeat the
same analysis every time a probe is hit.
A lot of code here is due to the improvement/suggestions/inputs
from Peter Zijlstra.
Changelog:
(v10):
- Add code to clear REX.B prefix as suggested by Denys Vlasenko
and Masami Hiramatsu.
(v9):
- Use insn_offset_modrm as suggested by Masami Hiramatsu.
(v7):
Handle comments from Peter Zijlstra:
- Dont take reference to inode. (expect inode to uprobe_register to be sane).
- Use PTR_ERR to set the return value.
- No need to take reference to inode.
- use PTR_ERR to return error value.
- register and uprobe_unregister share code.
(v5):
- Modified del_consumer as per comments from Peter.
- Drop reference to inode before dropping reference to uprobe.
- Use i_size_read(inode) instead of inode->i_size.
- Ensure uprobe->consumers is NULL, before __uprobe_unregister() is called.
- Includes errno.h as recommended by Stephen Rothwell to fix a build issue
on sparc defconfig
- Remove restrictions while unregistering.
- Earlier code leaked inode references under some conditions while
registering/unregistering.
- Continue the vma-rmap walk even if the intermediate vma doesnt
meet the requirements.
- Validate the vma found by find_vma before inserting/removing the
breakpoint
- Call del_consumer under mutex_lock.
- Use hash locks.
- Handle mremap.
- Introduce find_least_offset_node() instead of close match logic in
find_uprobe
- Uprobes no more depends on MM_OWNER; No reference to task_structs
while inserting/removing a probe.
- Uses read_mapping_page instead of grab_cache_page so that the pages
have valid content.
- pass NULL to get_user_pages for the task parameter.
- call SetPageUptodate on the new page allocated in write_opcode.
- fix leaking a reference to the new page under certain conditions.
- Include Instruction Decoder if Uprobes gets defined.
- Remove const attributes for instruction prefix arrays.
- Uses mm_context to know if the application is 32 bit.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Also-written-by: Jim Keniston <jkenisto@us.ibm.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Roland McGrath <roland@hack.frob.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Anton Arapov <anton@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Denys Vlasenko <vda.linux@googlemail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linux-mm <linux-mm@kvack.org>
Link: http://lkml.kernel.org/r/20120209092642.GE16600@linux.vnet.ibm.com
[ Made various small edits to the commit log ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-02-09 17:26:42 +08:00
|
|
|
}
|
|
|
|
|
2017-02-28 06:30:13 +08:00
|
|
|
if (!mmget_not_zero(vma->vm_mm))
|
2012-06-15 23:43:33 +08:00
|
|
|
continue;
|
2012-02-17 16:27:41 +08:00
|
|
|
|
2012-06-15 23:43:33 +08:00
|
|
|
info = prev;
|
|
|
|
prev = prev->next;
|
|
|
|
info->next = curr;
|
|
|
|
curr = info;
|
uprobes, mm, x86: Add the ability to install and remove uprobes breakpoints
Add uprobes support to the core kernel, with x86 support.
This commit adds the kernel facilities, the actual uprobes
user-space ABI and perf probe support comes in later commits.
General design:
Uprobes are maintained in an rb-tree indexed by inode and offset
(the offset here is from the start of the mapping). For a unique
(inode, offset) tuple, there can be at most one uprobe in the
rb-tree.
Since the (inode, offset) tuple identifies a unique uprobe, more
than one user may be interested in the same uprobe. This provides
the ability to connect multiple 'consumers' to the same uprobe.
Each consumer defines a handler and a filter (optional). The
'handler' is run every time the uprobe is hit, if it matches the
'filter' criteria.
The first consumer of a uprobe causes the breakpoint to be
inserted at the specified address and subsequent consumers are
appended to this list. On subsequent probes, the consumer gets
appended to the existing list of consumers. The breakpoint is
removed when the last consumer unregisters. For all other
unregisterations, the consumer is removed from the list of
consumers.
Given a inode, we get a list of the mms that have mapped the
inode. Do the actual registration if mm maps the page where a
probe needs to be inserted/removed.
We use a temporary list to walk through the vmas that map the
inode.
- The number of maps that map the inode, is not known before we
walk the rmap and keeps changing.
- extending vm_area_struct wasn't recommended, it's a
size-critical data structure.
- There can be more than one maps of the inode in the same mm.
We add callbacks to the mmap methods to keep an eye on text vmas
that are of interest to uprobes. When a vma of interest is mapped,
we insert the breakpoint at the right address.
Uprobe works by replacing the instruction at the address defined
by (inode, offset) with the arch specific breakpoint
instruction. We save a copy of the original instruction at the
uprobed address.
This is needed for:
a. executing the instruction out-of-line (xol).
b. instruction analysis for any subsequent fixups.
c. restoring the instruction back when the uprobe is unregistered.
We insert or delete a breakpoint instruction, and this
breakpoint instruction is assumed to be the smallest instruction
available on the platform. For fixed size instruction platforms
this is trivially true, for variable size instruction platforms
the breakpoint instruction is typically the smallest (often a
single byte).
Writing the instruction is done by COWing the page and changing
the instruction during the copy, this even though most platforms
allow atomic writes of the breakpoint instruction. This also
mirrors the behaviour of a ptrace() memory write to a PRIVATE
file map.
The core worker is derived from KSM's replace_page() logic.
In essence, similar to KSM:
a. allocate a new page and copy over contents of the page that
has the uprobed vaddr
b. modify the copy and insert the breakpoint at the required
address
c. switch the original page with the copy containing the
breakpoint
d. flush page tables.
replace_page() is being replicated here because of some minor
changes in the type of pages and also because Hugh Dickins had
plans to improve replace_page() for KSM specific work.
Instruction analysis on x86 is based on instruction decoder and
determines if an instruction can be probed and determines the
necessary fixups after singlestep. Instruction analysis is done
at probe insertion time so that we avoid having to repeat the
same analysis every time a probe is hit.
A lot of code here is due to the improvement/suggestions/inputs
from Peter Zijlstra.
Changelog:
(v10):
- Add code to clear REX.B prefix as suggested by Denys Vlasenko
and Masami Hiramatsu.
(v9):
- Use insn_offset_modrm as suggested by Masami Hiramatsu.
(v7):
Handle comments from Peter Zijlstra:
- Dont take reference to inode. (expect inode to uprobe_register to be sane).
- Use PTR_ERR to set the return value.
- No need to take reference to inode.
- use PTR_ERR to return error value.
- register and uprobe_unregister share code.
(v5):
- Modified del_consumer as per comments from Peter.
- Drop reference to inode before dropping reference to uprobe.
- Use i_size_read(inode) instead of inode->i_size.
- Ensure uprobe->consumers is NULL, before __uprobe_unregister() is called.
- Includes errno.h as recommended by Stephen Rothwell to fix a build issue
on sparc defconfig
- Remove restrictions while unregistering.
- Earlier code leaked inode references under some conditions while
registering/unregistering.
- Continue the vma-rmap walk even if the intermediate vma doesnt
meet the requirements.
- Validate the vma found by find_vma before inserting/removing the
breakpoint
- Call del_consumer under mutex_lock.
- Use hash locks.
- Handle mremap.
- Introduce find_least_offset_node() instead of close match logic in
find_uprobe
- Uprobes no more depends on MM_OWNER; No reference to task_structs
while inserting/removing a probe.
- Uses read_mapping_page instead of grab_cache_page so that the pages
have valid content.
- pass NULL to get_user_pages for the task parameter.
- call SetPageUptodate on the new page allocated in write_opcode.
- fix leaking a reference to the new page under certain conditions.
- Include Instruction Decoder if Uprobes gets defined.
- Remove const attributes for instruction prefix arrays.
- Uses mm_context to know if the application is 32 bit.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Also-written-by: Jim Keniston <jkenisto@us.ibm.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Roland McGrath <roland@hack.frob.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Anton Arapov <anton@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Denys Vlasenko <vda.linux@googlemail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linux-mm <linux-mm@kvack.org>
Link: http://lkml.kernel.org/r/20120209092642.GE16600@linux.vnet.ibm.com
[ Made various small edits to the commit log ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-02-09 17:26:42 +08:00
|
|
|
|
2012-06-15 23:43:33 +08:00
|
|
|
info->mm = vma->vm_mm;
|
2012-07-30 02:22:47 +08:00
|
|
|
info->vaddr = offset_to_vaddr(vma, offset);
|
2012-06-15 23:43:33 +08:00
|
|
|
}
|
2014-12-13 08:54:30 +08:00
|
|
|
i_mmap_unlock_read(mapping);
|
uprobes, mm, x86: Add the ability to install and remove uprobes breakpoints
Add uprobes support to the core kernel, with x86 support.
This commit adds the kernel facilities, the actual uprobes
user-space ABI and perf probe support comes in later commits.
General design:
Uprobes are maintained in an rb-tree indexed by inode and offset
(the offset here is from the start of the mapping). For a unique
(inode, offset) tuple, there can be at most one uprobe in the
rb-tree.
Since the (inode, offset) tuple identifies a unique uprobe, more
than one user may be interested in the same uprobe. This provides
the ability to connect multiple 'consumers' to the same uprobe.
Each consumer defines a handler and a filter (optional). The
'handler' is run every time the uprobe is hit, if it matches the
'filter' criteria.
The first consumer of a uprobe causes the breakpoint to be
inserted at the specified address and subsequent consumers are
appended to this list. On subsequent probes, the consumer gets
appended to the existing list of consumers. The breakpoint is
removed when the last consumer unregisters. For all other
unregisterations, the consumer is removed from the list of
consumers.
Given a inode, we get a list of the mms that have mapped the
inode. Do the actual registration if mm maps the page where a
probe needs to be inserted/removed.
We use a temporary list to walk through the vmas that map the
inode.
- The number of maps that map the inode, is not known before we
walk the rmap and keeps changing.
- extending vm_area_struct wasn't recommended, it's a
size-critical data structure.
- There can be more than one maps of the inode in the same mm.
We add callbacks to the mmap methods to keep an eye on text vmas
that are of interest to uprobes. When a vma of interest is mapped,
we insert the breakpoint at the right address.
Uprobe works by replacing the instruction at the address defined
by (inode, offset) with the arch specific breakpoint
instruction. We save a copy of the original instruction at the
uprobed address.
This is needed for:
a. executing the instruction out-of-line (xol).
b. instruction analysis for any subsequent fixups.
c. restoring the instruction back when the uprobe is unregistered.
We insert or delete a breakpoint instruction, and this
breakpoint instruction is assumed to be the smallest instruction
available on the platform. For fixed size instruction platforms
this is trivially true, for variable size instruction platforms
the breakpoint instruction is typically the smallest (often a
single byte).
Writing the instruction is done by COWing the page and changing
the instruction during the copy, this even though most platforms
allow atomic writes of the breakpoint instruction. This also
mirrors the behaviour of a ptrace() memory write to a PRIVATE
file map.
The core worker is derived from KSM's replace_page() logic.
In essence, similar to KSM:
a. allocate a new page and copy over contents of the page that
has the uprobed vaddr
b. modify the copy and insert the breakpoint at the required
address
c. switch the original page with the copy containing the
breakpoint
d. flush page tables.
replace_page() is being replicated here because of some minor
changes in the type of pages and also because Hugh Dickins had
plans to improve replace_page() for KSM specific work.
Instruction analysis on x86 is based on instruction decoder and
determines if an instruction can be probed and determines the
necessary fixups after singlestep. Instruction analysis is done
at probe insertion time so that we avoid having to repeat the
same analysis every time a probe is hit.
A lot of code here is due to the improvement/suggestions/inputs
from Peter Zijlstra.
Changelog:
(v10):
- Add code to clear REX.B prefix as suggested by Denys Vlasenko
and Masami Hiramatsu.
(v9):
- Use insn_offset_modrm as suggested by Masami Hiramatsu.
(v7):
Handle comments from Peter Zijlstra:
- Dont take reference to inode. (expect inode to uprobe_register to be sane).
- Use PTR_ERR to set the return value.
- No need to take reference to inode.
- use PTR_ERR to return error value.
- register and uprobe_unregister share code.
(v5):
- Modified del_consumer as per comments from Peter.
- Drop reference to inode before dropping reference to uprobe.
- Use i_size_read(inode) instead of inode->i_size.
- Ensure uprobe->consumers is NULL, before __uprobe_unregister() is called.
- Includes errno.h as recommended by Stephen Rothwell to fix a build issue
on sparc defconfig
- Remove restrictions while unregistering.
- Earlier code leaked inode references under some conditions while
registering/unregistering.
- Continue the vma-rmap walk even if the intermediate vma doesnt
meet the requirements.
- Validate the vma found by find_vma before inserting/removing the
breakpoint
- Call del_consumer under mutex_lock.
- Use hash locks.
- Handle mremap.
- Introduce find_least_offset_node() instead of close match logic in
find_uprobe
- Uprobes no more depends on MM_OWNER; No reference to task_structs
while inserting/removing a probe.
- Uses read_mapping_page instead of grab_cache_page so that the pages
have valid content.
- pass NULL to get_user_pages for the task parameter.
- call SetPageUptodate on the new page allocated in write_opcode.
- fix leaking a reference to the new page under certain conditions.
- Include Instruction Decoder if Uprobes gets defined.
- Remove const attributes for instruction prefix arrays.
- Uses mm_context to know if the application is 32 bit.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Also-written-by: Jim Keniston <jkenisto@us.ibm.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Roland McGrath <roland@hack.frob.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Anton Arapov <anton@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Denys Vlasenko <vda.linux@googlemail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linux-mm <linux-mm@kvack.org>
Link: http://lkml.kernel.org/r/20120209092642.GE16600@linux.vnet.ibm.com
[ Made various small edits to the commit log ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-02-09 17:26:42 +08:00
|
|
|
|
2012-06-15 23:43:33 +08:00
|
|
|
if (!more)
|
|
|
|
goto out;
|
|
|
|
|
|
|
|
prev = curr;
|
|
|
|
while (curr) {
|
|
|
|
mmput(curr->mm);
|
|
|
|
curr = curr->next;
|
|
|
|
}
|
2012-02-17 16:27:41 +08:00
|
|
|
|
2012-06-15 23:43:33 +08:00
|
|
|
do {
|
|
|
|
info = kmalloc(sizeof(struct map_info), GFP_KERNEL);
|
|
|
|
if (!info) {
|
|
|
|
curr = ERR_PTR(-ENOMEM);
|
|
|
|
goto out;
|
|
|
|
}
|
|
|
|
info->next = prev;
|
|
|
|
prev = info;
|
|
|
|
} while (--more);
|
|
|
|
|
|
|
|
goto again;
|
|
|
|
out:
|
|
|
|
while (prev)
|
|
|
|
prev = free_map_info(prev);
|
|
|
|
return curr;
|
uprobes, mm, x86: Add the ability to install and remove uprobes breakpoints
Add uprobes support to the core kernel, with x86 support.
This commit adds the kernel facilities, the actual uprobes
user-space ABI and perf probe support comes in later commits.
General design:
Uprobes are maintained in an rb-tree indexed by inode and offset
(the offset here is from the start of the mapping). For a unique
(inode, offset) tuple, there can be at most one uprobe in the
rb-tree.
Since the (inode, offset) tuple identifies a unique uprobe, more
than one user may be interested in the same uprobe. This provides
the ability to connect multiple 'consumers' to the same uprobe.
Each consumer defines a handler and a filter (optional). The
'handler' is run every time the uprobe is hit, if it matches the
'filter' criteria.
The first consumer of a uprobe causes the breakpoint to be
inserted at the specified address and subsequent consumers are
appended to this list. On subsequent probes, the consumer gets
appended to the existing list of consumers. The breakpoint is
removed when the last consumer unregisters. For all other
unregisterations, the consumer is removed from the list of
consumers.
Given a inode, we get a list of the mms that have mapped the
inode. Do the actual registration if mm maps the page where a
probe needs to be inserted/removed.
We use a temporary list to walk through the vmas that map the
inode.
- The number of maps that map the inode, is not known before we
walk the rmap and keeps changing.
- extending vm_area_struct wasn't recommended, it's a
size-critical data structure.
- There can be more than one maps of the inode in the same mm.
We add callbacks to the mmap methods to keep an eye on text vmas
that are of interest to uprobes. When a vma of interest is mapped,
we insert the breakpoint at the right address.
Uprobe works by replacing the instruction at the address defined
by (inode, offset) with the arch specific breakpoint
instruction. We save a copy of the original instruction at the
uprobed address.
This is needed for:
a. executing the instruction out-of-line (xol).
b. instruction analysis for any subsequent fixups.
c. restoring the instruction back when the uprobe is unregistered.
We insert or delete a breakpoint instruction, and this
breakpoint instruction is assumed to be the smallest instruction
available on the platform. For fixed size instruction platforms
this is trivially true, for variable size instruction platforms
the breakpoint instruction is typically the smallest (often a
single byte).
Writing the instruction is done by COWing the page and changing
the instruction during the copy, this even though most platforms
allow atomic writes of the breakpoint instruction. This also
mirrors the behaviour of a ptrace() memory write to a PRIVATE
file map.
The core worker is derived from KSM's replace_page() logic.
In essence, similar to KSM:
a. allocate a new page and copy over contents of the page that
has the uprobed vaddr
b. modify the copy and insert the breakpoint at the required
address
c. switch the original page with the copy containing the
breakpoint
d. flush page tables.
replace_page() is being replicated here because of some minor
changes in the type of pages and also because Hugh Dickins had
plans to improve replace_page() for KSM specific work.
Instruction analysis on x86 is based on instruction decoder and
determines if an instruction can be probed and determines the
necessary fixups after singlestep. Instruction analysis is done
at probe insertion time so that we avoid having to repeat the
same analysis every time a probe is hit.
A lot of code here is due to the improvement/suggestions/inputs
from Peter Zijlstra.
Changelog:
(v10):
- Add code to clear REX.B prefix as suggested by Denys Vlasenko
and Masami Hiramatsu.
(v9):
- Use insn_offset_modrm as suggested by Masami Hiramatsu.
(v7):
Handle comments from Peter Zijlstra:
- Dont take reference to inode. (expect inode to uprobe_register to be sane).
- Use PTR_ERR to set the return value.
- No need to take reference to inode.
- use PTR_ERR to return error value.
- register and uprobe_unregister share code.
(v5):
- Modified del_consumer as per comments from Peter.
- Drop reference to inode before dropping reference to uprobe.
- Use i_size_read(inode) instead of inode->i_size.
- Ensure uprobe->consumers is NULL, before __uprobe_unregister() is called.
- Includes errno.h as recommended by Stephen Rothwell to fix a build issue
on sparc defconfig
- Remove restrictions while unregistering.
- Earlier code leaked inode references under some conditions while
registering/unregistering.
- Continue the vma-rmap walk even if the intermediate vma doesnt
meet the requirements.
- Validate the vma found by find_vma before inserting/removing the
breakpoint
- Call del_consumer under mutex_lock.
- Use hash locks.
- Handle mremap.
- Introduce find_least_offset_node() instead of close match logic in
find_uprobe
- Uprobes no more depends on MM_OWNER; No reference to task_structs
while inserting/removing a probe.
- Uses read_mapping_page instead of grab_cache_page so that the pages
have valid content.
- pass NULL to get_user_pages for the task parameter.
- call SetPageUptodate on the new page allocated in write_opcode.
- fix leaking a reference to the new page under certain conditions.
- Include Instruction Decoder if Uprobes gets defined.
- Remove const attributes for instruction prefix arrays.
- Uses mm_context to know if the application is 32 bit.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Also-written-by: Jim Keniston <jkenisto@us.ibm.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Roland McGrath <roland@hack.frob.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Anton Arapov <anton@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Denys Vlasenko <vda.linux@googlemail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linux-mm <linux-mm@kvack.org>
Link: http://lkml.kernel.org/r/20120209092642.GE16600@linux.vnet.ibm.com
[ Made various small edits to the commit log ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-02-09 17:26:42 +08:00
|
|
|
}
|
|
|
|
|
uprobes: Introduce uprobe_apply()
Currently it is not possible to change the filtering constraints after
uprobe_register(), so a consumer can not, say, start to trace a task/mm
which was previously filtered out, or remove the no longer needed bp's.
Introduce uprobe_apply() which simply does register_for_each_vma() again
to consult uprobe_consumer->filter() and install/remove the breakpoints.
The only complication is that register_for_each_vma() can no longer
assume that uprobe->consumers should be consulter if is_register == T,
so we change it to accept "struct uprobe_consumer *new" instead.
Unlike uprobe_register(), uprobe_apply(true) doesn't do "unregister" if
register_for_each_vma() fails, it is up to caller to handle the error.
Note: we probably need to cleanup the current interface, it is strange
that uprobe_apply/unregister need inode/offset. We should either change
uprobe_register() to return "struct uprobe *", or add a private ->uprobe
member in uprobe_consumer. And in the long term uprobe_apply() should
take a single argument, uprobe or consumer, even "bool add" should go
away.
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
2013-02-04 02:21:12 +08:00
|
|
|
static int
|
|
|
|
register_for_each_vma(struct uprobe *uprobe, struct uprobe_consumer *new)
|
uprobes, mm, x86: Add the ability to install and remove uprobes breakpoints
Add uprobes support to the core kernel, with x86 support.
This commit adds the kernel facilities, the actual uprobes
user-space ABI and perf probe support comes in later commits.
General design:
Uprobes are maintained in an rb-tree indexed by inode and offset
(the offset here is from the start of the mapping). For a unique
(inode, offset) tuple, there can be at most one uprobe in the
rb-tree.
Since the (inode, offset) tuple identifies a unique uprobe, more
than one user may be interested in the same uprobe. This provides
the ability to connect multiple 'consumers' to the same uprobe.
Each consumer defines a handler and a filter (optional). The
'handler' is run every time the uprobe is hit, if it matches the
'filter' criteria.
The first consumer of a uprobe causes the breakpoint to be
inserted at the specified address and subsequent consumers are
appended to this list. On subsequent probes, the consumer gets
appended to the existing list of consumers. The breakpoint is
removed when the last consumer unregisters. For all other
unregisterations, the consumer is removed from the list of
consumers.
Given a inode, we get a list of the mms that have mapped the
inode. Do the actual registration if mm maps the page where a
probe needs to be inserted/removed.
We use a temporary list to walk through the vmas that map the
inode.
- The number of maps that map the inode, is not known before we
walk the rmap and keeps changing.
- extending vm_area_struct wasn't recommended, it's a
size-critical data structure.
- There can be more than one maps of the inode in the same mm.
We add callbacks to the mmap methods to keep an eye on text vmas
that are of interest to uprobes. When a vma of interest is mapped,
we insert the breakpoint at the right address.
Uprobe works by replacing the instruction at the address defined
by (inode, offset) with the arch specific breakpoint
instruction. We save a copy of the original instruction at the
uprobed address.
This is needed for:
a. executing the instruction out-of-line (xol).
b. instruction analysis for any subsequent fixups.
c. restoring the instruction back when the uprobe is unregistered.
We insert or delete a breakpoint instruction, and this
breakpoint instruction is assumed to be the smallest instruction
available on the platform. For fixed size instruction platforms
this is trivially true, for variable size instruction platforms
the breakpoint instruction is typically the smallest (often a
single byte).
Writing the instruction is done by COWing the page and changing
the instruction during the copy, this even though most platforms
allow atomic writes of the breakpoint instruction. This also
mirrors the behaviour of a ptrace() memory write to a PRIVATE
file map.
The core worker is derived from KSM's replace_page() logic.
In essence, similar to KSM:
a. allocate a new page and copy over contents of the page that
has the uprobed vaddr
b. modify the copy and insert the breakpoint at the required
address
c. switch the original page with the copy containing the
breakpoint
d. flush page tables.
replace_page() is being replicated here because of some minor
changes in the type of pages and also because Hugh Dickins had
plans to improve replace_page() for KSM specific work.
Instruction analysis on x86 is based on instruction decoder and
determines if an instruction can be probed and determines the
necessary fixups after singlestep. Instruction analysis is done
at probe insertion time so that we avoid having to repeat the
same analysis every time a probe is hit.
A lot of code here is due to the improvement/suggestions/inputs
from Peter Zijlstra.
Changelog:
(v10):
- Add code to clear REX.B prefix as suggested by Denys Vlasenko
and Masami Hiramatsu.
(v9):
- Use insn_offset_modrm as suggested by Masami Hiramatsu.
(v7):
Handle comments from Peter Zijlstra:
- Dont take reference to inode. (expect inode to uprobe_register to be sane).
- Use PTR_ERR to set the return value.
- No need to take reference to inode.
- use PTR_ERR to return error value.
- register and uprobe_unregister share code.
(v5):
- Modified del_consumer as per comments from Peter.
- Drop reference to inode before dropping reference to uprobe.
- Use i_size_read(inode) instead of inode->i_size.
- Ensure uprobe->consumers is NULL, before __uprobe_unregister() is called.
- Includes errno.h as recommended by Stephen Rothwell to fix a build issue
on sparc defconfig
- Remove restrictions while unregistering.
- Earlier code leaked inode references under some conditions while
registering/unregistering.
- Continue the vma-rmap walk even if the intermediate vma doesnt
meet the requirements.
- Validate the vma found by find_vma before inserting/removing the
breakpoint
- Call del_consumer under mutex_lock.
- Use hash locks.
- Handle mremap.
- Introduce find_least_offset_node() instead of close match logic in
find_uprobe
- Uprobes no more depends on MM_OWNER; No reference to task_structs
while inserting/removing a probe.
- Uses read_mapping_page instead of grab_cache_page so that the pages
have valid content.
- pass NULL to get_user_pages for the task parameter.
- call SetPageUptodate on the new page allocated in write_opcode.
- fix leaking a reference to the new page under certain conditions.
- Include Instruction Decoder if Uprobes gets defined.
- Remove const attributes for instruction prefix arrays.
- Uses mm_context to know if the application is 32 bit.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Also-written-by: Jim Keniston <jkenisto@us.ibm.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Roland McGrath <roland@hack.frob.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Anton Arapov <anton@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Denys Vlasenko <vda.linux@googlemail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linux-mm <linux-mm@kvack.org>
Link: http://lkml.kernel.org/r/20120209092642.GE16600@linux.vnet.ibm.com
[ Made various small edits to the commit log ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-02-09 17:26:42 +08:00
|
|
|
{
|
uprobes: Introduce uprobe_apply()
Currently it is not possible to change the filtering constraints after
uprobe_register(), so a consumer can not, say, start to trace a task/mm
which was previously filtered out, or remove the no longer needed bp's.
Introduce uprobe_apply() which simply does register_for_each_vma() again
to consult uprobe_consumer->filter() and install/remove the breakpoints.
The only complication is that register_for_each_vma() can no longer
assume that uprobe->consumers should be consulter if is_register == T,
so we change it to accept "struct uprobe_consumer *new" instead.
Unlike uprobe_register(), uprobe_apply(true) doesn't do "unregister" if
register_for_each_vma() fails, it is up to caller to handle the error.
Note: we probably need to cleanup the current interface, it is strange
that uprobe_apply/unregister need inode/offset. We should either change
uprobe_register() to return "struct uprobe *", or add a private ->uprobe
member in uprobe_consumer. And in the long term uprobe_apply() should
take a single argument, uprobe or consumer, even "bool add" should go
away.
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
2013-02-04 02:21:12 +08:00
|
|
|
bool is_register = !!new;
|
2012-06-15 23:43:33 +08:00
|
|
|
struct map_info *info;
|
|
|
|
int err = 0;
|
uprobes, mm, x86: Add the ability to install and remove uprobes breakpoints
Add uprobes support to the core kernel, with x86 support.
This commit adds the kernel facilities, the actual uprobes
user-space ABI and perf probe support comes in later commits.
General design:
Uprobes are maintained in an rb-tree indexed by inode and offset
(the offset here is from the start of the mapping). For a unique
(inode, offset) tuple, there can be at most one uprobe in the
rb-tree.
Since the (inode, offset) tuple identifies a unique uprobe, more
than one user may be interested in the same uprobe. This provides
the ability to connect multiple 'consumers' to the same uprobe.
Each consumer defines a handler and a filter (optional). The
'handler' is run every time the uprobe is hit, if it matches the
'filter' criteria.
The first consumer of a uprobe causes the breakpoint to be
inserted at the specified address and subsequent consumers are
appended to this list. On subsequent probes, the consumer gets
appended to the existing list of consumers. The breakpoint is
removed when the last consumer unregisters. For all other
unregisterations, the consumer is removed from the list of
consumers.
Given a inode, we get a list of the mms that have mapped the
inode. Do the actual registration if mm maps the page where a
probe needs to be inserted/removed.
We use a temporary list to walk through the vmas that map the
inode.
- The number of maps that map the inode, is not known before we
walk the rmap and keeps changing.
- extending vm_area_struct wasn't recommended, it's a
size-critical data structure.
- There can be more than one maps of the inode in the same mm.
We add callbacks to the mmap methods to keep an eye on text vmas
that are of interest to uprobes. When a vma of interest is mapped,
we insert the breakpoint at the right address.
Uprobe works by replacing the instruction at the address defined
by (inode, offset) with the arch specific breakpoint
instruction. We save a copy of the original instruction at the
uprobed address.
This is needed for:
a. executing the instruction out-of-line (xol).
b. instruction analysis for any subsequent fixups.
c. restoring the instruction back when the uprobe is unregistered.
We insert or delete a breakpoint instruction, and this
breakpoint instruction is assumed to be the smallest instruction
available on the platform. For fixed size instruction platforms
this is trivially true, for variable size instruction platforms
the breakpoint instruction is typically the smallest (often a
single byte).
Writing the instruction is done by COWing the page and changing
the instruction during the copy, this even though most platforms
allow atomic writes of the breakpoint instruction. This also
mirrors the behaviour of a ptrace() memory write to a PRIVATE
file map.
The core worker is derived from KSM's replace_page() logic.
In essence, similar to KSM:
a. allocate a new page and copy over contents of the page that
has the uprobed vaddr
b. modify the copy and insert the breakpoint at the required
address
c. switch the original page with the copy containing the
breakpoint
d. flush page tables.
replace_page() is being replicated here because of some minor
changes in the type of pages and also because Hugh Dickins had
plans to improve replace_page() for KSM specific work.
Instruction analysis on x86 is based on instruction decoder and
determines if an instruction can be probed and determines the
necessary fixups after singlestep. Instruction analysis is done
at probe insertion time so that we avoid having to repeat the
same analysis every time a probe is hit.
A lot of code here is due to the improvement/suggestions/inputs
from Peter Zijlstra.
Changelog:
(v10):
- Add code to clear REX.B prefix as suggested by Denys Vlasenko
and Masami Hiramatsu.
(v9):
- Use insn_offset_modrm as suggested by Masami Hiramatsu.
(v7):
Handle comments from Peter Zijlstra:
- Dont take reference to inode. (expect inode to uprobe_register to be sane).
- Use PTR_ERR to set the return value.
- No need to take reference to inode.
- use PTR_ERR to return error value.
- register and uprobe_unregister share code.
(v5):
- Modified del_consumer as per comments from Peter.
- Drop reference to inode before dropping reference to uprobe.
- Use i_size_read(inode) instead of inode->i_size.
- Ensure uprobe->consumers is NULL, before __uprobe_unregister() is called.
- Includes errno.h as recommended by Stephen Rothwell to fix a build issue
on sparc defconfig
- Remove restrictions while unregistering.
- Earlier code leaked inode references under some conditions while
registering/unregistering.
- Continue the vma-rmap walk even if the intermediate vma doesnt
meet the requirements.
- Validate the vma found by find_vma before inserting/removing the
breakpoint
- Call del_consumer under mutex_lock.
- Use hash locks.
- Handle mremap.
- Introduce find_least_offset_node() instead of close match logic in
find_uprobe
- Uprobes no more depends on MM_OWNER; No reference to task_structs
while inserting/removing a probe.
- Uses read_mapping_page instead of grab_cache_page so that the pages
have valid content.
- pass NULL to get_user_pages for the task parameter.
- call SetPageUptodate on the new page allocated in write_opcode.
- fix leaking a reference to the new page under certain conditions.
- Include Instruction Decoder if Uprobes gets defined.
- Remove const attributes for instruction prefix arrays.
- Uses mm_context to know if the application is 32 bit.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Also-written-by: Jim Keniston <jkenisto@us.ibm.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Roland McGrath <roland@hack.frob.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Anton Arapov <anton@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Denys Vlasenko <vda.linux@googlemail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linux-mm <linux-mm@kvack.org>
Link: http://lkml.kernel.org/r/20120209092642.GE16600@linux.vnet.ibm.com
[ Made various small edits to the commit log ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-02-09 17:26:42 +08:00
|
|
|
|
2012-11-15 02:03:42 +08:00
|
|
|
percpu_down_write(&dup_mmap_sem);
|
2012-06-15 23:43:33 +08:00
|
|
|
info = build_map_info(uprobe->inode->i_mapping,
|
|
|
|
uprobe->offset, is_register);
|
2012-11-15 02:03:42 +08:00
|
|
|
if (IS_ERR(info)) {
|
|
|
|
err = PTR_ERR(info);
|
|
|
|
goto out;
|
|
|
|
}
|
2012-02-17 16:27:41 +08:00
|
|
|
|
2012-06-15 23:43:33 +08:00
|
|
|
while (info) {
|
|
|
|
struct mm_struct *mm = info->mm;
|
|
|
|
struct vm_area_struct *vma;
|
2012-02-17 16:27:41 +08:00
|
|
|
|
2012-10-01 00:54:53 +08:00
|
|
|
if (err && is_register)
|
2012-06-15 23:43:33 +08:00
|
|
|
goto free;
|
2012-02-17 16:27:41 +08:00
|
|
|
|
2012-05-30 03:29:28 +08:00
|
|
|
down_write(&mm->mmap_sem);
|
2012-07-30 02:22:44 +08:00
|
|
|
vma = find_vma(mm, info->vaddr);
|
|
|
|
if (!vma || !valid_vma(vma, is_register) ||
|
2013-03-18 01:54:44 +08:00
|
|
|
file_inode(vma->vm_file) != uprobe->inode)
|
2012-06-15 23:43:33 +08:00
|
|
|
goto unlock;
|
|
|
|
|
2012-07-30 02:22:44 +08:00
|
|
|
if (vma->vm_start > info->vaddr ||
|
|
|
|
vaddr_to_offset(vma, info->vaddr) != uprobe->offset)
|
2012-06-15 23:43:33 +08:00
|
|
|
goto unlock;
|
uprobes, mm, x86: Add the ability to install and remove uprobes breakpoints
Add uprobes support to the core kernel, with x86 support.
This commit adds the kernel facilities, the actual uprobes
user-space ABI and perf probe support comes in later commits.
General design:
Uprobes are maintained in an rb-tree indexed by inode and offset
(the offset here is from the start of the mapping). For a unique
(inode, offset) tuple, there can be at most one uprobe in the
rb-tree.
Since the (inode, offset) tuple identifies a unique uprobe, more
than one user may be interested in the same uprobe. This provides
the ability to connect multiple 'consumers' to the same uprobe.
Each consumer defines a handler and a filter (optional). The
'handler' is run every time the uprobe is hit, if it matches the
'filter' criteria.
The first consumer of a uprobe causes the breakpoint to be
inserted at the specified address and subsequent consumers are
appended to this list. On subsequent probes, the consumer gets
appended to the existing list of consumers. The breakpoint is
removed when the last consumer unregisters. For all other
unregisterations, the consumer is removed from the list of
consumers.
Given a inode, we get a list of the mms that have mapped the
inode. Do the actual registration if mm maps the page where a
probe needs to be inserted/removed.
We use a temporary list to walk through the vmas that map the
inode.
- The number of maps that map the inode, is not known before we
walk the rmap and keeps changing.
- extending vm_area_struct wasn't recommended, it's a
size-critical data structure.
- There can be more than one maps of the inode in the same mm.
We add callbacks to the mmap methods to keep an eye on text vmas
that are of interest to uprobes. When a vma of interest is mapped,
we insert the breakpoint at the right address.
Uprobe works by replacing the instruction at the address defined
by (inode, offset) with the arch specific breakpoint
instruction. We save a copy of the original instruction at the
uprobed address.
This is needed for:
a. executing the instruction out-of-line (xol).
b. instruction analysis for any subsequent fixups.
c. restoring the instruction back when the uprobe is unregistered.
We insert or delete a breakpoint instruction, and this
breakpoint instruction is assumed to be the smallest instruction
available on the platform. For fixed size instruction platforms
this is trivially true, for variable size instruction platforms
the breakpoint instruction is typically the smallest (often a
single byte).
Writing the instruction is done by COWing the page and changing
the instruction during the copy, this even though most platforms
allow atomic writes of the breakpoint instruction. This also
mirrors the behaviour of a ptrace() memory write to a PRIVATE
file map.
The core worker is derived from KSM's replace_page() logic.
In essence, similar to KSM:
a. allocate a new page and copy over contents of the page that
has the uprobed vaddr
b. modify the copy and insert the breakpoint at the required
address
c. switch the original page with the copy containing the
breakpoint
d. flush page tables.
replace_page() is being replicated here because of some minor
changes in the type of pages and also because Hugh Dickins had
plans to improve replace_page() for KSM specific work.
Instruction analysis on x86 is based on instruction decoder and
determines if an instruction can be probed and determines the
necessary fixups after singlestep. Instruction analysis is done
at probe insertion time so that we avoid having to repeat the
same analysis every time a probe is hit.
A lot of code here is due to the improvement/suggestions/inputs
from Peter Zijlstra.
Changelog:
(v10):
- Add code to clear REX.B prefix as suggested by Denys Vlasenko
and Masami Hiramatsu.
(v9):
- Use insn_offset_modrm as suggested by Masami Hiramatsu.
(v7):
Handle comments from Peter Zijlstra:
- Dont take reference to inode. (expect inode to uprobe_register to be sane).
- Use PTR_ERR to set the return value.
- No need to take reference to inode.
- use PTR_ERR to return error value.
- register and uprobe_unregister share code.
(v5):
- Modified del_consumer as per comments from Peter.
- Drop reference to inode before dropping reference to uprobe.
- Use i_size_read(inode) instead of inode->i_size.
- Ensure uprobe->consumers is NULL, before __uprobe_unregister() is called.
- Includes errno.h as recommended by Stephen Rothwell to fix a build issue
on sparc defconfig
- Remove restrictions while unregistering.
- Earlier code leaked inode references under some conditions while
registering/unregistering.
- Continue the vma-rmap walk even if the intermediate vma doesnt
meet the requirements.
- Validate the vma found by find_vma before inserting/removing the
breakpoint
- Call del_consumer under mutex_lock.
- Use hash locks.
- Handle mremap.
- Introduce find_least_offset_node() instead of close match logic in
find_uprobe
- Uprobes no more depends on MM_OWNER; No reference to task_structs
while inserting/removing a probe.
- Uses read_mapping_page instead of grab_cache_page so that the pages
have valid content.
- pass NULL to get_user_pages for the task parameter.
- call SetPageUptodate on the new page allocated in write_opcode.
- fix leaking a reference to the new page under certain conditions.
- Include Instruction Decoder if Uprobes gets defined.
- Remove const attributes for instruction prefix arrays.
- Uses mm_context to know if the application is 32 bit.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Also-written-by: Jim Keniston <jkenisto@us.ibm.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Roland McGrath <roland@hack.frob.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Anton Arapov <anton@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Denys Vlasenko <vda.linux@googlemail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linux-mm <linux-mm@kvack.org>
Link: http://lkml.kernel.org/r/20120209092642.GE16600@linux.vnet.ibm.com
[ Made various small edits to the commit log ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-02-09 17:26:42 +08:00
|
|
|
|
2012-12-28 01:21:11 +08:00
|
|
|
if (is_register) {
|
|
|
|
/* consult only the "caller", new consumer. */
|
uprobes: Introduce uprobe_apply()
Currently it is not possible to change the filtering constraints after
uprobe_register(), so a consumer can not, say, start to trace a task/mm
which was previously filtered out, or remove the no longer needed bp's.
Introduce uprobe_apply() which simply does register_for_each_vma() again
to consult uprobe_consumer->filter() and install/remove the breakpoints.
The only complication is that register_for_each_vma() can no longer
assume that uprobe->consumers should be consulter if is_register == T,
so we change it to accept "struct uprobe_consumer *new" instead.
Unlike uprobe_register(), uprobe_apply(true) doesn't do "unregister" if
register_for_each_vma() fails, it is up to caller to handle the error.
Note: we probably need to cleanup the current interface, it is strange
that uprobe_apply/unregister need inode/offset. We should either change
uprobe_register() to return "struct uprobe *", or add a private ->uprobe
member in uprobe_consumer. And in the long term uprobe_apply() should
take a single argument, uprobe or consumer, even "bool add" should go
away.
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
2013-02-04 02:21:12 +08:00
|
|
|
if (consumer_filter(new,
|
2012-12-29 00:58:38 +08:00
|
|
|
UPROBE_FILTER_REGISTER, mm))
|
2012-12-28 01:21:11 +08:00
|
|
|
err = install_breakpoint(uprobe, mm, vma, info->vaddr);
|
|
|
|
} else if (test_bit(MMF_HAS_UPROBES, &mm->flags)) {
|
2012-12-29 00:58:38 +08:00
|
|
|
if (!filter_chain(uprobe,
|
|
|
|
UPROBE_FILTER_UNREGISTER, mm))
|
2012-12-28 01:21:11 +08:00
|
|
|
err |= remove_breakpoint(uprobe, mm, info->vaddr);
|
|
|
|
}
|
2012-08-08 23:35:08 +08:00
|
|
|
|
2012-06-15 23:43:33 +08:00
|
|
|
unlock:
|
|
|
|
up_write(&mm->mmap_sem);
|
|
|
|
free:
|
|
|
|
mmput(mm);
|
|
|
|
info = free_map_info(info);
|
uprobes, mm, x86: Add the ability to install and remove uprobes breakpoints
Add uprobes support to the core kernel, with x86 support.
This commit adds the kernel facilities, the actual uprobes
user-space ABI and perf probe support comes in later commits.
General design:
Uprobes are maintained in an rb-tree indexed by inode and offset
(the offset here is from the start of the mapping). For a unique
(inode, offset) tuple, there can be at most one uprobe in the
rb-tree.
Since the (inode, offset) tuple identifies a unique uprobe, more
than one user may be interested in the same uprobe. This provides
the ability to connect multiple 'consumers' to the same uprobe.
Each consumer defines a handler and a filter (optional). The
'handler' is run every time the uprobe is hit, if it matches the
'filter' criteria.
The first consumer of a uprobe causes the breakpoint to be
inserted at the specified address and subsequent consumers are
appended to this list. On subsequent probes, the consumer gets
appended to the existing list of consumers. The breakpoint is
removed when the last consumer unregisters. For all other
unregisterations, the consumer is removed from the list of
consumers.
Given a inode, we get a list of the mms that have mapped the
inode. Do the actual registration if mm maps the page where a
probe needs to be inserted/removed.
We use a temporary list to walk through the vmas that map the
inode.
- The number of maps that map the inode, is not known before we
walk the rmap and keeps changing.
- extending vm_area_struct wasn't recommended, it's a
size-critical data structure.
- There can be more than one maps of the inode in the same mm.
We add callbacks to the mmap methods to keep an eye on text vmas
that are of interest to uprobes. When a vma of interest is mapped,
we insert the breakpoint at the right address.
Uprobe works by replacing the instruction at the address defined
by (inode, offset) with the arch specific breakpoint
instruction. We save a copy of the original instruction at the
uprobed address.
This is needed for:
a. executing the instruction out-of-line (xol).
b. instruction analysis for any subsequent fixups.
c. restoring the instruction back when the uprobe is unregistered.
We insert or delete a breakpoint instruction, and this
breakpoint instruction is assumed to be the smallest instruction
available on the platform. For fixed size instruction platforms
this is trivially true, for variable size instruction platforms
the breakpoint instruction is typically the smallest (often a
single byte).
Writing the instruction is done by COWing the page and changing
the instruction during the copy, this even though most platforms
allow atomic writes of the breakpoint instruction. This also
mirrors the behaviour of a ptrace() memory write to a PRIVATE
file map.
The core worker is derived from KSM's replace_page() logic.
In essence, similar to KSM:
a. allocate a new page and copy over contents of the page that
has the uprobed vaddr
b. modify the copy and insert the breakpoint at the required
address
c. switch the original page with the copy containing the
breakpoint
d. flush page tables.
replace_page() is being replicated here because of some minor
changes in the type of pages and also because Hugh Dickins had
plans to improve replace_page() for KSM specific work.
Instruction analysis on x86 is based on instruction decoder and
determines if an instruction can be probed and determines the
necessary fixups after singlestep. Instruction analysis is done
at probe insertion time so that we avoid having to repeat the
same analysis every time a probe is hit.
A lot of code here is due to the improvement/suggestions/inputs
from Peter Zijlstra.
Changelog:
(v10):
- Add code to clear REX.B prefix as suggested by Denys Vlasenko
and Masami Hiramatsu.
(v9):
- Use insn_offset_modrm as suggested by Masami Hiramatsu.
(v7):
Handle comments from Peter Zijlstra:
- Dont take reference to inode. (expect inode to uprobe_register to be sane).
- Use PTR_ERR to set the return value.
- No need to take reference to inode.
- use PTR_ERR to return error value.
- register and uprobe_unregister share code.
(v5):
- Modified del_consumer as per comments from Peter.
- Drop reference to inode before dropping reference to uprobe.
- Use i_size_read(inode) instead of inode->i_size.
- Ensure uprobe->consumers is NULL, before __uprobe_unregister() is called.
- Includes errno.h as recommended by Stephen Rothwell to fix a build issue
on sparc defconfig
- Remove restrictions while unregistering.
- Earlier code leaked inode references under some conditions while
registering/unregistering.
- Continue the vma-rmap walk even if the intermediate vma doesnt
meet the requirements.
- Validate the vma found by find_vma before inserting/removing the
breakpoint
- Call del_consumer under mutex_lock.
- Use hash locks.
- Handle mremap.
- Introduce find_least_offset_node() instead of close match logic in
find_uprobe
- Uprobes no more depends on MM_OWNER; No reference to task_structs
while inserting/removing a probe.
- Uses read_mapping_page instead of grab_cache_page so that the pages
have valid content.
- pass NULL to get_user_pages for the task parameter.
- call SetPageUptodate on the new page allocated in write_opcode.
- fix leaking a reference to the new page under certain conditions.
- Include Instruction Decoder if Uprobes gets defined.
- Remove const attributes for instruction prefix arrays.
- Uses mm_context to know if the application is 32 bit.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Also-written-by: Jim Keniston <jkenisto@us.ibm.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Roland McGrath <roland@hack.frob.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Anton Arapov <anton@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Denys Vlasenko <vda.linux@googlemail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linux-mm <linux-mm@kvack.org>
Link: http://lkml.kernel.org/r/20120209092642.GE16600@linux.vnet.ibm.com
[ Made various small edits to the commit log ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-02-09 17:26:42 +08:00
|
|
|
}
|
2012-11-15 02:03:42 +08:00
|
|
|
out:
|
|
|
|
percpu_up_write(&dup_mmap_sem);
|
2012-06-15 23:43:33 +08:00
|
|
|
return err;
|
uprobes, mm, x86: Add the ability to install and remove uprobes breakpoints
Add uprobes support to the core kernel, with x86 support.
This commit adds the kernel facilities, the actual uprobes
user-space ABI and perf probe support comes in later commits.
General design:
Uprobes are maintained in an rb-tree indexed by inode and offset
(the offset here is from the start of the mapping). For a unique
(inode, offset) tuple, there can be at most one uprobe in the
rb-tree.
Since the (inode, offset) tuple identifies a unique uprobe, more
than one user may be interested in the same uprobe. This provides
the ability to connect multiple 'consumers' to the same uprobe.
Each consumer defines a handler and a filter (optional). The
'handler' is run every time the uprobe is hit, if it matches the
'filter' criteria.
The first consumer of a uprobe causes the breakpoint to be
inserted at the specified address and subsequent consumers are
appended to this list. On subsequent probes, the consumer gets
appended to the existing list of consumers. The breakpoint is
removed when the last consumer unregisters. For all other
unregisterations, the consumer is removed from the list of
consumers.
Given a inode, we get a list of the mms that have mapped the
inode. Do the actual registration if mm maps the page where a
probe needs to be inserted/removed.
We use a temporary list to walk through the vmas that map the
inode.
- The number of maps that map the inode, is not known before we
walk the rmap and keeps changing.
- extending vm_area_struct wasn't recommended, it's a
size-critical data structure.
- There can be more than one maps of the inode in the same mm.
We add callbacks to the mmap methods to keep an eye on text vmas
that are of interest to uprobes. When a vma of interest is mapped,
we insert the breakpoint at the right address.
Uprobe works by replacing the instruction at the address defined
by (inode, offset) with the arch specific breakpoint
instruction. We save a copy of the original instruction at the
uprobed address.
This is needed for:
a. executing the instruction out-of-line (xol).
b. instruction analysis for any subsequent fixups.
c. restoring the instruction back when the uprobe is unregistered.
We insert or delete a breakpoint instruction, and this
breakpoint instruction is assumed to be the smallest instruction
available on the platform. For fixed size instruction platforms
this is trivially true, for variable size instruction platforms
the breakpoint instruction is typically the smallest (often a
single byte).
Writing the instruction is done by COWing the page and changing
the instruction during the copy, this even though most platforms
allow atomic writes of the breakpoint instruction. This also
mirrors the behaviour of a ptrace() memory write to a PRIVATE
file map.
The core worker is derived from KSM's replace_page() logic.
In essence, similar to KSM:
a. allocate a new page and copy over contents of the page that
has the uprobed vaddr
b. modify the copy and insert the breakpoint at the required
address
c. switch the original page with the copy containing the
breakpoint
d. flush page tables.
replace_page() is being replicated here because of some minor
changes in the type of pages and also because Hugh Dickins had
plans to improve replace_page() for KSM specific work.
Instruction analysis on x86 is based on instruction decoder and
determines if an instruction can be probed and determines the
necessary fixups after singlestep. Instruction analysis is done
at probe insertion time so that we avoid having to repeat the
same analysis every time a probe is hit.
A lot of code here is due to the improvement/suggestions/inputs
from Peter Zijlstra.
Changelog:
(v10):
- Add code to clear REX.B prefix as suggested by Denys Vlasenko
and Masami Hiramatsu.
(v9):
- Use insn_offset_modrm as suggested by Masami Hiramatsu.
(v7):
Handle comments from Peter Zijlstra:
- Dont take reference to inode. (expect inode to uprobe_register to be sane).
- Use PTR_ERR to set the return value.
- No need to take reference to inode.
- use PTR_ERR to return error value.
- register and uprobe_unregister share code.
(v5):
- Modified del_consumer as per comments from Peter.
- Drop reference to inode before dropping reference to uprobe.
- Use i_size_read(inode) instead of inode->i_size.
- Ensure uprobe->consumers is NULL, before __uprobe_unregister() is called.
- Includes errno.h as recommended by Stephen Rothwell to fix a build issue
on sparc defconfig
- Remove restrictions while unregistering.
- Earlier code leaked inode references under some conditions while
registering/unregistering.
- Continue the vma-rmap walk even if the intermediate vma doesnt
meet the requirements.
- Validate the vma found by find_vma before inserting/removing the
breakpoint
- Call del_consumer under mutex_lock.
- Use hash locks.
- Handle mremap.
- Introduce find_least_offset_node() instead of close match logic in
find_uprobe
- Uprobes no more depends on MM_OWNER; No reference to task_structs
while inserting/removing a probe.
- Uses read_mapping_page instead of grab_cache_page so that the pages
have valid content.
- pass NULL to get_user_pages for the task parameter.
- call SetPageUptodate on the new page allocated in write_opcode.
- fix leaking a reference to the new page under certain conditions.
- Include Instruction Decoder if Uprobes gets defined.
- Remove const attributes for instruction prefix arrays.
- Uses mm_context to know if the application is 32 bit.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Also-written-by: Jim Keniston <jkenisto@us.ibm.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Roland McGrath <roland@hack.frob.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Anton Arapov <anton@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Denys Vlasenko <vda.linux@googlemail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linux-mm <linux-mm@kvack.org>
Link: http://lkml.kernel.org/r/20120209092642.GE16600@linux.vnet.ibm.com
[ Made various small edits to the commit log ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-02-09 17:26:42 +08:00
|
|
|
}
|
|
|
|
|
2018-08-09 12:18:51 +08:00
|
|
|
static void
|
|
|
|
__uprobe_unregister(struct uprobe *uprobe, struct uprobe_consumer *uc)
|
uprobes, mm, x86: Add the ability to install and remove uprobes breakpoints
Add uprobes support to the core kernel, with x86 support.
This commit adds the kernel facilities, the actual uprobes
user-space ABI and perf probe support comes in later commits.
General design:
Uprobes are maintained in an rb-tree indexed by inode and offset
(the offset here is from the start of the mapping). For a unique
(inode, offset) tuple, there can be at most one uprobe in the
rb-tree.
Since the (inode, offset) tuple identifies a unique uprobe, more
than one user may be interested in the same uprobe. This provides
the ability to connect multiple 'consumers' to the same uprobe.
Each consumer defines a handler and a filter (optional). The
'handler' is run every time the uprobe is hit, if it matches the
'filter' criteria.
The first consumer of a uprobe causes the breakpoint to be
inserted at the specified address and subsequent consumers are
appended to this list. On subsequent probes, the consumer gets
appended to the existing list of consumers. The breakpoint is
removed when the last consumer unregisters. For all other
unregisterations, the consumer is removed from the list of
consumers.
Given a inode, we get a list of the mms that have mapped the
inode. Do the actual registration if mm maps the page where a
probe needs to be inserted/removed.
We use a temporary list to walk through the vmas that map the
inode.
- The number of maps that map the inode, is not known before we
walk the rmap and keeps changing.
- extending vm_area_struct wasn't recommended, it's a
size-critical data structure.
- There can be more than one maps of the inode in the same mm.
We add callbacks to the mmap methods to keep an eye on text vmas
that are of interest to uprobes. When a vma of interest is mapped,
we insert the breakpoint at the right address.
Uprobe works by replacing the instruction at the address defined
by (inode, offset) with the arch specific breakpoint
instruction. We save a copy of the original instruction at the
uprobed address.
This is needed for:
a. executing the instruction out-of-line (xol).
b. instruction analysis for any subsequent fixups.
c. restoring the instruction back when the uprobe is unregistered.
We insert or delete a breakpoint instruction, and this
breakpoint instruction is assumed to be the smallest instruction
available on the platform. For fixed size instruction platforms
this is trivially true, for variable size instruction platforms
the breakpoint instruction is typically the smallest (often a
single byte).
Writing the instruction is done by COWing the page and changing
the instruction during the copy, this even though most platforms
allow atomic writes of the breakpoint instruction. This also
mirrors the behaviour of a ptrace() memory write to a PRIVATE
file map.
The core worker is derived from KSM's replace_page() logic.
In essence, similar to KSM:
a. allocate a new page and copy over contents of the page that
has the uprobed vaddr
b. modify the copy and insert the breakpoint at the required
address
c. switch the original page with the copy containing the
breakpoint
d. flush page tables.
replace_page() is being replicated here because of some minor
changes in the type of pages and also because Hugh Dickins had
plans to improve replace_page() for KSM specific work.
Instruction analysis on x86 is based on instruction decoder and
determines if an instruction can be probed and determines the
necessary fixups after singlestep. Instruction analysis is done
at probe insertion time so that we avoid having to repeat the
same analysis every time a probe is hit.
A lot of code here is due to the improvement/suggestions/inputs
from Peter Zijlstra.
Changelog:
(v10):
- Add code to clear REX.B prefix as suggested by Denys Vlasenko
and Masami Hiramatsu.
(v9):
- Use insn_offset_modrm as suggested by Masami Hiramatsu.
(v7):
Handle comments from Peter Zijlstra:
- Dont take reference to inode. (expect inode to uprobe_register to be sane).
- Use PTR_ERR to set the return value.
- No need to take reference to inode.
- use PTR_ERR to return error value.
- register and uprobe_unregister share code.
(v5):
- Modified del_consumer as per comments from Peter.
- Drop reference to inode before dropping reference to uprobe.
- Use i_size_read(inode) instead of inode->i_size.
- Ensure uprobe->consumers is NULL, before __uprobe_unregister() is called.
- Includes errno.h as recommended by Stephen Rothwell to fix a build issue
on sparc defconfig
- Remove restrictions while unregistering.
- Earlier code leaked inode references under some conditions while
registering/unregistering.
- Continue the vma-rmap walk even if the intermediate vma doesnt
meet the requirements.
- Validate the vma found by find_vma before inserting/removing the
breakpoint
- Call del_consumer under mutex_lock.
- Use hash locks.
- Handle mremap.
- Introduce find_least_offset_node() instead of close match logic in
find_uprobe
- Uprobes no more depends on MM_OWNER; No reference to task_structs
while inserting/removing a probe.
- Uses read_mapping_page instead of grab_cache_page so that the pages
have valid content.
- pass NULL to get_user_pages for the task parameter.
- call SetPageUptodate on the new page allocated in write_opcode.
- fix leaking a reference to the new page under certain conditions.
- Include Instruction Decoder if Uprobes gets defined.
- Remove const attributes for instruction prefix arrays.
- Uses mm_context to know if the application is 32 bit.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Also-written-by: Jim Keniston <jkenisto@us.ibm.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Roland McGrath <roland@hack.frob.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Anton Arapov <anton@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Denys Vlasenko <vda.linux@googlemail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linux-mm <linux-mm@kvack.org>
Link: http://lkml.kernel.org/r/20120209092642.GE16600@linux.vnet.ibm.com
[ Made various small edits to the commit log ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-02-09 17:26:42 +08:00
|
|
|
{
|
2012-11-24 02:43:50 +08:00
|
|
|
int err;
|
|
|
|
|
2014-06-28 01:01:40 +08:00
|
|
|
if (WARN_ON(!consumer_del(uprobe, uc)))
|
2012-11-24 02:43:50 +08:00
|
|
|
return;
|
uprobes, mm, x86: Add the ability to install and remove uprobes breakpoints
Add uprobes support to the core kernel, with x86 support.
This commit adds the kernel facilities, the actual uprobes
user-space ABI and perf probe support comes in later commits.
General design:
Uprobes are maintained in an rb-tree indexed by inode and offset
(the offset here is from the start of the mapping). For a unique
(inode, offset) tuple, there can be at most one uprobe in the
rb-tree.
Since the (inode, offset) tuple identifies a unique uprobe, more
than one user may be interested in the same uprobe. This provides
the ability to connect multiple 'consumers' to the same uprobe.
Each consumer defines a handler and a filter (optional). The
'handler' is run every time the uprobe is hit, if it matches the
'filter' criteria.
The first consumer of a uprobe causes the breakpoint to be
inserted at the specified address and subsequent consumers are
appended to this list. On subsequent probes, the consumer gets
appended to the existing list of consumers. The breakpoint is
removed when the last consumer unregisters. For all other
unregisterations, the consumer is removed from the list of
consumers.
Given a inode, we get a list of the mms that have mapped the
inode. Do the actual registration if mm maps the page where a
probe needs to be inserted/removed.
We use a temporary list to walk through the vmas that map the
inode.
- The number of maps that map the inode, is not known before we
walk the rmap and keeps changing.
- extending vm_area_struct wasn't recommended, it's a
size-critical data structure.
- There can be more than one maps of the inode in the same mm.
We add callbacks to the mmap methods to keep an eye on text vmas
that are of interest to uprobes. When a vma of interest is mapped,
we insert the breakpoint at the right address.
Uprobe works by replacing the instruction at the address defined
by (inode, offset) with the arch specific breakpoint
instruction. We save a copy of the original instruction at the
uprobed address.
This is needed for:
a. executing the instruction out-of-line (xol).
b. instruction analysis for any subsequent fixups.
c. restoring the instruction back when the uprobe is unregistered.
We insert or delete a breakpoint instruction, and this
breakpoint instruction is assumed to be the smallest instruction
available on the platform. For fixed size instruction platforms
this is trivially true, for variable size instruction platforms
the breakpoint instruction is typically the smallest (often a
single byte).
Writing the instruction is done by COWing the page and changing
the instruction during the copy, this even though most platforms
allow atomic writes of the breakpoint instruction. This also
mirrors the behaviour of a ptrace() memory write to a PRIVATE
file map.
The core worker is derived from KSM's replace_page() logic.
In essence, similar to KSM:
a. allocate a new page and copy over contents of the page that
has the uprobed vaddr
b. modify the copy and insert the breakpoint at the required
address
c. switch the original page with the copy containing the
breakpoint
d. flush page tables.
replace_page() is being replicated here because of some minor
changes in the type of pages and also because Hugh Dickins had
plans to improve replace_page() for KSM specific work.
Instruction analysis on x86 is based on instruction decoder and
determines if an instruction can be probed and determines the
necessary fixups after singlestep. Instruction analysis is done
at probe insertion time so that we avoid having to repeat the
same analysis every time a probe is hit.
A lot of code here is due to the improvement/suggestions/inputs
from Peter Zijlstra.
Changelog:
(v10):
- Add code to clear REX.B prefix as suggested by Denys Vlasenko
and Masami Hiramatsu.
(v9):
- Use insn_offset_modrm as suggested by Masami Hiramatsu.
(v7):
Handle comments from Peter Zijlstra:
- Dont take reference to inode. (expect inode to uprobe_register to be sane).
- Use PTR_ERR to set the return value.
- No need to take reference to inode.
- use PTR_ERR to return error value.
- register and uprobe_unregister share code.
(v5):
- Modified del_consumer as per comments from Peter.
- Drop reference to inode before dropping reference to uprobe.
- Use i_size_read(inode) instead of inode->i_size.
- Ensure uprobe->consumers is NULL, before __uprobe_unregister() is called.
- Includes errno.h as recommended by Stephen Rothwell to fix a build issue
on sparc defconfig
- Remove restrictions while unregistering.
- Earlier code leaked inode references under some conditions while
registering/unregistering.
- Continue the vma-rmap walk even if the intermediate vma doesnt
meet the requirements.
- Validate the vma found by find_vma before inserting/removing the
breakpoint
- Call del_consumer under mutex_lock.
- Use hash locks.
- Handle mremap.
- Introduce find_least_offset_node() instead of close match logic in
find_uprobe
- Uprobes no more depends on MM_OWNER; No reference to task_structs
while inserting/removing a probe.
- Uses read_mapping_page instead of grab_cache_page so that the pages
have valid content.
- pass NULL to get_user_pages for the task parameter.
- call SetPageUptodate on the new page allocated in write_opcode.
- fix leaking a reference to the new page under certain conditions.
- Include Instruction Decoder if Uprobes gets defined.
- Remove const attributes for instruction prefix arrays.
- Uses mm_context to know if the application is 32 bit.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Also-written-by: Jim Keniston <jkenisto@us.ibm.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Roland McGrath <roland@hack.frob.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Anton Arapov <anton@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Denys Vlasenko <vda.linux@googlemail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linux-mm <linux-mm@kvack.org>
Link: http://lkml.kernel.org/r/20120209092642.GE16600@linux.vnet.ibm.com
[ Made various small edits to the commit log ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-02-09 17:26:42 +08:00
|
|
|
|
uprobes: Introduce uprobe_apply()
Currently it is not possible to change the filtering constraints after
uprobe_register(), so a consumer can not, say, start to trace a task/mm
which was previously filtered out, or remove the no longer needed bp's.
Introduce uprobe_apply() which simply does register_for_each_vma() again
to consult uprobe_consumer->filter() and install/remove the breakpoints.
The only complication is that register_for_each_vma() can no longer
assume that uprobe->consumers should be consulter if is_register == T,
so we change it to accept "struct uprobe_consumer *new" instead.
Unlike uprobe_register(), uprobe_apply(true) doesn't do "unregister" if
register_for_each_vma() fails, it is up to caller to handle the error.
Note: we probably need to cleanup the current interface, it is strange
that uprobe_apply/unregister need inode/offset. We should either change
uprobe_register() to return "struct uprobe *", or add a private ->uprobe
member in uprobe_consumer. And in the long term uprobe_apply() should
take a single argument, uprobe or consumer, even "bool add" should go
away.
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
2013-02-04 02:21:12 +08:00
|
|
|
err = register_for_each_vma(uprobe, NULL);
|
2012-11-25 01:27:08 +08:00
|
|
|
/* TODO : cant unregister? schedule a worker thread */
|
|
|
|
if (!uprobe->consumers && !err)
|
|
|
|
delete_uprobe(uprobe);
|
uprobes, mm, x86: Add the ability to install and remove uprobes breakpoints
Add uprobes support to the core kernel, with x86 support.
This commit adds the kernel facilities, the actual uprobes
user-space ABI and perf probe support comes in later commits.
General design:
Uprobes are maintained in an rb-tree indexed by inode and offset
(the offset here is from the start of the mapping). For a unique
(inode, offset) tuple, there can be at most one uprobe in the
rb-tree.
Since the (inode, offset) tuple identifies a unique uprobe, more
than one user may be interested in the same uprobe. This provides
the ability to connect multiple 'consumers' to the same uprobe.
Each consumer defines a handler and a filter (optional). The
'handler' is run every time the uprobe is hit, if it matches the
'filter' criteria.
The first consumer of a uprobe causes the breakpoint to be
inserted at the specified address and subsequent consumers are
appended to this list. On subsequent probes, the consumer gets
appended to the existing list of consumers. The breakpoint is
removed when the last consumer unregisters. For all other
unregisterations, the consumer is removed from the list of
consumers.
Given a inode, we get a list of the mms that have mapped the
inode. Do the actual registration if mm maps the page where a
probe needs to be inserted/removed.
We use a temporary list to walk through the vmas that map the
inode.
- The number of maps that map the inode, is not known before we
walk the rmap and keeps changing.
- extending vm_area_struct wasn't recommended, it's a
size-critical data structure.
- There can be more than one maps of the inode in the same mm.
We add callbacks to the mmap methods to keep an eye on text vmas
that are of interest to uprobes. When a vma of interest is mapped,
we insert the breakpoint at the right address.
Uprobe works by replacing the instruction at the address defined
by (inode, offset) with the arch specific breakpoint
instruction. We save a copy of the original instruction at the
uprobed address.
This is needed for:
a. executing the instruction out-of-line (xol).
b. instruction analysis for any subsequent fixups.
c. restoring the instruction back when the uprobe is unregistered.
We insert or delete a breakpoint instruction, and this
breakpoint instruction is assumed to be the smallest instruction
available on the platform. For fixed size instruction platforms
this is trivially true, for variable size instruction platforms
the breakpoint instruction is typically the smallest (often a
single byte).
Writing the instruction is done by COWing the page and changing
the instruction during the copy, this even though most platforms
allow atomic writes of the breakpoint instruction. This also
mirrors the behaviour of a ptrace() memory write to a PRIVATE
file map.
The core worker is derived from KSM's replace_page() logic.
In essence, similar to KSM:
a. allocate a new page and copy over contents of the page that
has the uprobed vaddr
b. modify the copy and insert the breakpoint at the required
address
c. switch the original page with the copy containing the
breakpoint
d. flush page tables.
replace_page() is being replicated here because of some minor
changes in the type of pages and also because Hugh Dickins had
plans to improve replace_page() for KSM specific work.
Instruction analysis on x86 is based on instruction decoder and
determines if an instruction can be probed and determines the
necessary fixups after singlestep. Instruction analysis is done
at probe insertion time so that we avoid having to repeat the
same analysis every time a probe is hit.
A lot of code here is due to the improvement/suggestions/inputs
from Peter Zijlstra.
Changelog:
(v10):
- Add code to clear REX.B prefix as suggested by Denys Vlasenko
and Masami Hiramatsu.
(v9):
- Use insn_offset_modrm as suggested by Masami Hiramatsu.
(v7):
Handle comments from Peter Zijlstra:
- Dont take reference to inode. (expect inode to uprobe_register to be sane).
- Use PTR_ERR to set the return value.
- No need to take reference to inode.
- use PTR_ERR to return error value.
- register and uprobe_unregister share code.
(v5):
- Modified del_consumer as per comments from Peter.
- Drop reference to inode before dropping reference to uprobe.
- Use i_size_read(inode) instead of inode->i_size.
- Ensure uprobe->consumers is NULL, before __uprobe_unregister() is called.
- Includes errno.h as recommended by Stephen Rothwell to fix a build issue
on sparc defconfig
- Remove restrictions while unregistering.
- Earlier code leaked inode references under some conditions while
registering/unregistering.
- Continue the vma-rmap walk even if the intermediate vma doesnt
meet the requirements.
- Validate the vma found by find_vma before inserting/removing the
breakpoint
- Call del_consumer under mutex_lock.
- Use hash locks.
- Handle mremap.
- Introduce find_least_offset_node() instead of close match logic in
find_uprobe
- Uprobes no more depends on MM_OWNER; No reference to task_structs
while inserting/removing a probe.
- Uses read_mapping_page instead of grab_cache_page so that the pages
have valid content.
- pass NULL to get_user_pages for the task parameter.
- call SetPageUptodate on the new page allocated in write_opcode.
- fix leaking a reference to the new page under certain conditions.
- Include Instruction Decoder if Uprobes gets defined.
- Remove const attributes for instruction prefix arrays.
- Uses mm_context to know if the application is 32 bit.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Also-written-by: Jim Keniston <jkenisto@us.ibm.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Roland McGrath <roland@hack.frob.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Anton Arapov <anton@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Denys Vlasenko <vda.linux@googlemail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linux-mm <linux-mm@kvack.org>
Link: http://lkml.kernel.org/r/20120209092642.GE16600@linux.vnet.ibm.com
[ Made various small edits to the commit log ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-02-09 17:26:42 +08:00
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
Updates for v4.19:
- Restructure of lockdep and latency tracers
This is the biggest change. Joel Fernandes restructured the hooks
from irqs and preemption disabling and enabling. He got rid of
a lot of the preprocessor #ifdef mess that they caused.
He turned both lockdep and the latency tracers to use trace events
inserted in the preempt/irqs disabling paths. But unfortunately,
these started to cause issues in corner cases. Thus, parts of the
code was reverted back to where lockde and the latency tracers
just get called directly (without using the trace events).
But because the original change cleaned up the code very nicely
we kept that, as well as the trace events for preempt and irqs
disabling, but they are limited to not being called in NMIs.
- Have trace events use SRCU for "rcu idle" calls. This was required
for the preempt/irqs off trace events. But it also had to not
allow them to be called in NMI context. Waiting till Paul makes
an NMI safe SRCU API.
- New notrace SRCU API to allow trace events to use SRCU.
- Addition of mcount-nop option support
- SPDX headers replacing GPL templates.
- Various other fixes and clean ups.
- Some fixes are marked for stable, but were not fully tested
before the merge window opened.
-----BEGIN PGP SIGNATURE-----
iIoEABYIADIWIQRRSw7ePDh/lE+zeZMp5XQQmuv6qgUCW3ruhRQccm9zdGVkdEBn
b29kbWlzLm9yZwAKCRAp5XQQmuv6qiM7AP47NhYdSnCFCRUJfrt6PovXmQtuCHt3
c3QMoGGdvzh9YAEAqcSXwh7uLhpHUp1LjMAPkXdZVwNddf4zJQ1zyxQ+EAU=
=vgEr
-----END PGP SIGNATURE-----
Merge tag 'trace-v4.19' of git://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-trace
Pull tracing updates from Steven Rostedt:
- Restructure of lockdep and latency tracers
This is the biggest change. Joel Fernandes restructured the hooks
from irqs and preemption disabling and enabling. He got rid of a lot
of the preprocessor #ifdef mess that they caused.
He turned both lockdep and the latency tracers to use trace events
inserted in the preempt/irqs disabling paths. But unfortunately,
these started to cause issues in corner cases. Thus, parts of the
code was reverted back to where lockdep and the latency tracers just
get called directly (without using the trace events). But because the
original change cleaned up the code very nicely we kept that, as well
as the trace events for preempt and irqs disabling, but they are
limited to not being called in NMIs.
- Have trace events use SRCU for "rcu idle" calls. This was required
for the preempt/irqs off trace events. But it also had to not allow
them to be called in NMI context. Waiting till Paul makes an NMI safe
SRCU API.
- New notrace SRCU API to allow trace events to use SRCU.
- Addition of mcount-nop option support
- SPDX headers replacing GPL templates.
- Various other fixes and clean ups.
- Some fixes are marked for stable, but were not fully tested before
the merge window opened.
* tag 'trace-v4.19' of git://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-trace: (44 commits)
tracing: Fix SPDX format headers to use C++ style comments
tracing: Add SPDX License format tags to tracing files
tracing: Add SPDX License format to bpf_trace.c
blktrace: Add SPDX License format header
s390/ftrace: Add -mfentry and -mnop-mcount support
tracing: Add -mcount-nop option support
tracing: Avoid calling cc-option -mrecord-mcount for every Makefile
tracing: Handle CC_FLAGS_FTRACE more accurately
Uprobe: Additional argument arch_uprobe to uprobe_write_opcode()
Uprobes: Simplify uprobe_register() body
tracepoints: Free early tracepoints after RCU is initialized
uprobes: Use synchronize_rcu() not synchronize_sched()
tracing: Fix synchronizing to event changes with tracepoint_synchronize_unregister()
ftrace: Remove unused pointer ftrace_swapper_pid
tracing: More reverting of "tracing: Centralize preemptirq tracepoints and unify their usage"
tracing/irqsoff: Handle preempt_count for different configs
tracing: Partial revert of "tracing: Centralize preemptirq tracepoints and unify their usage"
tracing: irqsoff: Account for additional preempt_disable
trace: Use rcu_dereference_raw for hooks from trace-event subsystem
tracing/kprobes: Fix within_notrace_func() to check only notrace functions
...
2018-08-21 09:32:00 +08:00
|
|
|
* uprobe_unregister - unregister an already registered probe.
|
2018-08-09 12:18:51 +08:00
|
|
|
* @inode: the file in which the probe has to be removed.
|
|
|
|
* @offset: offset from the start of the file.
|
|
|
|
* @uc: identify which probe if multiple probes are colocated.
|
|
|
|
*/
|
|
|
|
void uprobe_unregister(struct inode *inode, loff_t offset, struct uprobe_consumer *uc)
|
|
|
|
{
|
|
|
|
struct uprobe *uprobe;
|
|
|
|
|
|
|
|
uprobe = find_uprobe(inode, offset);
|
|
|
|
if (WARN_ON(!uprobe))
|
|
|
|
return;
|
|
|
|
|
|
|
|
down_write(&uprobe->register_rwsem);
|
|
|
|
__uprobe_unregister(uprobe, uc);
|
|
|
|
up_write(&uprobe->register_rwsem);
|
|
|
|
put_uprobe(uprobe);
|
|
|
|
}
|
|
|
|
EXPORT_SYMBOL_GPL(uprobe_unregister);
|
|
|
|
|
|
|
|
/*
|
|
|
|
* __uprobe_register - register a probe
|
uprobes, mm, x86: Add the ability to install and remove uprobes breakpoints
Add uprobes support to the core kernel, with x86 support.
This commit adds the kernel facilities, the actual uprobes
user-space ABI and perf probe support comes in later commits.
General design:
Uprobes are maintained in an rb-tree indexed by inode and offset
(the offset here is from the start of the mapping). For a unique
(inode, offset) tuple, there can be at most one uprobe in the
rb-tree.
Since the (inode, offset) tuple identifies a unique uprobe, more
than one user may be interested in the same uprobe. This provides
the ability to connect multiple 'consumers' to the same uprobe.
Each consumer defines a handler and a filter (optional). The
'handler' is run every time the uprobe is hit, if it matches the
'filter' criteria.
The first consumer of a uprobe causes the breakpoint to be
inserted at the specified address and subsequent consumers are
appended to this list. On subsequent probes, the consumer gets
appended to the existing list of consumers. The breakpoint is
removed when the last consumer unregisters. For all other
unregisterations, the consumer is removed from the list of
consumers.
Given a inode, we get a list of the mms that have mapped the
inode. Do the actual registration if mm maps the page where a
probe needs to be inserted/removed.
We use a temporary list to walk through the vmas that map the
inode.
- The number of maps that map the inode, is not known before we
walk the rmap and keeps changing.
- extending vm_area_struct wasn't recommended, it's a
size-critical data structure.
- There can be more than one maps of the inode in the same mm.
We add callbacks to the mmap methods to keep an eye on text vmas
that are of interest to uprobes. When a vma of interest is mapped,
we insert the breakpoint at the right address.
Uprobe works by replacing the instruction at the address defined
by (inode, offset) with the arch specific breakpoint
instruction. We save a copy of the original instruction at the
uprobed address.
This is needed for:
a. executing the instruction out-of-line (xol).
b. instruction analysis for any subsequent fixups.
c. restoring the instruction back when the uprobe is unregistered.
We insert or delete a breakpoint instruction, and this
breakpoint instruction is assumed to be the smallest instruction
available on the platform. For fixed size instruction platforms
this is trivially true, for variable size instruction platforms
the breakpoint instruction is typically the smallest (often a
single byte).
Writing the instruction is done by COWing the page and changing
the instruction during the copy, this even though most platforms
allow atomic writes of the breakpoint instruction. This also
mirrors the behaviour of a ptrace() memory write to a PRIVATE
file map.
The core worker is derived from KSM's replace_page() logic.
In essence, similar to KSM:
a. allocate a new page and copy over contents of the page that
has the uprobed vaddr
b. modify the copy and insert the breakpoint at the required
address
c. switch the original page with the copy containing the
breakpoint
d. flush page tables.
replace_page() is being replicated here because of some minor
changes in the type of pages and also because Hugh Dickins had
plans to improve replace_page() for KSM specific work.
Instruction analysis on x86 is based on instruction decoder and
determines if an instruction can be probed and determines the
necessary fixups after singlestep. Instruction analysis is done
at probe insertion time so that we avoid having to repeat the
same analysis every time a probe is hit.
A lot of code here is due to the improvement/suggestions/inputs
from Peter Zijlstra.
Changelog:
(v10):
- Add code to clear REX.B prefix as suggested by Denys Vlasenko
and Masami Hiramatsu.
(v9):
- Use insn_offset_modrm as suggested by Masami Hiramatsu.
(v7):
Handle comments from Peter Zijlstra:
- Dont take reference to inode. (expect inode to uprobe_register to be sane).
- Use PTR_ERR to set the return value.
- No need to take reference to inode.
- use PTR_ERR to return error value.
- register and uprobe_unregister share code.
(v5):
- Modified del_consumer as per comments from Peter.
- Drop reference to inode before dropping reference to uprobe.
- Use i_size_read(inode) instead of inode->i_size.
- Ensure uprobe->consumers is NULL, before __uprobe_unregister() is called.
- Includes errno.h as recommended by Stephen Rothwell to fix a build issue
on sparc defconfig
- Remove restrictions while unregistering.
- Earlier code leaked inode references under some conditions while
registering/unregistering.
- Continue the vma-rmap walk even if the intermediate vma doesnt
meet the requirements.
- Validate the vma found by find_vma before inserting/removing the
breakpoint
- Call del_consumer under mutex_lock.
- Use hash locks.
- Handle mremap.
- Introduce find_least_offset_node() instead of close match logic in
find_uprobe
- Uprobes no more depends on MM_OWNER; No reference to task_structs
while inserting/removing a probe.
- Uses read_mapping_page instead of grab_cache_page so that the pages
have valid content.
- pass NULL to get_user_pages for the task parameter.
- call SetPageUptodate on the new page allocated in write_opcode.
- fix leaking a reference to the new page under certain conditions.
- Include Instruction Decoder if Uprobes gets defined.
- Remove const attributes for instruction prefix arrays.
- Uses mm_context to know if the application is 32 bit.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Also-written-by: Jim Keniston <jkenisto@us.ibm.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Roland McGrath <roland@hack.frob.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Anton Arapov <anton@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Denys Vlasenko <vda.linux@googlemail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linux-mm <linux-mm@kvack.org>
Link: http://lkml.kernel.org/r/20120209092642.GE16600@linux.vnet.ibm.com
[ Made various small edits to the commit log ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-02-09 17:26:42 +08:00
|
|
|
* @inode: the file in which the probe has to be placed.
|
|
|
|
* @offset: offset from the start of the file.
|
2012-03-12 17:25:30 +08:00
|
|
|
* @uc: information on howto handle the probe..
|
uprobes, mm, x86: Add the ability to install and remove uprobes breakpoints
Add uprobes support to the core kernel, with x86 support.
This commit adds the kernel facilities, the actual uprobes
user-space ABI and perf probe support comes in later commits.
General design:
Uprobes are maintained in an rb-tree indexed by inode and offset
(the offset here is from the start of the mapping). For a unique
(inode, offset) tuple, there can be at most one uprobe in the
rb-tree.
Since the (inode, offset) tuple identifies a unique uprobe, more
than one user may be interested in the same uprobe. This provides
the ability to connect multiple 'consumers' to the same uprobe.
Each consumer defines a handler and a filter (optional). The
'handler' is run every time the uprobe is hit, if it matches the
'filter' criteria.
The first consumer of a uprobe causes the breakpoint to be
inserted at the specified address and subsequent consumers are
appended to this list. On subsequent probes, the consumer gets
appended to the existing list of consumers. The breakpoint is
removed when the last consumer unregisters. For all other
unregisterations, the consumer is removed from the list of
consumers.
Given a inode, we get a list of the mms that have mapped the
inode. Do the actual registration if mm maps the page where a
probe needs to be inserted/removed.
We use a temporary list to walk through the vmas that map the
inode.
- The number of maps that map the inode, is not known before we
walk the rmap and keeps changing.
- extending vm_area_struct wasn't recommended, it's a
size-critical data structure.
- There can be more than one maps of the inode in the same mm.
We add callbacks to the mmap methods to keep an eye on text vmas
that are of interest to uprobes. When a vma of interest is mapped,
we insert the breakpoint at the right address.
Uprobe works by replacing the instruction at the address defined
by (inode, offset) with the arch specific breakpoint
instruction. We save a copy of the original instruction at the
uprobed address.
This is needed for:
a. executing the instruction out-of-line (xol).
b. instruction analysis for any subsequent fixups.
c. restoring the instruction back when the uprobe is unregistered.
We insert or delete a breakpoint instruction, and this
breakpoint instruction is assumed to be the smallest instruction
available on the platform. For fixed size instruction platforms
this is trivially true, for variable size instruction platforms
the breakpoint instruction is typically the smallest (often a
single byte).
Writing the instruction is done by COWing the page and changing
the instruction during the copy, this even though most platforms
allow atomic writes of the breakpoint instruction. This also
mirrors the behaviour of a ptrace() memory write to a PRIVATE
file map.
The core worker is derived from KSM's replace_page() logic.
In essence, similar to KSM:
a. allocate a new page and copy over contents of the page that
has the uprobed vaddr
b. modify the copy and insert the breakpoint at the required
address
c. switch the original page with the copy containing the
breakpoint
d. flush page tables.
replace_page() is being replicated here because of some minor
changes in the type of pages and also because Hugh Dickins had
plans to improve replace_page() for KSM specific work.
Instruction analysis on x86 is based on instruction decoder and
determines if an instruction can be probed and determines the
necessary fixups after singlestep. Instruction analysis is done
at probe insertion time so that we avoid having to repeat the
same analysis every time a probe is hit.
A lot of code here is due to the improvement/suggestions/inputs
from Peter Zijlstra.
Changelog:
(v10):
- Add code to clear REX.B prefix as suggested by Denys Vlasenko
and Masami Hiramatsu.
(v9):
- Use insn_offset_modrm as suggested by Masami Hiramatsu.
(v7):
Handle comments from Peter Zijlstra:
- Dont take reference to inode. (expect inode to uprobe_register to be sane).
- Use PTR_ERR to set the return value.
- No need to take reference to inode.
- use PTR_ERR to return error value.
- register and uprobe_unregister share code.
(v5):
- Modified del_consumer as per comments from Peter.
- Drop reference to inode before dropping reference to uprobe.
- Use i_size_read(inode) instead of inode->i_size.
- Ensure uprobe->consumers is NULL, before __uprobe_unregister() is called.
- Includes errno.h as recommended by Stephen Rothwell to fix a build issue
on sparc defconfig
- Remove restrictions while unregistering.
- Earlier code leaked inode references under some conditions while
registering/unregistering.
- Continue the vma-rmap walk even if the intermediate vma doesnt
meet the requirements.
- Validate the vma found by find_vma before inserting/removing the
breakpoint
- Call del_consumer under mutex_lock.
- Use hash locks.
- Handle mremap.
- Introduce find_least_offset_node() instead of close match logic in
find_uprobe
- Uprobes no more depends on MM_OWNER; No reference to task_structs
while inserting/removing a probe.
- Uses read_mapping_page instead of grab_cache_page so that the pages
have valid content.
- pass NULL to get_user_pages for the task parameter.
- call SetPageUptodate on the new page allocated in write_opcode.
- fix leaking a reference to the new page under certain conditions.
- Include Instruction Decoder if Uprobes gets defined.
- Remove const attributes for instruction prefix arrays.
- Uses mm_context to know if the application is 32 bit.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Also-written-by: Jim Keniston <jkenisto@us.ibm.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Roland McGrath <roland@hack.frob.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Anton Arapov <anton@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Denys Vlasenko <vda.linux@googlemail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linux-mm <linux-mm@kvack.org>
Link: http://lkml.kernel.org/r/20120209092642.GE16600@linux.vnet.ibm.com
[ Made various small edits to the commit log ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-02-09 17:26:42 +08:00
|
|
|
*
|
2018-08-09 12:18:51 +08:00
|
|
|
* Apart from the access refcount, __uprobe_register() takes a creation
|
uprobes, mm, x86: Add the ability to install and remove uprobes breakpoints
Add uprobes support to the core kernel, with x86 support.
This commit adds the kernel facilities, the actual uprobes
user-space ABI and perf probe support comes in later commits.
General design:
Uprobes are maintained in an rb-tree indexed by inode and offset
(the offset here is from the start of the mapping). For a unique
(inode, offset) tuple, there can be at most one uprobe in the
rb-tree.
Since the (inode, offset) tuple identifies a unique uprobe, more
than one user may be interested in the same uprobe. This provides
the ability to connect multiple 'consumers' to the same uprobe.
Each consumer defines a handler and a filter (optional). The
'handler' is run every time the uprobe is hit, if it matches the
'filter' criteria.
The first consumer of a uprobe causes the breakpoint to be
inserted at the specified address and subsequent consumers are
appended to this list. On subsequent probes, the consumer gets
appended to the existing list of consumers. The breakpoint is
removed when the last consumer unregisters. For all other
unregisterations, the consumer is removed from the list of
consumers.
Given a inode, we get a list of the mms that have mapped the
inode. Do the actual registration if mm maps the page where a
probe needs to be inserted/removed.
We use a temporary list to walk through the vmas that map the
inode.
- The number of maps that map the inode, is not known before we
walk the rmap and keeps changing.
- extending vm_area_struct wasn't recommended, it's a
size-critical data structure.
- There can be more than one maps of the inode in the same mm.
We add callbacks to the mmap methods to keep an eye on text vmas
that are of interest to uprobes. When a vma of interest is mapped,
we insert the breakpoint at the right address.
Uprobe works by replacing the instruction at the address defined
by (inode, offset) with the arch specific breakpoint
instruction. We save a copy of the original instruction at the
uprobed address.
This is needed for:
a. executing the instruction out-of-line (xol).
b. instruction analysis for any subsequent fixups.
c. restoring the instruction back when the uprobe is unregistered.
We insert or delete a breakpoint instruction, and this
breakpoint instruction is assumed to be the smallest instruction
available on the platform. For fixed size instruction platforms
this is trivially true, for variable size instruction platforms
the breakpoint instruction is typically the smallest (often a
single byte).
Writing the instruction is done by COWing the page and changing
the instruction during the copy, this even though most platforms
allow atomic writes of the breakpoint instruction. This also
mirrors the behaviour of a ptrace() memory write to a PRIVATE
file map.
The core worker is derived from KSM's replace_page() logic.
In essence, similar to KSM:
a. allocate a new page and copy over contents of the page that
has the uprobed vaddr
b. modify the copy and insert the breakpoint at the required
address
c. switch the original page with the copy containing the
breakpoint
d. flush page tables.
replace_page() is being replicated here because of some minor
changes in the type of pages and also because Hugh Dickins had
plans to improve replace_page() for KSM specific work.
Instruction analysis on x86 is based on instruction decoder and
determines if an instruction can be probed and determines the
necessary fixups after singlestep. Instruction analysis is done
at probe insertion time so that we avoid having to repeat the
same analysis every time a probe is hit.
A lot of code here is due to the improvement/suggestions/inputs
from Peter Zijlstra.
Changelog:
(v10):
- Add code to clear REX.B prefix as suggested by Denys Vlasenko
and Masami Hiramatsu.
(v9):
- Use insn_offset_modrm as suggested by Masami Hiramatsu.
(v7):
Handle comments from Peter Zijlstra:
- Dont take reference to inode. (expect inode to uprobe_register to be sane).
- Use PTR_ERR to set the return value.
- No need to take reference to inode.
- use PTR_ERR to return error value.
- register and uprobe_unregister share code.
(v5):
- Modified del_consumer as per comments from Peter.
- Drop reference to inode before dropping reference to uprobe.
- Use i_size_read(inode) instead of inode->i_size.
- Ensure uprobe->consumers is NULL, before __uprobe_unregister() is called.
- Includes errno.h as recommended by Stephen Rothwell to fix a build issue
on sparc defconfig
- Remove restrictions while unregistering.
- Earlier code leaked inode references under some conditions while
registering/unregistering.
- Continue the vma-rmap walk even if the intermediate vma doesnt
meet the requirements.
- Validate the vma found by find_vma before inserting/removing the
breakpoint
- Call del_consumer under mutex_lock.
- Use hash locks.
- Handle mremap.
- Introduce find_least_offset_node() instead of close match logic in
find_uprobe
- Uprobes no more depends on MM_OWNER; No reference to task_structs
while inserting/removing a probe.
- Uses read_mapping_page instead of grab_cache_page so that the pages
have valid content.
- pass NULL to get_user_pages for the task parameter.
- call SetPageUptodate on the new page allocated in write_opcode.
- fix leaking a reference to the new page under certain conditions.
- Include Instruction Decoder if Uprobes gets defined.
- Remove const attributes for instruction prefix arrays.
- Uses mm_context to know if the application is 32 bit.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Also-written-by: Jim Keniston <jkenisto@us.ibm.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Roland McGrath <roland@hack.frob.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Anton Arapov <anton@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Denys Vlasenko <vda.linux@googlemail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linux-mm <linux-mm@kvack.org>
Link: http://lkml.kernel.org/r/20120209092642.GE16600@linux.vnet.ibm.com
[ Made various small edits to the commit log ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-02-09 17:26:42 +08:00
|
|
|
* refcount (thro alloc_uprobe) if and only if this @uprobe is getting
|
|
|
|
* inserted into the rbtree (i.e first consumer for a @inode:@offset
|
2012-02-17 16:27:41 +08:00
|
|
|
* tuple). Creation refcount stops uprobe_unregister from freeing the
|
uprobes, mm, x86: Add the ability to install and remove uprobes breakpoints
Add uprobes support to the core kernel, with x86 support.
This commit adds the kernel facilities, the actual uprobes
user-space ABI and perf probe support comes in later commits.
General design:
Uprobes are maintained in an rb-tree indexed by inode and offset
(the offset here is from the start of the mapping). For a unique
(inode, offset) tuple, there can be at most one uprobe in the
rb-tree.
Since the (inode, offset) tuple identifies a unique uprobe, more
than one user may be interested in the same uprobe. This provides
the ability to connect multiple 'consumers' to the same uprobe.
Each consumer defines a handler and a filter (optional). The
'handler' is run every time the uprobe is hit, if it matches the
'filter' criteria.
The first consumer of a uprobe causes the breakpoint to be
inserted at the specified address and subsequent consumers are
appended to this list. On subsequent probes, the consumer gets
appended to the existing list of consumers. The breakpoint is
removed when the last consumer unregisters. For all other
unregisterations, the consumer is removed from the list of
consumers.
Given a inode, we get a list of the mms that have mapped the
inode. Do the actual registration if mm maps the page where a
probe needs to be inserted/removed.
We use a temporary list to walk through the vmas that map the
inode.
- The number of maps that map the inode, is not known before we
walk the rmap and keeps changing.
- extending vm_area_struct wasn't recommended, it's a
size-critical data structure.
- There can be more than one maps of the inode in the same mm.
We add callbacks to the mmap methods to keep an eye on text vmas
that are of interest to uprobes. When a vma of interest is mapped,
we insert the breakpoint at the right address.
Uprobe works by replacing the instruction at the address defined
by (inode, offset) with the arch specific breakpoint
instruction. We save a copy of the original instruction at the
uprobed address.
This is needed for:
a. executing the instruction out-of-line (xol).
b. instruction analysis for any subsequent fixups.
c. restoring the instruction back when the uprobe is unregistered.
We insert or delete a breakpoint instruction, and this
breakpoint instruction is assumed to be the smallest instruction
available on the platform. For fixed size instruction platforms
this is trivially true, for variable size instruction platforms
the breakpoint instruction is typically the smallest (often a
single byte).
Writing the instruction is done by COWing the page and changing
the instruction during the copy, this even though most platforms
allow atomic writes of the breakpoint instruction. This also
mirrors the behaviour of a ptrace() memory write to a PRIVATE
file map.
The core worker is derived from KSM's replace_page() logic.
In essence, similar to KSM:
a. allocate a new page and copy over contents of the page that
has the uprobed vaddr
b. modify the copy and insert the breakpoint at the required
address
c. switch the original page with the copy containing the
breakpoint
d. flush page tables.
replace_page() is being replicated here because of some minor
changes in the type of pages and also because Hugh Dickins had
plans to improve replace_page() for KSM specific work.
Instruction analysis on x86 is based on instruction decoder and
determines if an instruction can be probed and determines the
necessary fixups after singlestep. Instruction analysis is done
at probe insertion time so that we avoid having to repeat the
same analysis every time a probe is hit.
A lot of code here is due to the improvement/suggestions/inputs
from Peter Zijlstra.
Changelog:
(v10):
- Add code to clear REX.B prefix as suggested by Denys Vlasenko
and Masami Hiramatsu.
(v9):
- Use insn_offset_modrm as suggested by Masami Hiramatsu.
(v7):
Handle comments from Peter Zijlstra:
- Dont take reference to inode. (expect inode to uprobe_register to be sane).
- Use PTR_ERR to set the return value.
- No need to take reference to inode.
- use PTR_ERR to return error value.
- register and uprobe_unregister share code.
(v5):
- Modified del_consumer as per comments from Peter.
- Drop reference to inode before dropping reference to uprobe.
- Use i_size_read(inode) instead of inode->i_size.
- Ensure uprobe->consumers is NULL, before __uprobe_unregister() is called.
- Includes errno.h as recommended by Stephen Rothwell to fix a build issue
on sparc defconfig
- Remove restrictions while unregistering.
- Earlier code leaked inode references under some conditions while
registering/unregistering.
- Continue the vma-rmap walk even if the intermediate vma doesnt
meet the requirements.
- Validate the vma found by find_vma before inserting/removing the
breakpoint
- Call del_consumer under mutex_lock.
- Use hash locks.
- Handle mremap.
- Introduce find_least_offset_node() instead of close match logic in
find_uprobe
- Uprobes no more depends on MM_OWNER; No reference to task_structs
while inserting/removing a probe.
- Uses read_mapping_page instead of grab_cache_page so that the pages
have valid content.
- pass NULL to get_user_pages for the task parameter.
- call SetPageUptodate on the new page allocated in write_opcode.
- fix leaking a reference to the new page under certain conditions.
- Include Instruction Decoder if Uprobes gets defined.
- Remove const attributes for instruction prefix arrays.
- Uses mm_context to know if the application is 32 bit.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Also-written-by: Jim Keniston <jkenisto@us.ibm.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Roland McGrath <roland@hack.frob.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Anton Arapov <anton@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Denys Vlasenko <vda.linux@googlemail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linux-mm <linux-mm@kvack.org>
Link: http://lkml.kernel.org/r/20120209092642.GE16600@linux.vnet.ibm.com
[ Made various small edits to the commit log ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-02-09 17:26:42 +08:00
|
|
|
* @uprobe even before the register operation is complete. Creation
|
2012-03-12 17:25:30 +08:00
|
|
|
* refcount is released when the last @uc for the @uprobe
|
2018-08-09 12:18:51 +08:00
|
|
|
* unregisters. Caller of __uprobe_register() is required to keep @inode
|
2018-04-24 01:21:35 +08:00
|
|
|
* (and the containing mount) referenced.
|
uprobes, mm, x86: Add the ability to install and remove uprobes breakpoints
Add uprobes support to the core kernel, with x86 support.
This commit adds the kernel facilities, the actual uprobes
user-space ABI and perf probe support comes in later commits.
General design:
Uprobes are maintained in an rb-tree indexed by inode and offset
(the offset here is from the start of the mapping). For a unique
(inode, offset) tuple, there can be at most one uprobe in the
rb-tree.
Since the (inode, offset) tuple identifies a unique uprobe, more
than one user may be interested in the same uprobe. This provides
the ability to connect multiple 'consumers' to the same uprobe.
Each consumer defines a handler and a filter (optional). The
'handler' is run every time the uprobe is hit, if it matches the
'filter' criteria.
The first consumer of a uprobe causes the breakpoint to be
inserted at the specified address and subsequent consumers are
appended to this list. On subsequent probes, the consumer gets
appended to the existing list of consumers. The breakpoint is
removed when the last consumer unregisters. For all other
unregisterations, the consumer is removed from the list of
consumers.
Given a inode, we get a list of the mms that have mapped the
inode. Do the actual registration if mm maps the page where a
probe needs to be inserted/removed.
We use a temporary list to walk through the vmas that map the
inode.
- The number of maps that map the inode, is not known before we
walk the rmap and keeps changing.
- extending vm_area_struct wasn't recommended, it's a
size-critical data structure.
- There can be more than one maps of the inode in the same mm.
We add callbacks to the mmap methods to keep an eye on text vmas
that are of interest to uprobes. When a vma of interest is mapped,
we insert the breakpoint at the right address.
Uprobe works by replacing the instruction at the address defined
by (inode, offset) with the arch specific breakpoint
instruction. We save a copy of the original instruction at the
uprobed address.
This is needed for:
a. executing the instruction out-of-line (xol).
b. instruction analysis for any subsequent fixups.
c. restoring the instruction back when the uprobe is unregistered.
We insert or delete a breakpoint instruction, and this
breakpoint instruction is assumed to be the smallest instruction
available on the platform. For fixed size instruction platforms
this is trivially true, for variable size instruction platforms
the breakpoint instruction is typically the smallest (often a
single byte).
Writing the instruction is done by COWing the page and changing
the instruction during the copy, this even though most platforms
allow atomic writes of the breakpoint instruction. This also
mirrors the behaviour of a ptrace() memory write to a PRIVATE
file map.
The core worker is derived from KSM's replace_page() logic.
In essence, similar to KSM:
a. allocate a new page and copy over contents of the page that
has the uprobed vaddr
b. modify the copy and insert the breakpoint at the required
address
c. switch the original page with the copy containing the
breakpoint
d. flush page tables.
replace_page() is being replicated here because of some minor
changes in the type of pages and also because Hugh Dickins had
plans to improve replace_page() for KSM specific work.
Instruction analysis on x86 is based on instruction decoder and
determines if an instruction can be probed and determines the
necessary fixups after singlestep. Instruction analysis is done
at probe insertion time so that we avoid having to repeat the
same analysis every time a probe is hit.
A lot of code here is due to the improvement/suggestions/inputs
from Peter Zijlstra.
Changelog:
(v10):
- Add code to clear REX.B prefix as suggested by Denys Vlasenko
and Masami Hiramatsu.
(v9):
- Use insn_offset_modrm as suggested by Masami Hiramatsu.
(v7):
Handle comments from Peter Zijlstra:
- Dont take reference to inode. (expect inode to uprobe_register to be sane).
- Use PTR_ERR to set the return value.
- No need to take reference to inode.
- use PTR_ERR to return error value.
- register and uprobe_unregister share code.
(v5):
- Modified del_consumer as per comments from Peter.
- Drop reference to inode before dropping reference to uprobe.
- Use i_size_read(inode) instead of inode->i_size.
- Ensure uprobe->consumers is NULL, before __uprobe_unregister() is called.
- Includes errno.h as recommended by Stephen Rothwell to fix a build issue
on sparc defconfig
- Remove restrictions while unregistering.
- Earlier code leaked inode references under some conditions while
registering/unregistering.
- Continue the vma-rmap walk even if the intermediate vma doesnt
meet the requirements.
- Validate the vma found by find_vma before inserting/removing the
breakpoint
- Call del_consumer under mutex_lock.
- Use hash locks.
- Handle mremap.
- Introduce find_least_offset_node() instead of close match logic in
find_uprobe
- Uprobes no more depends on MM_OWNER; No reference to task_structs
while inserting/removing a probe.
- Uses read_mapping_page instead of grab_cache_page so that the pages
have valid content.
- pass NULL to get_user_pages for the task parameter.
- call SetPageUptodate on the new page allocated in write_opcode.
- fix leaking a reference to the new page under certain conditions.
- Include Instruction Decoder if Uprobes gets defined.
- Remove const attributes for instruction prefix arrays.
- Uses mm_context to know if the application is 32 bit.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Also-written-by: Jim Keniston <jkenisto@us.ibm.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Roland McGrath <roland@hack.frob.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Anton Arapov <anton@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Denys Vlasenko <vda.linux@googlemail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linux-mm <linux-mm@kvack.org>
Link: http://lkml.kernel.org/r/20120209092642.GE16600@linux.vnet.ibm.com
[ Made various small edits to the commit log ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-02-09 17:26:42 +08:00
|
|
|
*
|
|
|
|
* Return errno if it cannot successully install probes
|
|
|
|
* else return 0 (success)
|
|
|
|
*/
|
2018-08-09 12:18:51 +08:00
|
|
|
static int __uprobe_register(struct inode *inode, loff_t offset,
|
uprobes: Support SDT markers having reference count (semaphore)
Userspace Statically Defined Tracepoints[1] are dtrace style markers
inside userspace applications. Applications like PostgreSQL, MySQL,
Pthread, Perl, Python, Java, Ruby, Node.js, libvirt, QEMU, glib etc
have these markers embedded in them. These markers are added by developer
at important places in the code. Each marker source expands to a single
nop instruction in the compiled code but there may be additional
overhead for computing the marker arguments which expands to couple of
instructions. In case the overhead is more, execution of it can be
omitted by runtime if() condition when no one is tracing on the marker:
if (reference_counter > 0) {
Execute marker instructions;
}
Default value of reference counter is 0. Tracer has to increment the
reference counter before tracing on a marker and decrement it when
done with the tracing.
Implement the reference counter logic in core uprobe. User will be
able to use it from trace_uprobe as well as from kernel module. New
trace_uprobe definition with reference counter will now be:
<path>:<offset>[(ref_ctr_offset)]
where ref_ctr_offset is an optional field. For kernel module, new
variant of uprobe_register() has been introduced:
uprobe_register_refctr(inode, offset, ref_ctr_offset, consumer)
No new variant for uprobe_unregister() because it's assumed to have
only one reference counter for one uprobe.
[1] https://sourceware.org/systemtap/wiki/UserSpaceProbeImplementation
Note: 'reference counter' is called as 'semaphore' in original Dtrace
(or Systemtap, bcc and even in ELF) documentation and code. But the
term 'semaphore' is misleading in this context. This is just a counter
used to hold number of tracers tracing on a marker. This is not really
used for any synchronization. So we are calling it a 'reference counter'
in kernel / perf code.
Link: http://lkml.kernel.org/r/20180820044250.11659-2-ravi.bangoria@linux.ibm.com
Reviewed-by: Masami Hiramatsu <mhiramat@kernel.org>
[Only trace_uprobe.c]
Reviewed-by: Oleg Nesterov <oleg@redhat.com>
Reviewed-by: Song Liu <songliubraving@fb.com>
Tested-by: Song Liu <songliubraving@fb.com>
Signed-off-by: Ravi Bangoria <ravi.bangoria@linux.ibm.com>
Signed-off-by: Steven Rostedt (VMware) <rostedt@goodmis.org>
2018-08-20 12:42:47 +08:00
|
|
|
loff_t ref_ctr_offset, struct uprobe_consumer *uc)
|
uprobes, mm, x86: Add the ability to install and remove uprobes breakpoints
Add uprobes support to the core kernel, with x86 support.
This commit adds the kernel facilities, the actual uprobes
user-space ABI and perf probe support comes in later commits.
General design:
Uprobes are maintained in an rb-tree indexed by inode and offset
(the offset here is from the start of the mapping). For a unique
(inode, offset) tuple, there can be at most one uprobe in the
rb-tree.
Since the (inode, offset) tuple identifies a unique uprobe, more
than one user may be interested in the same uprobe. This provides
the ability to connect multiple 'consumers' to the same uprobe.
Each consumer defines a handler and a filter (optional). The
'handler' is run every time the uprobe is hit, if it matches the
'filter' criteria.
The first consumer of a uprobe causes the breakpoint to be
inserted at the specified address and subsequent consumers are
appended to this list. On subsequent probes, the consumer gets
appended to the existing list of consumers. The breakpoint is
removed when the last consumer unregisters. For all other
unregisterations, the consumer is removed from the list of
consumers.
Given a inode, we get a list of the mms that have mapped the
inode. Do the actual registration if mm maps the page where a
probe needs to be inserted/removed.
We use a temporary list to walk through the vmas that map the
inode.
- The number of maps that map the inode, is not known before we
walk the rmap and keeps changing.
- extending vm_area_struct wasn't recommended, it's a
size-critical data structure.
- There can be more than one maps of the inode in the same mm.
We add callbacks to the mmap methods to keep an eye on text vmas
that are of interest to uprobes. When a vma of interest is mapped,
we insert the breakpoint at the right address.
Uprobe works by replacing the instruction at the address defined
by (inode, offset) with the arch specific breakpoint
instruction. We save a copy of the original instruction at the
uprobed address.
This is needed for:
a. executing the instruction out-of-line (xol).
b. instruction analysis for any subsequent fixups.
c. restoring the instruction back when the uprobe is unregistered.
We insert or delete a breakpoint instruction, and this
breakpoint instruction is assumed to be the smallest instruction
available on the platform. For fixed size instruction platforms
this is trivially true, for variable size instruction platforms
the breakpoint instruction is typically the smallest (often a
single byte).
Writing the instruction is done by COWing the page and changing
the instruction during the copy, this even though most platforms
allow atomic writes of the breakpoint instruction. This also
mirrors the behaviour of a ptrace() memory write to a PRIVATE
file map.
The core worker is derived from KSM's replace_page() logic.
In essence, similar to KSM:
a. allocate a new page and copy over contents of the page that
has the uprobed vaddr
b. modify the copy and insert the breakpoint at the required
address
c. switch the original page with the copy containing the
breakpoint
d. flush page tables.
replace_page() is being replicated here because of some minor
changes in the type of pages and also because Hugh Dickins had
plans to improve replace_page() for KSM specific work.
Instruction analysis on x86 is based on instruction decoder and
determines if an instruction can be probed and determines the
necessary fixups after singlestep. Instruction analysis is done
at probe insertion time so that we avoid having to repeat the
same analysis every time a probe is hit.
A lot of code here is due to the improvement/suggestions/inputs
from Peter Zijlstra.
Changelog:
(v10):
- Add code to clear REX.B prefix as suggested by Denys Vlasenko
and Masami Hiramatsu.
(v9):
- Use insn_offset_modrm as suggested by Masami Hiramatsu.
(v7):
Handle comments from Peter Zijlstra:
- Dont take reference to inode. (expect inode to uprobe_register to be sane).
- Use PTR_ERR to set the return value.
- No need to take reference to inode.
- use PTR_ERR to return error value.
- register and uprobe_unregister share code.
(v5):
- Modified del_consumer as per comments from Peter.
- Drop reference to inode before dropping reference to uprobe.
- Use i_size_read(inode) instead of inode->i_size.
- Ensure uprobe->consumers is NULL, before __uprobe_unregister() is called.
- Includes errno.h as recommended by Stephen Rothwell to fix a build issue
on sparc defconfig
- Remove restrictions while unregistering.
- Earlier code leaked inode references under some conditions while
registering/unregistering.
- Continue the vma-rmap walk even if the intermediate vma doesnt
meet the requirements.
- Validate the vma found by find_vma before inserting/removing the
breakpoint
- Call del_consumer under mutex_lock.
- Use hash locks.
- Handle mremap.
- Introduce find_least_offset_node() instead of close match logic in
find_uprobe
- Uprobes no more depends on MM_OWNER; No reference to task_structs
while inserting/removing a probe.
- Uses read_mapping_page instead of grab_cache_page so that the pages
have valid content.
- pass NULL to get_user_pages for the task parameter.
- call SetPageUptodate on the new page allocated in write_opcode.
- fix leaking a reference to the new page under certain conditions.
- Include Instruction Decoder if Uprobes gets defined.
- Remove const attributes for instruction prefix arrays.
- Uses mm_context to know if the application is 32 bit.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Also-written-by: Jim Keniston <jkenisto@us.ibm.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Roland McGrath <roland@hack.frob.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Anton Arapov <anton@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Denys Vlasenko <vda.linux@googlemail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linux-mm <linux-mm@kvack.org>
Link: http://lkml.kernel.org/r/20120209092642.GE16600@linux.vnet.ibm.com
[ Made various small edits to the commit log ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-02-09 17:26:42 +08:00
|
|
|
{
|
|
|
|
struct uprobe *uprobe;
|
2012-02-17 16:27:41 +08:00
|
|
|
int ret;
|
uprobes, mm, x86: Add the ability to install and remove uprobes breakpoints
Add uprobes support to the core kernel, with x86 support.
This commit adds the kernel facilities, the actual uprobes
user-space ABI and perf probe support comes in later commits.
General design:
Uprobes are maintained in an rb-tree indexed by inode and offset
(the offset here is from the start of the mapping). For a unique
(inode, offset) tuple, there can be at most one uprobe in the
rb-tree.
Since the (inode, offset) tuple identifies a unique uprobe, more
than one user may be interested in the same uprobe. This provides
the ability to connect multiple 'consumers' to the same uprobe.
Each consumer defines a handler and a filter (optional). The
'handler' is run every time the uprobe is hit, if it matches the
'filter' criteria.
The first consumer of a uprobe causes the breakpoint to be
inserted at the specified address and subsequent consumers are
appended to this list. On subsequent probes, the consumer gets
appended to the existing list of consumers. The breakpoint is
removed when the last consumer unregisters. For all other
unregisterations, the consumer is removed from the list of
consumers.
Given a inode, we get a list of the mms that have mapped the
inode. Do the actual registration if mm maps the page where a
probe needs to be inserted/removed.
We use a temporary list to walk through the vmas that map the
inode.
- The number of maps that map the inode, is not known before we
walk the rmap and keeps changing.
- extending vm_area_struct wasn't recommended, it's a
size-critical data structure.
- There can be more than one maps of the inode in the same mm.
We add callbacks to the mmap methods to keep an eye on text vmas
that are of interest to uprobes. When a vma of interest is mapped,
we insert the breakpoint at the right address.
Uprobe works by replacing the instruction at the address defined
by (inode, offset) with the arch specific breakpoint
instruction. We save a copy of the original instruction at the
uprobed address.
This is needed for:
a. executing the instruction out-of-line (xol).
b. instruction analysis for any subsequent fixups.
c. restoring the instruction back when the uprobe is unregistered.
We insert or delete a breakpoint instruction, and this
breakpoint instruction is assumed to be the smallest instruction
available on the platform. For fixed size instruction platforms
this is trivially true, for variable size instruction platforms
the breakpoint instruction is typically the smallest (often a
single byte).
Writing the instruction is done by COWing the page and changing
the instruction during the copy, this even though most platforms
allow atomic writes of the breakpoint instruction. This also
mirrors the behaviour of a ptrace() memory write to a PRIVATE
file map.
The core worker is derived from KSM's replace_page() logic.
In essence, similar to KSM:
a. allocate a new page and copy over contents of the page that
has the uprobed vaddr
b. modify the copy and insert the breakpoint at the required
address
c. switch the original page with the copy containing the
breakpoint
d. flush page tables.
replace_page() is being replicated here because of some minor
changes in the type of pages and also because Hugh Dickins had
plans to improve replace_page() for KSM specific work.
Instruction analysis on x86 is based on instruction decoder and
determines if an instruction can be probed and determines the
necessary fixups after singlestep. Instruction analysis is done
at probe insertion time so that we avoid having to repeat the
same analysis every time a probe is hit.
A lot of code here is due to the improvement/suggestions/inputs
from Peter Zijlstra.
Changelog:
(v10):
- Add code to clear REX.B prefix as suggested by Denys Vlasenko
and Masami Hiramatsu.
(v9):
- Use insn_offset_modrm as suggested by Masami Hiramatsu.
(v7):
Handle comments from Peter Zijlstra:
- Dont take reference to inode. (expect inode to uprobe_register to be sane).
- Use PTR_ERR to set the return value.
- No need to take reference to inode.
- use PTR_ERR to return error value.
- register and uprobe_unregister share code.
(v5):
- Modified del_consumer as per comments from Peter.
- Drop reference to inode before dropping reference to uprobe.
- Use i_size_read(inode) instead of inode->i_size.
- Ensure uprobe->consumers is NULL, before __uprobe_unregister() is called.
- Includes errno.h as recommended by Stephen Rothwell to fix a build issue
on sparc defconfig
- Remove restrictions while unregistering.
- Earlier code leaked inode references under some conditions while
registering/unregistering.
- Continue the vma-rmap walk even if the intermediate vma doesnt
meet the requirements.
- Validate the vma found by find_vma before inserting/removing the
breakpoint
- Call del_consumer under mutex_lock.
- Use hash locks.
- Handle mremap.
- Introduce find_least_offset_node() instead of close match logic in
find_uprobe
- Uprobes no more depends on MM_OWNER; No reference to task_structs
while inserting/removing a probe.
- Uses read_mapping_page instead of grab_cache_page so that the pages
have valid content.
- pass NULL to get_user_pages for the task parameter.
- call SetPageUptodate on the new page allocated in write_opcode.
- fix leaking a reference to the new page under certain conditions.
- Include Instruction Decoder if Uprobes gets defined.
- Remove const attributes for instruction prefix arrays.
- Uses mm_context to know if the application is 32 bit.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Also-written-by: Jim Keniston <jkenisto@us.ibm.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Roland McGrath <roland@hack.frob.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Anton Arapov <anton@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Denys Vlasenko <vda.linux@googlemail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linux-mm <linux-mm@kvack.org>
Link: http://lkml.kernel.org/r/20120209092642.GE16600@linux.vnet.ibm.com
[ Made various small edits to the commit log ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-02-09 17:26:42 +08:00
|
|
|
|
2013-04-04 00:00:31 +08:00
|
|
|
/* Uprobe must have at least one set consumer */
|
|
|
|
if (!uc->handler && !uc->ret_handler)
|
|
|
|
return -EINVAL;
|
|
|
|
|
2014-05-20 02:41:36 +08:00
|
|
|
/* copy_insn() uses read_mapping_page() or shmem_read_mapping_page() */
|
|
|
|
if (!inode->i_mapping->a_ops->readpage && !shmem_mapping(inode->i_mapping))
|
2014-05-20 02:40:54 +08:00
|
|
|
return -EIO;
|
2012-11-22 01:01:43 +08:00
|
|
|
/* Racy, just to catch the obvious mistakes */
|
uprobes, mm, x86: Add the ability to install and remove uprobes breakpoints
Add uprobes support to the core kernel, with x86 support.
This commit adds the kernel facilities, the actual uprobes
user-space ABI and perf probe support comes in later commits.
General design:
Uprobes are maintained in an rb-tree indexed by inode and offset
(the offset here is from the start of the mapping). For a unique
(inode, offset) tuple, there can be at most one uprobe in the
rb-tree.
Since the (inode, offset) tuple identifies a unique uprobe, more
than one user may be interested in the same uprobe. This provides
the ability to connect multiple 'consumers' to the same uprobe.
Each consumer defines a handler and a filter (optional). The
'handler' is run every time the uprobe is hit, if it matches the
'filter' criteria.
The first consumer of a uprobe causes the breakpoint to be
inserted at the specified address and subsequent consumers are
appended to this list. On subsequent probes, the consumer gets
appended to the existing list of consumers. The breakpoint is
removed when the last consumer unregisters. For all other
unregisterations, the consumer is removed from the list of
consumers.
Given a inode, we get a list of the mms that have mapped the
inode. Do the actual registration if mm maps the page where a
probe needs to be inserted/removed.
We use a temporary list to walk through the vmas that map the
inode.
- The number of maps that map the inode, is not known before we
walk the rmap and keeps changing.
- extending vm_area_struct wasn't recommended, it's a
size-critical data structure.
- There can be more than one maps of the inode in the same mm.
We add callbacks to the mmap methods to keep an eye on text vmas
that are of interest to uprobes. When a vma of interest is mapped,
we insert the breakpoint at the right address.
Uprobe works by replacing the instruction at the address defined
by (inode, offset) with the arch specific breakpoint
instruction. We save a copy of the original instruction at the
uprobed address.
This is needed for:
a. executing the instruction out-of-line (xol).
b. instruction analysis for any subsequent fixups.
c. restoring the instruction back when the uprobe is unregistered.
We insert or delete a breakpoint instruction, and this
breakpoint instruction is assumed to be the smallest instruction
available on the platform. For fixed size instruction platforms
this is trivially true, for variable size instruction platforms
the breakpoint instruction is typically the smallest (often a
single byte).
Writing the instruction is done by COWing the page and changing
the instruction during the copy, this even though most platforms
allow atomic writes of the breakpoint instruction. This also
mirrors the behaviour of a ptrace() memory write to a PRIVATE
file map.
The core worker is derived from KSM's replace_page() logic.
In essence, similar to KSM:
a. allocate a new page and copy over contents of the page that
has the uprobed vaddr
b. modify the copy and insert the breakpoint at the required
address
c. switch the original page with the copy containing the
breakpoint
d. flush page tables.
replace_page() is being replicated here because of some minor
changes in the type of pages and also because Hugh Dickins had
plans to improve replace_page() for KSM specific work.
Instruction analysis on x86 is based on instruction decoder and
determines if an instruction can be probed and determines the
necessary fixups after singlestep. Instruction analysis is done
at probe insertion time so that we avoid having to repeat the
same analysis every time a probe is hit.
A lot of code here is due to the improvement/suggestions/inputs
from Peter Zijlstra.
Changelog:
(v10):
- Add code to clear REX.B prefix as suggested by Denys Vlasenko
and Masami Hiramatsu.
(v9):
- Use insn_offset_modrm as suggested by Masami Hiramatsu.
(v7):
Handle comments from Peter Zijlstra:
- Dont take reference to inode. (expect inode to uprobe_register to be sane).
- Use PTR_ERR to set the return value.
- No need to take reference to inode.
- use PTR_ERR to return error value.
- register and uprobe_unregister share code.
(v5):
- Modified del_consumer as per comments from Peter.
- Drop reference to inode before dropping reference to uprobe.
- Use i_size_read(inode) instead of inode->i_size.
- Ensure uprobe->consumers is NULL, before __uprobe_unregister() is called.
- Includes errno.h as recommended by Stephen Rothwell to fix a build issue
on sparc defconfig
- Remove restrictions while unregistering.
- Earlier code leaked inode references under some conditions while
registering/unregistering.
- Continue the vma-rmap walk even if the intermediate vma doesnt
meet the requirements.
- Validate the vma found by find_vma before inserting/removing the
breakpoint
- Call del_consumer under mutex_lock.
- Use hash locks.
- Handle mremap.
- Introduce find_least_offset_node() instead of close match logic in
find_uprobe
- Uprobes no more depends on MM_OWNER; No reference to task_structs
while inserting/removing a probe.
- Uses read_mapping_page instead of grab_cache_page so that the pages
have valid content.
- pass NULL to get_user_pages for the task parameter.
- call SetPageUptodate on the new page allocated in write_opcode.
- fix leaking a reference to the new page under certain conditions.
- Include Instruction Decoder if Uprobes gets defined.
- Remove const attributes for instruction prefix arrays.
- Uses mm_context to know if the application is 32 bit.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Also-written-by: Jim Keniston <jkenisto@us.ibm.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Roland McGrath <roland@hack.frob.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Anton Arapov <anton@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Denys Vlasenko <vda.linux@googlemail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linux-mm <linux-mm@kvack.org>
Link: http://lkml.kernel.org/r/20120209092642.GE16600@linux.vnet.ibm.com
[ Made various small edits to the commit log ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-02-09 17:26:42 +08:00
|
|
|
if (offset > i_size_read(inode))
|
2012-02-17 16:27:41 +08:00
|
|
|
return -EINVAL;
|
uprobes, mm, x86: Add the ability to install and remove uprobes breakpoints
Add uprobes support to the core kernel, with x86 support.
This commit adds the kernel facilities, the actual uprobes
user-space ABI and perf probe support comes in later commits.
General design:
Uprobes are maintained in an rb-tree indexed by inode and offset
(the offset here is from the start of the mapping). For a unique
(inode, offset) tuple, there can be at most one uprobe in the
rb-tree.
Since the (inode, offset) tuple identifies a unique uprobe, more
than one user may be interested in the same uprobe. This provides
the ability to connect multiple 'consumers' to the same uprobe.
Each consumer defines a handler and a filter (optional). The
'handler' is run every time the uprobe is hit, if it matches the
'filter' criteria.
The first consumer of a uprobe causes the breakpoint to be
inserted at the specified address and subsequent consumers are
appended to this list. On subsequent probes, the consumer gets
appended to the existing list of consumers. The breakpoint is
removed when the last consumer unregisters. For all other
unregisterations, the consumer is removed from the list of
consumers.
Given a inode, we get a list of the mms that have mapped the
inode. Do the actual registration if mm maps the page where a
probe needs to be inserted/removed.
We use a temporary list to walk through the vmas that map the
inode.
- The number of maps that map the inode, is not known before we
walk the rmap and keeps changing.
- extending vm_area_struct wasn't recommended, it's a
size-critical data structure.
- There can be more than one maps of the inode in the same mm.
We add callbacks to the mmap methods to keep an eye on text vmas
that are of interest to uprobes. When a vma of interest is mapped,
we insert the breakpoint at the right address.
Uprobe works by replacing the instruction at the address defined
by (inode, offset) with the arch specific breakpoint
instruction. We save a copy of the original instruction at the
uprobed address.
This is needed for:
a. executing the instruction out-of-line (xol).
b. instruction analysis for any subsequent fixups.
c. restoring the instruction back when the uprobe is unregistered.
We insert or delete a breakpoint instruction, and this
breakpoint instruction is assumed to be the smallest instruction
available on the platform. For fixed size instruction platforms
this is trivially true, for variable size instruction platforms
the breakpoint instruction is typically the smallest (often a
single byte).
Writing the instruction is done by COWing the page and changing
the instruction during the copy, this even though most platforms
allow atomic writes of the breakpoint instruction. This also
mirrors the behaviour of a ptrace() memory write to a PRIVATE
file map.
The core worker is derived from KSM's replace_page() logic.
In essence, similar to KSM:
a. allocate a new page and copy over contents of the page that
has the uprobed vaddr
b. modify the copy and insert the breakpoint at the required
address
c. switch the original page with the copy containing the
breakpoint
d. flush page tables.
replace_page() is being replicated here because of some minor
changes in the type of pages and also because Hugh Dickins had
plans to improve replace_page() for KSM specific work.
Instruction analysis on x86 is based on instruction decoder and
determines if an instruction can be probed and determines the
necessary fixups after singlestep. Instruction analysis is done
at probe insertion time so that we avoid having to repeat the
same analysis every time a probe is hit.
A lot of code here is due to the improvement/suggestions/inputs
from Peter Zijlstra.
Changelog:
(v10):
- Add code to clear REX.B prefix as suggested by Denys Vlasenko
and Masami Hiramatsu.
(v9):
- Use insn_offset_modrm as suggested by Masami Hiramatsu.
(v7):
Handle comments from Peter Zijlstra:
- Dont take reference to inode. (expect inode to uprobe_register to be sane).
- Use PTR_ERR to set the return value.
- No need to take reference to inode.
- use PTR_ERR to return error value.
- register and uprobe_unregister share code.
(v5):
- Modified del_consumer as per comments from Peter.
- Drop reference to inode before dropping reference to uprobe.
- Use i_size_read(inode) instead of inode->i_size.
- Ensure uprobe->consumers is NULL, before __uprobe_unregister() is called.
- Includes errno.h as recommended by Stephen Rothwell to fix a build issue
on sparc defconfig
- Remove restrictions while unregistering.
- Earlier code leaked inode references under some conditions while
registering/unregistering.
- Continue the vma-rmap walk even if the intermediate vma doesnt
meet the requirements.
- Validate the vma found by find_vma before inserting/removing the
breakpoint
- Call del_consumer under mutex_lock.
- Use hash locks.
- Handle mremap.
- Introduce find_least_offset_node() instead of close match logic in
find_uprobe
- Uprobes no more depends on MM_OWNER; No reference to task_structs
while inserting/removing a probe.
- Uses read_mapping_page instead of grab_cache_page so that the pages
have valid content.
- pass NULL to get_user_pages for the task parameter.
- call SetPageUptodate on the new page allocated in write_opcode.
- fix leaking a reference to the new page under certain conditions.
- Include Instruction Decoder if Uprobes gets defined.
- Remove const attributes for instruction prefix arrays.
- Uses mm_context to know if the application is 32 bit.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Also-written-by: Jim Keniston <jkenisto@us.ibm.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Roland McGrath <roland@hack.frob.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Anton Arapov <anton@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Denys Vlasenko <vda.linux@googlemail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linux-mm <linux-mm@kvack.org>
Link: http://lkml.kernel.org/r/20120209092642.GE16600@linux.vnet.ibm.com
[ Made various small edits to the commit log ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-02-09 17:26:42 +08:00
|
|
|
|
2012-11-26 05:48:37 +08:00
|
|
|
retry:
|
uprobes: Support SDT markers having reference count (semaphore)
Userspace Statically Defined Tracepoints[1] are dtrace style markers
inside userspace applications. Applications like PostgreSQL, MySQL,
Pthread, Perl, Python, Java, Ruby, Node.js, libvirt, QEMU, glib etc
have these markers embedded in them. These markers are added by developer
at important places in the code. Each marker source expands to a single
nop instruction in the compiled code but there may be additional
overhead for computing the marker arguments which expands to couple of
instructions. In case the overhead is more, execution of it can be
omitted by runtime if() condition when no one is tracing on the marker:
if (reference_counter > 0) {
Execute marker instructions;
}
Default value of reference counter is 0. Tracer has to increment the
reference counter before tracing on a marker and decrement it when
done with the tracing.
Implement the reference counter logic in core uprobe. User will be
able to use it from trace_uprobe as well as from kernel module. New
trace_uprobe definition with reference counter will now be:
<path>:<offset>[(ref_ctr_offset)]
where ref_ctr_offset is an optional field. For kernel module, new
variant of uprobe_register() has been introduced:
uprobe_register_refctr(inode, offset, ref_ctr_offset, consumer)
No new variant for uprobe_unregister() because it's assumed to have
only one reference counter for one uprobe.
[1] https://sourceware.org/systemtap/wiki/UserSpaceProbeImplementation
Note: 'reference counter' is called as 'semaphore' in original Dtrace
(or Systemtap, bcc and even in ELF) documentation and code. But the
term 'semaphore' is misleading in this context. This is just a counter
used to hold number of tracers tracing on a marker. This is not really
used for any synchronization. So we are calling it a 'reference counter'
in kernel / perf code.
Link: http://lkml.kernel.org/r/20180820044250.11659-2-ravi.bangoria@linux.ibm.com
Reviewed-by: Masami Hiramatsu <mhiramat@kernel.org>
[Only trace_uprobe.c]
Reviewed-by: Oleg Nesterov <oleg@redhat.com>
Reviewed-by: Song Liu <songliubraving@fb.com>
Tested-by: Song Liu <songliubraving@fb.com>
Signed-off-by: Ravi Bangoria <ravi.bangoria@linux.ibm.com>
Signed-off-by: Steven Rostedt (VMware) <rostedt@goodmis.org>
2018-08-20 12:42:47 +08:00
|
|
|
uprobe = alloc_uprobe(inode, offset, ref_ctr_offset);
|
2012-11-26 05:48:37 +08:00
|
|
|
if (!uprobe)
|
|
|
|
return -ENOMEM;
|
2018-08-20 12:42:48 +08:00
|
|
|
if (IS_ERR(uprobe))
|
|
|
|
return PTR_ERR(uprobe);
|
|
|
|
|
2012-11-26 05:48:37 +08:00
|
|
|
/*
|
|
|
|
* We can race with uprobe_unregister()->delete_uprobe().
|
|
|
|
* Check uprobe_is_active() and retry if it is false.
|
|
|
|
*/
|
|
|
|
down_write(&uprobe->register_rwsem);
|
|
|
|
ret = -EAGAIN;
|
|
|
|
if (likely(uprobe_is_active(uprobe))) {
|
2018-08-09 12:18:51 +08:00
|
|
|
consumer_add(uprobe, uc);
|
|
|
|
ret = register_for_each_vma(uprobe, uc);
|
2012-11-24 03:15:17 +08:00
|
|
|
if (ret)
|
2012-11-24 02:43:50 +08:00
|
|
|
__uprobe_unregister(uprobe, uc);
|
uprobes, mm, x86: Add the ability to install and remove uprobes breakpoints
Add uprobes support to the core kernel, with x86 support.
This commit adds the kernel facilities, the actual uprobes
user-space ABI and perf probe support comes in later commits.
General design:
Uprobes are maintained in an rb-tree indexed by inode and offset
(the offset here is from the start of the mapping). For a unique
(inode, offset) tuple, there can be at most one uprobe in the
rb-tree.
Since the (inode, offset) tuple identifies a unique uprobe, more
than one user may be interested in the same uprobe. This provides
the ability to connect multiple 'consumers' to the same uprobe.
Each consumer defines a handler and a filter (optional). The
'handler' is run every time the uprobe is hit, if it matches the
'filter' criteria.
The first consumer of a uprobe causes the breakpoint to be
inserted at the specified address and subsequent consumers are
appended to this list. On subsequent probes, the consumer gets
appended to the existing list of consumers. The breakpoint is
removed when the last consumer unregisters. For all other
unregisterations, the consumer is removed from the list of
consumers.
Given a inode, we get a list of the mms that have mapped the
inode. Do the actual registration if mm maps the page where a
probe needs to be inserted/removed.
We use a temporary list to walk through the vmas that map the
inode.
- The number of maps that map the inode, is not known before we
walk the rmap and keeps changing.
- extending vm_area_struct wasn't recommended, it's a
size-critical data structure.
- There can be more than one maps of the inode in the same mm.
We add callbacks to the mmap methods to keep an eye on text vmas
that are of interest to uprobes. When a vma of interest is mapped,
we insert the breakpoint at the right address.
Uprobe works by replacing the instruction at the address defined
by (inode, offset) with the arch specific breakpoint
instruction. We save a copy of the original instruction at the
uprobed address.
This is needed for:
a. executing the instruction out-of-line (xol).
b. instruction analysis for any subsequent fixups.
c. restoring the instruction back when the uprobe is unregistered.
We insert or delete a breakpoint instruction, and this
breakpoint instruction is assumed to be the smallest instruction
available on the platform. For fixed size instruction platforms
this is trivially true, for variable size instruction platforms
the breakpoint instruction is typically the smallest (often a
single byte).
Writing the instruction is done by COWing the page and changing
the instruction during the copy, this even though most platforms
allow atomic writes of the breakpoint instruction. This also
mirrors the behaviour of a ptrace() memory write to a PRIVATE
file map.
The core worker is derived from KSM's replace_page() logic.
In essence, similar to KSM:
a. allocate a new page and copy over contents of the page that
has the uprobed vaddr
b. modify the copy and insert the breakpoint at the required
address
c. switch the original page with the copy containing the
breakpoint
d. flush page tables.
replace_page() is being replicated here because of some minor
changes in the type of pages and also because Hugh Dickins had
plans to improve replace_page() for KSM specific work.
Instruction analysis on x86 is based on instruction decoder and
determines if an instruction can be probed and determines the
necessary fixups after singlestep. Instruction analysis is done
at probe insertion time so that we avoid having to repeat the
same analysis every time a probe is hit.
A lot of code here is due to the improvement/suggestions/inputs
from Peter Zijlstra.
Changelog:
(v10):
- Add code to clear REX.B prefix as suggested by Denys Vlasenko
and Masami Hiramatsu.
(v9):
- Use insn_offset_modrm as suggested by Masami Hiramatsu.
(v7):
Handle comments from Peter Zijlstra:
- Dont take reference to inode. (expect inode to uprobe_register to be sane).
- Use PTR_ERR to set the return value.
- No need to take reference to inode.
- use PTR_ERR to return error value.
- register and uprobe_unregister share code.
(v5):
- Modified del_consumer as per comments from Peter.
- Drop reference to inode before dropping reference to uprobe.
- Use i_size_read(inode) instead of inode->i_size.
- Ensure uprobe->consumers is NULL, before __uprobe_unregister() is called.
- Includes errno.h as recommended by Stephen Rothwell to fix a build issue
on sparc defconfig
- Remove restrictions while unregistering.
- Earlier code leaked inode references under some conditions while
registering/unregistering.
- Continue the vma-rmap walk even if the intermediate vma doesnt
meet the requirements.
- Validate the vma found by find_vma before inserting/removing the
breakpoint
- Call del_consumer under mutex_lock.
- Use hash locks.
- Handle mremap.
- Introduce find_least_offset_node() instead of close match logic in
find_uprobe
- Uprobes no more depends on MM_OWNER; No reference to task_structs
while inserting/removing a probe.
- Uses read_mapping_page instead of grab_cache_page so that the pages
have valid content.
- pass NULL to get_user_pages for the task parameter.
- call SetPageUptodate on the new page allocated in write_opcode.
- fix leaking a reference to the new page under certain conditions.
- Include Instruction Decoder if Uprobes gets defined.
- Remove const attributes for instruction prefix arrays.
- Uses mm_context to know if the application is 32 bit.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Also-written-by: Jim Keniston <jkenisto@us.ibm.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Roland McGrath <roland@hack.frob.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Anton Arapov <anton@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Denys Vlasenko <vda.linux@googlemail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linux-mm <linux-mm@kvack.org>
Link: http://lkml.kernel.org/r/20120209092642.GE16600@linux.vnet.ibm.com
[ Made various small edits to the commit log ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-02-09 17:26:42 +08:00
|
|
|
}
|
2012-11-26 05:48:37 +08:00
|
|
|
up_write(&uprobe->register_rwsem);
|
|
|
|
put_uprobe(uprobe);
|
uprobes, mm, x86: Add the ability to install and remove uprobes breakpoints
Add uprobes support to the core kernel, with x86 support.
This commit adds the kernel facilities, the actual uprobes
user-space ABI and perf probe support comes in later commits.
General design:
Uprobes are maintained in an rb-tree indexed by inode and offset
(the offset here is from the start of the mapping). For a unique
(inode, offset) tuple, there can be at most one uprobe in the
rb-tree.
Since the (inode, offset) tuple identifies a unique uprobe, more
than one user may be interested in the same uprobe. This provides
the ability to connect multiple 'consumers' to the same uprobe.
Each consumer defines a handler and a filter (optional). The
'handler' is run every time the uprobe is hit, if it matches the
'filter' criteria.
The first consumer of a uprobe causes the breakpoint to be
inserted at the specified address and subsequent consumers are
appended to this list. On subsequent probes, the consumer gets
appended to the existing list of consumers. The breakpoint is
removed when the last consumer unregisters. For all other
unregisterations, the consumer is removed from the list of
consumers.
Given a inode, we get a list of the mms that have mapped the
inode. Do the actual registration if mm maps the page where a
probe needs to be inserted/removed.
We use a temporary list to walk through the vmas that map the
inode.
- The number of maps that map the inode, is not known before we
walk the rmap and keeps changing.
- extending vm_area_struct wasn't recommended, it's a
size-critical data structure.
- There can be more than one maps of the inode in the same mm.
We add callbacks to the mmap methods to keep an eye on text vmas
that are of interest to uprobes. When a vma of interest is mapped,
we insert the breakpoint at the right address.
Uprobe works by replacing the instruction at the address defined
by (inode, offset) with the arch specific breakpoint
instruction. We save a copy of the original instruction at the
uprobed address.
This is needed for:
a. executing the instruction out-of-line (xol).
b. instruction analysis for any subsequent fixups.
c. restoring the instruction back when the uprobe is unregistered.
We insert or delete a breakpoint instruction, and this
breakpoint instruction is assumed to be the smallest instruction
available on the platform. For fixed size instruction platforms
this is trivially true, for variable size instruction platforms
the breakpoint instruction is typically the smallest (often a
single byte).
Writing the instruction is done by COWing the page and changing
the instruction during the copy, this even though most platforms
allow atomic writes of the breakpoint instruction. This also
mirrors the behaviour of a ptrace() memory write to a PRIVATE
file map.
The core worker is derived from KSM's replace_page() logic.
In essence, similar to KSM:
a. allocate a new page and copy over contents of the page that
has the uprobed vaddr
b. modify the copy and insert the breakpoint at the required
address
c. switch the original page with the copy containing the
breakpoint
d. flush page tables.
replace_page() is being replicated here because of some minor
changes in the type of pages and also because Hugh Dickins had
plans to improve replace_page() for KSM specific work.
Instruction analysis on x86 is based on instruction decoder and
determines if an instruction can be probed and determines the
necessary fixups after singlestep. Instruction analysis is done
at probe insertion time so that we avoid having to repeat the
same analysis every time a probe is hit.
A lot of code here is due to the improvement/suggestions/inputs
from Peter Zijlstra.
Changelog:
(v10):
- Add code to clear REX.B prefix as suggested by Denys Vlasenko
and Masami Hiramatsu.
(v9):
- Use insn_offset_modrm as suggested by Masami Hiramatsu.
(v7):
Handle comments from Peter Zijlstra:
- Dont take reference to inode. (expect inode to uprobe_register to be sane).
- Use PTR_ERR to set the return value.
- No need to take reference to inode.
- use PTR_ERR to return error value.
- register and uprobe_unregister share code.
(v5):
- Modified del_consumer as per comments from Peter.
- Drop reference to inode before dropping reference to uprobe.
- Use i_size_read(inode) instead of inode->i_size.
- Ensure uprobe->consumers is NULL, before __uprobe_unregister() is called.
- Includes errno.h as recommended by Stephen Rothwell to fix a build issue
on sparc defconfig
- Remove restrictions while unregistering.
- Earlier code leaked inode references under some conditions while
registering/unregistering.
- Continue the vma-rmap walk even if the intermediate vma doesnt
meet the requirements.
- Validate the vma found by find_vma before inserting/removing the
breakpoint
- Call del_consumer under mutex_lock.
- Use hash locks.
- Handle mremap.
- Introduce find_least_offset_node() instead of close match logic in
find_uprobe
- Uprobes no more depends on MM_OWNER; No reference to task_structs
while inserting/removing a probe.
- Uses read_mapping_page instead of grab_cache_page so that the pages
have valid content.
- pass NULL to get_user_pages for the task parameter.
- call SetPageUptodate on the new page allocated in write_opcode.
- fix leaking a reference to the new page under certain conditions.
- Include Instruction Decoder if Uprobes gets defined.
- Remove const attributes for instruction prefix arrays.
- Uses mm_context to know if the application is 32 bit.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Also-written-by: Jim Keniston <jkenisto@us.ibm.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Roland McGrath <roland@hack.frob.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Anton Arapov <anton@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Denys Vlasenko <vda.linux@googlemail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linux-mm <linux-mm@kvack.org>
Link: http://lkml.kernel.org/r/20120209092642.GE16600@linux.vnet.ibm.com
[ Made various small edits to the commit log ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-02-09 17:26:42 +08:00
|
|
|
|
2012-11-26 05:48:37 +08:00
|
|
|
if (unlikely(ret == -EAGAIN))
|
|
|
|
goto retry;
|
uprobes, mm, x86: Add the ability to install and remove uprobes breakpoints
Add uprobes support to the core kernel, with x86 support.
This commit adds the kernel facilities, the actual uprobes
user-space ABI and perf probe support comes in later commits.
General design:
Uprobes are maintained in an rb-tree indexed by inode and offset
(the offset here is from the start of the mapping). For a unique
(inode, offset) tuple, there can be at most one uprobe in the
rb-tree.
Since the (inode, offset) tuple identifies a unique uprobe, more
than one user may be interested in the same uprobe. This provides
the ability to connect multiple 'consumers' to the same uprobe.
Each consumer defines a handler and a filter (optional). The
'handler' is run every time the uprobe is hit, if it matches the
'filter' criteria.
The first consumer of a uprobe causes the breakpoint to be
inserted at the specified address and subsequent consumers are
appended to this list. On subsequent probes, the consumer gets
appended to the existing list of consumers. The breakpoint is
removed when the last consumer unregisters. For all other
unregisterations, the consumer is removed from the list of
consumers.
Given a inode, we get a list of the mms that have mapped the
inode. Do the actual registration if mm maps the page where a
probe needs to be inserted/removed.
We use a temporary list to walk through the vmas that map the
inode.
- The number of maps that map the inode, is not known before we
walk the rmap and keeps changing.
- extending vm_area_struct wasn't recommended, it's a
size-critical data structure.
- There can be more than one maps of the inode in the same mm.
We add callbacks to the mmap methods to keep an eye on text vmas
that are of interest to uprobes. When a vma of interest is mapped,
we insert the breakpoint at the right address.
Uprobe works by replacing the instruction at the address defined
by (inode, offset) with the arch specific breakpoint
instruction. We save a copy of the original instruction at the
uprobed address.
This is needed for:
a. executing the instruction out-of-line (xol).
b. instruction analysis for any subsequent fixups.
c. restoring the instruction back when the uprobe is unregistered.
We insert or delete a breakpoint instruction, and this
breakpoint instruction is assumed to be the smallest instruction
available on the platform. For fixed size instruction platforms
this is trivially true, for variable size instruction platforms
the breakpoint instruction is typically the smallest (often a
single byte).
Writing the instruction is done by COWing the page and changing
the instruction during the copy, this even though most platforms
allow atomic writes of the breakpoint instruction. This also
mirrors the behaviour of a ptrace() memory write to a PRIVATE
file map.
The core worker is derived from KSM's replace_page() logic.
In essence, similar to KSM:
a. allocate a new page and copy over contents of the page that
has the uprobed vaddr
b. modify the copy and insert the breakpoint at the required
address
c. switch the original page with the copy containing the
breakpoint
d. flush page tables.
replace_page() is being replicated here because of some minor
changes in the type of pages and also because Hugh Dickins had
plans to improve replace_page() for KSM specific work.
Instruction analysis on x86 is based on instruction decoder and
determines if an instruction can be probed and determines the
necessary fixups after singlestep. Instruction analysis is done
at probe insertion time so that we avoid having to repeat the
same analysis every time a probe is hit.
A lot of code here is due to the improvement/suggestions/inputs
from Peter Zijlstra.
Changelog:
(v10):
- Add code to clear REX.B prefix as suggested by Denys Vlasenko
and Masami Hiramatsu.
(v9):
- Use insn_offset_modrm as suggested by Masami Hiramatsu.
(v7):
Handle comments from Peter Zijlstra:
- Dont take reference to inode. (expect inode to uprobe_register to be sane).
- Use PTR_ERR to set the return value.
- No need to take reference to inode.
- use PTR_ERR to return error value.
- register and uprobe_unregister share code.
(v5):
- Modified del_consumer as per comments from Peter.
- Drop reference to inode before dropping reference to uprobe.
- Use i_size_read(inode) instead of inode->i_size.
- Ensure uprobe->consumers is NULL, before __uprobe_unregister() is called.
- Includes errno.h as recommended by Stephen Rothwell to fix a build issue
on sparc defconfig
- Remove restrictions while unregistering.
- Earlier code leaked inode references under some conditions while
registering/unregistering.
- Continue the vma-rmap walk even if the intermediate vma doesnt
meet the requirements.
- Validate the vma found by find_vma before inserting/removing the
breakpoint
- Call del_consumer under mutex_lock.
- Use hash locks.
- Handle mremap.
- Introduce find_least_offset_node() instead of close match logic in
find_uprobe
- Uprobes no more depends on MM_OWNER; No reference to task_structs
while inserting/removing a probe.
- Uses read_mapping_page instead of grab_cache_page so that the pages
have valid content.
- pass NULL to get_user_pages for the task parameter.
- call SetPageUptodate on the new page allocated in write_opcode.
- fix leaking a reference to the new page under certain conditions.
- Include Instruction Decoder if Uprobes gets defined.
- Remove const attributes for instruction prefix arrays.
- Uses mm_context to know if the application is 32 bit.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Also-written-by: Jim Keniston <jkenisto@us.ibm.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Roland McGrath <roland@hack.frob.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Anton Arapov <anton@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Denys Vlasenko <vda.linux@googlemail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linux-mm <linux-mm@kvack.org>
Link: http://lkml.kernel.org/r/20120209092642.GE16600@linux.vnet.ibm.com
[ Made various small edits to the commit log ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-02-09 17:26:42 +08:00
|
|
|
return ret;
|
|
|
|
}
|
2018-08-09 12:18:51 +08:00
|
|
|
|
|
|
|
int uprobe_register(struct inode *inode, loff_t offset,
|
|
|
|
struct uprobe_consumer *uc)
|
|
|
|
{
|
uprobes: Support SDT markers having reference count (semaphore)
Userspace Statically Defined Tracepoints[1] are dtrace style markers
inside userspace applications. Applications like PostgreSQL, MySQL,
Pthread, Perl, Python, Java, Ruby, Node.js, libvirt, QEMU, glib etc
have these markers embedded in them. These markers are added by developer
at important places in the code. Each marker source expands to a single
nop instruction in the compiled code but there may be additional
overhead for computing the marker arguments which expands to couple of
instructions. In case the overhead is more, execution of it can be
omitted by runtime if() condition when no one is tracing on the marker:
if (reference_counter > 0) {
Execute marker instructions;
}
Default value of reference counter is 0. Tracer has to increment the
reference counter before tracing on a marker and decrement it when
done with the tracing.
Implement the reference counter logic in core uprobe. User will be
able to use it from trace_uprobe as well as from kernel module. New
trace_uprobe definition with reference counter will now be:
<path>:<offset>[(ref_ctr_offset)]
where ref_ctr_offset is an optional field. For kernel module, new
variant of uprobe_register() has been introduced:
uprobe_register_refctr(inode, offset, ref_ctr_offset, consumer)
No new variant for uprobe_unregister() because it's assumed to have
only one reference counter for one uprobe.
[1] https://sourceware.org/systemtap/wiki/UserSpaceProbeImplementation
Note: 'reference counter' is called as 'semaphore' in original Dtrace
(or Systemtap, bcc and even in ELF) documentation and code. But the
term 'semaphore' is misleading in this context. This is just a counter
used to hold number of tracers tracing on a marker. This is not really
used for any synchronization. So we are calling it a 'reference counter'
in kernel / perf code.
Link: http://lkml.kernel.org/r/20180820044250.11659-2-ravi.bangoria@linux.ibm.com
Reviewed-by: Masami Hiramatsu <mhiramat@kernel.org>
[Only trace_uprobe.c]
Reviewed-by: Oleg Nesterov <oleg@redhat.com>
Reviewed-by: Song Liu <songliubraving@fb.com>
Tested-by: Song Liu <songliubraving@fb.com>
Signed-off-by: Ravi Bangoria <ravi.bangoria@linux.ibm.com>
Signed-off-by: Steven Rostedt (VMware) <rostedt@goodmis.org>
2018-08-20 12:42:47 +08:00
|
|
|
return __uprobe_register(inode, offset, 0, uc);
|
2018-08-09 12:18:51 +08:00
|
|
|
}
|
2013-01-14 02:03:34 +08:00
|
|
|
EXPORT_SYMBOL_GPL(uprobe_register);
|
uprobes, mm, x86: Add the ability to install and remove uprobes breakpoints
Add uprobes support to the core kernel, with x86 support.
This commit adds the kernel facilities, the actual uprobes
user-space ABI and perf probe support comes in later commits.
General design:
Uprobes are maintained in an rb-tree indexed by inode and offset
(the offset here is from the start of the mapping). For a unique
(inode, offset) tuple, there can be at most one uprobe in the
rb-tree.
Since the (inode, offset) tuple identifies a unique uprobe, more
than one user may be interested in the same uprobe. This provides
the ability to connect multiple 'consumers' to the same uprobe.
Each consumer defines a handler and a filter (optional). The
'handler' is run every time the uprobe is hit, if it matches the
'filter' criteria.
The first consumer of a uprobe causes the breakpoint to be
inserted at the specified address and subsequent consumers are
appended to this list. On subsequent probes, the consumer gets
appended to the existing list of consumers. The breakpoint is
removed when the last consumer unregisters. For all other
unregisterations, the consumer is removed from the list of
consumers.
Given a inode, we get a list of the mms that have mapped the
inode. Do the actual registration if mm maps the page where a
probe needs to be inserted/removed.
We use a temporary list to walk through the vmas that map the
inode.
- The number of maps that map the inode, is not known before we
walk the rmap and keeps changing.
- extending vm_area_struct wasn't recommended, it's a
size-critical data structure.
- There can be more than one maps of the inode in the same mm.
We add callbacks to the mmap methods to keep an eye on text vmas
that are of interest to uprobes. When a vma of interest is mapped,
we insert the breakpoint at the right address.
Uprobe works by replacing the instruction at the address defined
by (inode, offset) with the arch specific breakpoint
instruction. We save a copy of the original instruction at the
uprobed address.
This is needed for:
a. executing the instruction out-of-line (xol).
b. instruction analysis for any subsequent fixups.
c. restoring the instruction back when the uprobe is unregistered.
We insert or delete a breakpoint instruction, and this
breakpoint instruction is assumed to be the smallest instruction
available on the platform. For fixed size instruction platforms
this is trivially true, for variable size instruction platforms
the breakpoint instruction is typically the smallest (often a
single byte).
Writing the instruction is done by COWing the page and changing
the instruction during the copy, this even though most platforms
allow atomic writes of the breakpoint instruction. This also
mirrors the behaviour of a ptrace() memory write to a PRIVATE
file map.
The core worker is derived from KSM's replace_page() logic.
In essence, similar to KSM:
a. allocate a new page and copy over contents of the page that
has the uprobed vaddr
b. modify the copy and insert the breakpoint at the required
address
c. switch the original page with the copy containing the
breakpoint
d. flush page tables.
replace_page() is being replicated here because of some minor
changes in the type of pages and also because Hugh Dickins had
plans to improve replace_page() for KSM specific work.
Instruction analysis on x86 is based on instruction decoder and
determines if an instruction can be probed and determines the
necessary fixups after singlestep. Instruction analysis is done
at probe insertion time so that we avoid having to repeat the
same analysis every time a probe is hit.
A lot of code here is due to the improvement/suggestions/inputs
from Peter Zijlstra.
Changelog:
(v10):
- Add code to clear REX.B prefix as suggested by Denys Vlasenko
and Masami Hiramatsu.
(v9):
- Use insn_offset_modrm as suggested by Masami Hiramatsu.
(v7):
Handle comments from Peter Zijlstra:
- Dont take reference to inode. (expect inode to uprobe_register to be sane).
- Use PTR_ERR to set the return value.
- No need to take reference to inode.
- use PTR_ERR to return error value.
- register and uprobe_unregister share code.
(v5):
- Modified del_consumer as per comments from Peter.
- Drop reference to inode before dropping reference to uprobe.
- Use i_size_read(inode) instead of inode->i_size.
- Ensure uprobe->consumers is NULL, before __uprobe_unregister() is called.
- Includes errno.h as recommended by Stephen Rothwell to fix a build issue
on sparc defconfig
- Remove restrictions while unregistering.
- Earlier code leaked inode references under some conditions while
registering/unregistering.
- Continue the vma-rmap walk even if the intermediate vma doesnt
meet the requirements.
- Validate the vma found by find_vma before inserting/removing the
breakpoint
- Call del_consumer under mutex_lock.
- Use hash locks.
- Handle mremap.
- Introduce find_least_offset_node() instead of close match logic in
find_uprobe
- Uprobes no more depends on MM_OWNER; No reference to task_structs
while inserting/removing a probe.
- Uses read_mapping_page instead of grab_cache_page so that the pages
have valid content.
- pass NULL to get_user_pages for the task parameter.
- call SetPageUptodate on the new page allocated in write_opcode.
- fix leaking a reference to the new page under certain conditions.
- Include Instruction Decoder if Uprobes gets defined.
- Remove const attributes for instruction prefix arrays.
- Uses mm_context to know if the application is 32 bit.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Also-written-by: Jim Keniston <jkenisto@us.ibm.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Roland McGrath <roland@hack.frob.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Anton Arapov <anton@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Denys Vlasenko <vda.linux@googlemail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linux-mm <linux-mm@kvack.org>
Link: http://lkml.kernel.org/r/20120209092642.GE16600@linux.vnet.ibm.com
[ Made various small edits to the commit log ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-02-09 17:26:42 +08:00
|
|
|
|
uprobes: Support SDT markers having reference count (semaphore)
Userspace Statically Defined Tracepoints[1] are dtrace style markers
inside userspace applications. Applications like PostgreSQL, MySQL,
Pthread, Perl, Python, Java, Ruby, Node.js, libvirt, QEMU, glib etc
have these markers embedded in them. These markers are added by developer
at important places in the code. Each marker source expands to a single
nop instruction in the compiled code but there may be additional
overhead for computing the marker arguments which expands to couple of
instructions. In case the overhead is more, execution of it can be
omitted by runtime if() condition when no one is tracing on the marker:
if (reference_counter > 0) {
Execute marker instructions;
}
Default value of reference counter is 0. Tracer has to increment the
reference counter before tracing on a marker and decrement it when
done with the tracing.
Implement the reference counter logic in core uprobe. User will be
able to use it from trace_uprobe as well as from kernel module. New
trace_uprobe definition with reference counter will now be:
<path>:<offset>[(ref_ctr_offset)]
where ref_ctr_offset is an optional field. For kernel module, new
variant of uprobe_register() has been introduced:
uprobe_register_refctr(inode, offset, ref_ctr_offset, consumer)
No new variant for uprobe_unregister() because it's assumed to have
only one reference counter for one uprobe.
[1] https://sourceware.org/systemtap/wiki/UserSpaceProbeImplementation
Note: 'reference counter' is called as 'semaphore' in original Dtrace
(or Systemtap, bcc and even in ELF) documentation and code. But the
term 'semaphore' is misleading in this context. This is just a counter
used to hold number of tracers tracing on a marker. This is not really
used for any synchronization. So we are calling it a 'reference counter'
in kernel / perf code.
Link: http://lkml.kernel.org/r/20180820044250.11659-2-ravi.bangoria@linux.ibm.com
Reviewed-by: Masami Hiramatsu <mhiramat@kernel.org>
[Only trace_uprobe.c]
Reviewed-by: Oleg Nesterov <oleg@redhat.com>
Reviewed-by: Song Liu <songliubraving@fb.com>
Tested-by: Song Liu <songliubraving@fb.com>
Signed-off-by: Ravi Bangoria <ravi.bangoria@linux.ibm.com>
Signed-off-by: Steven Rostedt (VMware) <rostedt@goodmis.org>
2018-08-20 12:42:47 +08:00
|
|
|
int uprobe_register_refctr(struct inode *inode, loff_t offset,
|
|
|
|
loff_t ref_ctr_offset, struct uprobe_consumer *uc)
|
|
|
|
{
|
|
|
|
return __uprobe_register(inode, offset, ref_ctr_offset, uc);
|
|
|
|
}
|
|
|
|
EXPORT_SYMBOL_GPL(uprobe_register_refctr);
|
|
|
|
|
uprobes: Introduce uprobe_apply()
Currently it is not possible to change the filtering constraints after
uprobe_register(), so a consumer can not, say, start to trace a task/mm
which was previously filtered out, or remove the no longer needed bp's.
Introduce uprobe_apply() which simply does register_for_each_vma() again
to consult uprobe_consumer->filter() and install/remove the breakpoints.
The only complication is that register_for_each_vma() can no longer
assume that uprobe->consumers should be consulter if is_register == T,
so we change it to accept "struct uprobe_consumer *new" instead.
Unlike uprobe_register(), uprobe_apply(true) doesn't do "unregister" if
register_for_each_vma() fails, it is up to caller to handle the error.
Note: we probably need to cleanup the current interface, it is strange
that uprobe_apply/unregister need inode/offset. We should either change
uprobe_register() to return "struct uprobe *", or add a private ->uprobe
member in uprobe_consumer. And in the long term uprobe_apply() should
take a single argument, uprobe or consumer, even "bool add" should go
away.
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
2013-02-04 02:21:12 +08:00
|
|
|
/*
|
2018-07-09 18:57:15 +08:00
|
|
|
* uprobe_apply - unregister an already registered probe.
|
uprobes: Introduce uprobe_apply()
Currently it is not possible to change the filtering constraints after
uprobe_register(), so a consumer can not, say, start to trace a task/mm
which was previously filtered out, or remove the no longer needed bp's.
Introduce uprobe_apply() which simply does register_for_each_vma() again
to consult uprobe_consumer->filter() and install/remove the breakpoints.
The only complication is that register_for_each_vma() can no longer
assume that uprobe->consumers should be consulter if is_register == T,
so we change it to accept "struct uprobe_consumer *new" instead.
Unlike uprobe_register(), uprobe_apply(true) doesn't do "unregister" if
register_for_each_vma() fails, it is up to caller to handle the error.
Note: we probably need to cleanup the current interface, it is strange
that uprobe_apply/unregister need inode/offset. We should either change
uprobe_register() to return "struct uprobe *", or add a private ->uprobe
member in uprobe_consumer. And in the long term uprobe_apply() should
take a single argument, uprobe or consumer, even "bool add" should go
away.
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
2013-02-04 02:21:12 +08:00
|
|
|
* @inode: the file in which the probe has to be removed.
|
|
|
|
* @offset: offset from the start of the file.
|
|
|
|
* @uc: consumer which wants to add more or remove some breakpoints
|
|
|
|
* @add: add or remove the breakpoints
|
|
|
|
*/
|
|
|
|
int uprobe_apply(struct inode *inode, loff_t offset,
|
|
|
|
struct uprobe_consumer *uc, bool add)
|
|
|
|
{
|
|
|
|
struct uprobe *uprobe;
|
|
|
|
struct uprobe_consumer *con;
|
|
|
|
int ret = -ENOENT;
|
|
|
|
|
|
|
|
uprobe = find_uprobe(inode, offset);
|
2014-06-28 01:01:40 +08:00
|
|
|
if (WARN_ON(!uprobe))
|
uprobes: Introduce uprobe_apply()
Currently it is not possible to change the filtering constraints after
uprobe_register(), so a consumer can not, say, start to trace a task/mm
which was previously filtered out, or remove the no longer needed bp's.
Introduce uprobe_apply() which simply does register_for_each_vma() again
to consult uprobe_consumer->filter() and install/remove the breakpoints.
The only complication is that register_for_each_vma() can no longer
assume that uprobe->consumers should be consulter if is_register == T,
so we change it to accept "struct uprobe_consumer *new" instead.
Unlike uprobe_register(), uprobe_apply(true) doesn't do "unregister" if
register_for_each_vma() fails, it is up to caller to handle the error.
Note: we probably need to cleanup the current interface, it is strange
that uprobe_apply/unregister need inode/offset. We should either change
uprobe_register() to return "struct uprobe *", or add a private ->uprobe
member in uprobe_consumer. And in the long term uprobe_apply() should
take a single argument, uprobe or consumer, even "bool add" should go
away.
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
2013-02-04 02:21:12 +08:00
|
|
|
return ret;
|
|
|
|
|
|
|
|
down_write(&uprobe->register_rwsem);
|
|
|
|
for (con = uprobe->consumers; con && con != uc ; con = con->next)
|
|
|
|
;
|
|
|
|
if (con)
|
|
|
|
ret = register_for_each_vma(uprobe, add ? uc : NULL);
|
|
|
|
up_write(&uprobe->register_rwsem);
|
|
|
|
put_uprobe(uprobe);
|
|
|
|
|
|
|
|
return ret;
|
|
|
|
}
|
|
|
|
|
2012-12-30 00:49:11 +08:00
|
|
|
static int unapply_uprobe(struct uprobe *uprobe, struct mm_struct *mm)
|
|
|
|
{
|
|
|
|
struct vm_area_struct *vma;
|
|
|
|
int err = 0;
|
|
|
|
|
|
|
|
down_read(&mm->mmap_sem);
|
|
|
|
for (vma = mm->mmap; vma; vma = vma->vm_next) {
|
|
|
|
unsigned long vaddr;
|
|
|
|
loff_t offset;
|
|
|
|
|
|
|
|
if (!valid_vma(vma, false) ||
|
2013-03-18 01:54:44 +08:00
|
|
|
file_inode(vma->vm_file) != uprobe->inode)
|
2012-12-30 00:49:11 +08:00
|
|
|
continue;
|
|
|
|
|
|
|
|
offset = (loff_t)vma->vm_pgoff << PAGE_SHIFT;
|
|
|
|
if (uprobe->offset < offset ||
|
|
|
|
uprobe->offset >= offset + vma->vm_end - vma->vm_start)
|
|
|
|
continue;
|
|
|
|
|
|
|
|
vaddr = offset_to_vaddr(vma, uprobe->offset);
|
|
|
|
err |= remove_breakpoint(uprobe, mm, vaddr);
|
|
|
|
}
|
|
|
|
up_read(&mm->mmap_sem);
|
|
|
|
|
|
|
|
return err;
|
|
|
|
}
|
|
|
|
|
2012-07-30 02:22:40 +08:00
|
|
|
static struct rb_node *
|
|
|
|
find_node_in_range(struct inode *inode, loff_t min, loff_t max)
|
uprobes, mm, x86: Add the ability to install and remove uprobes breakpoints
Add uprobes support to the core kernel, with x86 support.
This commit adds the kernel facilities, the actual uprobes
user-space ABI and perf probe support comes in later commits.
General design:
Uprobes are maintained in an rb-tree indexed by inode and offset
(the offset here is from the start of the mapping). For a unique
(inode, offset) tuple, there can be at most one uprobe in the
rb-tree.
Since the (inode, offset) tuple identifies a unique uprobe, more
than one user may be interested in the same uprobe. This provides
the ability to connect multiple 'consumers' to the same uprobe.
Each consumer defines a handler and a filter (optional). The
'handler' is run every time the uprobe is hit, if it matches the
'filter' criteria.
The first consumer of a uprobe causes the breakpoint to be
inserted at the specified address and subsequent consumers are
appended to this list. On subsequent probes, the consumer gets
appended to the existing list of consumers. The breakpoint is
removed when the last consumer unregisters. For all other
unregisterations, the consumer is removed from the list of
consumers.
Given a inode, we get a list of the mms that have mapped the
inode. Do the actual registration if mm maps the page where a
probe needs to be inserted/removed.
We use a temporary list to walk through the vmas that map the
inode.
- The number of maps that map the inode, is not known before we
walk the rmap and keeps changing.
- extending vm_area_struct wasn't recommended, it's a
size-critical data structure.
- There can be more than one maps of the inode in the same mm.
We add callbacks to the mmap methods to keep an eye on text vmas
that are of interest to uprobes. When a vma of interest is mapped,
we insert the breakpoint at the right address.
Uprobe works by replacing the instruction at the address defined
by (inode, offset) with the arch specific breakpoint
instruction. We save a copy of the original instruction at the
uprobed address.
This is needed for:
a. executing the instruction out-of-line (xol).
b. instruction analysis for any subsequent fixups.
c. restoring the instruction back when the uprobe is unregistered.
We insert or delete a breakpoint instruction, and this
breakpoint instruction is assumed to be the smallest instruction
available on the platform. For fixed size instruction platforms
this is trivially true, for variable size instruction platforms
the breakpoint instruction is typically the smallest (often a
single byte).
Writing the instruction is done by COWing the page and changing
the instruction during the copy, this even though most platforms
allow atomic writes of the breakpoint instruction. This also
mirrors the behaviour of a ptrace() memory write to a PRIVATE
file map.
The core worker is derived from KSM's replace_page() logic.
In essence, similar to KSM:
a. allocate a new page and copy over contents of the page that
has the uprobed vaddr
b. modify the copy and insert the breakpoint at the required
address
c. switch the original page with the copy containing the
breakpoint
d. flush page tables.
replace_page() is being replicated here because of some minor
changes in the type of pages and also because Hugh Dickins had
plans to improve replace_page() for KSM specific work.
Instruction analysis on x86 is based on instruction decoder and
determines if an instruction can be probed and determines the
necessary fixups after singlestep. Instruction analysis is done
at probe insertion time so that we avoid having to repeat the
same analysis every time a probe is hit.
A lot of code here is due to the improvement/suggestions/inputs
from Peter Zijlstra.
Changelog:
(v10):
- Add code to clear REX.B prefix as suggested by Denys Vlasenko
and Masami Hiramatsu.
(v9):
- Use insn_offset_modrm as suggested by Masami Hiramatsu.
(v7):
Handle comments from Peter Zijlstra:
- Dont take reference to inode. (expect inode to uprobe_register to be sane).
- Use PTR_ERR to set the return value.
- No need to take reference to inode.
- use PTR_ERR to return error value.
- register and uprobe_unregister share code.
(v5):
- Modified del_consumer as per comments from Peter.
- Drop reference to inode before dropping reference to uprobe.
- Use i_size_read(inode) instead of inode->i_size.
- Ensure uprobe->consumers is NULL, before __uprobe_unregister() is called.
- Includes errno.h as recommended by Stephen Rothwell to fix a build issue
on sparc defconfig
- Remove restrictions while unregistering.
- Earlier code leaked inode references under some conditions while
registering/unregistering.
- Continue the vma-rmap walk even if the intermediate vma doesnt
meet the requirements.
- Validate the vma found by find_vma before inserting/removing the
breakpoint
- Call del_consumer under mutex_lock.
- Use hash locks.
- Handle mremap.
- Introduce find_least_offset_node() instead of close match logic in
find_uprobe
- Uprobes no more depends on MM_OWNER; No reference to task_structs
while inserting/removing a probe.
- Uses read_mapping_page instead of grab_cache_page so that the pages
have valid content.
- pass NULL to get_user_pages for the task parameter.
- call SetPageUptodate on the new page allocated in write_opcode.
- fix leaking a reference to the new page under certain conditions.
- Include Instruction Decoder if Uprobes gets defined.
- Remove const attributes for instruction prefix arrays.
- Uses mm_context to know if the application is 32 bit.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Also-written-by: Jim Keniston <jkenisto@us.ibm.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Roland McGrath <roland@hack.frob.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Anton Arapov <anton@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Denys Vlasenko <vda.linux@googlemail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linux-mm <linux-mm@kvack.org>
Link: http://lkml.kernel.org/r/20120209092642.GE16600@linux.vnet.ibm.com
[ Made various small edits to the commit log ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-02-09 17:26:42 +08:00
|
|
|
{
|
|
|
|
struct rb_node *n = uprobes_tree.rb_node;
|
|
|
|
|
|
|
|
while (n) {
|
2012-07-30 02:22:40 +08:00
|
|
|
struct uprobe *u = rb_entry(n, struct uprobe, rb_node);
|
uprobes, mm, x86: Add the ability to install and remove uprobes breakpoints
Add uprobes support to the core kernel, with x86 support.
This commit adds the kernel facilities, the actual uprobes
user-space ABI and perf probe support comes in later commits.
General design:
Uprobes are maintained in an rb-tree indexed by inode and offset
(the offset here is from the start of the mapping). For a unique
(inode, offset) tuple, there can be at most one uprobe in the
rb-tree.
Since the (inode, offset) tuple identifies a unique uprobe, more
than one user may be interested in the same uprobe. This provides
the ability to connect multiple 'consumers' to the same uprobe.
Each consumer defines a handler and a filter (optional). The
'handler' is run every time the uprobe is hit, if it matches the
'filter' criteria.
The first consumer of a uprobe causes the breakpoint to be
inserted at the specified address and subsequent consumers are
appended to this list. On subsequent probes, the consumer gets
appended to the existing list of consumers. The breakpoint is
removed when the last consumer unregisters. For all other
unregisterations, the consumer is removed from the list of
consumers.
Given a inode, we get a list of the mms that have mapped the
inode. Do the actual registration if mm maps the page where a
probe needs to be inserted/removed.
We use a temporary list to walk through the vmas that map the
inode.
- The number of maps that map the inode, is not known before we
walk the rmap and keeps changing.
- extending vm_area_struct wasn't recommended, it's a
size-critical data structure.
- There can be more than one maps of the inode in the same mm.
We add callbacks to the mmap methods to keep an eye on text vmas
that are of interest to uprobes. When a vma of interest is mapped,
we insert the breakpoint at the right address.
Uprobe works by replacing the instruction at the address defined
by (inode, offset) with the arch specific breakpoint
instruction. We save a copy of the original instruction at the
uprobed address.
This is needed for:
a. executing the instruction out-of-line (xol).
b. instruction analysis for any subsequent fixups.
c. restoring the instruction back when the uprobe is unregistered.
We insert or delete a breakpoint instruction, and this
breakpoint instruction is assumed to be the smallest instruction
available on the platform. For fixed size instruction platforms
this is trivially true, for variable size instruction platforms
the breakpoint instruction is typically the smallest (often a
single byte).
Writing the instruction is done by COWing the page and changing
the instruction during the copy, this even though most platforms
allow atomic writes of the breakpoint instruction. This also
mirrors the behaviour of a ptrace() memory write to a PRIVATE
file map.
The core worker is derived from KSM's replace_page() logic.
In essence, similar to KSM:
a. allocate a new page and copy over contents of the page that
has the uprobed vaddr
b. modify the copy and insert the breakpoint at the required
address
c. switch the original page with the copy containing the
breakpoint
d. flush page tables.
replace_page() is being replicated here because of some minor
changes in the type of pages and also because Hugh Dickins had
plans to improve replace_page() for KSM specific work.
Instruction analysis on x86 is based on instruction decoder and
determines if an instruction can be probed and determines the
necessary fixups after singlestep. Instruction analysis is done
at probe insertion time so that we avoid having to repeat the
same analysis every time a probe is hit.
A lot of code here is due to the improvement/suggestions/inputs
from Peter Zijlstra.
Changelog:
(v10):
- Add code to clear REX.B prefix as suggested by Denys Vlasenko
and Masami Hiramatsu.
(v9):
- Use insn_offset_modrm as suggested by Masami Hiramatsu.
(v7):
Handle comments from Peter Zijlstra:
- Dont take reference to inode. (expect inode to uprobe_register to be sane).
- Use PTR_ERR to set the return value.
- No need to take reference to inode.
- use PTR_ERR to return error value.
- register and uprobe_unregister share code.
(v5):
- Modified del_consumer as per comments from Peter.
- Drop reference to inode before dropping reference to uprobe.
- Use i_size_read(inode) instead of inode->i_size.
- Ensure uprobe->consumers is NULL, before __uprobe_unregister() is called.
- Includes errno.h as recommended by Stephen Rothwell to fix a build issue
on sparc defconfig
- Remove restrictions while unregistering.
- Earlier code leaked inode references under some conditions while
registering/unregistering.
- Continue the vma-rmap walk even if the intermediate vma doesnt
meet the requirements.
- Validate the vma found by find_vma before inserting/removing the
breakpoint
- Call del_consumer under mutex_lock.
- Use hash locks.
- Handle mremap.
- Introduce find_least_offset_node() instead of close match logic in
find_uprobe
- Uprobes no more depends on MM_OWNER; No reference to task_structs
while inserting/removing a probe.
- Uses read_mapping_page instead of grab_cache_page so that the pages
have valid content.
- pass NULL to get_user_pages for the task parameter.
- call SetPageUptodate on the new page allocated in write_opcode.
- fix leaking a reference to the new page under certain conditions.
- Include Instruction Decoder if Uprobes gets defined.
- Remove const attributes for instruction prefix arrays.
- Uses mm_context to know if the application is 32 bit.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Also-written-by: Jim Keniston <jkenisto@us.ibm.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Roland McGrath <roland@hack.frob.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Anton Arapov <anton@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Denys Vlasenko <vda.linux@googlemail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linux-mm <linux-mm@kvack.org>
Link: http://lkml.kernel.org/r/20120209092642.GE16600@linux.vnet.ibm.com
[ Made various small edits to the commit log ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-02-09 17:26:42 +08:00
|
|
|
|
2012-07-30 02:22:40 +08:00
|
|
|
if (inode < u->inode) {
|
uprobes, mm, x86: Add the ability to install and remove uprobes breakpoints
Add uprobes support to the core kernel, with x86 support.
This commit adds the kernel facilities, the actual uprobes
user-space ABI and perf probe support comes in later commits.
General design:
Uprobes are maintained in an rb-tree indexed by inode and offset
(the offset here is from the start of the mapping). For a unique
(inode, offset) tuple, there can be at most one uprobe in the
rb-tree.
Since the (inode, offset) tuple identifies a unique uprobe, more
than one user may be interested in the same uprobe. This provides
the ability to connect multiple 'consumers' to the same uprobe.
Each consumer defines a handler and a filter (optional). The
'handler' is run every time the uprobe is hit, if it matches the
'filter' criteria.
The first consumer of a uprobe causes the breakpoint to be
inserted at the specified address and subsequent consumers are
appended to this list. On subsequent probes, the consumer gets
appended to the existing list of consumers. The breakpoint is
removed when the last consumer unregisters. For all other
unregisterations, the consumer is removed from the list of
consumers.
Given a inode, we get a list of the mms that have mapped the
inode. Do the actual registration if mm maps the page where a
probe needs to be inserted/removed.
We use a temporary list to walk through the vmas that map the
inode.
- The number of maps that map the inode, is not known before we
walk the rmap and keeps changing.
- extending vm_area_struct wasn't recommended, it's a
size-critical data structure.
- There can be more than one maps of the inode in the same mm.
We add callbacks to the mmap methods to keep an eye on text vmas
that are of interest to uprobes. When a vma of interest is mapped,
we insert the breakpoint at the right address.
Uprobe works by replacing the instruction at the address defined
by (inode, offset) with the arch specific breakpoint
instruction. We save a copy of the original instruction at the
uprobed address.
This is needed for:
a. executing the instruction out-of-line (xol).
b. instruction analysis for any subsequent fixups.
c. restoring the instruction back when the uprobe is unregistered.
We insert or delete a breakpoint instruction, and this
breakpoint instruction is assumed to be the smallest instruction
available on the platform. For fixed size instruction platforms
this is trivially true, for variable size instruction platforms
the breakpoint instruction is typically the smallest (often a
single byte).
Writing the instruction is done by COWing the page and changing
the instruction during the copy, this even though most platforms
allow atomic writes of the breakpoint instruction. This also
mirrors the behaviour of a ptrace() memory write to a PRIVATE
file map.
The core worker is derived from KSM's replace_page() logic.
In essence, similar to KSM:
a. allocate a new page and copy over contents of the page that
has the uprobed vaddr
b. modify the copy and insert the breakpoint at the required
address
c. switch the original page with the copy containing the
breakpoint
d. flush page tables.
replace_page() is being replicated here because of some minor
changes in the type of pages and also because Hugh Dickins had
plans to improve replace_page() for KSM specific work.
Instruction analysis on x86 is based on instruction decoder and
determines if an instruction can be probed and determines the
necessary fixups after singlestep. Instruction analysis is done
at probe insertion time so that we avoid having to repeat the
same analysis every time a probe is hit.
A lot of code here is due to the improvement/suggestions/inputs
from Peter Zijlstra.
Changelog:
(v10):
- Add code to clear REX.B prefix as suggested by Denys Vlasenko
and Masami Hiramatsu.
(v9):
- Use insn_offset_modrm as suggested by Masami Hiramatsu.
(v7):
Handle comments from Peter Zijlstra:
- Dont take reference to inode. (expect inode to uprobe_register to be sane).
- Use PTR_ERR to set the return value.
- No need to take reference to inode.
- use PTR_ERR to return error value.
- register and uprobe_unregister share code.
(v5):
- Modified del_consumer as per comments from Peter.
- Drop reference to inode before dropping reference to uprobe.
- Use i_size_read(inode) instead of inode->i_size.
- Ensure uprobe->consumers is NULL, before __uprobe_unregister() is called.
- Includes errno.h as recommended by Stephen Rothwell to fix a build issue
on sparc defconfig
- Remove restrictions while unregistering.
- Earlier code leaked inode references under some conditions while
registering/unregistering.
- Continue the vma-rmap walk even if the intermediate vma doesnt
meet the requirements.
- Validate the vma found by find_vma before inserting/removing the
breakpoint
- Call del_consumer under mutex_lock.
- Use hash locks.
- Handle mremap.
- Introduce find_least_offset_node() instead of close match logic in
find_uprobe
- Uprobes no more depends on MM_OWNER; No reference to task_structs
while inserting/removing a probe.
- Uses read_mapping_page instead of grab_cache_page so that the pages
have valid content.
- pass NULL to get_user_pages for the task parameter.
- call SetPageUptodate on the new page allocated in write_opcode.
- fix leaking a reference to the new page under certain conditions.
- Include Instruction Decoder if Uprobes gets defined.
- Remove const attributes for instruction prefix arrays.
- Uses mm_context to know if the application is 32 bit.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Also-written-by: Jim Keniston <jkenisto@us.ibm.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Roland McGrath <roland@hack.frob.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Anton Arapov <anton@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Denys Vlasenko <vda.linux@googlemail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linux-mm <linux-mm@kvack.org>
Link: http://lkml.kernel.org/r/20120209092642.GE16600@linux.vnet.ibm.com
[ Made various small edits to the commit log ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-02-09 17:26:42 +08:00
|
|
|
n = n->rb_left;
|
2012-07-30 02:22:40 +08:00
|
|
|
} else if (inode > u->inode) {
|
uprobes, mm, x86: Add the ability to install and remove uprobes breakpoints
Add uprobes support to the core kernel, with x86 support.
This commit adds the kernel facilities, the actual uprobes
user-space ABI and perf probe support comes in later commits.
General design:
Uprobes are maintained in an rb-tree indexed by inode and offset
(the offset here is from the start of the mapping). For a unique
(inode, offset) tuple, there can be at most one uprobe in the
rb-tree.
Since the (inode, offset) tuple identifies a unique uprobe, more
than one user may be interested in the same uprobe. This provides
the ability to connect multiple 'consumers' to the same uprobe.
Each consumer defines a handler and a filter (optional). The
'handler' is run every time the uprobe is hit, if it matches the
'filter' criteria.
The first consumer of a uprobe causes the breakpoint to be
inserted at the specified address and subsequent consumers are
appended to this list. On subsequent probes, the consumer gets
appended to the existing list of consumers. The breakpoint is
removed when the last consumer unregisters. For all other
unregisterations, the consumer is removed from the list of
consumers.
Given a inode, we get a list of the mms that have mapped the
inode. Do the actual registration if mm maps the page where a
probe needs to be inserted/removed.
We use a temporary list to walk through the vmas that map the
inode.
- The number of maps that map the inode, is not known before we
walk the rmap and keeps changing.
- extending vm_area_struct wasn't recommended, it's a
size-critical data structure.
- There can be more than one maps of the inode in the same mm.
We add callbacks to the mmap methods to keep an eye on text vmas
that are of interest to uprobes. When a vma of interest is mapped,
we insert the breakpoint at the right address.
Uprobe works by replacing the instruction at the address defined
by (inode, offset) with the arch specific breakpoint
instruction. We save a copy of the original instruction at the
uprobed address.
This is needed for:
a. executing the instruction out-of-line (xol).
b. instruction analysis for any subsequent fixups.
c. restoring the instruction back when the uprobe is unregistered.
We insert or delete a breakpoint instruction, and this
breakpoint instruction is assumed to be the smallest instruction
available on the platform. For fixed size instruction platforms
this is trivially true, for variable size instruction platforms
the breakpoint instruction is typically the smallest (often a
single byte).
Writing the instruction is done by COWing the page and changing
the instruction during the copy, this even though most platforms
allow atomic writes of the breakpoint instruction. This also
mirrors the behaviour of a ptrace() memory write to a PRIVATE
file map.
The core worker is derived from KSM's replace_page() logic.
In essence, similar to KSM:
a. allocate a new page and copy over contents of the page that
has the uprobed vaddr
b. modify the copy and insert the breakpoint at the required
address
c. switch the original page with the copy containing the
breakpoint
d. flush page tables.
replace_page() is being replicated here because of some minor
changes in the type of pages and also because Hugh Dickins had
plans to improve replace_page() for KSM specific work.
Instruction analysis on x86 is based on instruction decoder and
determines if an instruction can be probed and determines the
necessary fixups after singlestep. Instruction analysis is done
at probe insertion time so that we avoid having to repeat the
same analysis every time a probe is hit.
A lot of code here is due to the improvement/suggestions/inputs
from Peter Zijlstra.
Changelog:
(v10):
- Add code to clear REX.B prefix as suggested by Denys Vlasenko
and Masami Hiramatsu.
(v9):
- Use insn_offset_modrm as suggested by Masami Hiramatsu.
(v7):
Handle comments from Peter Zijlstra:
- Dont take reference to inode. (expect inode to uprobe_register to be sane).
- Use PTR_ERR to set the return value.
- No need to take reference to inode.
- use PTR_ERR to return error value.
- register and uprobe_unregister share code.
(v5):
- Modified del_consumer as per comments from Peter.
- Drop reference to inode before dropping reference to uprobe.
- Use i_size_read(inode) instead of inode->i_size.
- Ensure uprobe->consumers is NULL, before __uprobe_unregister() is called.
- Includes errno.h as recommended by Stephen Rothwell to fix a build issue
on sparc defconfig
- Remove restrictions while unregistering.
- Earlier code leaked inode references under some conditions while
registering/unregistering.
- Continue the vma-rmap walk even if the intermediate vma doesnt
meet the requirements.
- Validate the vma found by find_vma before inserting/removing the
breakpoint
- Call del_consumer under mutex_lock.
- Use hash locks.
- Handle mremap.
- Introduce find_least_offset_node() instead of close match logic in
find_uprobe
- Uprobes no more depends on MM_OWNER; No reference to task_structs
while inserting/removing a probe.
- Uses read_mapping_page instead of grab_cache_page so that the pages
have valid content.
- pass NULL to get_user_pages for the task parameter.
- call SetPageUptodate on the new page allocated in write_opcode.
- fix leaking a reference to the new page under certain conditions.
- Include Instruction Decoder if Uprobes gets defined.
- Remove const attributes for instruction prefix arrays.
- Uses mm_context to know if the application is 32 bit.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Also-written-by: Jim Keniston <jkenisto@us.ibm.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Roland McGrath <roland@hack.frob.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Anton Arapov <anton@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Denys Vlasenko <vda.linux@googlemail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linux-mm <linux-mm@kvack.org>
Link: http://lkml.kernel.org/r/20120209092642.GE16600@linux.vnet.ibm.com
[ Made various small edits to the commit log ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-02-09 17:26:42 +08:00
|
|
|
n = n->rb_right;
|
2012-07-30 02:22:40 +08:00
|
|
|
} else {
|
|
|
|
if (max < u->offset)
|
|
|
|
n = n->rb_left;
|
|
|
|
else if (min > u->offset)
|
|
|
|
n = n->rb_right;
|
|
|
|
else
|
|
|
|
break;
|
|
|
|
}
|
uprobes, mm, x86: Add the ability to install and remove uprobes breakpoints
Add uprobes support to the core kernel, with x86 support.
This commit adds the kernel facilities, the actual uprobes
user-space ABI and perf probe support comes in later commits.
General design:
Uprobes are maintained in an rb-tree indexed by inode and offset
(the offset here is from the start of the mapping). For a unique
(inode, offset) tuple, there can be at most one uprobe in the
rb-tree.
Since the (inode, offset) tuple identifies a unique uprobe, more
than one user may be interested in the same uprobe. This provides
the ability to connect multiple 'consumers' to the same uprobe.
Each consumer defines a handler and a filter (optional). The
'handler' is run every time the uprobe is hit, if it matches the
'filter' criteria.
The first consumer of a uprobe causes the breakpoint to be
inserted at the specified address and subsequent consumers are
appended to this list. On subsequent probes, the consumer gets
appended to the existing list of consumers. The breakpoint is
removed when the last consumer unregisters. For all other
unregisterations, the consumer is removed from the list of
consumers.
Given a inode, we get a list of the mms that have mapped the
inode. Do the actual registration if mm maps the page where a
probe needs to be inserted/removed.
We use a temporary list to walk through the vmas that map the
inode.
- The number of maps that map the inode, is not known before we
walk the rmap and keeps changing.
- extending vm_area_struct wasn't recommended, it's a
size-critical data structure.
- There can be more than one maps of the inode in the same mm.
We add callbacks to the mmap methods to keep an eye on text vmas
that are of interest to uprobes. When a vma of interest is mapped,
we insert the breakpoint at the right address.
Uprobe works by replacing the instruction at the address defined
by (inode, offset) with the arch specific breakpoint
instruction. We save a copy of the original instruction at the
uprobed address.
This is needed for:
a. executing the instruction out-of-line (xol).
b. instruction analysis for any subsequent fixups.
c. restoring the instruction back when the uprobe is unregistered.
We insert or delete a breakpoint instruction, and this
breakpoint instruction is assumed to be the smallest instruction
available on the platform. For fixed size instruction platforms
this is trivially true, for variable size instruction platforms
the breakpoint instruction is typically the smallest (often a
single byte).
Writing the instruction is done by COWing the page and changing
the instruction during the copy, this even though most platforms
allow atomic writes of the breakpoint instruction. This also
mirrors the behaviour of a ptrace() memory write to a PRIVATE
file map.
The core worker is derived from KSM's replace_page() logic.
In essence, similar to KSM:
a. allocate a new page and copy over contents of the page that
has the uprobed vaddr
b. modify the copy and insert the breakpoint at the required
address
c. switch the original page with the copy containing the
breakpoint
d. flush page tables.
replace_page() is being replicated here because of some minor
changes in the type of pages and also because Hugh Dickins had
plans to improve replace_page() for KSM specific work.
Instruction analysis on x86 is based on instruction decoder and
determines if an instruction can be probed and determines the
necessary fixups after singlestep. Instruction analysis is done
at probe insertion time so that we avoid having to repeat the
same analysis every time a probe is hit.
A lot of code here is due to the improvement/suggestions/inputs
from Peter Zijlstra.
Changelog:
(v10):
- Add code to clear REX.B prefix as suggested by Denys Vlasenko
and Masami Hiramatsu.
(v9):
- Use insn_offset_modrm as suggested by Masami Hiramatsu.
(v7):
Handle comments from Peter Zijlstra:
- Dont take reference to inode. (expect inode to uprobe_register to be sane).
- Use PTR_ERR to set the return value.
- No need to take reference to inode.
- use PTR_ERR to return error value.
- register and uprobe_unregister share code.
(v5):
- Modified del_consumer as per comments from Peter.
- Drop reference to inode before dropping reference to uprobe.
- Use i_size_read(inode) instead of inode->i_size.
- Ensure uprobe->consumers is NULL, before __uprobe_unregister() is called.
- Includes errno.h as recommended by Stephen Rothwell to fix a build issue
on sparc defconfig
- Remove restrictions while unregistering.
- Earlier code leaked inode references under some conditions while
registering/unregistering.
- Continue the vma-rmap walk even if the intermediate vma doesnt
meet the requirements.
- Validate the vma found by find_vma before inserting/removing the
breakpoint
- Call del_consumer under mutex_lock.
- Use hash locks.
- Handle mremap.
- Introduce find_least_offset_node() instead of close match logic in
find_uprobe
- Uprobes no more depends on MM_OWNER; No reference to task_structs
while inserting/removing a probe.
- Uses read_mapping_page instead of grab_cache_page so that the pages
have valid content.
- pass NULL to get_user_pages for the task parameter.
- call SetPageUptodate on the new page allocated in write_opcode.
- fix leaking a reference to the new page under certain conditions.
- Include Instruction Decoder if Uprobes gets defined.
- Remove const attributes for instruction prefix arrays.
- Uses mm_context to know if the application is 32 bit.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Also-written-by: Jim Keniston <jkenisto@us.ibm.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Roland McGrath <roland@hack.frob.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Anton Arapov <anton@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Denys Vlasenko <vda.linux@googlemail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linux-mm <linux-mm@kvack.org>
Link: http://lkml.kernel.org/r/20120209092642.GE16600@linux.vnet.ibm.com
[ Made various small edits to the commit log ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-02-09 17:26:42 +08:00
|
|
|
}
|
2012-02-17 16:27:41 +08:00
|
|
|
|
2012-07-30 02:22:40 +08:00
|
|
|
return n;
|
uprobes, mm, x86: Add the ability to install and remove uprobes breakpoints
Add uprobes support to the core kernel, with x86 support.
This commit adds the kernel facilities, the actual uprobes
user-space ABI and perf probe support comes in later commits.
General design:
Uprobes are maintained in an rb-tree indexed by inode and offset
(the offset here is from the start of the mapping). For a unique
(inode, offset) tuple, there can be at most one uprobe in the
rb-tree.
Since the (inode, offset) tuple identifies a unique uprobe, more
than one user may be interested in the same uprobe. This provides
the ability to connect multiple 'consumers' to the same uprobe.
Each consumer defines a handler and a filter (optional). The
'handler' is run every time the uprobe is hit, if it matches the
'filter' criteria.
The first consumer of a uprobe causes the breakpoint to be
inserted at the specified address and subsequent consumers are
appended to this list. On subsequent probes, the consumer gets
appended to the existing list of consumers. The breakpoint is
removed when the last consumer unregisters. For all other
unregisterations, the consumer is removed from the list of
consumers.
Given a inode, we get a list of the mms that have mapped the
inode. Do the actual registration if mm maps the page where a
probe needs to be inserted/removed.
We use a temporary list to walk through the vmas that map the
inode.
- The number of maps that map the inode, is not known before we
walk the rmap and keeps changing.
- extending vm_area_struct wasn't recommended, it's a
size-critical data structure.
- There can be more than one maps of the inode in the same mm.
We add callbacks to the mmap methods to keep an eye on text vmas
that are of interest to uprobes. When a vma of interest is mapped,
we insert the breakpoint at the right address.
Uprobe works by replacing the instruction at the address defined
by (inode, offset) with the arch specific breakpoint
instruction. We save a copy of the original instruction at the
uprobed address.
This is needed for:
a. executing the instruction out-of-line (xol).
b. instruction analysis for any subsequent fixups.
c. restoring the instruction back when the uprobe is unregistered.
We insert or delete a breakpoint instruction, and this
breakpoint instruction is assumed to be the smallest instruction
available on the platform. For fixed size instruction platforms
this is trivially true, for variable size instruction platforms
the breakpoint instruction is typically the smallest (often a
single byte).
Writing the instruction is done by COWing the page and changing
the instruction during the copy, this even though most platforms
allow atomic writes of the breakpoint instruction. This also
mirrors the behaviour of a ptrace() memory write to a PRIVATE
file map.
The core worker is derived from KSM's replace_page() logic.
In essence, similar to KSM:
a. allocate a new page and copy over contents of the page that
has the uprobed vaddr
b. modify the copy and insert the breakpoint at the required
address
c. switch the original page with the copy containing the
breakpoint
d. flush page tables.
replace_page() is being replicated here because of some minor
changes in the type of pages and also because Hugh Dickins had
plans to improve replace_page() for KSM specific work.
Instruction analysis on x86 is based on instruction decoder and
determines if an instruction can be probed and determines the
necessary fixups after singlestep. Instruction analysis is done
at probe insertion time so that we avoid having to repeat the
same analysis every time a probe is hit.
A lot of code here is due to the improvement/suggestions/inputs
from Peter Zijlstra.
Changelog:
(v10):
- Add code to clear REX.B prefix as suggested by Denys Vlasenko
and Masami Hiramatsu.
(v9):
- Use insn_offset_modrm as suggested by Masami Hiramatsu.
(v7):
Handle comments from Peter Zijlstra:
- Dont take reference to inode. (expect inode to uprobe_register to be sane).
- Use PTR_ERR to set the return value.
- No need to take reference to inode.
- use PTR_ERR to return error value.
- register and uprobe_unregister share code.
(v5):
- Modified del_consumer as per comments from Peter.
- Drop reference to inode before dropping reference to uprobe.
- Use i_size_read(inode) instead of inode->i_size.
- Ensure uprobe->consumers is NULL, before __uprobe_unregister() is called.
- Includes errno.h as recommended by Stephen Rothwell to fix a build issue
on sparc defconfig
- Remove restrictions while unregistering.
- Earlier code leaked inode references under some conditions while
registering/unregistering.
- Continue the vma-rmap walk even if the intermediate vma doesnt
meet the requirements.
- Validate the vma found by find_vma before inserting/removing the
breakpoint
- Call del_consumer under mutex_lock.
- Use hash locks.
- Handle mremap.
- Introduce find_least_offset_node() instead of close match logic in
find_uprobe
- Uprobes no more depends on MM_OWNER; No reference to task_structs
while inserting/removing a probe.
- Uses read_mapping_page instead of grab_cache_page so that the pages
have valid content.
- pass NULL to get_user_pages for the task parameter.
- call SetPageUptodate on the new page allocated in write_opcode.
- fix leaking a reference to the new page under certain conditions.
- Include Instruction Decoder if Uprobes gets defined.
- Remove const attributes for instruction prefix arrays.
- Uses mm_context to know if the application is 32 bit.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Also-written-by: Jim Keniston <jkenisto@us.ibm.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Roland McGrath <roland@hack.frob.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Anton Arapov <anton@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Denys Vlasenko <vda.linux@googlemail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linux-mm <linux-mm@kvack.org>
Link: http://lkml.kernel.org/r/20120209092642.GE16600@linux.vnet.ibm.com
[ Made various small edits to the commit log ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-02-09 17:26:42 +08:00
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
2012-07-30 02:22:40 +08:00
|
|
|
* For a given range in vma, build a list of probes that need to be inserted.
|
uprobes, mm, x86: Add the ability to install and remove uprobes breakpoints
Add uprobes support to the core kernel, with x86 support.
This commit adds the kernel facilities, the actual uprobes
user-space ABI and perf probe support comes in later commits.
General design:
Uprobes are maintained in an rb-tree indexed by inode and offset
(the offset here is from the start of the mapping). For a unique
(inode, offset) tuple, there can be at most one uprobe in the
rb-tree.
Since the (inode, offset) tuple identifies a unique uprobe, more
than one user may be interested in the same uprobe. This provides
the ability to connect multiple 'consumers' to the same uprobe.
Each consumer defines a handler and a filter (optional). The
'handler' is run every time the uprobe is hit, if it matches the
'filter' criteria.
The first consumer of a uprobe causes the breakpoint to be
inserted at the specified address and subsequent consumers are
appended to this list. On subsequent probes, the consumer gets
appended to the existing list of consumers. The breakpoint is
removed when the last consumer unregisters. For all other
unregisterations, the consumer is removed from the list of
consumers.
Given a inode, we get a list of the mms that have mapped the
inode. Do the actual registration if mm maps the page where a
probe needs to be inserted/removed.
We use a temporary list to walk through the vmas that map the
inode.
- The number of maps that map the inode, is not known before we
walk the rmap and keeps changing.
- extending vm_area_struct wasn't recommended, it's a
size-critical data structure.
- There can be more than one maps of the inode in the same mm.
We add callbacks to the mmap methods to keep an eye on text vmas
that are of interest to uprobes. When a vma of interest is mapped,
we insert the breakpoint at the right address.
Uprobe works by replacing the instruction at the address defined
by (inode, offset) with the arch specific breakpoint
instruction. We save a copy of the original instruction at the
uprobed address.
This is needed for:
a. executing the instruction out-of-line (xol).
b. instruction analysis for any subsequent fixups.
c. restoring the instruction back when the uprobe is unregistered.
We insert or delete a breakpoint instruction, and this
breakpoint instruction is assumed to be the smallest instruction
available on the platform. For fixed size instruction platforms
this is trivially true, for variable size instruction platforms
the breakpoint instruction is typically the smallest (often a
single byte).
Writing the instruction is done by COWing the page and changing
the instruction during the copy, this even though most platforms
allow atomic writes of the breakpoint instruction. This also
mirrors the behaviour of a ptrace() memory write to a PRIVATE
file map.
The core worker is derived from KSM's replace_page() logic.
In essence, similar to KSM:
a. allocate a new page and copy over contents of the page that
has the uprobed vaddr
b. modify the copy and insert the breakpoint at the required
address
c. switch the original page with the copy containing the
breakpoint
d. flush page tables.
replace_page() is being replicated here because of some minor
changes in the type of pages and also because Hugh Dickins had
plans to improve replace_page() for KSM specific work.
Instruction analysis on x86 is based on instruction decoder and
determines if an instruction can be probed and determines the
necessary fixups after singlestep. Instruction analysis is done
at probe insertion time so that we avoid having to repeat the
same analysis every time a probe is hit.
A lot of code here is due to the improvement/suggestions/inputs
from Peter Zijlstra.
Changelog:
(v10):
- Add code to clear REX.B prefix as suggested by Denys Vlasenko
and Masami Hiramatsu.
(v9):
- Use insn_offset_modrm as suggested by Masami Hiramatsu.
(v7):
Handle comments from Peter Zijlstra:
- Dont take reference to inode. (expect inode to uprobe_register to be sane).
- Use PTR_ERR to set the return value.
- No need to take reference to inode.
- use PTR_ERR to return error value.
- register and uprobe_unregister share code.
(v5):
- Modified del_consumer as per comments from Peter.
- Drop reference to inode before dropping reference to uprobe.
- Use i_size_read(inode) instead of inode->i_size.
- Ensure uprobe->consumers is NULL, before __uprobe_unregister() is called.
- Includes errno.h as recommended by Stephen Rothwell to fix a build issue
on sparc defconfig
- Remove restrictions while unregistering.
- Earlier code leaked inode references under some conditions while
registering/unregistering.
- Continue the vma-rmap walk even if the intermediate vma doesnt
meet the requirements.
- Validate the vma found by find_vma before inserting/removing the
breakpoint
- Call del_consumer under mutex_lock.
- Use hash locks.
- Handle mremap.
- Introduce find_least_offset_node() instead of close match logic in
find_uprobe
- Uprobes no more depends on MM_OWNER; No reference to task_structs
while inserting/removing a probe.
- Uses read_mapping_page instead of grab_cache_page so that the pages
have valid content.
- pass NULL to get_user_pages for the task parameter.
- call SetPageUptodate on the new page allocated in write_opcode.
- fix leaking a reference to the new page under certain conditions.
- Include Instruction Decoder if Uprobes gets defined.
- Remove const attributes for instruction prefix arrays.
- Uses mm_context to know if the application is 32 bit.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Also-written-by: Jim Keniston <jkenisto@us.ibm.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Roland McGrath <roland@hack.frob.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Anton Arapov <anton@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Denys Vlasenko <vda.linux@googlemail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linux-mm <linux-mm@kvack.org>
Link: http://lkml.kernel.org/r/20120209092642.GE16600@linux.vnet.ibm.com
[ Made various small edits to the commit log ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-02-09 17:26:42 +08:00
|
|
|
*/
|
2012-07-30 02:22:40 +08:00
|
|
|
static void build_probe_list(struct inode *inode,
|
|
|
|
struct vm_area_struct *vma,
|
|
|
|
unsigned long start, unsigned long end,
|
|
|
|
struct list_head *head)
|
uprobes, mm, x86: Add the ability to install and remove uprobes breakpoints
Add uprobes support to the core kernel, with x86 support.
This commit adds the kernel facilities, the actual uprobes
user-space ABI and perf probe support comes in later commits.
General design:
Uprobes are maintained in an rb-tree indexed by inode and offset
(the offset here is from the start of the mapping). For a unique
(inode, offset) tuple, there can be at most one uprobe in the
rb-tree.
Since the (inode, offset) tuple identifies a unique uprobe, more
than one user may be interested in the same uprobe. This provides
the ability to connect multiple 'consumers' to the same uprobe.
Each consumer defines a handler and a filter (optional). The
'handler' is run every time the uprobe is hit, if it matches the
'filter' criteria.
The first consumer of a uprobe causes the breakpoint to be
inserted at the specified address and subsequent consumers are
appended to this list. On subsequent probes, the consumer gets
appended to the existing list of consumers. The breakpoint is
removed when the last consumer unregisters. For all other
unregisterations, the consumer is removed from the list of
consumers.
Given a inode, we get a list of the mms that have mapped the
inode. Do the actual registration if mm maps the page where a
probe needs to be inserted/removed.
We use a temporary list to walk through the vmas that map the
inode.
- The number of maps that map the inode, is not known before we
walk the rmap and keeps changing.
- extending vm_area_struct wasn't recommended, it's a
size-critical data structure.
- There can be more than one maps of the inode in the same mm.
We add callbacks to the mmap methods to keep an eye on text vmas
that are of interest to uprobes. When a vma of interest is mapped,
we insert the breakpoint at the right address.
Uprobe works by replacing the instruction at the address defined
by (inode, offset) with the arch specific breakpoint
instruction. We save a copy of the original instruction at the
uprobed address.
This is needed for:
a. executing the instruction out-of-line (xol).
b. instruction analysis for any subsequent fixups.
c. restoring the instruction back when the uprobe is unregistered.
We insert or delete a breakpoint instruction, and this
breakpoint instruction is assumed to be the smallest instruction
available on the platform. For fixed size instruction platforms
this is trivially true, for variable size instruction platforms
the breakpoint instruction is typically the smallest (often a
single byte).
Writing the instruction is done by COWing the page and changing
the instruction during the copy, this even though most platforms
allow atomic writes of the breakpoint instruction. This also
mirrors the behaviour of a ptrace() memory write to a PRIVATE
file map.
The core worker is derived from KSM's replace_page() logic.
In essence, similar to KSM:
a. allocate a new page and copy over contents of the page that
has the uprobed vaddr
b. modify the copy and insert the breakpoint at the required
address
c. switch the original page with the copy containing the
breakpoint
d. flush page tables.
replace_page() is being replicated here because of some minor
changes in the type of pages and also because Hugh Dickins had
plans to improve replace_page() for KSM specific work.
Instruction analysis on x86 is based on instruction decoder and
determines if an instruction can be probed and determines the
necessary fixups after singlestep. Instruction analysis is done
at probe insertion time so that we avoid having to repeat the
same analysis every time a probe is hit.
A lot of code here is due to the improvement/suggestions/inputs
from Peter Zijlstra.
Changelog:
(v10):
- Add code to clear REX.B prefix as suggested by Denys Vlasenko
and Masami Hiramatsu.
(v9):
- Use insn_offset_modrm as suggested by Masami Hiramatsu.
(v7):
Handle comments from Peter Zijlstra:
- Dont take reference to inode. (expect inode to uprobe_register to be sane).
- Use PTR_ERR to set the return value.
- No need to take reference to inode.
- use PTR_ERR to return error value.
- register and uprobe_unregister share code.
(v5):
- Modified del_consumer as per comments from Peter.
- Drop reference to inode before dropping reference to uprobe.
- Use i_size_read(inode) instead of inode->i_size.
- Ensure uprobe->consumers is NULL, before __uprobe_unregister() is called.
- Includes errno.h as recommended by Stephen Rothwell to fix a build issue
on sparc defconfig
- Remove restrictions while unregistering.
- Earlier code leaked inode references under some conditions while
registering/unregistering.
- Continue the vma-rmap walk even if the intermediate vma doesnt
meet the requirements.
- Validate the vma found by find_vma before inserting/removing the
breakpoint
- Call del_consumer under mutex_lock.
- Use hash locks.
- Handle mremap.
- Introduce find_least_offset_node() instead of close match logic in
find_uprobe
- Uprobes no more depends on MM_OWNER; No reference to task_structs
while inserting/removing a probe.
- Uses read_mapping_page instead of grab_cache_page so that the pages
have valid content.
- pass NULL to get_user_pages for the task parameter.
- call SetPageUptodate on the new page allocated in write_opcode.
- fix leaking a reference to the new page under certain conditions.
- Include Instruction Decoder if Uprobes gets defined.
- Remove const attributes for instruction prefix arrays.
- Uses mm_context to know if the application is 32 bit.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Also-written-by: Jim Keniston <jkenisto@us.ibm.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Roland McGrath <roland@hack.frob.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Anton Arapov <anton@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Denys Vlasenko <vda.linux@googlemail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linux-mm <linux-mm@kvack.org>
Link: http://lkml.kernel.org/r/20120209092642.GE16600@linux.vnet.ibm.com
[ Made various small edits to the commit log ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-02-09 17:26:42 +08:00
|
|
|
{
|
2012-07-30 02:22:40 +08:00
|
|
|
loff_t min, max;
|
|
|
|
struct rb_node *n, *t;
|
|
|
|
struct uprobe *u;
|
2012-02-17 16:27:41 +08:00
|
|
|
|
2012-07-30 02:22:40 +08:00
|
|
|
INIT_LIST_HEAD(head);
|
2012-07-30 02:22:42 +08:00
|
|
|
min = vaddr_to_offset(vma, start);
|
2012-07-30 02:22:40 +08:00
|
|
|
max = min + (end - start) - 1;
|
uprobes, mm, x86: Add the ability to install and remove uprobes breakpoints
Add uprobes support to the core kernel, with x86 support.
This commit adds the kernel facilities, the actual uprobes
user-space ABI and perf probe support comes in later commits.
General design:
Uprobes are maintained in an rb-tree indexed by inode and offset
(the offset here is from the start of the mapping). For a unique
(inode, offset) tuple, there can be at most one uprobe in the
rb-tree.
Since the (inode, offset) tuple identifies a unique uprobe, more
than one user may be interested in the same uprobe. This provides
the ability to connect multiple 'consumers' to the same uprobe.
Each consumer defines a handler and a filter (optional). The
'handler' is run every time the uprobe is hit, if it matches the
'filter' criteria.
The first consumer of a uprobe causes the breakpoint to be
inserted at the specified address and subsequent consumers are
appended to this list. On subsequent probes, the consumer gets
appended to the existing list of consumers. The breakpoint is
removed when the last consumer unregisters. For all other
unregisterations, the consumer is removed from the list of
consumers.
Given a inode, we get a list of the mms that have mapped the
inode. Do the actual registration if mm maps the page where a
probe needs to be inserted/removed.
We use a temporary list to walk through the vmas that map the
inode.
- The number of maps that map the inode, is not known before we
walk the rmap and keeps changing.
- extending vm_area_struct wasn't recommended, it's a
size-critical data structure.
- There can be more than one maps of the inode in the same mm.
We add callbacks to the mmap methods to keep an eye on text vmas
that are of interest to uprobes. When a vma of interest is mapped,
we insert the breakpoint at the right address.
Uprobe works by replacing the instruction at the address defined
by (inode, offset) with the arch specific breakpoint
instruction. We save a copy of the original instruction at the
uprobed address.
This is needed for:
a. executing the instruction out-of-line (xol).
b. instruction analysis for any subsequent fixups.
c. restoring the instruction back when the uprobe is unregistered.
We insert or delete a breakpoint instruction, and this
breakpoint instruction is assumed to be the smallest instruction
available on the platform. For fixed size instruction platforms
this is trivially true, for variable size instruction platforms
the breakpoint instruction is typically the smallest (often a
single byte).
Writing the instruction is done by COWing the page and changing
the instruction during the copy, this even though most platforms
allow atomic writes of the breakpoint instruction. This also
mirrors the behaviour of a ptrace() memory write to a PRIVATE
file map.
The core worker is derived from KSM's replace_page() logic.
In essence, similar to KSM:
a. allocate a new page and copy over contents of the page that
has the uprobed vaddr
b. modify the copy and insert the breakpoint at the required
address
c. switch the original page with the copy containing the
breakpoint
d. flush page tables.
replace_page() is being replicated here because of some minor
changes in the type of pages and also because Hugh Dickins had
plans to improve replace_page() for KSM specific work.
Instruction analysis on x86 is based on instruction decoder and
determines if an instruction can be probed and determines the
necessary fixups after singlestep. Instruction analysis is done
at probe insertion time so that we avoid having to repeat the
same analysis every time a probe is hit.
A lot of code here is due to the improvement/suggestions/inputs
from Peter Zijlstra.
Changelog:
(v10):
- Add code to clear REX.B prefix as suggested by Denys Vlasenko
and Masami Hiramatsu.
(v9):
- Use insn_offset_modrm as suggested by Masami Hiramatsu.
(v7):
Handle comments from Peter Zijlstra:
- Dont take reference to inode. (expect inode to uprobe_register to be sane).
- Use PTR_ERR to set the return value.
- No need to take reference to inode.
- use PTR_ERR to return error value.
- register and uprobe_unregister share code.
(v5):
- Modified del_consumer as per comments from Peter.
- Drop reference to inode before dropping reference to uprobe.
- Use i_size_read(inode) instead of inode->i_size.
- Ensure uprobe->consumers is NULL, before __uprobe_unregister() is called.
- Includes errno.h as recommended by Stephen Rothwell to fix a build issue
on sparc defconfig
- Remove restrictions while unregistering.
- Earlier code leaked inode references under some conditions while
registering/unregistering.
- Continue the vma-rmap walk even if the intermediate vma doesnt
meet the requirements.
- Validate the vma found by find_vma before inserting/removing the
breakpoint
- Call del_consumer under mutex_lock.
- Use hash locks.
- Handle mremap.
- Introduce find_least_offset_node() instead of close match logic in
find_uprobe
- Uprobes no more depends on MM_OWNER; No reference to task_structs
while inserting/removing a probe.
- Uses read_mapping_page instead of grab_cache_page so that the pages
have valid content.
- pass NULL to get_user_pages for the task parameter.
- call SetPageUptodate on the new page allocated in write_opcode.
- fix leaking a reference to the new page under certain conditions.
- Include Instruction Decoder if Uprobes gets defined.
- Remove const attributes for instruction prefix arrays.
- Uses mm_context to know if the application is 32 bit.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Also-written-by: Jim Keniston <jkenisto@us.ibm.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Roland McGrath <roland@hack.frob.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Anton Arapov <anton@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Denys Vlasenko <vda.linux@googlemail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linux-mm <linux-mm@kvack.org>
Link: http://lkml.kernel.org/r/20120209092642.GE16600@linux.vnet.ibm.com
[ Made various small edits to the commit log ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-02-09 17:26:42 +08:00
|
|
|
|
2012-08-18 23:01:57 +08:00
|
|
|
spin_lock(&uprobes_treelock);
|
2012-07-30 02:22:40 +08:00
|
|
|
n = find_node_in_range(inode, min, max);
|
|
|
|
if (n) {
|
|
|
|
for (t = n; t; t = rb_prev(t)) {
|
|
|
|
u = rb_entry(t, struct uprobe, rb_node);
|
|
|
|
if (u->inode != inode || u->offset < min)
|
|
|
|
break;
|
|
|
|
list_add(&u->pending_list, head);
|
2015-07-21 21:40:03 +08:00
|
|
|
get_uprobe(u);
|
2012-07-30 02:22:40 +08:00
|
|
|
}
|
|
|
|
for (t = n; (t = rb_next(t)); ) {
|
|
|
|
u = rb_entry(t, struct uprobe, rb_node);
|
|
|
|
if (u->inode != inode || u->offset > max)
|
|
|
|
break;
|
|
|
|
list_add(&u->pending_list, head);
|
2015-07-21 21:40:03 +08:00
|
|
|
get_uprobe(u);
|
2012-07-30 02:22:40 +08:00
|
|
|
}
|
uprobes, mm, x86: Add the ability to install and remove uprobes breakpoints
Add uprobes support to the core kernel, with x86 support.
This commit adds the kernel facilities, the actual uprobes
user-space ABI and perf probe support comes in later commits.
General design:
Uprobes are maintained in an rb-tree indexed by inode and offset
(the offset here is from the start of the mapping). For a unique
(inode, offset) tuple, there can be at most one uprobe in the
rb-tree.
Since the (inode, offset) tuple identifies a unique uprobe, more
than one user may be interested in the same uprobe. This provides
the ability to connect multiple 'consumers' to the same uprobe.
Each consumer defines a handler and a filter (optional). The
'handler' is run every time the uprobe is hit, if it matches the
'filter' criteria.
The first consumer of a uprobe causes the breakpoint to be
inserted at the specified address and subsequent consumers are
appended to this list. On subsequent probes, the consumer gets
appended to the existing list of consumers. The breakpoint is
removed when the last consumer unregisters. For all other
unregisterations, the consumer is removed from the list of
consumers.
Given a inode, we get a list of the mms that have mapped the
inode. Do the actual registration if mm maps the page where a
probe needs to be inserted/removed.
We use a temporary list to walk through the vmas that map the
inode.
- The number of maps that map the inode, is not known before we
walk the rmap and keeps changing.
- extending vm_area_struct wasn't recommended, it's a
size-critical data structure.
- There can be more than one maps of the inode in the same mm.
We add callbacks to the mmap methods to keep an eye on text vmas
that are of interest to uprobes. When a vma of interest is mapped,
we insert the breakpoint at the right address.
Uprobe works by replacing the instruction at the address defined
by (inode, offset) with the arch specific breakpoint
instruction. We save a copy of the original instruction at the
uprobed address.
This is needed for:
a. executing the instruction out-of-line (xol).
b. instruction analysis for any subsequent fixups.
c. restoring the instruction back when the uprobe is unregistered.
We insert or delete a breakpoint instruction, and this
breakpoint instruction is assumed to be the smallest instruction
available on the platform. For fixed size instruction platforms
this is trivially true, for variable size instruction platforms
the breakpoint instruction is typically the smallest (often a
single byte).
Writing the instruction is done by COWing the page and changing
the instruction during the copy, this even though most platforms
allow atomic writes of the breakpoint instruction. This also
mirrors the behaviour of a ptrace() memory write to a PRIVATE
file map.
The core worker is derived from KSM's replace_page() logic.
In essence, similar to KSM:
a. allocate a new page and copy over contents of the page that
has the uprobed vaddr
b. modify the copy and insert the breakpoint at the required
address
c. switch the original page with the copy containing the
breakpoint
d. flush page tables.
replace_page() is being replicated here because of some minor
changes in the type of pages and also because Hugh Dickins had
plans to improve replace_page() for KSM specific work.
Instruction analysis on x86 is based on instruction decoder and
determines if an instruction can be probed and determines the
necessary fixups after singlestep. Instruction analysis is done
at probe insertion time so that we avoid having to repeat the
same analysis every time a probe is hit.
A lot of code here is due to the improvement/suggestions/inputs
from Peter Zijlstra.
Changelog:
(v10):
- Add code to clear REX.B prefix as suggested by Denys Vlasenko
and Masami Hiramatsu.
(v9):
- Use insn_offset_modrm as suggested by Masami Hiramatsu.
(v7):
Handle comments from Peter Zijlstra:
- Dont take reference to inode. (expect inode to uprobe_register to be sane).
- Use PTR_ERR to set the return value.
- No need to take reference to inode.
- use PTR_ERR to return error value.
- register and uprobe_unregister share code.
(v5):
- Modified del_consumer as per comments from Peter.
- Drop reference to inode before dropping reference to uprobe.
- Use i_size_read(inode) instead of inode->i_size.
- Ensure uprobe->consumers is NULL, before __uprobe_unregister() is called.
- Includes errno.h as recommended by Stephen Rothwell to fix a build issue
on sparc defconfig
- Remove restrictions while unregistering.
- Earlier code leaked inode references under some conditions while
registering/unregistering.
- Continue the vma-rmap walk even if the intermediate vma doesnt
meet the requirements.
- Validate the vma found by find_vma before inserting/removing the
breakpoint
- Call del_consumer under mutex_lock.
- Use hash locks.
- Handle mremap.
- Introduce find_least_offset_node() instead of close match logic in
find_uprobe
- Uprobes no more depends on MM_OWNER; No reference to task_structs
while inserting/removing a probe.
- Uses read_mapping_page instead of grab_cache_page so that the pages
have valid content.
- pass NULL to get_user_pages for the task parameter.
- call SetPageUptodate on the new page allocated in write_opcode.
- fix leaking a reference to the new page under certain conditions.
- Include Instruction Decoder if Uprobes gets defined.
- Remove const attributes for instruction prefix arrays.
- Uses mm_context to know if the application is 32 bit.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Also-written-by: Jim Keniston <jkenisto@us.ibm.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Roland McGrath <roland@hack.frob.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Anton Arapov <anton@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Denys Vlasenko <vda.linux@googlemail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linux-mm <linux-mm@kvack.org>
Link: http://lkml.kernel.org/r/20120209092642.GE16600@linux.vnet.ibm.com
[ Made various small edits to the commit log ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-02-09 17:26:42 +08:00
|
|
|
}
|
2012-08-18 23:01:57 +08:00
|
|
|
spin_unlock(&uprobes_treelock);
|
uprobes, mm, x86: Add the ability to install and remove uprobes breakpoints
Add uprobes support to the core kernel, with x86 support.
This commit adds the kernel facilities, the actual uprobes
user-space ABI and perf probe support comes in later commits.
General design:
Uprobes are maintained in an rb-tree indexed by inode and offset
(the offset here is from the start of the mapping). For a unique
(inode, offset) tuple, there can be at most one uprobe in the
rb-tree.
Since the (inode, offset) tuple identifies a unique uprobe, more
than one user may be interested in the same uprobe. This provides
the ability to connect multiple 'consumers' to the same uprobe.
Each consumer defines a handler and a filter (optional). The
'handler' is run every time the uprobe is hit, if it matches the
'filter' criteria.
The first consumer of a uprobe causes the breakpoint to be
inserted at the specified address and subsequent consumers are
appended to this list. On subsequent probes, the consumer gets
appended to the existing list of consumers. The breakpoint is
removed when the last consumer unregisters. For all other
unregisterations, the consumer is removed from the list of
consumers.
Given a inode, we get a list of the mms that have mapped the
inode. Do the actual registration if mm maps the page where a
probe needs to be inserted/removed.
We use a temporary list to walk through the vmas that map the
inode.
- The number of maps that map the inode, is not known before we
walk the rmap and keeps changing.
- extending vm_area_struct wasn't recommended, it's a
size-critical data structure.
- There can be more than one maps of the inode in the same mm.
We add callbacks to the mmap methods to keep an eye on text vmas
that are of interest to uprobes. When a vma of interest is mapped,
we insert the breakpoint at the right address.
Uprobe works by replacing the instruction at the address defined
by (inode, offset) with the arch specific breakpoint
instruction. We save a copy of the original instruction at the
uprobed address.
This is needed for:
a. executing the instruction out-of-line (xol).
b. instruction analysis for any subsequent fixups.
c. restoring the instruction back when the uprobe is unregistered.
We insert or delete a breakpoint instruction, and this
breakpoint instruction is assumed to be the smallest instruction
available on the platform. For fixed size instruction platforms
this is trivially true, for variable size instruction platforms
the breakpoint instruction is typically the smallest (often a
single byte).
Writing the instruction is done by COWing the page and changing
the instruction during the copy, this even though most platforms
allow atomic writes of the breakpoint instruction. This also
mirrors the behaviour of a ptrace() memory write to a PRIVATE
file map.
The core worker is derived from KSM's replace_page() logic.
In essence, similar to KSM:
a. allocate a new page and copy over contents of the page that
has the uprobed vaddr
b. modify the copy and insert the breakpoint at the required
address
c. switch the original page with the copy containing the
breakpoint
d. flush page tables.
replace_page() is being replicated here because of some minor
changes in the type of pages and also because Hugh Dickins had
plans to improve replace_page() for KSM specific work.
Instruction analysis on x86 is based on instruction decoder and
determines if an instruction can be probed and determines the
necessary fixups after singlestep. Instruction analysis is done
at probe insertion time so that we avoid having to repeat the
same analysis every time a probe is hit.
A lot of code here is due to the improvement/suggestions/inputs
from Peter Zijlstra.
Changelog:
(v10):
- Add code to clear REX.B prefix as suggested by Denys Vlasenko
and Masami Hiramatsu.
(v9):
- Use insn_offset_modrm as suggested by Masami Hiramatsu.
(v7):
Handle comments from Peter Zijlstra:
- Dont take reference to inode. (expect inode to uprobe_register to be sane).
- Use PTR_ERR to set the return value.
- No need to take reference to inode.
- use PTR_ERR to return error value.
- register and uprobe_unregister share code.
(v5):
- Modified del_consumer as per comments from Peter.
- Drop reference to inode before dropping reference to uprobe.
- Use i_size_read(inode) instead of inode->i_size.
- Ensure uprobe->consumers is NULL, before __uprobe_unregister() is called.
- Includes errno.h as recommended by Stephen Rothwell to fix a build issue
on sparc defconfig
- Remove restrictions while unregistering.
- Earlier code leaked inode references under some conditions while
registering/unregistering.
- Continue the vma-rmap walk even if the intermediate vma doesnt
meet the requirements.
- Validate the vma found by find_vma before inserting/removing the
breakpoint
- Call del_consumer under mutex_lock.
- Use hash locks.
- Handle mremap.
- Introduce find_least_offset_node() instead of close match logic in
find_uprobe
- Uprobes no more depends on MM_OWNER; No reference to task_structs
while inserting/removing a probe.
- Uses read_mapping_page instead of grab_cache_page so that the pages
have valid content.
- pass NULL to get_user_pages for the task parameter.
- call SetPageUptodate on the new page allocated in write_opcode.
- fix leaking a reference to the new page under certain conditions.
- Include Instruction Decoder if Uprobes gets defined.
- Remove const attributes for instruction prefix arrays.
- Uses mm_context to know if the application is 32 bit.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Also-written-by: Jim Keniston <jkenisto@us.ibm.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Roland McGrath <roland@hack.frob.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Anton Arapov <anton@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Denys Vlasenko <vda.linux@googlemail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linux-mm <linux-mm@kvack.org>
Link: http://lkml.kernel.org/r/20120209092642.GE16600@linux.vnet.ibm.com
[ Made various small edits to the commit log ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-02-09 17:26:42 +08:00
|
|
|
}
|
|
|
|
|
uprobes: Support SDT markers having reference count (semaphore)
Userspace Statically Defined Tracepoints[1] are dtrace style markers
inside userspace applications. Applications like PostgreSQL, MySQL,
Pthread, Perl, Python, Java, Ruby, Node.js, libvirt, QEMU, glib etc
have these markers embedded in them. These markers are added by developer
at important places in the code. Each marker source expands to a single
nop instruction in the compiled code but there may be additional
overhead for computing the marker arguments which expands to couple of
instructions. In case the overhead is more, execution of it can be
omitted by runtime if() condition when no one is tracing on the marker:
if (reference_counter > 0) {
Execute marker instructions;
}
Default value of reference counter is 0. Tracer has to increment the
reference counter before tracing on a marker and decrement it when
done with the tracing.
Implement the reference counter logic in core uprobe. User will be
able to use it from trace_uprobe as well as from kernel module. New
trace_uprobe definition with reference counter will now be:
<path>:<offset>[(ref_ctr_offset)]
where ref_ctr_offset is an optional field. For kernel module, new
variant of uprobe_register() has been introduced:
uprobe_register_refctr(inode, offset, ref_ctr_offset, consumer)
No new variant for uprobe_unregister() because it's assumed to have
only one reference counter for one uprobe.
[1] https://sourceware.org/systemtap/wiki/UserSpaceProbeImplementation
Note: 'reference counter' is called as 'semaphore' in original Dtrace
(or Systemtap, bcc and even in ELF) documentation and code. But the
term 'semaphore' is misleading in this context. This is just a counter
used to hold number of tracers tracing on a marker. This is not really
used for any synchronization. So we are calling it a 'reference counter'
in kernel / perf code.
Link: http://lkml.kernel.org/r/20180820044250.11659-2-ravi.bangoria@linux.ibm.com
Reviewed-by: Masami Hiramatsu <mhiramat@kernel.org>
[Only trace_uprobe.c]
Reviewed-by: Oleg Nesterov <oleg@redhat.com>
Reviewed-by: Song Liu <songliubraving@fb.com>
Tested-by: Song Liu <songliubraving@fb.com>
Signed-off-by: Ravi Bangoria <ravi.bangoria@linux.ibm.com>
Signed-off-by: Steven Rostedt (VMware) <rostedt@goodmis.org>
2018-08-20 12:42:47 +08:00
|
|
|
/* @vma contains reference counter, not the probed instruction. */
|
|
|
|
static int delayed_ref_ctr_inc(struct vm_area_struct *vma)
|
|
|
|
{
|
|
|
|
struct list_head *pos, *q;
|
|
|
|
struct delayed_uprobe *du;
|
|
|
|
unsigned long vaddr;
|
|
|
|
int ret = 0, err = 0;
|
|
|
|
|
|
|
|
mutex_lock(&delayed_uprobe_lock);
|
|
|
|
list_for_each_safe(pos, q, &delayed_uprobe_list) {
|
|
|
|
du = list_entry(pos, struct delayed_uprobe, list);
|
|
|
|
|
|
|
|
if (du->mm != vma->vm_mm ||
|
|
|
|
!valid_ref_ctr_vma(du->uprobe, vma))
|
|
|
|
continue;
|
|
|
|
|
|
|
|
vaddr = offset_to_vaddr(vma, du->uprobe->ref_ctr_offset);
|
|
|
|
ret = __update_ref_ctr(vma->vm_mm, vaddr, 1);
|
|
|
|
if (ret) {
|
|
|
|
update_ref_ctr_warn(du->uprobe, vma->vm_mm, 1);
|
|
|
|
if (!err)
|
|
|
|
err = ret;
|
|
|
|
}
|
|
|
|
delayed_uprobe_delete(du);
|
|
|
|
}
|
|
|
|
mutex_unlock(&delayed_uprobe_lock);
|
|
|
|
return err;
|
|
|
|
}
|
|
|
|
|
uprobes, mm, x86: Add the ability to install and remove uprobes breakpoints
Add uprobes support to the core kernel, with x86 support.
This commit adds the kernel facilities, the actual uprobes
user-space ABI and perf probe support comes in later commits.
General design:
Uprobes are maintained in an rb-tree indexed by inode and offset
(the offset here is from the start of the mapping). For a unique
(inode, offset) tuple, there can be at most one uprobe in the
rb-tree.
Since the (inode, offset) tuple identifies a unique uprobe, more
than one user may be interested in the same uprobe. This provides
the ability to connect multiple 'consumers' to the same uprobe.
Each consumer defines a handler and a filter (optional). The
'handler' is run every time the uprobe is hit, if it matches the
'filter' criteria.
The first consumer of a uprobe causes the breakpoint to be
inserted at the specified address and subsequent consumers are
appended to this list. On subsequent probes, the consumer gets
appended to the existing list of consumers. The breakpoint is
removed when the last consumer unregisters. For all other
unregisterations, the consumer is removed from the list of
consumers.
Given a inode, we get a list of the mms that have mapped the
inode. Do the actual registration if mm maps the page where a
probe needs to be inserted/removed.
We use a temporary list to walk through the vmas that map the
inode.
- The number of maps that map the inode, is not known before we
walk the rmap and keeps changing.
- extending vm_area_struct wasn't recommended, it's a
size-critical data structure.
- There can be more than one maps of the inode in the same mm.
We add callbacks to the mmap methods to keep an eye on text vmas
that are of interest to uprobes. When a vma of interest is mapped,
we insert the breakpoint at the right address.
Uprobe works by replacing the instruction at the address defined
by (inode, offset) with the arch specific breakpoint
instruction. We save a copy of the original instruction at the
uprobed address.
This is needed for:
a. executing the instruction out-of-line (xol).
b. instruction analysis for any subsequent fixups.
c. restoring the instruction back when the uprobe is unregistered.
We insert or delete a breakpoint instruction, and this
breakpoint instruction is assumed to be the smallest instruction
available on the platform. For fixed size instruction platforms
this is trivially true, for variable size instruction platforms
the breakpoint instruction is typically the smallest (often a
single byte).
Writing the instruction is done by COWing the page and changing
the instruction during the copy, this even though most platforms
allow atomic writes of the breakpoint instruction. This also
mirrors the behaviour of a ptrace() memory write to a PRIVATE
file map.
The core worker is derived from KSM's replace_page() logic.
In essence, similar to KSM:
a. allocate a new page and copy over contents of the page that
has the uprobed vaddr
b. modify the copy and insert the breakpoint at the required
address
c. switch the original page with the copy containing the
breakpoint
d. flush page tables.
replace_page() is being replicated here because of some minor
changes in the type of pages and also because Hugh Dickins had
plans to improve replace_page() for KSM specific work.
Instruction analysis on x86 is based on instruction decoder and
determines if an instruction can be probed and determines the
necessary fixups after singlestep. Instruction analysis is done
at probe insertion time so that we avoid having to repeat the
same analysis every time a probe is hit.
A lot of code here is due to the improvement/suggestions/inputs
from Peter Zijlstra.
Changelog:
(v10):
- Add code to clear REX.B prefix as suggested by Denys Vlasenko
and Masami Hiramatsu.
(v9):
- Use insn_offset_modrm as suggested by Masami Hiramatsu.
(v7):
Handle comments from Peter Zijlstra:
- Dont take reference to inode. (expect inode to uprobe_register to be sane).
- Use PTR_ERR to set the return value.
- No need to take reference to inode.
- use PTR_ERR to return error value.
- register and uprobe_unregister share code.
(v5):
- Modified del_consumer as per comments from Peter.
- Drop reference to inode before dropping reference to uprobe.
- Use i_size_read(inode) instead of inode->i_size.
- Ensure uprobe->consumers is NULL, before __uprobe_unregister() is called.
- Includes errno.h as recommended by Stephen Rothwell to fix a build issue
on sparc defconfig
- Remove restrictions while unregistering.
- Earlier code leaked inode references under some conditions while
registering/unregistering.
- Continue the vma-rmap walk even if the intermediate vma doesnt
meet the requirements.
- Validate the vma found by find_vma before inserting/removing the
breakpoint
- Call del_consumer under mutex_lock.
- Use hash locks.
- Handle mremap.
- Introduce find_least_offset_node() instead of close match logic in
find_uprobe
- Uprobes no more depends on MM_OWNER; No reference to task_structs
while inserting/removing a probe.
- Uses read_mapping_page instead of grab_cache_page so that the pages
have valid content.
- pass NULL to get_user_pages for the task parameter.
- call SetPageUptodate on the new page allocated in write_opcode.
- fix leaking a reference to the new page under certain conditions.
- Include Instruction Decoder if Uprobes gets defined.
- Remove const attributes for instruction prefix arrays.
- Uses mm_context to know if the application is 32 bit.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Also-written-by: Jim Keniston <jkenisto@us.ibm.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Roland McGrath <roland@hack.frob.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Anton Arapov <anton@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Denys Vlasenko <vda.linux@googlemail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linux-mm <linux-mm@kvack.org>
Link: http://lkml.kernel.org/r/20120209092642.GE16600@linux.vnet.ibm.com
[ Made various small edits to the commit log ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-02-09 17:26:42 +08:00
|
|
|
/*
|
2012-08-06 20:49:56 +08:00
|
|
|
* Called from mmap_region/vma_adjust with mm->mmap_sem acquired.
|
uprobes, mm, x86: Add the ability to install and remove uprobes breakpoints
Add uprobes support to the core kernel, with x86 support.
This commit adds the kernel facilities, the actual uprobes
user-space ABI and perf probe support comes in later commits.
General design:
Uprobes are maintained in an rb-tree indexed by inode and offset
(the offset here is from the start of the mapping). For a unique
(inode, offset) tuple, there can be at most one uprobe in the
rb-tree.
Since the (inode, offset) tuple identifies a unique uprobe, more
than one user may be interested in the same uprobe. This provides
the ability to connect multiple 'consumers' to the same uprobe.
Each consumer defines a handler and a filter (optional). The
'handler' is run every time the uprobe is hit, if it matches the
'filter' criteria.
The first consumer of a uprobe causes the breakpoint to be
inserted at the specified address and subsequent consumers are
appended to this list. On subsequent probes, the consumer gets
appended to the existing list of consumers. The breakpoint is
removed when the last consumer unregisters. For all other
unregisterations, the consumer is removed from the list of
consumers.
Given a inode, we get a list of the mms that have mapped the
inode. Do the actual registration if mm maps the page where a
probe needs to be inserted/removed.
We use a temporary list to walk through the vmas that map the
inode.
- The number of maps that map the inode, is not known before we
walk the rmap and keeps changing.
- extending vm_area_struct wasn't recommended, it's a
size-critical data structure.
- There can be more than one maps of the inode in the same mm.
We add callbacks to the mmap methods to keep an eye on text vmas
that are of interest to uprobes. When a vma of interest is mapped,
we insert the breakpoint at the right address.
Uprobe works by replacing the instruction at the address defined
by (inode, offset) with the arch specific breakpoint
instruction. We save a copy of the original instruction at the
uprobed address.
This is needed for:
a. executing the instruction out-of-line (xol).
b. instruction analysis for any subsequent fixups.
c. restoring the instruction back when the uprobe is unregistered.
We insert or delete a breakpoint instruction, and this
breakpoint instruction is assumed to be the smallest instruction
available on the platform. For fixed size instruction platforms
this is trivially true, for variable size instruction platforms
the breakpoint instruction is typically the smallest (often a
single byte).
Writing the instruction is done by COWing the page and changing
the instruction during the copy, this even though most platforms
allow atomic writes of the breakpoint instruction. This also
mirrors the behaviour of a ptrace() memory write to a PRIVATE
file map.
The core worker is derived from KSM's replace_page() logic.
In essence, similar to KSM:
a. allocate a new page and copy over contents of the page that
has the uprobed vaddr
b. modify the copy and insert the breakpoint at the required
address
c. switch the original page with the copy containing the
breakpoint
d. flush page tables.
replace_page() is being replicated here because of some minor
changes in the type of pages and also because Hugh Dickins had
plans to improve replace_page() for KSM specific work.
Instruction analysis on x86 is based on instruction decoder and
determines if an instruction can be probed and determines the
necessary fixups after singlestep. Instruction analysis is done
at probe insertion time so that we avoid having to repeat the
same analysis every time a probe is hit.
A lot of code here is due to the improvement/suggestions/inputs
from Peter Zijlstra.
Changelog:
(v10):
- Add code to clear REX.B prefix as suggested by Denys Vlasenko
and Masami Hiramatsu.
(v9):
- Use insn_offset_modrm as suggested by Masami Hiramatsu.
(v7):
Handle comments from Peter Zijlstra:
- Dont take reference to inode. (expect inode to uprobe_register to be sane).
- Use PTR_ERR to set the return value.
- No need to take reference to inode.
- use PTR_ERR to return error value.
- register and uprobe_unregister share code.
(v5):
- Modified del_consumer as per comments from Peter.
- Drop reference to inode before dropping reference to uprobe.
- Use i_size_read(inode) instead of inode->i_size.
- Ensure uprobe->consumers is NULL, before __uprobe_unregister() is called.
- Includes errno.h as recommended by Stephen Rothwell to fix a build issue
on sparc defconfig
- Remove restrictions while unregistering.
- Earlier code leaked inode references under some conditions while
registering/unregistering.
- Continue the vma-rmap walk even if the intermediate vma doesnt
meet the requirements.
- Validate the vma found by find_vma before inserting/removing the
breakpoint
- Call del_consumer under mutex_lock.
- Use hash locks.
- Handle mremap.
- Introduce find_least_offset_node() instead of close match logic in
find_uprobe
- Uprobes no more depends on MM_OWNER; No reference to task_structs
while inserting/removing a probe.
- Uses read_mapping_page instead of grab_cache_page so that the pages
have valid content.
- pass NULL to get_user_pages for the task parameter.
- call SetPageUptodate on the new page allocated in write_opcode.
- fix leaking a reference to the new page under certain conditions.
- Include Instruction Decoder if Uprobes gets defined.
- Remove const attributes for instruction prefix arrays.
- Uses mm_context to know if the application is 32 bit.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Also-written-by: Jim Keniston <jkenisto@us.ibm.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Roland McGrath <roland@hack.frob.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Anton Arapov <anton@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Denys Vlasenko <vda.linux@googlemail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linux-mm <linux-mm@kvack.org>
Link: http://lkml.kernel.org/r/20120209092642.GE16600@linux.vnet.ibm.com
[ Made various small edits to the commit log ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-02-09 17:26:42 +08:00
|
|
|
*
|
2012-08-06 20:49:56 +08:00
|
|
|
* Currently we ignore all errors and always return 0, the callers
|
|
|
|
* can't handle the failure anyway.
|
uprobes, mm, x86: Add the ability to install and remove uprobes breakpoints
Add uprobes support to the core kernel, with x86 support.
This commit adds the kernel facilities, the actual uprobes
user-space ABI and perf probe support comes in later commits.
General design:
Uprobes are maintained in an rb-tree indexed by inode and offset
(the offset here is from the start of the mapping). For a unique
(inode, offset) tuple, there can be at most one uprobe in the
rb-tree.
Since the (inode, offset) tuple identifies a unique uprobe, more
than one user may be interested in the same uprobe. This provides
the ability to connect multiple 'consumers' to the same uprobe.
Each consumer defines a handler and a filter (optional). The
'handler' is run every time the uprobe is hit, if it matches the
'filter' criteria.
The first consumer of a uprobe causes the breakpoint to be
inserted at the specified address and subsequent consumers are
appended to this list. On subsequent probes, the consumer gets
appended to the existing list of consumers. The breakpoint is
removed when the last consumer unregisters. For all other
unregisterations, the consumer is removed from the list of
consumers.
Given a inode, we get a list of the mms that have mapped the
inode. Do the actual registration if mm maps the page where a
probe needs to be inserted/removed.
We use a temporary list to walk through the vmas that map the
inode.
- The number of maps that map the inode, is not known before we
walk the rmap and keeps changing.
- extending vm_area_struct wasn't recommended, it's a
size-critical data structure.
- There can be more than one maps of the inode in the same mm.
We add callbacks to the mmap methods to keep an eye on text vmas
that are of interest to uprobes. When a vma of interest is mapped,
we insert the breakpoint at the right address.
Uprobe works by replacing the instruction at the address defined
by (inode, offset) with the arch specific breakpoint
instruction. We save a copy of the original instruction at the
uprobed address.
This is needed for:
a. executing the instruction out-of-line (xol).
b. instruction analysis for any subsequent fixups.
c. restoring the instruction back when the uprobe is unregistered.
We insert or delete a breakpoint instruction, and this
breakpoint instruction is assumed to be the smallest instruction
available on the platform. For fixed size instruction platforms
this is trivially true, for variable size instruction platforms
the breakpoint instruction is typically the smallest (often a
single byte).
Writing the instruction is done by COWing the page and changing
the instruction during the copy, this even though most platforms
allow atomic writes of the breakpoint instruction. This also
mirrors the behaviour of a ptrace() memory write to a PRIVATE
file map.
The core worker is derived from KSM's replace_page() logic.
In essence, similar to KSM:
a. allocate a new page and copy over contents of the page that
has the uprobed vaddr
b. modify the copy and insert the breakpoint at the required
address
c. switch the original page with the copy containing the
breakpoint
d. flush page tables.
replace_page() is being replicated here because of some minor
changes in the type of pages and also because Hugh Dickins had
plans to improve replace_page() for KSM specific work.
Instruction analysis on x86 is based on instruction decoder and
determines if an instruction can be probed and determines the
necessary fixups after singlestep. Instruction analysis is done
at probe insertion time so that we avoid having to repeat the
same analysis every time a probe is hit.
A lot of code here is due to the improvement/suggestions/inputs
from Peter Zijlstra.
Changelog:
(v10):
- Add code to clear REX.B prefix as suggested by Denys Vlasenko
and Masami Hiramatsu.
(v9):
- Use insn_offset_modrm as suggested by Masami Hiramatsu.
(v7):
Handle comments from Peter Zijlstra:
- Dont take reference to inode. (expect inode to uprobe_register to be sane).
- Use PTR_ERR to set the return value.
- No need to take reference to inode.
- use PTR_ERR to return error value.
- register and uprobe_unregister share code.
(v5):
- Modified del_consumer as per comments from Peter.
- Drop reference to inode before dropping reference to uprobe.
- Use i_size_read(inode) instead of inode->i_size.
- Ensure uprobe->consumers is NULL, before __uprobe_unregister() is called.
- Includes errno.h as recommended by Stephen Rothwell to fix a build issue
on sparc defconfig
- Remove restrictions while unregistering.
- Earlier code leaked inode references under some conditions while
registering/unregistering.
- Continue the vma-rmap walk even if the intermediate vma doesnt
meet the requirements.
- Validate the vma found by find_vma before inserting/removing the
breakpoint
- Call del_consumer under mutex_lock.
- Use hash locks.
- Handle mremap.
- Introduce find_least_offset_node() instead of close match logic in
find_uprobe
- Uprobes no more depends on MM_OWNER; No reference to task_structs
while inserting/removing a probe.
- Uses read_mapping_page instead of grab_cache_page so that the pages
have valid content.
- pass NULL to get_user_pages for the task parameter.
- call SetPageUptodate on the new page allocated in write_opcode.
- fix leaking a reference to the new page under certain conditions.
- Include Instruction Decoder if Uprobes gets defined.
- Remove const attributes for instruction prefix arrays.
- Uses mm_context to know if the application is 32 bit.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Also-written-by: Jim Keniston <jkenisto@us.ibm.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Roland McGrath <roland@hack.frob.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Anton Arapov <anton@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Denys Vlasenko <vda.linux@googlemail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linux-mm <linux-mm@kvack.org>
Link: http://lkml.kernel.org/r/20120209092642.GE16600@linux.vnet.ibm.com
[ Made various small edits to the commit log ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-02-09 17:26:42 +08:00
|
|
|
*/
|
2012-02-17 16:27:41 +08:00
|
|
|
int uprobe_mmap(struct vm_area_struct *vma)
|
uprobes, mm, x86: Add the ability to install and remove uprobes breakpoints
Add uprobes support to the core kernel, with x86 support.
This commit adds the kernel facilities, the actual uprobes
user-space ABI and perf probe support comes in later commits.
General design:
Uprobes are maintained in an rb-tree indexed by inode and offset
(the offset here is from the start of the mapping). For a unique
(inode, offset) tuple, there can be at most one uprobe in the
rb-tree.
Since the (inode, offset) tuple identifies a unique uprobe, more
than one user may be interested in the same uprobe. This provides
the ability to connect multiple 'consumers' to the same uprobe.
Each consumer defines a handler and a filter (optional). The
'handler' is run every time the uprobe is hit, if it matches the
'filter' criteria.
The first consumer of a uprobe causes the breakpoint to be
inserted at the specified address and subsequent consumers are
appended to this list. On subsequent probes, the consumer gets
appended to the existing list of consumers. The breakpoint is
removed when the last consumer unregisters. For all other
unregisterations, the consumer is removed from the list of
consumers.
Given a inode, we get a list of the mms that have mapped the
inode. Do the actual registration if mm maps the page where a
probe needs to be inserted/removed.
We use a temporary list to walk through the vmas that map the
inode.
- The number of maps that map the inode, is not known before we
walk the rmap and keeps changing.
- extending vm_area_struct wasn't recommended, it's a
size-critical data structure.
- There can be more than one maps of the inode in the same mm.
We add callbacks to the mmap methods to keep an eye on text vmas
that are of interest to uprobes. When a vma of interest is mapped,
we insert the breakpoint at the right address.
Uprobe works by replacing the instruction at the address defined
by (inode, offset) with the arch specific breakpoint
instruction. We save a copy of the original instruction at the
uprobed address.
This is needed for:
a. executing the instruction out-of-line (xol).
b. instruction analysis for any subsequent fixups.
c. restoring the instruction back when the uprobe is unregistered.
We insert or delete a breakpoint instruction, and this
breakpoint instruction is assumed to be the smallest instruction
available on the platform. For fixed size instruction platforms
this is trivially true, for variable size instruction platforms
the breakpoint instruction is typically the smallest (often a
single byte).
Writing the instruction is done by COWing the page and changing
the instruction during the copy, this even though most platforms
allow atomic writes of the breakpoint instruction. This also
mirrors the behaviour of a ptrace() memory write to a PRIVATE
file map.
The core worker is derived from KSM's replace_page() logic.
In essence, similar to KSM:
a. allocate a new page and copy over contents of the page that
has the uprobed vaddr
b. modify the copy and insert the breakpoint at the required
address
c. switch the original page with the copy containing the
breakpoint
d. flush page tables.
replace_page() is being replicated here because of some minor
changes in the type of pages and also because Hugh Dickins had
plans to improve replace_page() for KSM specific work.
Instruction analysis on x86 is based on instruction decoder and
determines if an instruction can be probed and determines the
necessary fixups after singlestep. Instruction analysis is done
at probe insertion time so that we avoid having to repeat the
same analysis every time a probe is hit.
A lot of code here is due to the improvement/suggestions/inputs
from Peter Zijlstra.
Changelog:
(v10):
- Add code to clear REX.B prefix as suggested by Denys Vlasenko
and Masami Hiramatsu.
(v9):
- Use insn_offset_modrm as suggested by Masami Hiramatsu.
(v7):
Handle comments from Peter Zijlstra:
- Dont take reference to inode. (expect inode to uprobe_register to be sane).
- Use PTR_ERR to set the return value.
- No need to take reference to inode.
- use PTR_ERR to return error value.
- register and uprobe_unregister share code.
(v5):
- Modified del_consumer as per comments from Peter.
- Drop reference to inode before dropping reference to uprobe.
- Use i_size_read(inode) instead of inode->i_size.
- Ensure uprobe->consumers is NULL, before __uprobe_unregister() is called.
- Includes errno.h as recommended by Stephen Rothwell to fix a build issue
on sparc defconfig
- Remove restrictions while unregistering.
- Earlier code leaked inode references under some conditions while
registering/unregistering.
- Continue the vma-rmap walk even if the intermediate vma doesnt
meet the requirements.
- Validate the vma found by find_vma before inserting/removing the
breakpoint
- Call del_consumer under mutex_lock.
- Use hash locks.
- Handle mremap.
- Introduce find_least_offset_node() instead of close match logic in
find_uprobe
- Uprobes no more depends on MM_OWNER; No reference to task_structs
while inserting/removing a probe.
- Uses read_mapping_page instead of grab_cache_page so that the pages
have valid content.
- pass NULL to get_user_pages for the task parameter.
- call SetPageUptodate on the new page allocated in write_opcode.
- fix leaking a reference to the new page under certain conditions.
- Include Instruction Decoder if Uprobes gets defined.
- Remove const attributes for instruction prefix arrays.
- Uses mm_context to know if the application is 32 bit.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Also-written-by: Jim Keniston <jkenisto@us.ibm.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Roland McGrath <roland@hack.frob.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Anton Arapov <anton@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Denys Vlasenko <vda.linux@googlemail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linux-mm <linux-mm@kvack.org>
Link: http://lkml.kernel.org/r/20120209092642.GE16600@linux.vnet.ibm.com
[ Made various small edits to the commit log ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-02-09 17:26:42 +08:00
|
|
|
{
|
|
|
|
struct list_head tmp_list;
|
2012-07-30 02:22:29 +08:00
|
|
|
struct uprobe *uprobe, *u;
|
uprobes, mm, x86: Add the ability to install and remove uprobes breakpoints
Add uprobes support to the core kernel, with x86 support.
This commit adds the kernel facilities, the actual uprobes
user-space ABI and perf probe support comes in later commits.
General design:
Uprobes are maintained in an rb-tree indexed by inode and offset
(the offset here is from the start of the mapping). For a unique
(inode, offset) tuple, there can be at most one uprobe in the
rb-tree.
Since the (inode, offset) tuple identifies a unique uprobe, more
than one user may be interested in the same uprobe. This provides
the ability to connect multiple 'consumers' to the same uprobe.
Each consumer defines a handler and a filter (optional). The
'handler' is run every time the uprobe is hit, if it matches the
'filter' criteria.
The first consumer of a uprobe causes the breakpoint to be
inserted at the specified address and subsequent consumers are
appended to this list. On subsequent probes, the consumer gets
appended to the existing list of consumers. The breakpoint is
removed when the last consumer unregisters. For all other
unregisterations, the consumer is removed from the list of
consumers.
Given a inode, we get a list of the mms that have mapped the
inode. Do the actual registration if mm maps the page where a
probe needs to be inserted/removed.
We use a temporary list to walk through the vmas that map the
inode.
- The number of maps that map the inode, is not known before we
walk the rmap and keeps changing.
- extending vm_area_struct wasn't recommended, it's a
size-critical data structure.
- There can be more than one maps of the inode in the same mm.
We add callbacks to the mmap methods to keep an eye on text vmas
that are of interest to uprobes. When a vma of interest is mapped,
we insert the breakpoint at the right address.
Uprobe works by replacing the instruction at the address defined
by (inode, offset) with the arch specific breakpoint
instruction. We save a copy of the original instruction at the
uprobed address.
This is needed for:
a. executing the instruction out-of-line (xol).
b. instruction analysis for any subsequent fixups.
c. restoring the instruction back when the uprobe is unregistered.
We insert or delete a breakpoint instruction, and this
breakpoint instruction is assumed to be the smallest instruction
available on the platform. For fixed size instruction platforms
this is trivially true, for variable size instruction platforms
the breakpoint instruction is typically the smallest (often a
single byte).
Writing the instruction is done by COWing the page and changing
the instruction during the copy, this even though most platforms
allow atomic writes of the breakpoint instruction. This also
mirrors the behaviour of a ptrace() memory write to a PRIVATE
file map.
The core worker is derived from KSM's replace_page() logic.
In essence, similar to KSM:
a. allocate a new page and copy over contents of the page that
has the uprobed vaddr
b. modify the copy and insert the breakpoint at the required
address
c. switch the original page with the copy containing the
breakpoint
d. flush page tables.
replace_page() is being replicated here because of some minor
changes in the type of pages and also because Hugh Dickins had
plans to improve replace_page() for KSM specific work.
Instruction analysis on x86 is based on instruction decoder and
determines if an instruction can be probed and determines the
necessary fixups after singlestep. Instruction analysis is done
at probe insertion time so that we avoid having to repeat the
same analysis every time a probe is hit.
A lot of code here is due to the improvement/suggestions/inputs
from Peter Zijlstra.
Changelog:
(v10):
- Add code to clear REX.B prefix as suggested by Denys Vlasenko
and Masami Hiramatsu.
(v9):
- Use insn_offset_modrm as suggested by Masami Hiramatsu.
(v7):
Handle comments from Peter Zijlstra:
- Dont take reference to inode. (expect inode to uprobe_register to be sane).
- Use PTR_ERR to set the return value.
- No need to take reference to inode.
- use PTR_ERR to return error value.
- register and uprobe_unregister share code.
(v5):
- Modified del_consumer as per comments from Peter.
- Drop reference to inode before dropping reference to uprobe.
- Use i_size_read(inode) instead of inode->i_size.
- Ensure uprobe->consumers is NULL, before __uprobe_unregister() is called.
- Includes errno.h as recommended by Stephen Rothwell to fix a build issue
on sparc defconfig
- Remove restrictions while unregistering.
- Earlier code leaked inode references under some conditions while
registering/unregistering.
- Continue the vma-rmap walk even if the intermediate vma doesnt
meet the requirements.
- Validate the vma found by find_vma before inserting/removing the
breakpoint
- Call del_consumer under mutex_lock.
- Use hash locks.
- Handle mremap.
- Introduce find_least_offset_node() instead of close match logic in
find_uprobe
- Uprobes no more depends on MM_OWNER; No reference to task_structs
while inserting/removing a probe.
- Uses read_mapping_page instead of grab_cache_page so that the pages
have valid content.
- pass NULL to get_user_pages for the task parameter.
- call SetPageUptodate on the new page allocated in write_opcode.
- fix leaking a reference to the new page under certain conditions.
- Include Instruction Decoder if Uprobes gets defined.
- Remove const attributes for instruction prefix arrays.
- Uses mm_context to know if the application is 32 bit.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Also-written-by: Jim Keniston <jkenisto@us.ibm.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Roland McGrath <roland@hack.frob.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Anton Arapov <anton@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Denys Vlasenko <vda.linux@googlemail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linux-mm <linux-mm@kvack.org>
Link: http://lkml.kernel.org/r/20120209092642.GE16600@linux.vnet.ibm.com
[ Made various small edits to the commit log ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-02-09 17:26:42 +08:00
|
|
|
struct inode *inode;
|
|
|
|
|
uprobes: Support SDT markers having reference count (semaphore)
Userspace Statically Defined Tracepoints[1] are dtrace style markers
inside userspace applications. Applications like PostgreSQL, MySQL,
Pthread, Perl, Python, Java, Ruby, Node.js, libvirt, QEMU, glib etc
have these markers embedded in them. These markers are added by developer
at important places in the code. Each marker source expands to a single
nop instruction in the compiled code but there may be additional
overhead for computing the marker arguments which expands to couple of
instructions. In case the overhead is more, execution of it can be
omitted by runtime if() condition when no one is tracing on the marker:
if (reference_counter > 0) {
Execute marker instructions;
}
Default value of reference counter is 0. Tracer has to increment the
reference counter before tracing on a marker and decrement it when
done with the tracing.
Implement the reference counter logic in core uprobe. User will be
able to use it from trace_uprobe as well as from kernel module. New
trace_uprobe definition with reference counter will now be:
<path>:<offset>[(ref_ctr_offset)]
where ref_ctr_offset is an optional field. For kernel module, new
variant of uprobe_register() has been introduced:
uprobe_register_refctr(inode, offset, ref_ctr_offset, consumer)
No new variant for uprobe_unregister() because it's assumed to have
only one reference counter for one uprobe.
[1] https://sourceware.org/systemtap/wiki/UserSpaceProbeImplementation
Note: 'reference counter' is called as 'semaphore' in original Dtrace
(or Systemtap, bcc and even in ELF) documentation and code. But the
term 'semaphore' is misleading in this context. This is just a counter
used to hold number of tracers tracing on a marker. This is not really
used for any synchronization. So we are calling it a 'reference counter'
in kernel / perf code.
Link: http://lkml.kernel.org/r/20180820044250.11659-2-ravi.bangoria@linux.ibm.com
Reviewed-by: Masami Hiramatsu <mhiramat@kernel.org>
[Only trace_uprobe.c]
Reviewed-by: Oleg Nesterov <oleg@redhat.com>
Reviewed-by: Song Liu <songliubraving@fb.com>
Tested-by: Song Liu <songliubraving@fb.com>
Signed-off-by: Ravi Bangoria <ravi.bangoria@linux.ibm.com>
Signed-off-by: Steven Rostedt (VMware) <rostedt@goodmis.org>
2018-08-20 12:42:47 +08:00
|
|
|
if (no_uprobe_events())
|
|
|
|
return 0;
|
|
|
|
|
|
|
|
if (vma->vm_file &&
|
|
|
|
(vma->vm_flags & (VM_WRITE|VM_SHARED)) == VM_WRITE &&
|
|
|
|
test_bit(MMF_HAS_UPROBES, &vma->vm_mm->flags))
|
|
|
|
delayed_ref_ctr_inc(vma);
|
|
|
|
|
|
|
|
if (!valid_vma(vma, true))
|
2012-02-17 16:27:41 +08:00
|
|
|
return 0;
|
uprobes, mm, x86: Add the ability to install and remove uprobes breakpoints
Add uprobes support to the core kernel, with x86 support.
This commit adds the kernel facilities, the actual uprobes
user-space ABI and perf probe support comes in later commits.
General design:
Uprobes are maintained in an rb-tree indexed by inode and offset
(the offset here is from the start of the mapping). For a unique
(inode, offset) tuple, there can be at most one uprobe in the
rb-tree.
Since the (inode, offset) tuple identifies a unique uprobe, more
than one user may be interested in the same uprobe. This provides
the ability to connect multiple 'consumers' to the same uprobe.
Each consumer defines a handler and a filter (optional). The
'handler' is run every time the uprobe is hit, if it matches the
'filter' criteria.
The first consumer of a uprobe causes the breakpoint to be
inserted at the specified address and subsequent consumers are
appended to this list. On subsequent probes, the consumer gets
appended to the existing list of consumers. The breakpoint is
removed when the last consumer unregisters. For all other
unregisterations, the consumer is removed from the list of
consumers.
Given a inode, we get a list of the mms that have mapped the
inode. Do the actual registration if mm maps the page where a
probe needs to be inserted/removed.
We use a temporary list to walk through the vmas that map the
inode.
- The number of maps that map the inode, is not known before we
walk the rmap and keeps changing.
- extending vm_area_struct wasn't recommended, it's a
size-critical data structure.
- There can be more than one maps of the inode in the same mm.
We add callbacks to the mmap methods to keep an eye on text vmas
that are of interest to uprobes. When a vma of interest is mapped,
we insert the breakpoint at the right address.
Uprobe works by replacing the instruction at the address defined
by (inode, offset) with the arch specific breakpoint
instruction. We save a copy of the original instruction at the
uprobed address.
This is needed for:
a. executing the instruction out-of-line (xol).
b. instruction analysis for any subsequent fixups.
c. restoring the instruction back when the uprobe is unregistered.
We insert or delete a breakpoint instruction, and this
breakpoint instruction is assumed to be the smallest instruction
available on the platform. For fixed size instruction platforms
this is trivially true, for variable size instruction platforms
the breakpoint instruction is typically the smallest (often a
single byte).
Writing the instruction is done by COWing the page and changing
the instruction during the copy, this even though most platforms
allow atomic writes of the breakpoint instruction. This also
mirrors the behaviour of a ptrace() memory write to a PRIVATE
file map.
The core worker is derived from KSM's replace_page() logic.
In essence, similar to KSM:
a. allocate a new page and copy over contents of the page that
has the uprobed vaddr
b. modify the copy and insert the breakpoint at the required
address
c. switch the original page with the copy containing the
breakpoint
d. flush page tables.
replace_page() is being replicated here because of some minor
changes in the type of pages and also because Hugh Dickins had
plans to improve replace_page() for KSM specific work.
Instruction analysis on x86 is based on instruction decoder and
determines if an instruction can be probed and determines the
necessary fixups after singlestep. Instruction analysis is done
at probe insertion time so that we avoid having to repeat the
same analysis every time a probe is hit.
A lot of code here is due to the improvement/suggestions/inputs
from Peter Zijlstra.
Changelog:
(v10):
- Add code to clear REX.B prefix as suggested by Denys Vlasenko
and Masami Hiramatsu.
(v9):
- Use insn_offset_modrm as suggested by Masami Hiramatsu.
(v7):
Handle comments from Peter Zijlstra:
- Dont take reference to inode. (expect inode to uprobe_register to be sane).
- Use PTR_ERR to set the return value.
- No need to take reference to inode.
- use PTR_ERR to return error value.
- register and uprobe_unregister share code.
(v5):
- Modified del_consumer as per comments from Peter.
- Drop reference to inode before dropping reference to uprobe.
- Use i_size_read(inode) instead of inode->i_size.
- Ensure uprobe->consumers is NULL, before __uprobe_unregister() is called.
- Includes errno.h as recommended by Stephen Rothwell to fix a build issue
on sparc defconfig
- Remove restrictions while unregistering.
- Earlier code leaked inode references under some conditions while
registering/unregistering.
- Continue the vma-rmap walk even if the intermediate vma doesnt
meet the requirements.
- Validate the vma found by find_vma before inserting/removing the
breakpoint
- Call del_consumer under mutex_lock.
- Use hash locks.
- Handle mremap.
- Introduce find_least_offset_node() instead of close match logic in
find_uprobe
- Uprobes no more depends on MM_OWNER; No reference to task_structs
while inserting/removing a probe.
- Uses read_mapping_page instead of grab_cache_page so that the pages
have valid content.
- pass NULL to get_user_pages for the task parameter.
- call SetPageUptodate on the new page allocated in write_opcode.
- fix leaking a reference to the new page under certain conditions.
- Include Instruction Decoder if Uprobes gets defined.
- Remove const attributes for instruction prefix arrays.
- Uses mm_context to know if the application is 32 bit.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Also-written-by: Jim Keniston <jkenisto@us.ibm.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Roland McGrath <roland@hack.frob.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Anton Arapov <anton@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Denys Vlasenko <vda.linux@googlemail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linux-mm <linux-mm@kvack.org>
Link: http://lkml.kernel.org/r/20120209092642.GE16600@linux.vnet.ibm.com
[ Made various small edits to the commit log ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-02-09 17:26:42 +08:00
|
|
|
|
2013-03-18 01:54:44 +08:00
|
|
|
inode = file_inode(vma->vm_file);
|
uprobes, mm, x86: Add the ability to install and remove uprobes breakpoints
Add uprobes support to the core kernel, with x86 support.
This commit adds the kernel facilities, the actual uprobes
user-space ABI and perf probe support comes in later commits.
General design:
Uprobes are maintained in an rb-tree indexed by inode and offset
(the offset here is from the start of the mapping). For a unique
(inode, offset) tuple, there can be at most one uprobe in the
rb-tree.
Since the (inode, offset) tuple identifies a unique uprobe, more
than one user may be interested in the same uprobe. This provides
the ability to connect multiple 'consumers' to the same uprobe.
Each consumer defines a handler and a filter (optional). The
'handler' is run every time the uprobe is hit, if it matches the
'filter' criteria.
The first consumer of a uprobe causes the breakpoint to be
inserted at the specified address and subsequent consumers are
appended to this list. On subsequent probes, the consumer gets
appended to the existing list of consumers. The breakpoint is
removed when the last consumer unregisters. For all other
unregisterations, the consumer is removed from the list of
consumers.
Given a inode, we get a list of the mms that have mapped the
inode. Do the actual registration if mm maps the page where a
probe needs to be inserted/removed.
We use a temporary list to walk through the vmas that map the
inode.
- The number of maps that map the inode, is not known before we
walk the rmap and keeps changing.
- extending vm_area_struct wasn't recommended, it's a
size-critical data structure.
- There can be more than one maps of the inode in the same mm.
We add callbacks to the mmap methods to keep an eye on text vmas
that are of interest to uprobes. When a vma of interest is mapped,
we insert the breakpoint at the right address.
Uprobe works by replacing the instruction at the address defined
by (inode, offset) with the arch specific breakpoint
instruction. We save a copy of the original instruction at the
uprobed address.
This is needed for:
a. executing the instruction out-of-line (xol).
b. instruction analysis for any subsequent fixups.
c. restoring the instruction back when the uprobe is unregistered.
We insert or delete a breakpoint instruction, and this
breakpoint instruction is assumed to be the smallest instruction
available on the platform. For fixed size instruction platforms
this is trivially true, for variable size instruction platforms
the breakpoint instruction is typically the smallest (often a
single byte).
Writing the instruction is done by COWing the page and changing
the instruction during the copy, this even though most platforms
allow atomic writes of the breakpoint instruction. This also
mirrors the behaviour of a ptrace() memory write to a PRIVATE
file map.
The core worker is derived from KSM's replace_page() logic.
In essence, similar to KSM:
a. allocate a new page and copy over contents of the page that
has the uprobed vaddr
b. modify the copy and insert the breakpoint at the required
address
c. switch the original page with the copy containing the
breakpoint
d. flush page tables.
replace_page() is being replicated here because of some minor
changes in the type of pages and also because Hugh Dickins had
plans to improve replace_page() for KSM specific work.
Instruction analysis on x86 is based on instruction decoder and
determines if an instruction can be probed and determines the
necessary fixups after singlestep. Instruction analysis is done
at probe insertion time so that we avoid having to repeat the
same analysis every time a probe is hit.
A lot of code here is due to the improvement/suggestions/inputs
from Peter Zijlstra.
Changelog:
(v10):
- Add code to clear REX.B prefix as suggested by Denys Vlasenko
and Masami Hiramatsu.
(v9):
- Use insn_offset_modrm as suggested by Masami Hiramatsu.
(v7):
Handle comments from Peter Zijlstra:
- Dont take reference to inode. (expect inode to uprobe_register to be sane).
- Use PTR_ERR to set the return value.
- No need to take reference to inode.
- use PTR_ERR to return error value.
- register and uprobe_unregister share code.
(v5):
- Modified del_consumer as per comments from Peter.
- Drop reference to inode before dropping reference to uprobe.
- Use i_size_read(inode) instead of inode->i_size.
- Ensure uprobe->consumers is NULL, before __uprobe_unregister() is called.
- Includes errno.h as recommended by Stephen Rothwell to fix a build issue
on sparc defconfig
- Remove restrictions while unregistering.
- Earlier code leaked inode references under some conditions while
registering/unregistering.
- Continue the vma-rmap walk even if the intermediate vma doesnt
meet the requirements.
- Validate the vma found by find_vma before inserting/removing the
breakpoint
- Call del_consumer under mutex_lock.
- Use hash locks.
- Handle mremap.
- Introduce find_least_offset_node() instead of close match logic in
find_uprobe
- Uprobes no more depends on MM_OWNER; No reference to task_structs
while inserting/removing a probe.
- Uses read_mapping_page instead of grab_cache_page so that the pages
have valid content.
- pass NULL to get_user_pages for the task parameter.
- call SetPageUptodate on the new page allocated in write_opcode.
- fix leaking a reference to the new page under certain conditions.
- Include Instruction Decoder if Uprobes gets defined.
- Remove const attributes for instruction prefix arrays.
- Uses mm_context to know if the application is 32 bit.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Also-written-by: Jim Keniston <jkenisto@us.ibm.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Roland McGrath <roland@hack.frob.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Anton Arapov <anton@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Denys Vlasenko <vda.linux@googlemail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linux-mm <linux-mm@kvack.org>
Link: http://lkml.kernel.org/r/20120209092642.GE16600@linux.vnet.ibm.com
[ Made various small edits to the commit log ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-02-09 17:26:42 +08:00
|
|
|
if (!inode)
|
2012-02-17 16:27:41 +08:00
|
|
|
return 0;
|
uprobes, mm, x86: Add the ability to install and remove uprobes breakpoints
Add uprobes support to the core kernel, with x86 support.
This commit adds the kernel facilities, the actual uprobes
user-space ABI and perf probe support comes in later commits.
General design:
Uprobes are maintained in an rb-tree indexed by inode and offset
(the offset here is from the start of the mapping). For a unique
(inode, offset) tuple, there can be at most one uprobe in the
rb-tree.
Since the (inode, offset) tuple identifies a unique uprobe, more
than one user may be interested in the same uprobe. This provides
the ability to connect multiple 'consumers' to the same uprobe.
Each consumer defines a handler and a filter (optional). The
'handler' is run every time the uprobe is hit, if it matches the
'filter' criteria.
The first consumer of a uprobe causes the breakpoint to be
inserted at the specified address and subsequent consumers are
appended to this list. On subsequent probes, the consumer gets
appended to the existing list of consumers. The breakpoint is
removed when the last consumer unregisters. For all other
unregisterations, the consumer is removed from the list of
consumers.
Given a inode, we get a list of the mms that have mapped the
inode. Do the actual registration if mm maps the page where a
probe needs to be inserted/removed.
We use a temporary list to walk through the vmas that map the
inode.
- The number of maps that map the inode, is not known before we
walk the rmap and keeps changing.
- extending vm_area_struct wasn't recommended, it's a
size-critical data structure.
- There can be more than one maps of the inode in the same mm.
We add callbacks to the mmap methods to keep an eye on text vmas
that are of interest to uprobes. When a vma of interest is mapped,
we insert the breakpoint at the right address.
Uprobe works by replacing the instruction at the address defined
by (inode, offset) with the arch specific breakpoint
instruction. We save a copy of the original instruction at the
uprobed address.
This is needed for:
a. executing the instruction out-of-line (xol).
b. instruction analysis for any subsequent fixups.
c. restoring the instruction back when the uprobe is unregistered.
We insert or delete a breakpoint instruction, and this
breakpoint instruction is assumed to be the smallest instruction
available on the platform. For fixed size instruction platforms
this is trivially true, for variable size instruction platforms
the breakpoint instruction is typically the smallest (often a
single byte).
Writing the instruction is done by COWing the page and changing
the instruction during the copy, this even though most platforms
allow atomic writes of the breakpoint instruction. This also
mirrors the behaviour of a ptrace() memory write to a PRIVATE
file map.
The core worker is derived from KSM's replace_page() logic.
In essence, similar to KSM:
a. allocate a new page and copy over contents of the page that
has the uprobed vaddr
b. modify the copy and insert the breakpoint at the required
address
c. switch the original page with the copy containing the
breakpoint
d. flush page tables.
replace_page() is being replicated here because of some minor
changes in the type of pages and also because Hugh Dickins had
plans to improve replace_page() for KSM specific work.
Instruction analysis on x86 is based on instruction decoder and
determines if an instruction can be probed and determines the
necessary fixups after singlestep. Instruction analysis is done
at probe insertion time so that we avoid having to repeat the
same analysis every time a probe is hit.
A lot of code here is due to the improvement/suggestions/inputs
from Peter Zijlstra.
Changelog:
(v10):
- Add code to clear REX.B prefix as suggested by Denys Vlasenko
and Masami Hiramatsu.
(v9):
- Use insn_offset_modrm as suggested by Masami Hiramatsu.
(v7):
Handle comments from Peter Zijlstra:
- Dont take reference to inode. (expect inode to uprobe_register to be sane).
- Use PTR_ERR to set the return value.
- No need to take reference to inode.
- use PTR_ERR to return error value.
- register and uprobe_unregister share code.
(v5):
- Modified del_consumer as per comments from Peter.
- Drop reference to inode before dropping reference to uprobe.
- Use i_size_read(inode) instead of inode->i_size.
- Ensure uprobe->consumers is NULL, before __uprobe_unregister() is called.
- Includes errno.h as recommended by Stephen Rothwell to fix a build issue
on sparc defconfig
- Remove restrictions while unregistering.
- Earlier code leaked inode references under some conditions while
registering/unregistering.
- Continue the vma-rmap walk even if the intermediate vma doesnt
meet the requirements.
- Validate the vma found by find_vma before inserting/removing the
breakpoint
- Call del_consumer under mutex_lock.
- Use hash locks.
- Handle mremap.
- Introduce find_least_offset_node() instead of close match logic in
find_uprobe
- Uprobes no more depends on MM_OWNER; No reference to task_structs
while inserting/removing a probe.
- Uses read_mapping_page instead of grab_cache_page so that the pages
have valid content.
- pass NULL to get_user_pages for the task parameter.
- call SetPageUptodate on the new page allocated in write_opcode.
- fix leaking a reference to the new page under certain conditions.
- Include Instruction Decoder if Uprobes gets defined.
- Remove const attributes for instruction prefix arrays.
- Uses mm_context to know if the application is 32 bit.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Also-written-by: Jim Keniston <jkenisto@us.ibm.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Roland McGrath <roland@hack.frob.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Anton Arapov <anton@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Denys Vlasenko <vda.linux@googlemail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linux-mm <linux-mm@kvack.org>
Link: http://lkml.kernel.org/r/20120209092642.GE16600@linux.vnet.ibm.com
[ Made various small edits to the commit log ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-02-09 17:26:42 +08:00
|
|
|
|
|
|
|
mutex_lock(uprobes_mmap_hash(inode));
|
2012-07-30 02:22:40 +08:00
|
|
|
build_probe_list(inode, vma, vma->vm_start, vma->vm_end, &tmp_list);
|
2012-12-28 01:21:11 +08:00
|
|
|
/*
|
|
|
|
* We can race with uprobe_unregister(), this uprobe can be already
|
|
|
|
* removed. But in this case filter_chain() must return false, all
|
|
|
|
* consumers have gone away.
|
|
|
|
*/
|
2012-07-30 02:22:29 +08:00
|
|
|
list_for_each_entry_safe(uprobe, u, &tmp_list, pending_list) {
|
2012-12-28 01:21:11 +08:00
|
|
|
if (!fatal_signal_pending(current) &&
|
2012-12-29 00:58:38 +08:00
|
|
|
filter_chain(uprobe, UPROBE_FILTER_MMAP, vma->vm_mm)) {
|
2012-07-30 02:22:47 +08:00
|
|
|
unsigned long vaddr = offset_to_vaddr(vma, uprobe->offset);
|
2012-08-06 20:49:56 +08:00
|
|
|
install_breakpoint(uprobe, vma->vm_mm, vma, vaddr);
|
uprobes, mm, x86: Add the ability to install and remove uprobes breakpoints
Add uprobes support to the core kernel, with x86 support.
This commit adds the kernel facilities, the actual uprobes
user-space ABI and perf probe support comes in later commits.
General design:
Uprobes are maintained in an rb-tree indexed by inode and offset
(the offset here is from the start of the mapping). For a unique
(inode, offset) tuple, there can be at most one uprobe in the
rb-tree.
Since the (inode, offset) tuple identifies a unique uprobe, more
than one user may be interested in the same uprobe. This provides
the ability to connect multiple 'consumers' to the same uprobe.
Each consumer defines a handler and a filter (optional). The
'handler' is run every time the uprobe is hit, if it matches the
'filter' criteria.
The first consumer of a uprobe causes the breakpoint to be
inserted at the specified address and subsequent consumers are
appended to this list. On subsequent probes, the consumer gets
appended to the existing list of consumers. The breakpoint is
removed when the last consumer unregisters. For all other
unregisterations, the consumer is removed from the list of
consumers.
Given a inode, we get a list of the mms that have mapped the
inode. Do the actual registration if mm maps the page where a
probe needs to be inserted/removed.
We use a temporary list to walk through the vmas that map the
inode.
- The number of maps that map the inode, is not known before we
walk the rmap and keeps changing.
- extending vm_area_struct wasn't recommended, it's a
size-critical data structure.
- There can be more than one maps of the inode in the same mm.
We add callbacks to the mmap methods to keep an eye on text vmas
that are of interest to uprobes. When a vma of interest is mapped,
we insert the breakpoint at the right address.
Uprobe works by replacing the instruction at the address defined
by (inode, offset) with the arch specific breakpoint
instruction. We save a copy of the original instruction at the
uprobed address.
This is needed for:
a. executing the instruction out-of-line (xol).
b. instruction analysis for any subsequent fixups.
c. restoring the instruction back when the uprobe is unregistered.
We insert or delete a breakpoint instruction, and this
breakpoint instruction is assumed to be the smallest instruction
available on the platform. For fixed size instruction platforms
this is trivially true, for variable size instruction platforms
the breakpoint instruction is typically the smallest (often a
single byte).
Writing the instruction is done by COWing the page and changing
the instruction during the copy, this even though most platforms
allow atomic writes of the breakpoint instruction. This also
mirrors the behaviour of a ptrace() memory write to a PRIVATE
file map.
The core worker is derived from KSM's replace_page() logic.
In essence, similar to KSM:
a. allocate a new page and copy over contents of the page that
has the uprobed vaddr
b. modify the copy and insert the breakpoint at the required
address
c. switch the original page with the copy containing the
breakpoint
d. flush page tables.
replace_page() is being replicated here because of some minor
changes in the type of pages and also because Hugh Dickins had
plans to improve replace_page() for KSM specific work.
Instruction analysis on x86 is based on instruction decoder and
determines if an instruction can be probed and determines the
necessary fixups after singlestep. Instruction analysis is done
at probe insertion time so that we avoid having to repeat the
same analysis every time a probe is hit.
A lot of code here is due to the improvement/suggestions/inputs
from Peter Zijlstra.
Changelog:
(v10):
- Add code to clear REX.B prefix as suggested by Denys Vlasenko
and Masami Hiramatsu.
(v9):
- Use insn_offset_modrm as suggested by Masami Hiramatsu.
(v7):
Handle comments from Peter Zijlstra:
- Dont take reference to inode. (expect inode to uprobe_register to be sane).
- Use PTR_ERR to set the return value.
- No need to take reference to inode.
- use PTR_ERR to return error value.
- register and uprobe_unregister share code.
(v5):
- Modified del_consumer as per comments from Peter.
- Drop reference to inode before dropping reference to uprobe.
- Use i_size_read(inode) instead of inode->i_size.
- Ensure uprobe->consumers is NULL, before __uprobe_unregister() is called.
- Includes errno.h as recommended by Stephen Rothwell to fix a build issue
on sparc defconfig
- Remove restrictions while unregistering.
- Earlier code leaked inode references under some conditions while
registering/unregistering.
- Continue the vma-rmap walk even if the intermediate vma doesnt
meet the requirements.
- Validate the vma found by find_vma before inserting/removing the
breakpoint
- Call del_consumer under mutex_lock.
- Use hash locks.
- Handle mremap.
- Introduce find_least_offset_node() instead of close match logic in
find_uprobe
- Uprobes no more depends on MM_OWNER; No reference to task_structs
while inserting/removing a probe.
- Uses read_mapping_page instead of grab_cache_page so that the pages
have valid content.
- pass NULL to get_user_pages for the task parameter.
- call SetPageUptodate on the new page allocated in write_opcode.
- fix leaking a reference to the new page under certain conditions.
- Include Instruction Decoder if Uprobes gets defined.
- Remove const attributes for instruction prefix arrays.
- Uses mm_context to know if the application is 32 bit.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Also-written-by: Jim Keniston <jkenisto@us.ibm.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Roland McGrath <roland@hack.frob.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Anton Arapov <anton@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Denys Vlasenko <vda.linux@googlemail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linux-mm <linux-mm@kvack.org>
Link: http://lkml.kernel.org/r/20120209092642.GE16600@linux.vnet.ibm.com
[ Made various small edits to the commit log ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-02-09 17:26:42 +08:00
|
|
|
}
|
|
|
|
put_uprobe(uprobe);
|
|
|
|
}
|
|
|
|
mutex_unlock(uprobes_mmap_hash(inode));
|
|
|
|
|
2012-08-06 20:49:56 +08:00
|
|
|
return 0;
|
uprobes, mm, x86: Add the ability to install and remove uprobes breakpoints
Add uprobes support to the core kernel, with x86 support.
This commit adds the kernel facilities, the actual uprobes
user-space ABI and perf probe support comes in later commits.
General design:
Uprobes are maintained in an rb-tree indexed by inode and offset
(the offset here is from the start of the mapping). For a unique
(inode, offset) tuple, there can be at most one uprobe in the
rb-tree.
Since the (inode, offset) tuple identifies a unique uprobe, more
than one user may be interested in the same uprobe. This provides
the ability to connect multiple 'consumers' to the same uprobe.
Each consumer defines a handler and a filter (optional). The
'handler' is run every time the uprobe is hit, if it matches the
'filter' criteria.
The first consumer of a uprobe causes the breakpoint to be
inserted at the specified address and subsequent consumers are
appended to this list. On subsequent probes, the consumer gets
appended to the existing list of consumers. The breakpoint is
removed when the last consumer unregisters. For all other
unregisterations, the consumer is removed from the list of
consumers.
Given a inode, we get a list of the mms that have mapped the
inode. Do the actual registration if mm maps the page where a
probe needs to be inserted/removed.
We use a temporary list to walk through the vmas that map the
inode.
- The number of maps that map the inode, is not known before we
walk the rmap and keeps changing.
- extending vm_area_struct wasn't recommended, it's a
size-critical data structure.
- There can be more than one maps of the inode in the same mm.
We add callbacks to the mmap methods to keep an eye on text vmas
that are of interest to uprobes. When a vma of interest is mapped,
we insert the breakpoint at the right address.
Uprobe works by replacing the instruction at the address defined
by (inode, offset) with the arch specific breakpoint
instruction. We save a copy of the original instruction at the
uprobed address.
This is needed for:
a. executing the instruction out-of-line (xol).
b. instruction analysis for any subsequent fixups.
c. restoring the instruction back when the uprobe is unregistered.
We insert or delete a breakpoint instruction, and this
breakpoint instruction is assumed to be the smallest instruction
available on the platform. For fixed size instruction platforms
this is trivially true, for variable size instruction platforms
the breakpoint instruction is typically the smallest (often a
single byte).
Writing the instruction is done by COWing the page and changing
the instruction during the copy, this even though most platforms
allow atomic writes of the breakpoint instruction. This also
mirrors the behaviour of a ptrace() memory write to a PRIVATE
file map.
The core worker is derived from KSM's replace_page() logic.
In essence, similar to KSM:
a. allocate a new page and copy over contents of the page that
has the uprobed vaddr
b. modify the copy and insert the breakpoint at the required
address
c. switch the original page with the copy containing the
breakpoint
d. flush page tables.
replace_page() is being replicated here because of some minor
changes in the type of pages and also because Hugh Dickins had
plans to improve replace_page() for KSM specific work.
Instruction analysis on x86 is based on instruction decoder and
determines if an instruction can be probed and determines the
necessary fixups after singlestep. Instruction analysis is done
at probe insertion time so that we avoid having to repeat the
same analysis every time a probe is hit.
A lot of code here is due to the improvement/suggestions/inputs
from Peter Zijlstra.
Changelog:
(v10):
- Add code to clear REX.B prefix as suggested by Denys Vlasenko
and Masami Hiramatsu.
(v9):
- Use insn_offset_modrm as suggested by Masami Hiramatsu.
(v7):
Handle comments from Peter Zijlstra:
- Dont take reference to inode. (expect inode to uprobe_register to be sane).
- Use PTR_ERR to set the return value.
- No need to take reference to inode.
- use PTR_ERR to return error value.
- register and uprobe_unregister share code.
(v5):
- Modified del_consumer as per comments from Peter.
- Drop reference to inode before dropping reference to uprobe.
- Use i_size_read(inode) instead of inode->i_size.
- Ensure uprobe->consumers is NULL, before __uprobe_unregister() is called.
- Includes errno.h as recommended by Stephen Rothwell to fix a build issue
on sparc defconfig
- Remove restrictions while unregistering.
- Earlier code leaked inode references under some conditions while
registering/unregistering.
- Continue the vma-rmap walk even if the intermediate vma doesnt
meet the requirements.
- Validate the vma found by find_vma before inserting/removing the
breakpoint
- Call del_consumer under mutex_lock.
- Use hash locks.
- Handle mremap.
- Introduce find_least_offset_node() instead of close match logic in
find_uprobe
- Uprobes no more depends on MM_OWNER; No reference to task_structs
while inserting/removing a probe.
- Uses read_mapping_page instead of grab_cache_page so that the pages
have valid content.
- pass NULL to get_user_pages for the task parameter.
- call SetPageUptodate on the new page allocated in write_opcode.
- fix leaking a reference to the new page under certain conditions.
- Include Instruction Decoder if Uprobes gets defined.
- Remove const attributes for instruction prefix arrays.
- Uses mm_context to know if the application is 32 bit.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Also-written-by: Jim Keniston <jkenisto@us.ibm.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Roland McGrath <roland@hack.frob.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Anton Arapov <anton@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Denys Vlasenko <vda.linux@googlemail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linux-mm <linux-mm@kvack.org>
Link: http://lkml.kernel.org/r/20120209092642.GE16600@linux.vnet.ibm.com
[ Made various small edits to the commit log ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-02-09 17:26:42 +08:00
|
|
|
}
|
|
|
|
|
2012-08-19 22:15:09 +08:00
|
|
|
static bool
|
|
|
|
vma_has_uprobes(struct vm_area_struct *vma, unsigned long start, unsigned long end)
|
|
|
|
{
|
|
|
|
loff_t min, max;
|
|
|
|
struct inode *inode;
|
|
|
|
struct rb_node *n;
|
|
|
|
|
2013-03-18 01:54:44 +08:00
|
|
|
inode = file_inode(vma->vm_file);
|
2012-08-19 22:15:09 +08:00
|
|
|
|
|
|
|
min = vaddr_to_offset(vma, start);
|
|
|
|
max = min + (end - start) - 1;
|
|
|
|
|
|
|
|
spin_lock(&uprobes_treelock);
|
|
|
|
n = find_node_in_range(inode, min, max);
|
|
|
|
spin_unlock(&uprobes_treelock);
|
|
|
|
|
|
|
|
return !!n;
|
|
|
|
}
|
|
|
|
|
2012-03-31 02:26:46 +08:00
|
|
|
/*
|
|
|
|
* Called in context of a munmap of a vma.
|
|
|
|
*/
|
2012-04-11 18:35:27 +08:00
|
|
|
void uprobe_munmap(struct vm_area_struct *vma, unsigned long start, unsigned long end)
|
2012-03-31 02:26:46 +08:00
|
|
|
{
|
2012-11-26 02:54:29 +08:00
|
|
|
if (no_uprobe_events() || !valid_vma(vma, false))
|
2012-03-31 02:26:46 +08:00
|
|
|
return;
|
|
|
|
|
2012-07-30 02:22:31 +08:00
|
|
|
if (!atomic_read(&vma->vm_mm->mm_users)) /* called by mmput() ? */
|
|
|
|
return;
|
|
|
|
|
2012-08-19 22:15:09 +08:00
|
|
|
if (!test_bit(MMF_HAS_UPROBES, &vma->vm_mm->flags) ||
|
|
|
|
test_bit(MMF_RECALC_UPROBES, &vma->vm_mm->flags))
|
2012-08-08 23:11:42 +08:00
|
|
|
return;
|
|
|
|
|
2012-08-19 22:15:09 +08:00
|
|
|
if (vma_has_uprobes(vma, start, end))
|
|
|
|
set_bit(MMF_RECALC_UPROBES, &vma->vm_mm->flags);
|
2012-03-31 02:26:46 +08:00
|
|
|
}
|
|
|
|
|
2012-03-31 02:26:31 +08:00
|
|
|
/* Slot allocation for XOL */
|
2013-10-14 03:18:35 +08:00
|
|
|
static int xol_add_vma(struct mm_struct *mm, struct xol_area *area)
|
2012-03-31 02:26:31 +08:00
|
|
|
{
|
2015-07-21 21:40:33 +08:00
|
|
|
struct vm_area_struct *vma;
|
|
|
|
int ret;
|
2012-03-31 02:26:31 +08:00
|
|
|
|
2016-05-24 07:26:08 +08:00
|
|
|
if (down_write_killable(&mm->mmap_sem))
|
|
|
|
return -EINTR;
|
|
|
|
|
2015-07-21 21:40:33 +08:00
|
|
|
if (mm->uprobes_state.xol_area) {
|
|
|
|
ret = -EALREADY;
|
2012-03-31 02:26:31 +08:00
|
|
|
goto fail;
|
2015-07-21 21:40:33 +08:00
|
|
|
}
|
2012-03-31 02:26:31 +08:00
|
|
|
|
2013-10-14 03:18:38 +08:00
|
|
|
if (!area->vaddr) {
|
|
|
|
/* Try to map as high as possible, this is only a hint. */
|
|
|
|
area->vaddr = get_unmapped_area(NULL, TASK_SIZE - PAGE_SIZE,
|
|
|
|
PAGE_SIZE, 0, 0);
|
|
|
|
if (area->vaddr & ~PAGE_MASK) {
|
|
|
|
ret = area->vaddr;
|
|
|
|
goto fail;
|
|
|
|
}
|
2012-03-31 02:26:31 +08:00
|
|
|
}
|
|
|
|
|
2015-07-21 21:40:33 +08:00
|
|
|
vma = _install_special_mapping(mm, area->vaddr, PAGE_SIZE,
|
|
|
|
VM_EXEC|VM_MAYEXEC|VM_DONTCOPY|VM_IO,
|
|
|
|
&area->xol_mapping);
|
|
|
|
if (IS_ERR(vma)) {
|
|
|
|
ret = PTR_ERR(vma);
|
2012-03-31 02:26:31 +08:00
|
|
|
goto fail;
|
2015-07-21 21:40:33 +08:00
|
|
|
}
|
2012-03-31 02:26:31 +08:00
|
|
|
|
2015-07-21 21:40:33 +08:00
|
|
|
ret = 0;
|
2017-10-10 02:08:53 +08:00
|
|
|
/* pairs with get_xol_area() */
|
|
|
|
smp_store_release(&mm->uprobes_state.xol_area, area); /* ^^^ */
|
2012-12-31 00:40:39 +08:00
|
|
|
fail:
|
2012-03-31 02:26:31 +08:00
|
|
|
up_write(&mm->mmap_sem);
|
|
|
|
|
|
|
|
return ret;
|
|
|
|
}
|
|
|
|
|
2013-10-14 03:18:38 +08:00
|
|
|
static struct xol_area *__create_xol_area(unsigned long vaddr)
|
2012-03-31 02:26:31 +08:00
|
|
|
{
|
2012-12-31 23:39:49 +08:00
|
|
|
struct mm_struct *mm = current->mm;
|
2013-04-04 00:00:32 +08:00
|
|
|
uprobe_opcode_t insn = UPROBE_SWBP_INSN;
|
2013-10-14 03:18:35 +08:00
|
|
|
struct xol_area *area;
|
2012-12-31 23:39:49 +08:00
|
|
|
|
2013-10-14 03:18:38 +08:00
|
|
|
area = kmalloc(sizeof(*area), GFP_KERNEL);
|
2012-03-31 02:26:31 +08:00
|
|
|
if (unlikely(!area))
|
2012-12-31 00:40:39 +08:00
|
|
|
goto out;
|
2012-03-31 02:26:31 +08:00
|
|
|
|
treewide: kzalloc() -> kcalloc()
The kzalloc() function has a 2-factor argument form, kcalloc(). This
patch replaces cases of:
kzalloc(a * b, gfp)
with:
kcalloc(a * b, gfp)
as well as handling cases of:
kzalloc(a * b * c, gfp)
with:
kzalloc(array3_size(a, b, c), gfp)
as it's slightly less ugly than:
kzalloc_array(array_size(a, b), c, gfp)
This does, however, attempt to ignore constant size factors like:
kzalloc(4 * 1024, gfp)
though any constants defined via macros get caught up in the conversion.
Any factors with a sizeof() of "unsigned char", "char", and "u8" were
dropped, since they're redundant.
The Coccinelle script used for this was:
// Fix redundant parens around sizeof().
@@
type TYPE;
expression THING, E;
@@
(
kzalloc(
- (sizeof(TYPE)) * E
+ sizeof(TYPE) * E
, ...)
|
kzalloc(
- (sizeof(THING)) * E
+ sizeof(THING) * E
, ...)
)
// Drop single-byte sizes and redundant parens.
@@
expression COUNT;
typedef u8;
typedef __u8;
@@
(
kzalloc(
- sizeof(u8) * (COUNT)
+ COUNT
, ...)
|
kzalloc(
- sizeof(__u8) * (COUNT)
+ COUNT
, ...)
|
kzalloc(
- sizeof(char) * (COUNT)
+ COUNT
, ...)
|
kzalloc(
- sizeof(unsigned char) * (COUNT)
+ COUNT
, ...)
|
kzalloc(
- sizeof(u8) * COUNT
+ COUNT
, ...)
|
kzalloc(
- sizeof(__u8) * COUNT
+ COUNT
, ...)
|
kzalloc(
- sizeof(char) * COUNT
+ COUNT
, ...)
|
kzalloc(
- sizeof(unsigned char) * COUNT
+ COUNT
, ...)
)
// 2-factor product with sizeof(type/expression) and identifier or constant.
@@
type TYPE;
expression THING;
identifier COUNT_ID;
constant COUNT_CONST;
@@
(
- kzalloc
+ kcalloc
(
- sizeof(TYPE) * (COUNT_ID)
+ COUNT_ID, sizeof(TYPE)
, ...)
|
- kzalloc
+ kcalloc
(
- sizeof(TYPE) * COUNT_ID
+ COUNT_ID, sizeof(TYPE)
, ...)
|
- kzalloc
+ kcalloc
(
- sizeof(TYPE) * (COUNT_CONST)
+ COUNT_CONST, sizeof(TYPE)
, ...)
|
- kzalloc
+ kcalloc
(
- sizeof(TYPE) * COUNT_CONST
+ COUNT_CONST, sizeof(TYPE)
, ...)
|
- kzalloc
+ kcalloc
(
- sizeof(THING) * (COUNT_ID)
+ COUNT_ID, sizeof(THING)
, ...)
|
- kzalloc
+ kcalloc
(
- sizeof(THING) * COUNT_ID
+ COUNT_ID, sizeof(THING)
, ...)
|
- kzalloc
+ kcalloc
(
- sizeof(THING) * (COUNT_CONST)
+ COUNT_CONST, sizeof(THING)
, ...)
|
- kzalloc
+ kcalloc
(
- sizeof(THING) * COUNT_CONST
+ COUNT_CONST, sizeof(THING)
, ...)
)
// 2-factor product, only identifiers.
@@
identifier SIZE, COUNT;
@@
- kzalloc
+ kcalloc
(
- SIZE * COUNT
+ COUNT, SIZE
, ...)
// 3-factor product with 1 sizeof(type) or sizeof(expression), with
// redundant parens removed.
@@
expression THING;
identifier STRIDE, COUNT;
type TYPE;
@@
(
kzalloc(
- sizeof(TYPE) * (COUNT) * (STRIDE)
+ array3_size(COUNT, STRIDE, sizeof(TYPE))
, ...)
|
kzalloc(
- sizeof(TYPE) * (COUNT) * STRIDE
+ array3_size(COUNT, STRIDE, sizeof(TYPE))
, ...)
|
kzalloc(
- sizeof(TYPE) * COUNT * (STRIDE)
+ array3_size(COUNT, STRIDE, sizeof(TYPE))
, ...)
|
kzalloc(
- sizeof(TYPE) * COUNT * STRIDE
+ array3_size(COUNT, STRIDE, sizeof(TYPE))
, ...)
|
kzalloc(
- sizeof(THING) * (COUNT) * (STRIDE)
+ array3_size(COUNT, STRIDE, sizeof(THING))
, ...)
|
kzalloc(
- sizeof(THING) * (COUNT) * STRIDE
+ array3_size(COUNT, STRIDE, sizeof(THING))
, ...)
|
kzalloc(
- sizeof(THING) * COUNT * (STRIDE)
+ array3_size(COUNT, STRIDE, sizeof(THING))
, ...)
|
kzalloc(
- sizeof(THING) * COUNT * STRIDE
+ array3_size(COUNT, STRIDE, sizeof(THING))
, ...)
)
// 3-factor product with 2 sizeof(variable), with redundant parens removed.
@@
expression THING1, THING2;
identifier COUNT;
type TYPE1, TYPE2;
@@
(
kzalloc(
- sizeof(TYPE1) * sizeof(TYPE2) * COUNT
+ array3_size(COUNT, sizeof(TYPE1), sizeof(TYPE2))
, ...)
|
kzalloc(
- sizeof(TYPE1) * sizeof(THING2) * (COUNT)
+ array3_size(COUNT, sizeof(TYPE1), sizeof(TYPE2))
, ...)
|
kzalloc(
- sizeof(THING1) * sizeof(THING2) * COUNT
+ array3_size(COUNT, sizeof(THING1), sizeof(THING2))
, ...)
|
kzalloc(
- sizeof(THING1) * sizeof(THING2) * (COUNT)
+ array3_size(COUNT, sizeof(THING1), sizeof(THING2))
, ...)
|
kzalloc(
- sizeof(TYPE1) * sizeof(THING2) * COUNT
+ array3_size(COUNT, sizeof(TYPE1), sizeof(THING2))
, ...)
|
kzalloc(
- sizeof(TYPE1) * sizeof(THING2) * (COUNT)
+ array3_size(COUNT, sizeof(TYPE1), sizeof(THING2))
, ...)
)
// 3-factor product, only identifiers, with redundant parens removed.
@@
identifier STRIDE, SIZE, COUNT;
@@
(
kzalloc(
- (COUNT) * STRIDE * SIZE
+ array3_size(COUNT, STRIDE, SIZE)
, ...)
|
kzalloc(
- COUNT * (STRIDE) * SIZE
+ array3_size(COUNT, STRIDE, SIZE)
, ...)
|
kzalloc(
- COUNT * STRIDE * (SIZE)
+ array3_size(COUNT, STRIDE, SIZE)
, ...)
|
kzalloc(
- (COUNT) * (STRIDE) * SIZE
+ array3_size(COUNT, STRIDE, SIZE)
, ...)
|
kzalloc(
- COUNT * (STRIDE) * (SIZE)
+ array3_size(COUNT, STRIDE, SIZE)
, ...)
|
kzalloc(
- (COUNT) * STRIDE * (SIZE)
+ array3_size(COUNT, STRIDE, SIZE)
, ...)
|
kzalloc(
- (COUNT) * (STRIDE) * (SIZE)
+ array3_size(COUNT, STRIDE, SIZE)
, ...)
|
kzalloc(
- COUNT * STRIDE * SIZE
+ array3_size(COUNT, STRIDE, SIZE)
, ...)
)
// Any remaining multi-factor products, first at least 3-factor products,
// when they're not all constants...
@@
expression E1, E2, E3;
constant C1, C2, C3;
@@
(
kzalloc(C1 * C2 * C3, ...)
|
kzalloc(
- (E1) * E2 * E3
+ array3_size(E1, E2, E3)
, ...)
|
kzalloc(
- (E1) * (E2) * E3
+ array3_size(E1, E2, E3)
, ...)
|
kzalloc(
- (E1) * (E2) * (E3)
+ array3_size(E1, E2, E3)
, ...)
|
kzalloc(
- E1 * E2 * E3
+ array3_size(E1, E2, E3)
, ...)
)
// And then all remaining 2 factors products when they're not all constants,
// keeping sizeof() as the second factor argument.
@@
expression THING, E1, E2;
type TYPE;
constant C1, C2, C3;
@@
(
kzalloc(sizeof(THING) * C2, ...)
|
kzalloc(sizeof(TYPE) * C2, ...)
|
kzalloc(C1 * C2 * C3, ...)
|
kzalloc(C1 * C2, ...)
|
- kzalloc
+ kcalloc
(
- sizeof(TYPE) * (E2)
+ E2, sizeof(TYPE)
, ...)
|
- kzalloc
+ kcalloc
(
- sizeof(TYPE) * E2
+ E2, sizeof(TYPE)
, ...)
|
- kzalloc
+ kcalloc
(
- sizeof(THING) * (E2)
+ E2, sizeof(THING)
, ...)
|
- kzalloc
+ kcalloc
(
- sizeof(THING) * E2
+ E2, sizeof(THING)
, ...)
|
- kzalloc
+ kcalloc
(
- (E1) * E2
+ E1, E2
, ...)
|
- kzalloc
+ kcalloc
(
- (E1) * (E2)
+ E1, E2
, ...)
|
- kzalloc
+ kcalloc
(
- E1 * E2
+ E1, E2
, ...)
)
Signed-off-by: Kees Cook <keescook@chromium.org>
2018-06-13 05:03:40 +08:00
|
|
|
area->bitmap = kcalloc(BITS_TO_LONGS(UINSNS_PER_PAGE), sizeof(long),
|
|
|
|
GFP_KERNEL);
|
2012-03-31 02:26:31 +08:00
|
|
|
if (!area->bitmap)
|
2012-12-31 00:40:39 +08:00
|
|
|
goto free_area;
|
|
|
|
|
2015-07-21 21:40:33 +08:00
|
|
|
area->xol_mapping.name = "[uprobes]";
|
2016-02-28 06:11:28 +08:00
|
|
|
area->xol_mapping.fault = NULL;
|
2015-07-21 21:40:33 +08:00
|
|
|
area->xol_mapping.pages = area->pages;
|
2015-07-21 21:40:31 +08:00
|
|
|
area->pages[0] = alloc_page(GFP_HIGHUSER);
|
|
|
|
if (!area->pages[0])
|
2012-12-31 00:40:39 +08:00
|
|
|
goto free_bitmap;
|
2015-07-21 21:40:31 +08:00
|
|
|
area->pages[1] = NULL;
|
2012-03-31 02:26:31 +08:00
|
|
|
|
2013-10-14 03:18:38 +08:00
|
|
|
area->vaddr = vaddr;
|
2013-10-14 03:18:35 +08:00
|
|
|
init_waitqueue_head(&area->wq);
|
|
|
|
/* Reserve the 1st slot for get_trampoline_vaddr() */
|
2013-04-04 00:00:32 +08:00
|
|
|
set_bit(0, area->bitmap);
|
|
|
|
atomic_set(&area->slot_count, 1);
|
2016-12-13 18:40:57 +08:00
|
|
|
arch_uprobe_copy_ixol(area->pages[0], 0, &insn, UPROBE_SWBP_INSN_SIZE);
|
2013-04-04 00:00:32 +08:00
|
|
|
|
2013-10-14 03:18:35 +08:00
|
|
|
if (!xol_add_vma(mm, area))
|
2012-03-31 02:26:31 +08:00
|
|
|
return area;
|
|
|
|
|
2015-07-21 21:40:31 +08:00
|
|
|
__free_page(area->pages[0]);
|
2012-12-31 00:40:39 +08:00
|
|
|
free_bitmap:
|
2012-03-31 02:26:31 +08:00
|
|
|
kfree(area->bitmap);
|
2012-12-31 00:40:39 +08:00
|
|
|
free_area:
|
2012-03-31 02:26:31 +08:00
|
|
|
kfree(area);
|
2012-12-31 00:40:39 +08:00
|
|
|
out:
|
2013-10-14 03:18:35 +08:00
|
|
|
return NULL;
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* get_xol_area - Allocate process's xol_area if necessary.
|
|
|
|
* This area will be used for storing instructions for execution out of line.
|
|
|
|
*
|
|
|
|
* Returns the allocated area or NULL.
|
|
|
|
*/
|
|
|
|
static struct xol_area *get_xol_area(void)
|
|
|
|
{
|
|
|
|
struct mm_struct *mm = current->mm;
|
|
|
|
struct xol_area *area;
|
|
|
|
|
|
|
|
if (!mm->uprobes_state.xol_area)
|
2013-10-14 03:18:38 +08:00
|
|
|
__create_xol_area(0);
|
2013-10-14 03:18:35 +08:00
|
|
|
|
2017-10-10 02:08:53 +08:00
|
|
|
/* Pairs with xol_add_vma() smp_store_release() */
|
|
|
|
area = READ_ONCE(mm->uprobes_state.xol_area); /* ^^^ */
|
2012-12-31 23:39:49 +08:00
|
|
|
return area;
|
2012-03-31 02:26:31 +08:00
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* uprobe_clear_state - Free the area allocated for slots.
|
|
|
|
*/
|
|
|
|
void uprobe_clear_state(struct mm_struct *mm)
|
|
|
|
{
|
|
|
|
struct xol_area *area = mm->uprobes_state.xol_area;
|
|
|
|
|
uprobes: Support SDT markers having reference count (semaphore)
Userspace Statically Defined Tracepoints[1] are dtrace style markers
inside userspace applications. Applications like PostgreSQL, MySQL,
Pthread, Perl, Python, Java, Ruby, Node.js, libvirt, QEMU, glib etc
have these markers embedded in them. These markers are added by developer
at important places in the code. Each marker source expands to a single
nop instruction in the compiled code but there may be additional
overhead for computing the marker arguments which expands to couple of
instructions. In case the overhead is more, execution of it can be
omitted by runtime if() condition when no one is tracing on the marker:
if (reference_counter > 0) {
Execute marker instructions;
}
Default value of reference counter is 0. Tracer has to increment the
reference counter before tracing on a marker and decrement it when
done with the tracing.
Implement the reference counter logic in core uprobe. User will be
able to use it from trace_uprobe as well as from kernel module. New
trace_uprobe definition with reference counter will now be:
<path>:<offset>[(ref_ctr_offset)]
where ref_ctr_offset is an optional field. For kernel module, new
variant of uprobe_register() has been introduced:
uprobe_register_refctr(inode, offset, ref_ctr_offset, consumer)
No new variant for uprobe_unregister() because it's assumed to have
only one reference counter for one uprobe.
[1] https://sourceware.org/systemtap/wiki/UserSpaceProbeImplementation
Note: 'reference counter' is called as 'semaphore' in original Dtrace
(or Systemtap, bcc and even in ELF) documentation and code. But the
term 'semaphore' is misleading in this context. This is just a counter
used to hold number of tracers tracing on a marker. This is not really
used for any synchronization. So we are calling it a 'reference counter'
in kernel / perf code.
Link: http://lkml.kernel.org/r/20180820044250.11659-2-ravi.bangoria@linux.ibm.com
Reviewed-by: Masami Hiramatsu <mhiramat@kernel.org>
[Only trace_uprobe.c]
Reviewed-by: Oleg Nesterov <oleg@redhat.com>
Reviewed-by: Song Liu <songliubraving@fb.com>
Tested-by: Song Liu <songliubraving@fb.com>
Signed-off-by: Ravi Bangoria <ravi.bangoria@linux.ibm.com>
Signed-off-by: Steven Rostedt (VMware) <rostedt@goodmis.org>
2018-08-20 12:42:47 +08:00
|
|
|
mutex_lock(&delayed_uprobe_lock);
|
|
|
|
delayed_uprobe_remove(NULL, mm);
|
|
|
|
mutex_unlock(&delayed_uprobe_lock);
|
|
|
|
|
2012-03-31 02:26:31 +08:00
|
|
|
if (!area)
|
|
|
|
return;
|
|
|
|
|
2015-07-21 21:40:31 +08:00
|
|
|
put_page(area->pages[0]);
|
2012-03-31 02:26:31 +08:00
|
|
|
kfree(area->bitmap);
|
|
|
|
kfree(area);
|
|
|
|
}
|
|
|
|
|
2012-11-15 02:03:42 +08:00
|
|
|
void uprobe_start_dup_mmap(void)
|
|
|
|
{
|
|
|
|
percpu_down_read(&dup_mmap_sem);
|
|
|
|
}
|
|
|
|
|
|
|
|
void uprobe_end_dup_mmap(void)
|
|
|
|
{
|
|
|
|
percpu_up_read(&dup_mmap_sem);
|
|
|
|
}
|
|
|
|
|
2012-08-08 23:11:42 +08:00
|
|
|
void uprobe_dup_mmap(struct mm_struct *oldmm, struct mm_struct *newmm)
|
|
|
|
{
|
2012-08-19 22:15:09 +08:00
|
|
|
if (test_bit(MMF_HAS_UPROBES, &oldmm->flags)) {
|
2012-08-08 23:11:42 +08:00
|
|
|
set_bit(MMF_HAS_UPROBES, &newmm->flags);
|
2012-08-19 22:15:09 +08:00
|
|
|
/* unconditionally, dup_mmap() skips VM_DONTCOPY vmas */
|
|
|
|
set_bit(MMF_RECALC_UPROBES, &newmm->flags);
|
|
|
|
}
|
2012-08-08 23:11:42 +08:00
|
|
|
}
|
|
|
|
|
2012-03-31 02:26:31 +08:00
|
|
|
/*
|
|
|
|
* - search for a free slot.
|
|
|
|
*/
|
|
|
|
static unsigned long xol_take_insn_slot(struct xol_area *area)
|
|
|
|
{
|
|
|
|
unsigned long slot_addr;
|
|
|
|
int slot_nr;
|
|
|
|
|
|
|
|
do {
|
|
|
|
slot_nr = find_first_zero_bit(area->bitmap, UINSNS_PER_PAGE);
|
|
|
|
if (slot_nr < UINSNS_PER_PAGE) {
|
|
|
|
if (!test_and_set_bit(slot_nr, area->bitmap))
|
|
|
|
break;
|
|
|
|
|
|
|
|
slot_nr = UINSNS_PER_PAGE;
|
|
|
|
continue;
|
|
|
|
}
|
|
|
|
wait_event(area->wq, (atomic_read(&area->slot_count) < UINSNS_PER_PAGE));
|
|
|
|
} while (slot_nr >= UINSNS_PER_PAGE);
|
|
|
|
|
|
|
|
slot_addr = area->vaddr + (slot_nr * UPROBE_XOL_SLOT_BYTES);
|
|
|
|
atomic_inc(&area->slot_count);
|
|
|
|
|
|
|
|
return slot_addr;
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
2013-01-01 01:00:06 +08:00
|
|
|
* xol_get_insn_slot - allocate a slot for xol.
|
2012-03-31 02:26:31 +08:00
|
|
|
* Returns the allocated slot address or 0.
|
|
|
|
*/
|
2013-01-01 01:00:06 +08:00
|
|
|
static unsigned long xol_get_insn_slot(struct uprobe *uprobe)
|
2012-03-31 02:26:31 +08:00
|
|
|
{
|
|
|
|
struct xol_area *area;
|
2013-01-01 01:00:06 +08:00
|
|
|
unsigned long xol_vaddr;
|
2012-03-31 02:26:31 +08:00
|
|
|
|
2012-12-31 23:39:49 +08:00
|
|
|
area = get_xol_area();
|
|
|
|
if (!area)
|
|
|
|
return 0;
|
2012-03-31 02:26:31 +08:00
|
|
|
|
2013-01-01 01:00:06 +08:00
|
|
|
xol_vaddr = xol_take_insn_slot(area);
|
|
|
|
if (unlikely(!xol_vaddr))
|
2012-03-31 02:26:31 +08:00
|
|
|
return 0;
|
|
|
|
|
2015-07-21 21:40:31 +08:00
|
|
|
arch_uprobe_copy_ixol(area->pages[0], xol_vaddr,
|
2014-04-29 11:20:52 +08:00
|
|
|
&uprobe->arch.ixol, sizeof(uprobe->arch.ixol));
|
2012-03-31 02:26:31 +08:00
|
|
|
|
2013-01-01 01:00:06 +08:00
|
|
|
return xol_vaddr;
|
2012-03-31 02:26:31 +08:00
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* xol_free_insn_slot - If slot was earlier allocated by
|
|
|
|
* @xol_get_insn_slot(), make the slot available for
|
|
|
|
* subsequent requests.
|
|
|
|
*/
|
|
|
|
static void xol_free_insn_slot(struct task_struct *tsk)
|
|
|
|
{
|
|
|
|
struct xol_area *area;
|
|
|
|
unsigned long vma_end;
|
|
|
|
unsigned long slot_addr;
|
|
|
|
|
|
|
|
if (!tsk->mm || !tsk->mm->uprobes_state.xol_area || !tsk->utask)
|
|
|
|
return;
|
|
|
|
|
|
|
|
slot_addr = tsk->utask->xol_vaddr;
|
2013-01-01 01:37:11 +08:00
|
|
|
if (unlikely(!slot_addr))
|
2012-03-31 02:26:31 +08:00
|
|
|
return;
|
|
|
|
|
|
|
|
area = tsk->mm->uprobes_state.xol_area;
|
|
|
|
vma_end = area->vaddr + PAGE_SIZE;
|
|
|
|
if (area->vaddr <= slot_addr && slot_addr < vma_end) {
|
|
|
|
unsigned long offset;
|
|
|
|
int slot_nr;
|
|
|
|
|
|
|
|
offset = slot_addr - area->vaddr;
|
|
|
|
slot_nr = offset / UPROBE_XOL_SLOT_BYTES;
|
|
|
|
if (slot_nr >= UINSNS_PER_PAGE)
|
|
|
|
return;
|
|
|
|
|
|
|
|
clear_bit(slot_nr, area->bitmap);
|
|
|
|
atomic_dec(&area->slot_count);
|
2015-07-21 21:40:36 +08:00
|
|
|
smp_mb__after_atomic(); /* pairs with prepare_to_wait() */
|
2012-03-31 02:26:31 +08:00
|
|
|
if (waitqueue_active(&area->wq))
|
|
|
|
wake_up(&area->wq);
|
|
|
|
|
|
|
|
tsk->utask->xol_vaddr = 0;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2014-04-29 11:20:52 +08:00
|
|
|
void __weak arch_uprobe_copy_ixol(struct page *page, unsigned long vaddr,
|
|
|
|
void *src, unsigned long len)
|
|
|
|
{
|
|
|
|
/* Initialize the slot */
|
|
|
|
copy_to_page(page, vaddr, src, len);
|
|
|
|
|
|
|
|
/*
|
|
|
|
* We probably need flush_icache_user_range() but it needs vma.
|
|
|
|
* This should work on most of architectures by default. If
|
|
|
|
* architecture needs to do something different it can define
|
|
|
|
* its own version of the function.
|
|
|
|
*/
|
|
|
|
flush_dcache_page(page);
|
|
|
|
}
|
|
|
|
|
uprobes/core: Handle breakpoint and singlestep exceptions
Uprobes uses exception notifiers to get to know if a thread hit
a breakpoint or a singlestep exception.
When a thread hits a uprobe or is singlestepping post a uprobe
hit, the uprobe exception notifier sets its TIF_UPROBE bit,
which will then be checked on its return to userspace path
(do_notify_resume() ->uprobe_notify_resume()), where the
consumers handlers are run (in task context) based on the
defined filters.
Uprobe hits are thread specific and hence we need to maintain
information about if a task hit a uprobe, what uprobe was hit,
the slot where the original instruction was copied for xol so
that it can be singlestepped with appropriate fixups.
In some cases, special care is needed for instructions that are
executed out of line (xol). These are architecture specific
artefacts, such as handling RIP relative instructions on x86_64.
Since the instruction at which the uprobe was inserted is
executed out of line, architecture specific fixups are added so
that the thread continues normal execution in the presence of a
uprobe.
Postpone the signals until we execute the probed insn.
post_xol() path does a recalc_sigpending() before return to
user-mode, this ensures the signal can't be lost.
Uprobes relies on DIE_DEBUG notification to notify if a
singlestep is complete.
Adds x86 specific uprobe exception notifiers and appropriate
hooks needed to determine a uprobe hit and subsequent post
processing.
Add requisite x86 fixups for xol for uprobes. Specific cases
needing fixups include relative jumps (x86_64), calls, etc.
Where possible, we check and skip singlestepping the
breakpointed instructions. For now we skip single byte as well
as few multibyte nop instructions. However this can be extended
to other instructions too.
Credits to Oleg Nesterov for suggestions/patches related to
signal, breakpoint, singlestep handling code.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Jim Keniston <jkenisto@linux.vnet.ibm.com>
Cc: Linux-mm <linux-mm@kvack.org>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/20120313180011.29771.89027.sendpatchset@srdronam.in.ibm.com
[ Performed various cleanliness edits ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-03-14 02:00:11 +08:00
|
|
|
/**
|
|
|
|
* uprobe_get_swbp_addr - compute address of swbp given post-swbp regs
|
|
|
|
* @regs: Reflects the saved state of the task after it has hit a breakpoint
|
|
|
|
* instruction.
|
|
|
|
* Return the address of the breakpoint instruction.
|
|
|
|
*/
|
|
|
|
unsigned long __weak uprobe_get_swbp_addr(struct pt_regs *regs)
|
|
|
|
{
|
|
|
|
return instruction_pointer(regs) - UPROBE_SWBP_INSN_SIZE;
|
|
|
|
}
|
|
|
|
|
uprobes/x86: Fix the wrong ->si_addr when xol triggers a trap
If the probed insn triggers a trap, ->si_addr = regs->ip is technically
correct, but this is not what the signal handler wants; we need to pass
the address of the probed insn, not the address of xol slot.
Add the new arch-agnostic helper, uprobe_get_trap_addr(), and change
fill_trap_info() and math_error() to use it. !CONFIG_UPROBES case in
uprobes.h uses a macro to avoid include hell and ensure that it can be
compiled even if an architecture doesn't define instruction_pointer().
Test-case:
#include <signal.h>
#include <stdio.h>
#include <unistd.h>
extern void probe_div(void);
void sigh(int sig, siginfo_t *info, void *c)
{
int passed = (info->si_addr == probe_div);
printf(passed ? "PASS\n" : "FAIL\n");
_exit(!passed);
}
int main(void)
{
struct sigaction sa = {
.sa_sigaction = sigh,
.sa_flags = SA_SIGINFO,
};
sigaction(SIGFPE, &sa, NULL);
asm (
"xor %ecx,%ecx\n"
".globl probe_div; probe_div:\n"
"idiv %ecx\n"
);
return 0;
}
it fails if probe_div() is probed.
Note: show_unhandled_signals users should probably use this helper too,
but we need to cleanup them first.
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Reviewed-by: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
2014-05-13 00:24:45 +08:00
|
|
|
unsigned long uprobe_get_trap_addr(struct pt_regs *regs)
|
|
|
|
{
|
|
|
|
struct uprobe_task *utask = current->utask;
|
|
|
|
|
|
|
|
if (unlikely(utask && utask->active_uprobe))
|
|
|
|
return utask->vaddr;
|
|
|
|
|
|
|
|
return instruction_pointer(regs);
|
|
|
|
}
|
|
|
|
|
2015-07-21 21:40:06 +08:00
|
|
|
static struct return_instance *free_ret_instance(struct return_instance *ri)
|
|
|
|
{
|
|
|
|
struct return_instance *next = ri->next;
|
|
|
|
put_uprobe(ri->uprobe);
|
|
|
|
kfree(ri);
|
|
|
|
return next;
|
|
|
|
}
|
|
|
|
|
uprobes/core: Handle breakpoint and singlestep exceptions
Uprobes uses exception notifiers to get to know if a thread hit
a breakpoint or a singlestep exception.
When a thread hits a uprobe or is singlestepping post a uprobe
hit, the uprobe exception notifier sets its TIF_UPROBE bit,
which will then be checked on its return to userspace path
(do_notify_resume() ->uprobe_notify_resume()), where the
consumers handlers are run (in task context) based on the
defined filters.
Uprobe hits are thread specific and hence we need to maintain
information about if a task hit a uprobe, what uprobe was hit,
the slot where the original instruction was copied for xol so
that it can be singlestepped with appropriate fixups.
In some cases, special care is needed for instructions that are
executed out of line (xol). These are architecture specific
artefacts, such as handling RIP relative instructions on x86_64.
Since the instruction at which the uprobe was inserted is
executed out of line, architecture specific fixups are added so
that the thread continues normal execution in the presence of a
uprobe.
Postpone the signals until we execute the probed insn.
post_xol() path does a recalc_sigpending() before return to
user-mode, this ensures the signal can't be lost.
Uprobes relies on DIE_DEBUG notification to notify if a
singlestep is complete.
Adds x86 specific uprobe exception notifiers and appropriate
hooks needed to determine a uprobe hit and subsequent post
processing.
Add requisite x86 fixups for xol for uprobes. Specific cases
needing fixups include relative jumps (x86_64), calls, etc.
Where possible, we check and skip singlestepping the
breakpointed instructions. For now we skip single byte as well
as few multibyte nop instructions. However this can be extended
to other instructions too.
Credits to Oleg Nesterov for suggestions/patches related to
signal, breakpoint, singlestep handling code.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Jim Keniston <jkenisto@linux.vnet.ibm.com>
Cc: Linux-mm <linux-mm@kvack.org>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/20120313180011.29771.89027.sendpatchset@srdronam.in.ibm.com
[ Performed various cleanliness edits ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-03-14 02:00:11 +08:00
|
|
|
/*
|
|
|
|
* Called with no locks held.
|
2018-07-09 18:57:15 +08:00
|
|
|
* Called in context of an exiting or an exec-ing thread.
|
uprobes/core: Handle breakpoint and singlestep exceptions
Uprobes uses exception notifiers to get to know if a thread hit
a breakpoint or a singlestep exception.
When a thread hits a uprobe or is singlestepping post a uprobe
hit, the uprobe exception notifier sets its TIF_UPROBE bit,
which will then be checked on its return to userspace path
(do_notify_resume() ->uprobe_notify_resume()), where the
consumers handlers are run (in task context) based on the
defined filters.
Uprobe hits are thread specific and hence we need to maintain
information about if a task hit a uprobe, what uprobe was hit,
the slot where the original instruction was copied for xol so
that it can be singlestepped with appropriate fixups.
In some cases, special care is needed for instructions that are
executed out of line (xol). These are architecture specific
artefacts, such as handling RIP relative instructions on x86_64.
Since the instruction at which the uprobe was inserted is
executed out of line, architecture specific fixups are added so
that the thread continues normal execution in the presence of a
uprobe.
Postpone the signals until we execute the probed insn.
post_xol() path does a recalc_sigpending() before return to
user-mode, this ensures the signal can't be lost.
Uprobes relies on DIE_DEBUG notification to notify if a
singlestep is complete.
Adds x86 specific uprobe exception notifiers and appropriate
hooks needed to determine a uprobe hit and subsequent post
processing.
Add requisite x86 fixups for xol for uprobes. Specific cases
needing fixups include relative jumps (x86_64), calls, etc.
Where possible, we check and skip singlestepping the
breakpointed instructions. For now we skip single byte as well
as few multibyte nop instructions. However this can be extended
to other instructions too.
Credits to Oleg Nesterov for suggestions/patches related to
signal, breakpoint, singlestep handling code.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Jim Keniston <jkenisto@linux.vnet.ibm.com>
Cc: Linux-mm <linux-mm@kvack.org>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/20120313180011.29771.89027.sendpatchset@srdronam.in.ibm.com
[ Performed various cleanliness edits ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-03-14 02:00:11 +08:00
|
|
|
*/
|
|
|
|
void uprobe_free_utask(struct task_struct *t)
|
|
|
|
{
|
|
|
|
struct uprobe_task *utask = t->utask;
|
2015-07-21 21:40:06 +08:00
|
|
|
struct return_instance *ri;
|
uprobes/core: Handle breakpoint and singlestep exceptions
Uprobes uses exception notifiers to get to know if a thread hit
a breakpoint or a singlestep exception.
When a thread hits a uprobe or is singlestepping post a uprobe
hit, the uprobe exception notifier sets its TIF_UPROBE bit,
which will then be checked on its return to userspace path
(do_notify_resume() ->uprobe_notify_resume()), where the
consumers handlers are run (in task context) based on the
defined filters.
Uprobe hits are thread specific and hence we need to maintain
information about if a task hit a uprobe, what uprobe was hit,
the slot where the original instruction was copied for xol so
that it can be singlestepped with appropriate fixups.
In some cases, special care is needed for instructions that are
executed out of line (xol). These are architecture specific
artefacts, such as handling RIP relative instructions on x86_64.
Since the instruction at which the uprobe was inserted is
executed out of line, architecture specific fixups are added so
that the thread continues normal execution in the presence of a
uprobe.
Postpone the signals until we execute the probed insn.
post_xol() path does a recalc_sigpending() before return to
user-mode, this ensures the signal can't be lost.
Uprobes relies on DIE_DEBUG notification to notify if a
singlestep is complete.
Adds x86 specific uprobe exception notifiers and appropriate
hooks needed to determine a uprobe hit and subsequent post
processing.
Add requisite x86 fixups for xol for uprobes. Specific cases
needing fixups include relative jumps (x86_64), calls, etc.
Where possible, we check and skip singlestepping the
breakpointed instructions. For now we skip single byte as well
as few multibyte nop instructions. However this can be extended
to other instructions too.
Credits to Oleg Nesterov for suggestions/patches related to
signal, breakpoint, singlestep handling code.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Jim Keniston <jkenisto@linux.vnet.ibm.com>
Cc: Linux-mm <linux-mm@kvack.org>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/20120313180011.29771.89027.sendpatchset@srdronam.in.ibm.com
[ Performed various cleanliness edits ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-03-14 02:00:11 +08:00
|
|
|
|
|
|
|
if (!utask)
|
|
|
|
return;
|
|
|
|
|
|
|
|
if (utask->active_uprobe)
|
|
|
|
put_uprobe(utask->active_uprobe);
|
|
|
|
|
2013-04-04 00:00:35 +08:00
|
|
|
ri = utask->return_instances;
|
2015-07-21 21:40:06 +08:00
|
|
|
while (ri)
|
|
|
|
ri = free_ret_instance(ri);
|
2013-04-04 00:00:35 +08:00
|
|
|
|
2012-03-31 02:26:31 +08:00
|
|
|
xol_free_insn_slot(t);
|
uprobes/core: Handle breakpoint and singlestep exceptions
Uprobes uses exception notifiers to get to know if a thread hit
a breakpoint or a singlestep exception.
When a thread hits a uprobe or is singlestepping post a uprobe
hit, the uprobe exception notifier sets its TIF_UPROBE bit,
which will then be checked on its return to userspace path
(do_notify_resume() ->uprobe_notify_resume()), where the
consumers handlers are run (in task context) based on the
defined filters.
Uprobe hits are thread specific and hence we need to maintain
information about if a task hit a uprobe, what uprobe was hit,
the slot where the original instruction was copied for xol so
that it can be singlestepped with appropriate fixups.
In some cases, special care is needed for instructions that are
executed out of line (xol). These are architecture specific
artefacts, such as handling RIP relative instructions on x86_64.
Since the instruction at which the uprobe was inserted is
executed out of line, architecture specific fixups are added so
that the thread continues normal execution in the presence of a
uprobe.
Postpone the signals until we execute the probed insn.
post_xol() path does a recalc_sigpending() before return to
user-mode, this ensures the signal can't be lost.
Uprobes relies on DIE_DEBUG notification to notify if a
singlestep is complete.
Adds x86 specific uprobe exception notifiers and appropriate
hooks needed to determine a uprobe hit and subsequent post
processing.
Add requisite x86 fixups for xol for uprobes. Specific cases
needing fixups include relative jumps (x86_64), calls, etc.
Where possible, we check and skip singlestepping the
breakpointed instructions. For now we skip single byte as well
as few multibyte nop instructions. However this can be extended
to other instructions too.
Credits to Oleg Nesterov for suggestions/patches related to
signal, breakpoint, singlestep handling code.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Jim Keniston <jkenisto@linux.vnet.ibm.com>
Cc: Linux-mm <linux-mm@kvack.org>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/20120313180011.29771.89027.sendpatchset@srdronam.in.ibm.com
[ Performed various cleanliness edits ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-03-14 02:00:11 +08:00
|
|
|
kfree(utask);
|
|
|
|
t->utask = NULL;
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
2013-01-01 00:03:32 +08:00
|
|
|
* Allocate a uprobe_task object for the task if if necessary.
|
|
|
|
* Called when the thread hits a breakpoint.
|
uprobes/core: Handle breakpoint and singlestep exceptions
Uprobes uses exception notifiers to get to know if a thread hit
a breakpoint or a singlestep exception.
When a thread hits a uprobe or is singlestepping post a uprobe
hit, the uprobe exception notifier sets its TIF_UPROBE bit,
which will then be checked on its return to userspace path
(do_notify_resume() ->uprobe_notify_resume()), where the
consumers handlers are run (in task context) based on the
defined filters.
Uprobe hits are thread specific and hence we need to maintain
information about if a task hit a uprobe, what uprobe was hit,
the slot where the original instruction was copied for xol so
that it can be singlestepped with appropriate fixups.
In some cases, special care is needed for instructions that are
executed out of line (xol). These are architecture specific
artefacts, such as handling RIP relative instructions on x86_64.
Since the instruction at which the uprobe was inserted is
executed out of line, architecture specific fixups are added so
that the thread continues normal execution in the presence of a
uprobe.
Postpone the signals until we execute the probed insn.
post_xol() path does a recalc_sigpending() before return to
user-mode, this ensures the signal can't be lost.
Uprobes relies on DIE_DEBUG notification to notify if a
singlestep is complete.
Adds x86 specific uprobe exception notifiers and appropriate
hooks needed to determine a uprobe hit and subsequent post
processing.
Add requisite x86 fixups for xol for uprobes. Specific cases
needing fixups include relative jumps (x86_64), calls, etc.
Where possible, we check and skip singlestepping the
breakpointed instructions. For now we skip single byte as well
as few multibyte nop instructions. However this can be extended
to other instructions too.
Credits to Oleg Nesterov for suggestions/patches related to
signal, breakpoint, singlestep handling code.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Jim Keniston <jkenisto@linux.vnet.ibm.com>
Cc: Linux-mm <linux-mm@kvack.org>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/20120313180011.29771.89027.sendpatchset@srdronam.in.ibm.com
[ Performed various cleanliness edits ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-03-14 02:00:11 +08:00
|
|
|
*
|
|
|
|
* Returns:
|
|
|
|
* - pointer to new uprobe_task on success
|
|
|
|
* - NULL otherwise
|
|
|
|
*/
|
2013-01-01 00:03:32 +08:00
|
|
|
static struct uprobe_task *get_utask(void)
|
uprobes/core: Handle breakpoint and singlestep exceptions
Uprobes uses exception notifiers to get to know if a thread hit
a breakpoint or a singlestep exception.
When a thread hits a uprobe or is singlestepping post a uprobe
hit, the uprobe exception notifier sets its TIF_UPROBE bit,
which will then be checked on its return to userspace path
(do_notify_resume() ->uprobe_notify_resume()), where the
consumers handlers are run (in task context) based on the
defined filters.
Uprobe hits are thread specific and hence we need to maintain
information about if a task hit a uprobe, what uprobe was hit,
the slot where the original instruction was copied for xol so
that it can be singlestepped with appropriate fixups.
In some cases, special care is needed for instructions that are
executed out of line (xol). These are architecture specific
artefacts, such as handling RIP relative instructions on x86_64.
Since the instruction at which the uprobe was inserted is
executed out of line, architecture specific fixups are added so
that the thread continues normal execution in the presence of a
uprobe.
Postpone the signals until we execute the probed insn.
post_xol() path does a recalc_sigpending() before return to
user-mode, this ensures the signal can't be lost.
Uprobes relies on DIE_DEBUG notification to notify if a
singlestep is complete.
Adds x86 specific uprobe exception notifiers and appropriate
hooks needed to determine a uprobe hit and subsequent post
processing.
Add requisite x86 fixups for xol for uprobes. Specific cases
needing fixups include relative jumps (x86_64), calls, etc.
Where possible, we check and skip singlestepping the
breakpointed instructions. For now we skip single byte as well
as few multibyte nop instructions. However this can be extended
to other instructions too.
Credits to Oleg Nesterov for suggestions/patches related to
signal, breakpoint, singlestep handling code.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Jim Keniston <jkenisto@linux.vnet.ibm.com>
Cc: Linux-mm <linux-mm@kvack.org>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/20120313180011.29771.89027.sendpatchset@srdronam.in.ibm.com
[ Performed various cleanliness edits ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-03-14 02:00:11 +08:00
|
|
|
{
|
2013-01-01 00:03:32 +08:00
|
|
|
if (!current->utask)
|
|
|
|
current->utask = kzalloc(sizeof(struct uprobe_task), GFP_KERNEL);
|
|
|
|
return current->utask;
|
uprobes/core: Handle breakpoint and singlestep exceptions
Uprobes uses exception notifiers to get to know if a thread hit
a breakpoint or a singlestep exception.
When a thread hits a uprobe or is singlestepping post a uprobe
hit, the uprobe exception notifier sets its TIF_UPROBE bit,
which will then be checked on its return to userspace path
(do_notify_resume() ->uprobe_notify_resume()), where the
consumers handlers are run (in task context) based on the
defined filters.
Uprobe hits are thread specific and hence we need to maintain
information about if a task hit a uprobe, what uprobe was hit,
the slot where the original instruction was copied for xol so
that it can be singlestepped with appropriate fixups.
In some cases, special care is needed for instructions that are
executed out of line (xol). These are architecture specific
artefacts, such as handling RIP relative instructions on x86_64.
Since the instruction at which the uprobe was inserted is
executed out of line, architecture specific fixups are added so
that the thread continues normal execution in the presence of a
uprobe.
Postpone the signals until we execute the probed insn.
post_xol() path does a recalc_sigpending() before return to
user-mode, this ensures the signal can't be lost.
Uprobes relies on DIE_DEBUG notification to notify if a
singlestep is complete.
Adds x86 specific uprobe exception notifiers and appropriate
hooks needed to determine a uprobe hit and subsequent post
processing.
Add requisite x86 fixups for xol for uprobes. Specific cases
needing fixups include relative jumps (x86_64), calls, etc.
Where possible, we check and skip singlestepping the
breakpointed instructions. For now we skip single byte as well
as few multibyte nop instructions. However this can be extended
to other instructions too.
Credits to Oleg Nesterov for suggestions/patches related to
signal, breakpoint, singlestep handling code.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Jim Keniston <jkenisto@linux.vnet.ibm.com>
Cc: Linux-mm <linux-mm@kvack.org>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/20120313180011.29771.89027.sendpatchset@srdronam.in.ibm.com
[ Performed various cleanliness edits ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-03-14 02:00:11 +08:00
|
|
|
}
|
|
|
|
|
2013-10-14 03:18:41 +08:00
|
|
|
static int dup_utask(struct task_struct *t, struct uprobe_task *o_utask)
|
|
|
|
{
|
|
|
|
struct uprobe_task *n_utask;
|
|
|
|
struct return_instance **p, *o, *n;
|
|
|
|
|
|
|
|
n_utask = kzalloc(sizeof(struct uprobe_task), GFP_KERNEL);
|
|
|
|
if (!n_utask)
|
|
|
|
return -ENOMEM;
|
|
|
|
t->utask = n_utask;
|
|
|
|
|
|
|
|
p = &n_utask->return_instances;
|
|
|
|
for (o = o_utask->return_instances; o; o = o->next) {
|
|
|
|
n = kmalloc(sizeof(struct return_instance), GFP_KERNEL);
|
|
|
|
if (!n)
|
|
|
|
return -ENOMEM;
|
|
|
|
|
|
|
|
*n = *o;
|
2015-07-21 21:40:03 +08:00
|
|
|
get_uprobe(n->uprobe);
|
2013-10-14 03:18:41 +08:00
|
|
|
n->next = NULL;
|
|
|
|
|
|
|
|
*p = n;
|
|
|
|
p = &n->next;
|
|
|
|
n_utask->depth++;
|
|
|
|
}
|
|
|
|
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
static void uprobe_warn(struct task_struct *t, const char *msg)
|
|
|
|
{
|
|
|
|
pr_warn("uprobe: %s:%d failed to %s\n",
|
|
|
|
current->comm, current->pid, msg);
|
|
|
|
}
|
|
|
|
|
uprobes: Change uprobe_copy_process() to dup xol_area
This finally fixes the serious bug in uretprobes: a forked child
crashes if the parent called fork() with the pending ret probe.
Trivial test-case:
# perf probe -x /lib/libc.so.6 __fork%return
# perf record -e probe_libc:__fork perl -le 'fork || print "OK"'
(the child doesn't print "OK", it is killed by SIGSEGV)
If the child returns from the probed function it actually returns
to trampoline_vaddr, because it got the copy of parent's stack
mangled by prepare_uretprobe() when the parent entered this func.
It crashes because a) this address is not mapped and b) until the
previous change it doesn't have the proper->return_instances info.
This means that uprobe_copy_process() has to create xol_area which
has the trampoline slot, and its vaddr should be equal to parent's
xol_area->vaddr.
Unfortunately, uprobe_copy_process() can not simply do
__create_xol_area(child, xol_area->vaddr). This could actually work
but perf_event_mmap() doesn't expect the usage of foreign ->mm. So
we offload this to task_work_run(), and pass the argument via not
yet used utask->vaddr.
We know that this vaddr is fine for install_special_mapping(), the
necessary hole was recently "created" by dup_mmap() which skips the
parent's VM_DONTCOPY area, and nobody else could use the new mm.
Unfortunately, this also means that we can not handle the errors
properly, we obviously can not abort the already completed fork().
So we simply print the warning if GFP_KERNEL allocation (the only
possible reason) fails.
Reported-by: Martin Cermak <mcermak@redhat.com>
Reported-by: David Smith <dsmith@redhat.com>
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Acked-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
2013-10-14 03:18:44 +08:00
|
|
|
static void dup_xol_work(struct callback_head *work)
|
|
|
|
{
|
|
|
|
if (current->flags & PF_EXITING)
|
|
|
|
return;
|
|
|
|
|
2016-05-24 07:26:08 +08:00
|
|
|
if (!__create_xol_area(current->utask->dup_xol_addr) &&
|
|
|
|
!fatal_signal_pending(current))
|
uprobes: Change uprobe_copy_process() to dup xol_area
This finally fixes the serious bug in uretprobes: a forked child
crashes if the parent called fork() with the pending ret probe.
Trivial test-case:
# perf probe -x /lib/libc.so.6 __fork%return
# perf record -e probe_libc:__fork perl -le 'fork || print "OK"'
(the child doesn't print "OK", it is killed by SIGSEGV)
If the child returns from the probed function it actually returns
to trampoline_vaddr, because it got the copy of parent's stack
mangled by prepare_uretprobe() when the parent entered this func.
It crashes because a) this address is not mapped and b) until the
previous change it doesn't have the proper->return_instances info.
This means that uprobe_copy_process() has to create xol_area which
has the trampoline slot, and its vaddr should be equal to parent's
xol_area->vaddr.
Unfortunately, uprobe_copy_process() can not simply do
__create_xol_area(child, xol_area->vaddr). This could actually work
but perf_event_mmap() doesn't expect the usage of foreign ->mm. So
we offload this to task_work_run(), and pass the argument via not
yet used utask->vaddr.
We know that this vaddr is fine for install_special_mapping(), the
necessary hole was recently "created" by dup_mmap() which skips the
parent's VM_DONTCOPY area, and nobody else could use the new mm.
Unfortunately, this also means that we can not handle the errors
properly, we obviously can not abort the already completed fork().
So we simply print the warning if GFP_KERNEL allocation (the only
possible reason) fails.
Reported-by: Martin Cermak <mcermak@redhat.com>
Reported-by: David Smith <dsmith@redhat.com>
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Acked-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
2013-10-14 03:18:44 +08:00
|
|
|
uprobe_warn(current, "dup xol area");
|
|
|
|
}
|
|
|
|
|
2013-10-14 03:18:31 +08:00
|
|
|
/*
|
|
|
|
* Called in context of a new clone/fork from copy_process.
|
|
|
|
*/
|
2013-10-17 01:39:37 +08:00
|
|
|
void uprobe_copy_process(struct task_struct *t, unsigned long flags)
|
2013-10-14 03:18:31 +08:00
|
|
|
{
|
2013-10-14 03:18:41 +08:00
|
|
|
struct uprobe_task *utask = current->utask;
|
|
|
|
struct mm_struct *mm = current->mm;
|
uprobes: Change uprobe_copy_process() to dup xol_area
This finally fixes the serious bug in uretprobes: a forked child
crashes if the parent called fork() with the pending ret probe.
Trivial test-case:
# perf probe -x /lib/libc.so.6 __fork%return
# perf record -e probe_libc:__fork perl -le 'fork || print "OK"'
(the child doesn't print "OK", it is killed by SIGSEGV)
If the child returns from the probed function it actually returns
to trampoline_vaddr, because it got the copy of parent's stack
mangled by prepare_uretprobe() when the parent entered this func.
It crashes because a) this address is not mapped and b) until the
previous change it doesn't have the proper->return_instances info.
This means that uprobe_copy_process() has to create xol_area which
has the trampoline slot, and its vaddr should be equal to parent's
xol_area->vaddr.
Unfortunately, uprobe_copy_process() can not simply do
__create_xol_area(child, xol_area->vaddr). This could actually work
but perf_event_mmap() doesn't expect the usage of foreign ->mm. So
we offload this to task_work_run(), and pass the argument via not
yet used utask->vaddr.
We know that this vaddr is fine for install_special_mapping(), the
necessary hole was recently "created" by dup_mmap() which skips the
parent's VM_DONTCOPY area, and nobody else could use the new mm.
Unfortunately, this also means that we can not handle the errors
properly, we obviously can not abort the already completed fork().
So we simply print the warning if GFP_KERNEL allocation (the only
possible reason) fails.
Reported-by: Martin Cermak <mcermak@redhat.com>
Reported-by: David Smith <dsmith@redhat.com>
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Acked-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
2013-10-14 03:18:44 +08:00
|
|
|
struct xol_area *area;
|
2013-10-14 03:18:41 +08:00
|
|
|
|
2013-10-14 03:18:31 +08:00
|
|
|
t->utask = NULL;
|
2013-10-14 03:18:41 +08:00
|
|
|
|
2013-10-17 01:39:37 +08:00
|
|
|
if (!utask || !utask->return_instances)
|
|
|
|
return;
|
|
|
|
|
|
|
|
if (mm == t->mm && !(flags & CLONE_VFORK))
|
2013-10-14 03:18:41 +08:00
|
|
|
return;
|
|
|
|
|
|
|
|
if (dup_utask(t, utask))
|
|
|
|
return uprobe_warn(t, "dup ret instances");
|
uprobes: Change uprobe_copy_process() to dup xol_area
This finally fixes the serious bug in uretprobes: a forked child
crashes if the parent called fork() with the pending ret probe.
Trivial test-case:
# perf probe -x /lib/libc.so.6 __fork%return
# perf record -e probe_libc:__fork perl -le 'fork || print "OK"'
(the child doesn't print "OK", it is killed by SIGSEGV)
If the child returns from the probed function it actually returns
to trampoline_vaddr, because it got the copy of parent's stack
mangled by prepare_uretprobe() when the parent entered this func.
It crashes because a) this address is not mapped and b) until the
previous change it doesn't have the proper->return_instances info.
This means that uprobe_copy_process() has to create xol_area which
has the trampoline slot, and its vaddr should be equal to parent's
xol_area->vaddr.
Unfortunately, uprobe_copy_process() can not simply do
__create_xol_area(child, xol_area->vaddr). This could actually work
but perf_event_mmap() doesn't expect the usage of foreign ->mm. So
we offload this to task_work_run(), and pass the argument via not
yet used utask->vaddr.
We know that this vaddr is fine for install_special_mapping(), the
necessary hole was recently "created" by dup_mmap() which skips the
parent's VM_DONTCOPY area, and nobody else could use the new mm.
Unfortunately, this also means that we can not handle the errors
properly, we obviously can not abort the already completed fork().
So we simply print the warning if GFP_KERNEL allocation (the only
possible reason) fails.
Reported-by: Martin Cermak <mcermak@redhat.com>
Reported-by: David Smith <dsmith@redhat.com>
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Acked-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
2013-10-14 03:18:44 +08:00
|
|
|
|
|
|
|
/* The task can fork() after dup_xol_work() fails */
|
|
|
|
area = mm->uprobes_state.xol_area;
|
|
|
|
if (!area)
|
|
|
|
return uprobe_warn(t, "dup xol area");
|
|
|
|
|
2013-10-17 01:39:37 +08:00
|
|
|
if (mm == t->mm)
|
|
|
|
return;
|
|
|
|
|
2013-11-09 01:52:21 +08:00
|
|
|
t->utask->dup_xol_addr = area->vaddr;
|
|
|
|
init_task_work(&t->utask->dup_xol_work, dup_xol_work);
|
|
|
|
task_work_add(t, &t->utask->dup_xol_work, true);
|
2013-10-14 03:18:31 +08:00
|
|
|
}
|
|
|
|
|
2013-04-04 00:00:32 +08:00
|
|
|
/*
|
|
|
|
* Current area->vaddr notion assume the trampoline address is always
|
|
|
|
* equal area->vaddr.
|
|
|
|
*
|
|
|
|
* Returns -1 in case the xol_area is not allocated.
|
|
|
|
*/
|
|
|
|
static unsigned long get_trampoline_vaddr(void)
|
|
|
|
{
|
|
|
|
struct xol_area *area;
|
|
|
|
unsigned long trampoline_vaddr = -1;
|
|
|
|
|
2017-10-10 02:08:53 +08:00
|
|
|
/* Pairs with xol_add_vma() smp_store_release() */
|
|
|
|
area = READ_ONCE(current->mm->uprobes_state.xol_area); /* ^^^ */
|
2013-04-04 00:00:32 +08:00
|
|
|
if (area)
|
|
|
|
trampoline_vaddr = area->vaddr;
|
|
|
|
|
|
|
|
return trampoline_vaddr;
|
|
|
|
}
|
|
|
|
|
2015-07-21 21:40:28 +08:00
|
|
|
static void cleanup_return_instances(struct uprobe_task *utask, bool chained,
|
|
|
|
struct pt_regs *regs)
|
2015-07-21 21:40:23 +08:00
|
|
|
{
|
|
|
|
struct return_instance *ri = utask->return_instances;
|
2015-07-21 21:40:28 +08:00
|
|
|
enum rp_check ctx = chained ? RP_CHECK_CHAIN_CALL : RP_CHECK_CALL;
|
2015-07-21 21:40:26 +08:00
|
|
|
|
|
|
|
while (ri && !arch_uretprobe_is_alive(ri, ctx, regs)) {
|
2015-07-21 21:40:23 +08:00
|
|
|
ri = free_ret_instance(ri);
|
|
|
|
utask->depth--;
|
|
|
|
}
|
|
|
|
utask->return_instances = ri;
|
|
|
|
}
|
|
|
|
|
2013-04-04 00:00:35 +08:00
|
|
|
static void prepare_uretprobe(struct uprobe *uprobe, struct pt_regs *regs)
|
|
|
|
{
|
|
|
|
struct return_instance *ri;
|
|
|
|
struct uprobe_task *utask;
|
|
|
|
unsigned long orig_ret_vaddr, trampoline_vaddr;
|
2015-07-21 21:40:28 +08:00
|
|
|
bool chained;
|
2013-04-04 00:00:35 +08:00
|
|
|
|
|
|
|
if (!get_xol_area())
|
|
|
|
return;
|
|
|
|
|
|
|
|
utask = get_utask();
|
|
|
|
if (!utask)
|
|
|
|
return;
|
|
|
|
|
2013-04-04 00:00:37 +08:00
|
|
|
if (utask->depth >= MAX_URETPROBE_DEPTH) {
|
|
|
|
printk_ratelimited(KERN_INFO "uprobe: omit uretprobe due to"
|
|
|
|
" nestedness limit pid/tgid=%d/%d\n",
|
|
|
|
current->pid, current->tgid);
|
|
|
|
return;
|
|
|
|
}
|
|
|
|
|
2015-07-21 21:40:10 +08:00
|
|
|
ri = kmalloc(sizeof(struct return_instance), GFP_KERNEL);
|
2013-04-04 00:00:35 +08:00
|
|
|
if (!ri)
|
2015-07-21 21:40:10 +08:00
|
|
|
return;
|
2013-04-04 00:00:35 +08:00
|
|
|
|
|
|
|
trampoline_vaddr = get_trampoline_vaddr();
|
|
|
|
orig_ret_vaddr = arch_uretprobe_hijack_return_addr(trampoline_vaddr, regs);
|
|
|
|
if (orig_ret_vaddr == -1)
|
|
|
|
goto fail;
|
|
|
|
|
2015-07-21 21:40:23 +08:00
|
|
|
/* drop the entries invalidated by longjmp() */
|
2015-07-21 21:40:28 +08:00
|
|
|
chained = (orig_ret_vaddr == trampoline_vaddr);
|
|
|
|
cleanup_return_instances(utask, chained, regs);
|
2015-07-21 21:40:23 +08:00
|
|
|
|
2013-04-04 00:00:35 +08:00
|
|
|
/*
|
|
|
|
* We don't want to keep trampoline address in stack, rather keep the
|
|
|
|
* original return address of first caller thru all the consequent
|
|
|
|
* instances. This also makes breakpoint unwrapping easier.
|
|
|
|
*/
|
2015-07-21 21:40:28 +08:00
|
|
|
if (chained) {
|
2013-04-04 00:00:35 +08:00
|
|
|
if (!utask->return_instances) {
|
|
|
|
/*
|
|
|
|
* This situation is not possible. Likely we have an
|
|
|
|
* attack from user-space.
|
|
|
|
*/
|
2015-07-21 21:40:10 +08:00
|
|
|
uprobe_warn(current, "handle tail call");
|
2013-04-04 00:00:35 +08:00
|
|
|
goto fail;
|
|
|
|
}
|
|
|
|
orig_ret_vaddr = utask->return_instances->orig_ret_vaddr;
|
|
|
|
}
|
|
|
|
|
2015-07-21 21:40:03 +08:00
|
|
|
ri->uprobe = get_uprobe(uprobe);
|
2013-04-04 00:00:35 +08:00
|
|
|
ri->func = instruction_pointer(regs);
|
2015-07-21 21:40:18 +08:00
|
|
|
ri->stack = user_stack_pointer(regs);
|
2013-04-04 00:00:35 +08:00
|
|
|
ri->orig_ret_vaddr = orig_ret_vaddr;
|
|
|
|
ri->chained = chained;
|
|
|
|
|
2013-04-04 00:00:37 +08:00
|
|
|
utask->depth++;
|
2013-04-04 00:00:35 +08:00
|
|
|
ri->next = utask->return_instances;
|
|
|
|
utask->return_instances = ri;
|
|
|
|
|
|
|
|
return;
|
|
|
|
fail:
|
|
|
|
kfree(ri);
|
|
|
|
}
|
|
|
|
|
uprobes/core: Handle breakpoint and singlestep exceptions
Uprobes uses exception notifiers to get to know if a thread hit
a breakpoint or a singlestep exception.
When a thread hits a uprobe or is singlestepping post a uprobe
hit, the uprobe exception notifier sets its TIF_UPROBE bit,
which will then be checked on its return to userspace path
(do_notify_resume() ->uprobe_notify_resume()), where the
consumers handlers are run (in task context) based on the
defined filters.
Uprobe hits are thread specific and hence we need to maintain
information about if a task hit a uprobe, what uprobe was hit,
the slot where the original instruction was copied for xol so
that it can be singlestepped with appropriate fixups.
In some cases, special care is needed for instructions that are
executed out of line (xol). These are architecture specific
artefacts, such as handling RIP relative instructions on x86_64.
Since the instruction at which the uprobe was inserted is
executed out of line, architecture specific fixups are added so
that the thread continues normal execution in the presence of a
uprobe.
Postpone the signals until we execute the probed insn.
post_xol() path does a recalc_sigpending() before return to
user-mode, this ensures the signal can't be lost.
Uprobes relies on DIE_DEBUG notification to notify if a
singlestep is complete.
Adds x86 specific uprobe exception notifiers and appropriate
hooks needed to determine a uprobe hit and subsequent post
processing.
Add requisite x86 fixups for xol for uprobes. Specific cases
needing fixups include relative jumps (x86_64), calls, etc.
Where possible, we check and skip singlestepping the
breakpointed instructions. For now we skip single byte as well
as few multibyte nop instructions. However this can be extended
to other instructions too.
Credits to Oleg Nesterov for suggestions/patches related to
signal, breakpoint, singlestep handling code.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Jim Keniston <jkenisto@linux.vnet.ibm.com>
Cc: Linux-mm <linux-mm@kvack.org>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/20120313180011.29771.89027.sendpatchset@srdronam.in.ibm.com
[ Performed various cleanliness edits ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-03-14 02:00:11 +08:00
|
|
|
/* Prepare to single-step probed instruction out of line. */
|
|
|
|
static int
|
2013-01-01 01:00:06 +08:00
|
|
|
pre_ssout(struct uprobe *uprobe, struct pt_regs *regs, unsigned long bp_vaddr)
|
uprobes/core: Handle breakpoint and singlestep exceptions
Uprobes uses exception notifiers to get to know if a thread hit
a breakpoint or a singlestep exception.
When a thread hits a uprobe or is singlestepping post a uprobe
hit, the uprobe exception notifier sets its TIF_UPROBE bit,
which will then be checked on its return to userspace path
(do_notify_resume() ->uprobe_notify_resume()), where the
consumers handlers are run (in task context) based on the
defined filters.
Uprobe hits are thread specific and hence we need to maintain
information about if a task hit a uprobe, what uprobe was hit,
the slot where the original instruction was copied for xol so
that it can be singlestepped with appropriate fixups.
In some cases, special care is needed for instructions that are
executed out of line (xol). These are architecture specific
artefacts, such as handling RIP relative instructions on x86_64.
Since the instruction at which the uprobe was inserted is
executed out of line, architecture specific fixups are added so
that the thread continues normal execution in the presence of a
uprobe.
Postpone the signals until we execute the probed insn.
post_xol() path does a recalc_sigpending() before return to
user-mode, this ensures the signal can't be lost.
Uprobes relies on DIE_DEBUG notification to notify if a
singlestep is complete.
Adds x86 specific uprobe exception notifiers and appropriate
hooks needed to determine a uprobe hit and subsequent post
processing.
Add requisite x86 fixups for xol for uprobes. Specific cases
needing fixups include relative jumps (x86_64), calls, etc.
Where possible, we check and skip singlestepping the
breakpointed instructions. For now we skip single byte as well
as few multibyte nop instructions. However this can be extended
to other instructions too.
Credits to Oleg Nesterov for suggestions/patches related to
signal, breakpoint, singlestep handling code.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Jim Keniston <jkenisto@linux.vnet.ibm.com>
Cc: Linux-mm <linux-mm@kvack.org>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/20120313180011.29771.89027.sendpatchset@srdronam.in.ibm.com
[ Performed various cleanliness edits ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-03-14 02:00:11 +08:00
|
|
|
{
|
2013-01-01 01:00:06 +08:00
|
|
|
struct uprobe_task *utask;
|
|
|
|
unsigned long xol_vaddr;
|
2013-01-01 01:12:48 +08:00
|
|
|
int err;
|
2013-01-01 01:00:06 +08:00
|
|
|
|
2013-01-01 01:20:42 +08:00
|
|
|
utask = get_utask();
|
|
|
|
if (!utask)
|
|
|
|
return -ENOMEM;
|
2013-01-01 01:00:06 +08:00
|
|
|
|
|
|
|
xol_vaddr = xol_get_insn_slot(uprobe);
|
|
|
|
if (!xol_vaddr)
|
|
|
|
return -ENOMEM;
|
|
|
|
|
|
|
|
utask->xol_vaddr = xol_vaddr;
|
|
|
|
utask->vaddr = bp_vaddr;
|
2012-03-31 02:26:31 +08:00
|
|
|
|
2013-01-01 01:12:48 +08:00
|
|
|
err = arch_uprobe_pre_xol(&uprobe->arch, regs);
|
|
|
|
if (unlikely(err)) {
|
|
|
|
xol_free_insn_slot(current);
|
|
|
|
return err;
|
|
|
|
}
|
|
|
|
|
2013-01-01 01:20:42 +08:00
|
|
|
utask->active_uprobe = uprobe;
|
|
|
|
utask->state = UTASK_SSTEP;
|
2013-01-01 01:12:48 +08:00
|
|
|
return 0;
|
uprobes/core: Handle breakpoint and singlestep exceptions
Uprobes uses exception notifiers to get to know if a thread hit
a breakpoint or a singlestep exception.
When a thread hits a uprobe or is singlestepping post a uprobe
hit, the uprobe exception notifier sets its TIF_UPROBE bit,
which will then be checked on its return to userspace path
(do_notify_resume() ->uprobe_notify_resume()), where the
consumers handlers are run (in task context) based on the
defined filters.
Uprobe hits are thread specific and hence we need to maintain
information about if a task hit a uprobe, what uprobe was hit,
the slot where the original instruction was copied for xol so
that it can be singlestepped with appropriate fixups.
In some cases, special care is needed for instructions that are
executed out of line (xol). These are architecture specific
artefacts, such as handling RIP relative instructions on x86_64.
Since the instruction at which the uprobe was inserted is
executed out of line, architecture specific fixups are added so
that the thread continues normal execution in the presence of a
uprobe.
Postpone the signals until we execute the probed insn.
post_xol() path does a recalc_sigpending() before return to
user-mode, this ensures the signal can't be lost.
Uprobes relies on DIE_DEBUG notification to notify if a
singlestep is complete.
Adds x86 specific uprobe exception notifiers and appropriate
hooks needed to determine a uprobe hit and subsequent post
processing.
Add requisite x86 fixups for xol for uprobes. Specific cases
needing fixups include relative jumps (x86_64), calls, etc.
Where possible, we check and skip singlestepping the
breakpointed instructions. For now we skip single byte as well
as few multibyte nop instructions. However this can be extended
to other instructions too.
Credits to Oleg Nesterov for suggestions/patches related to
signal, breakpoint, singlestep handling code.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Jim Keniston <jkenisto@linux.vnet.ibm.com>
Cc: Linux-mm <linux-mm@kvack.org>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/20120313180011.29771.89027.sendpatchset@srdronam.in.ibm.com
[ Performed various cleanliness edits ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-03-14 02:00:11 +08:00
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* If we are singlestepping, then ensure this thread is not connected to
|
|
|
|
* non-fatal signals until completion of singlestep. When xol insn itself
|
|
|
|
* triggers the signal, restart the original insn even if the task is
|
|
|
|
* already SIGKILL'ed (since coredump should report the correct ip). This
|
|
|
|
* is even more important if the task has a handler for SIGSEGV/etc, The
|
|
|
|
* _same_ instruction should be repeated again after return from the signal
|
|
|
|
* handler, and SSTEP can never finish in this case.
|
|
|
|
*/
|
|
|
|
bool uprobe_deny_signal(void)
|
|
|
|
{
|
|
|
|
struct task_struct *t = current;
|
|
|
|
struct uprobe_task *utask = t->utask;
|
|
|
|
|
|
|
|
if (likely(!utask || !utask->active_uprobe))
|
|
|
|
return false;
|
|
|
|
|
|
|
|
WARN_ON_ONCE(utask->state != UTASK_SSTEP);
|
|
|
|
|
|
|
|
if (signal_pending(t)) {
|
|
|
|
spin_lock_irq(&t->sighand->siglock);
|
|
|
|
clear_tsk_thread_flag(t, TIF_SIGPENDING);
|
|
|
|
spin_unlock_irq(&t->sighand->siglock);
|
|
|
|
|
|
|
|
if (__fatal_signal_pending(t) || arch_uprobe_xol_was_trapped(t)) {
|
|
|
|
utask->state = UTASK_SSTEP_TRAPPED;
|
|
|
|
set_tsk_thread_flag(t, TIF_UPROBE);
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
return true;
|
|
|
|
}
|
|
|
|
|
2012-08-19 23:41:34 +08:00
|
|
|
static void mmf_recalc_uprobes(struct mm_struct *mm)
|
|
|
|
{
|
|
|
|
struct vm_area_struct *vma;
|
|
|
|
|
|
|
|
for (vma = mm->mmap; vma; vma = vma->vm_next) {
|
|
|
|
if (!valid_vma(vma, false))
|
|
|
|
continue;
|
|
|
|
/*
|
|
|
|
* This is not strictly accurate, we can race with
|
|
|
|
* uprobe_unregister() and see the already removed
|
|
|
|
* uprobe if delete_uprobe() was not yet called.
|
2012-11-23 01:30:15 +08:00
|
|
|
* Or this uprobe can be filtered out.
|
2012-08-19 23:41:34 +08:00
|
|
|
*/
|
|
|
|
if (vma_has_uprobes(vma, vma->vm_start, vma->vm_end))
|
|
|
|
return;
|
|
|
|
}
|
|
|
|
|
|
|
|
clear_bit(MMF_HAS_UPROBES, &mm->flags);
|
|
|
|
}
|
|
|
|
|
2013-03-22 23:16:27 +08:00
|
|
|
static int is_trap_at_addr(struct mm_struct *mm, unsigned long vaddr)
|
2012-09-24 03:55:19 +08:00
|
|
|
{
|
|
|
|
struct page *page;
|
|
|
|
uprobe_opcode_t opcode;
|
|
|
|
int result;
|
|
|
|
|
|
|
|
pagefault_disable();
|
2016-05-23 08:21:27 +08:00
|
|
|
result = __get_user(opcode, (uprobe_opcode_t __user *)vaddr);
|
2012-09-24 03:55:19 +08:00
|
|
|
pagefault_enable();
|
|
|
|
|
|
|
|
if (likely(result == 0))
|
|
|
|
goto out;
|
|
|
|
|
2016-02-13 05:01:54 +08:00
|
|
|
/*
|
|
|
|
* The NULL 'tsk' here ensures that any faults that occur here
|
|
|
|
* will not be accounted to the task. 'mm' *is* current->mm,
|
|
|
|
* but we treat this as a 'remote' access since it is
|
|
|
|
* essentially a kernel access to the memory.
|
|
|
|
*/
|
2016-10-13 08:20:17 +08:00
|
|
|
result = get_user_pages_remote(NULL, mm, vaddr, 1, FOLL_FORCE, &page,
|
2016-12-15 07:06:52 +08:00
|
|
|
NULL, NULL);
|
2012-09-24 03:55:19 +08:00
|
|
|
if (result < 0)
|
|
|
|
return result;
|
|
|
|
|
2013-03-25 01:24:37 +08:00
|
|
|
copy_from_page(page, vaddr, &opcode, UPROBE_SWBP_INSN_SIZE);
|
2012-09-24 03:55:19 +08:00
|
|
|
put_page(page);
|
|
|
|
out:
|
2013-03-22 23:16:27 +08:00
|
|
|
/* This needs to return true for any variant of the trap insn */
|
|
|
|
return is_trap_insn(&opcode);
|
2012-09-24 03:55:19 +08:00
|
|
|
}
|
|
|
|
|
uprobes: Teach find_active_uprobe() to provide the "is_swbp" info
A separate patch to simplify the review, and for the
documentation.
The patch adds another "int *is_swbp" argument to
find_active_uprobe(), so far its only caller doesn't use this
info.
With this patch find_active_uprobe() additionally does:
- if find_vma() + ->vm_start check fails, *is_swbp = -EFAULT
- otherwise, if valid_vma() + find_uprobe() fails, it holds
the result of is_swbp_at_addr(), can be negative too. The
latter is only possible if we raced with another thread
which did munmap/etc after we hit this bp.
IOW. If find_active_uprobe(&is_swbp) returns NULL, the caller
can look at is_swbp to figure out whether the current insn is bp
or not, or detect the race with another thread if it is
negative.
Note: I think that performance-wise this change is fine. This
adds is_swbp_at_addr(), but only if we raced with
uprobe_unregister() or if we hit the "normal" int3 but this mm
has uprobes as well. And even in this case the slow
read_opcode() path is very unlikely, this insn recently
triggered do_int3(), __copy_from_user_inatomic() shouldn't fail
in the likely case.
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Acked-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Anton Arapov <anton@redhat.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/20120529192914.GD8057@redhat.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
2012-05-30 03:29:14 +08:00
|
|
|
static struct uprobe *find_active_uprobe(unsigned long bp_vaddr, int *is_swbp)
|
uprobes/core: Handle breakpoint and singlestep exceptions
Uprobes uses exception notifiers to get to know if a thread hit
a breakpoint or a singlestep exception.
When a thread hits a uprobe or is singlestepping post a uprobe
hit, the uprobe exception notifier sets its TIF_UPROBE bit,
which will then be checked on its return to userspace path
(do_notify_resume() ->uprobe_notify_resume()), where the
consumers handlers are run (in task context) based on the
defined filters.
Uprobe hits are thread specific and hence we need to maintain
information about if a task hit a uprobe, what uprobe was hit,
the slot where the original instruction was copied for xol so
that it can be singlestepped with appropriate fixups.
In some cases, special care is needed for instructions that are
executed out of line (xol). These are architecture specific
artefacts, such as handling RIP relative instructions on x86_64.
Since the instruction at which the uprobe was inserted is
executed out of line, architecture specific fixups are added so
that the thread continues normal execution in the presence of a
uprobe.
Postpone the signals until we execute the probed insn.
post_xol() path does a recalc_sigpending() before return to
user-mode, this ensures the signal can't be lost.
Uprobes relies on DIE_DEBUG notification to notify if a
singlestep is complete.
Adds x86 specific uprobe exception notifiers and appropriate
hooks needed to determine a uprobe hit and subsequent post
processing.
Add requisite x86 fixups for xol for uprobes. Specific cases
needing fixups include relative jumps (x86_64), calls, etc.
Where possible, we check and skip singlestepping the
breakpointed instructions. For now we skip single byte as well
as few multibyte nop instructions. However this can be extended
to other instructions too.
Credits to Oleg Nesterov for suggestions/patches related to
signal, breakpoint, singlestep handling code.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Jim Keniston <jkenisto@linux.vnet.ibm.com>
Cc: Linux-mm <linux-mm@kvack.org>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/20120313180011.29771.89027.sendpatchset@srdronam.in.ibm.com
[ Performed various cleanliness edits ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-03-14 02:00:11 +08:00
|
|
|
{
|
2012-05-30 03:28:57 +08:00
|
|
|
struct mm_struct *mm = current->mm;
|
|
|
|
struct uprobe *uprobe = NULL;
|
uprobes/core: Handle breakpoint and singlestep exceptions
Uprobes uses exception notifiers to get to know if a thread hit
a breakpoint or a singlestep exception.
When a thread hits a uprobe or is singlestepping post a uprobe
hit, the uprobe exception notifier sets its TIF_UPROBE bit,
which will then be checked on its return to userspace path
(do_notify_resume() ->uprobe_notify_resume()), where the
consumers handlers are run (in task context) based on the
defined filters.
Uprobe hits are thread specific and hence we need to maintain
information about if a task hit a uprobe, what uprobe was hit,
the slot where the original instruction was copied for xol so
that it can be singlestepped with appropriate fixups.
In some cases, special care is needed for instructions that are
executed out of line (xol). These are architecture specific
artefacts, such as handling RIP relative instructions on x86_64.
Since the instruction at which the uprobe was inserted is
executed out of line, architecture specific fixups are added so
that the thread continues normal execution in the presence of a
uprobe.
Postpone the signals until we execute the probed insn.
post_xol() path does a recalc_sigpending() before return to
user-mode, this ensures the signal can't be lost.
Uprobes relies on DIE_DEBUG notification to notify if a
singlestep is complete.
Adds x86 specific uprobe exception notifiers and appropriate
hooks needed to determine a uprobe hit and subsequent post
processing.
Add requisite x86 fixups for xol for uprobes. Specific cases
needing fixups include relative jumps (x86_64), calls, etc.
Where possible, we check and skip singlestepping the
breakpointed instructions. For now we skip single byte as well
as few multibyte nop instructions. However this can be extended
to other instructions too.
Credits to Oleg Nesterov for suggestions/patches related to
signal, breakpoint, singlestep handling code.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Jim Keniston <jkenisto@linux.vnet.ibm.com>
Cc: Linux-mm <linux-mm@kvack.org>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/20120313180011.29771.89027.sendpatchset@srdronam.in.ibm.com
[ Performed various cleanliness edits ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-03-14 02:00:11 +08:00
|
|
|
struct vm_area_struct *vma;
|
|
|
|
|
|
|
|
down_read(&mm->mmap_sem);
|
|
|
|
vma = find_vma(mm, bp_vaddr);
|
2012-05-30 03:28:57 +08:00
|
|
|
if (vma && vma->vm_start <= bp_vaddr) {
|
|
|
|
if (valid_vma(vma, false)) {
|
2013-03-18 01:54:44 +08:00
|
|
|
struct inode *inode = file_inode(vma->vm_file);
|
2012-07-30 02:22:42 +08:00
|
|
|
loff_t offset = vaddr_to_offset(vma, bp_vaddr);
|
uprobes/core: Handle breakpoint and singlestep exceptions
Uprobes uses exception notifiers to get to know if a thread hit
a breakpoint or a singlestep exception.
When a thread hits a uprobe or is singlestepping post a uprobe
hit, the uprobe exception notifier sets its TIF_UPROBE bit,
which will then be checked on its return to userspace path
(do_notify_resume() ->uprobe_notify_resume()), where the
consumers handlers are run (in task context) based on the
defined filters.
Uprobe hits are thread specific and hence we need to maintain
information about if a task hit a uprobe, what uprobe was hit,
the slot where the original instruction was copied for xol so
that it can be singlestepped with appropriate fixups.
In some cases, special care is needed for instructions that are
executed out of line (xol). These are architecture specific
artefacts, such as handling RIP relative instructions on x86_64.
Since the instruction at which the uprobe was inserted is
executed out of line, architecture specific fixups are added so
that the thread continues normal execution in the presence of a
uprobe.
Postpone the signals until we execute the probed insn.
post_xol() path does a recalc_sigpending() before return to
user-mode, this ensures the signal can't be lost.
Uprobes relies on DIE_DEBUG notification to notify if a
singlestep is complete.
Adds x86 specific uprobe exception notifiers and appropriate
hooks needed to determine a uprobe hit and subsequent post
processing.
Add requisite x86 fixups for xol for uprobes. Specific cases
needing fixups include relative jumps (x86_64), calls, etc.
Where possible, we check and skip singlestepping the
breakpointed instructions. For now we skip single byte as well
as few multibyte nop instructions. However this can be extended
to other instructions too.
Credits to Oleg Nesterov for suggestions/patches related to
signal, breakpoint, singlestep handling code.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Jim Keniston <jkenisto@linux.vnet.ibm.com>
Cc: Linux-mm <linux-mm@kvack.org>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/20120313180011.29771.89027.sendpatchset@srdronam.in.ibm.com
[ Performed various cleanliness edits ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-03-14 02:00:11 +08:00
|
|
|
|
2012-05-30 03:28:57 +08:00
|
|
|
uprobe = find_uprobe(inode, offset);
|
|
|
|
}
|
uprobes: Teach find_active_uprobe() to provide the "is_swbp" info
A separate patch to simplify the review, and for the
documentation.
The patch adds another "int *is_swbp" argument to
find_active_uprobe(), so far its only caller doesn't use this
info.
With this patch find_active_uprobe() additionally does:
- if find_vma() + ->vm_start check fails, *is_swbp = -EFAULT
- otherwise, if valid_vma() + find_uprobe() fails, it holds
the result of is_swbp_at_addr(), can be negative too. The
latter is only possible if we raced with another thread
which did munmap/etc after we hit this bp.
IOW. If find_active_uprobe(&is_swbp) returns NULL, the caller
can look at is_swbp to figure out whether the current insn is bp
or not, or detect the race with another thread if it is
negative.
Note: I think that performance-wise this change is fine. This
adds is_swbp_at_addr(), but only if we raced with
uprobe_unregister() or if we hit the "normal" int3 but this mm
has uprobes as well. And even in this case the slow
read_opcode() path is very unlikely, this insn recently
triggered do_int3(), __copy_from_user_inatomic() shouldn't fail
in the likely case.
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Acked-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Anton Arapov <anton@redhat.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/20120529192914.GD8057@redhat.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
2012-05-30 03:29:14 +08:00
|
|
|
|
|
|
|
if (!uprobe)
|
2013-03-22 23:16:27 +08:00
|
|
|
*is_swbp = is_trap_at_addr(mm, bp_vaddr);
|
uprobes: Teach find_active_uprobe() to provide the "is_swbp" info
A separate patch to simplify the review, and for the
documentation.
The patch adds another "int *is_swbp" argument to
find_active_uprobe(), so far its only caller doesn't use this
info.
With this patch find_active_uprobe() additionally does:
- if find_vma() + ->vm_start check fails, *is_swbp = -EFAULT
- otherwise, if valid_vma() + find_uprobe() fails, it holds
the result of is_swbp_at_addr(), can be negative too. The
latter is only possible if we raced with another thread
which did munmap/etc after we hit this bp.
IOW. If find_active_uprobe(&is_swbp) returns NULL, the caller
can look at is_swbp to figure out whether the current insn is bp
or not, or detect the race with another thread if it is
negative.
Note: I think that performance-wise this change is fine. This
adds is_swbp_at_addr(), but only if we raced with
uprobe_unregister() or if we hit the "normal" int3 but this mm
has uprobes as well. And even in this case the slow
read_opcode() path is very unlikely, this insn recently
triggered do_int3(), __copy_from_user_inatomic() shouldn't fail
in the likely case.
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Acked-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Anton Arapov <anton@redhat.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/20120529192914.GD8057@redhat.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
2012-05-30 03:29:14 +08:00
|
|
|
} else {
|
|
|
|
*is_swbp = -EFAULT;
|
uprobes/core: Handle breakpoint and singlestep exceptions
Uprobes uses exception notifiers to get to know if a thread hit
a breakpoint or a singlestep exception.
When a thread hits a uprobe or is singlestepping post a uprobe
hit, the uprobe exception notifier sets its TIF_UPROBE bit,
which will then be checked on its return to userspace path
(do_notify_resume() ->uprobe_notify_resume()), where the
consumers handlers are run (in task context) based on the
defined filters.
Uprobe hits are thread specific and hence we need to maintain
information about if a task hit a uprobe, what uprobe was hit,
the slot where the original instruction was copied for xol so
that it can be singlestepped with appropriate fixups.
In some cases, special care is needed for instructions that are
executed out of line (xol). These are architecture specific
artefacts, such as handling RIP relative instructions on x86_64.
Since the instruction at which the uprobe was inserted is
executed out of line, architecture specific fixups are added so
that the thread continues normal execution in the presence of a
uprobe.
Postpone the signals until we execute the probed insn.
post_xol() path does a recalc_sigpending() before return to
user-mode, this ensures the signal can't be lost.
Uprobes relies on DIE_DEBUG notification to notify if a
singlestep is complete.
Adds x86 specific uprobe exception notifiers and appropriate
hooks needed to determine a uprobe hit and subsequent post
processing.
Add requisite x86 fixups for xol for uprobes. Specific cases
needing fixups include relative jumps (x86_64), calls, etc.
Where possible, we check and skip singlestepping the
breakpointed instructions. For now we skip single byte as well
as few multibyte nop instructions. However this can be extended
to other instructions too.
Credits to Oleg Nesterov for suggestions/patches related to
signal, breakpoint, singlestep handling code.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Jim Keniston <jkenisto@linux.vnet.ibm.com>
Cc: Linux-mm <linux-mm@kvack.org>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/20120313180011.29771.89027.sendpatchset@srdronam.in.ibm.com
[ Performed various cleanliness edits ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-03-14 02:00:11 +08:00
|
|
|
}
|
2012-08-19 23:41:34 +08:00
|
|
|
|
|
|
|
if (!uprobe && test_and_clear_bit(MMF_RECALC_UPROBES, &mm->flags))
|
|
|
|
mmf_recalc_uprobes(mm);
|
uprobes/core: Handle breakpoint and singlestep exceptions
Uprobes uses exception notifiers to get to know if a thread hit
a breakpoint or a singlestep exception.
When a thread hits a uprobe or is singlestepping post a uprobe
hit, the uprobe exception notifier sets its TIF_UPROBE bit,
which will then be checked on its return to userspace path
(do_notify_resume() ->uprobe_notify_resume()), where the
consumers handlers are run (in task context) based on the
defined filters.
Uprobe hits are thread specific and hence we need to maintain
information about if a task hit a uprobe, what uprobe was hit,
the slot where the original instruction was copied for xol so
that it can be singlestepped with appropriate fixups.
In some cases, special care is needed for instructions that are
executed out of line (xol). These are architecture specific
artefacts, such as handling RIP relative instructions on x86_64.
Since the instruction at which the uprobe was inserted is
executed out of line, architecture specific fixups are added so
that the thread continues normal execution in the presence of a
uprobe.
Postpone the signals until we execute the probed insn.
post_xol() path does a recalc_sigpending() before return to
user-mode, this ensures the signal can't be lost.
Uprobes relies on DIE_DEBUG notification to notify if a
singlestep is complete.
Adds x86 specific uprobe exception notifiers and appropriate
hooks needed to determine a uprobe hit and subsequent post
processing.
Add requisite x86 fixups for xol for uprobes. Specific cases
needing fixups include relative jumps (x86_64), calls, etc.
Where possible, we check and skip singlestepping the
breakpointed instructions. For now we skip single byte as well
as few multibyte nop instructions. However this can be extended
to other instructions too.
Credits to Oleg Nesterov for suggestions/patches related to
signal, breakpoint, singlestep handling code.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Jim Keniston <jkenisto@linux.vnet.ibm.com>
Cc: Linux-mm <linux-mm@kvack.org>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/20120313180011.29771.89027.sendpatchset@srdronam.in.ibm.com
[ Performed various cleanliness edits ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-03-14 02:00:11 +08:00
|
|
|
up_read(&mm->mmap_sem);
|
|
|
|
|
2012-05-30 03:28:57 +08:00
|
|
|
return uprobe;
|
|
|
|
}
|
|
|
|
|
2012-12-30 00:49:11 +08:00
|
|
|
static void handler_chain(struct uprobe *uprobe, struct pt_regs *regs)
|
|
|
|
{
|
|
|
|
struct uprobe_consumer *uc;
|
|
|
|
int remove = UPROBE_HANDLER_REMOVE;
|
2013-04-04 00:00:35 +08:00
|
|
|
bool need_prep = false; /* prepare return uprobe, when needed */
|
2012-12-30 00:49:11 +08:00
|
|
|
|
|
|
|
down_read(&uprobe->register_rwsem);
|
|
|
|
for (uc = uprobe->consumers; uc; uc = uc->next) {
|
2013-04-04 00:00:31 +08:00
|
|
|
int rc = 0;
|
2012-12-30 00:49:11 +08:00
|
|
|
|
2013-04-04 00:00:31 +08:00
|
|
|
if (uc->handler) {
|
|
|
|
rc = uc->handler(uc, regs);
|
|
|
|
WARN(rc & ~UPROBE_HANDLER_MASK,
|
|
|
|
"bad rc=0x%x from %pf()\n", rc, uc->handler);
|
|
|
|
}
|
2013-04-04 00:00:35 +08:00
|
|
|
|
|
|
|
if (uc->ret_handler)
|
|
|
|
need_prep = true;
|
|
|
|
|
2012-12-30 00:49:11 +08:00
|
|
|
remove &= rc;
|
|
|
|
}
|
|
|
|
|
2013-04-04 00:00:35 +08:00
|
|
|
if (need_prep && !remove)
|
|
|
|
prepare_uretprobe(uprobe, regs); /* put bp at return */
|
|
|
|
|
2012-12-30 00:49:11 +08:00
|
|
|
if (remove && uprobe->consumers) {
|
|
|
|
WARN_ON(!uprobe_is_active(uprobe));
|
|
|
|
unapply_uprobe(uprobe, current->mm);
|
|
|
|
}
|
|
|
|
up_read(&uprobe->register_rwsem);
|
|
|
|
}
|
|
|
|
|
2013-04-04 00:00:36 +08:00
|
|
|
static void
|
|
|
|
handle_uretprobe_chain(struct return_instance *ri, struct pt_regs *regs)
|
|
|
|
{
|
|
|
|
struct uprobe *uprobe = ri->uprobe;
|
|
|
|
struct uprobe_consumer *uc;
|
|
|
|
|
|
|
|
down_read(&uprobe->register_rwsem);
|
|
|
|
for (uc = uprobe->consumers; uc; uc = uc->next) {
|
|
|
|
if (uc->ret_handler)
|
|
|
|
uc->ret_handler(uc, ri->func, regs);
|
|
|
|
}
|
|
|
|
up_read(&uprobe->register_rwsem);
|
|
|
|
}
|
|
|
|
|
2015-07-21 21:40:13 +08:00
|
|
|
static struct return_instance *find_next_ret_chain(struct return_instance *ri)
|
|
|
|
{
|
|
|
|
bool chained;
|
|
|
|
|
|
|
|
do {
|
|
|
|
chained = ri->chained;
|
|
|
|
ri = ri->next; /* can't be NULL if chained */
|
|
|
|
} while (chained);
|
|
|
|
|
|
|
|
return ri;
|
|
|
|
}
|
|
|
|
|
2015-07-21 21:40:08 +08:00
|
|
|
static void handle_trampoline(struct pt_regs *regs)
|
2013-04-04 00:00:36 +08:00
|
|
|
{
|
|
|
|
struct uprobe_task *utask;
|
2015-07-21 21:40:13 +08:00
|
|
|
struct return_instance *ri, *next;
|
2015-07-21 21:40:21 +08:00
|
|
|
bool valid;
|
2013-04-04 00:00:36 +08:00
|
|
|
|
|
|
|
utask = current->utask;
|
|
|
|
if (!utask)
|
2015-07-21 21:40:08 +08:00
|
|
|
goto sigill;
|
2013-04-04 00:00:36 +08:00
|
|
|
|
|
|
|
ri = utask->return_instances;
|
|
|
|
if (!ri)
|
2015-07-21 21:40:08 +08:00
|
|
|
goto sigill;
|
2013-04-04 00:00:36 +08:00
|
|
|
|
2015-07-21 21:40:13 +08:00
|
|
|
do {
|
2015-07-21 21:40:21 +08:00
|
|
|
/*
|
|
|
|
* We should throw out the frames invalidated by longjmp().
|
|
|
|
* If this chain is valid, then the next one should be alive
|
|
|
|
* or NULL; the latter case means that nobody but ri->func
|
|
|
|
* could hit this trampoline on return. TODO: sigaltstack().
|
|
|
|
*/
|
|
|
|
next = find_next_ret_chain(ri);
|
2015-07-21 21:40:26 +08:00
|
|
|
valid = !next || arch_uretprobe_is_alive(next, RP_CHECK_RET, regs);
|
2015-07-21 21:40:21 +08:00
|
|
|
|
|
|
|
instruction_pointer_set(regs, ri->orig_ret_vaddr);
|
|
|
|
do {
|
|
|
|
if (valid)
|
|
|
|
handle_uretprobe_chain(ri, regs);
|
|
|
|
ri = free_ret_instance(ri);
|
|
|
|
utask->depth--;
|
|
|
|
} while (ri != next);
|
|
|
|
} while (!valid);
|
2013-04-04 00:00:36 +08:00
|
|
|
|
|
|
|
utask->return_instances = ri;
|
2015-07-21 21:40:08 +08:00
|
|
|
return;
|
|
|
|
|
|
|
|
sigill:
|
|
|
|
uprobe_warn(current, "handle uretprobe, sending SIGILL.");
|
2018-07-20 09:33:53 +08:00
|
|
|
force_sig(SIGILL, current);
|
2013-04-04 00:00:36 +08:00
|
|
|
|
|
|
|
}
|
|
|
|
|
2014-02-04 03:25:49 +08:00
|
|
|
bool __weak arch_uprobe_ignore(struct arch_uprobe *aup, struct pt_regs *regs)
|
|
|
|
{
|
|
|
|
return false;
|
|
|
|
}
|
|
|
|
|
2015-07-21 21:40:26 +08:00
|
|
|
bool __weak arch_uretprobe_is_alive(struct return_instance *ret, enum rp_check ctx,
|
|
|
|
struct pt_regs *regs)
|
2015-07-21 21:40:16 +08:00
|
|
|
{
|
|
|
|
return true;
|
|
|
|
}
|
|
|
|
|
2012-05-30 03:28:57 +08:00
|
|
|
/*
|
|
|
|
* Run handler and ask thread to singlestep.
|
|
|
|
* Ensure all non-fatal signals cannot interrupt thread while it singlesteps.
|
|
|
|
*/
|
|
|
|
static void handle_swbp(struct pt_regs *regs)
|
|
|
|
{
|
|
|
|
struct uprobe *uprobe;
|
|
|
|
unsigned long bp_vaddr;
|
uprobes: Teach handle_swbp() to rely on "is_swbp" rather than uprobes_srcu
Currently handle_swbp() assumes that it can't race with
unregister, so it roughly does:
if (find_uprobe(vaddr))
process_uprobe();
else
send_sig(SIGTRAP);
This relies on the not-really-working uprobes_srcu code we are
going to remove, see the next patch.
With this patch we rely on the result of
is_swbp_at_addr(bp_vaddr) if find_uprobe() fails.
If is_swbp == 1, then we hit the normal int3, we should send
SIGTRAP.
If is_swbp == 0, we raced with uprobe_unregister(), we simply
restart this insn again.
The "difficult" case is is_swbp == -EFAULT, when we can't read
this memory. In this case I think we should restart too, and
this is more correct compared to the current code which sends
SIGTRAP.
Ignoring ENOMEM/etc from get_user_pages(), this can only happen
if another thread unmaps this memory before find_active_uprobe()
takes mmap_sem. It would be better to pretend it was unmapped
before this insn was executed, restart, and get SIGSEGV.
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Acked-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Anton Arapov <anton@redhat.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/20120529192947.GF8057@redhat.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
2012-05-30 03:29:47 +08:00
|
|
|
int uninitialized_var(is_swbp);
|
2012-05-30 03:28:57 +08:00
|
|
|
|
|
|
|
bp_vaddr = uprobe_get_swbp_addr(regs);
|
2015-07-21 21:40:08 +08:00
|
|
|
if (bp_vaddr == get_trampoline_vaddr())
|
|
|
|
return handle_trampoline(regs);
|
2013-04-04 00:00:36 +08:00
|
|
|
|
|
|
|
uprobe = find_active_uprobe(bp_vaddr, &is_swbp);
|
uprobes/core: Handle breakpoint and singlestep exceptions
Uprobes uses exception notifiers to get to know if a thread hit
a breakpoint or a singlestep exception.
When a thread hits a uprobe or is singlestepping post a uprobe
hit, the uprobe exception notifier sets its TIF_UPROBE bit,
which will then be checked on its return to userspace path
(do_notify_resume() ->uprobe_notify_resume()), where the
consumers handlers are run (in task context) based on the
defined filters.
Uprobe hits are thread specific and hence we need to maintain
information about if a task hit a uprobe, what uprobe was hit,
the slot where the original instruction was copied for xol so
that it can be singlestepped with appropriate fixups.
In some cases, special care is needed for instructions that are
executed out of line (xol). These are architecture specific
artefacts, such as handling RIP relative instructions on x86_64.
Since the instruction at which the uprobe was inserted is
executed out of line, architecture specific fixups are added so
that the thread continues normal execution in the presence of a
uprobe.
Postpone the signals until we execute the probed insn.
post_xol() path does a recalc_sigpending() before return to
user-mode, this ensures the signal can't be lost.
Uprobes relies on DIE_DEBUG notification to notify if a
singlestep is complete.
Adds x86 specific uprobe exception notifiers and appropriate
hooks needed to determine a uprobe hit and subsequent post
processing.
Add requisite x86 fixups for xol for uprobes. Specific cases
needing fixups include relative jumps (x86_64), calls, etc.
Where possible, we check and skip singlestepping the
breakpointed instructions. For now we skip single byte as well
as few multibyte nop instructions. However this can be extended
to other instructions too.
Credits to Oleg Nesterov for suggestions/patches related to
signal, breakpoint, singlestep handling code.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Jim Keniston <jkenisto@linux.vnet.ibm.com>
Cc: Linux-mm <linux-mm@kvack.org>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/20120313180011.29771.89027.sendpatchset@srdronam.in.ibm.com
[ Performed various cleanliness edits ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-03-14 02:00:11 +08:00
|
|
|
if (!uprobe) {
|
uprobes: Teach handle_swbp() to rely on "is_swbp" rather than uprobes_srcu
Currently handle_swbp() assumes that it can't race with
unregister, so it roughly does:
if (find_uprobe(vaddr))
process_uprobe();
else
send_sig(SIGTRAP);
This relies on the not-really-working uprobes_srcu code we are
going to remove, see the next patch.
With this patch we rely on the result of
is_swbp_at_addr(bp_vaddr) if find_uprobe() fails.
If is_swbp == 1, then we hit the normal int3, we should send
SIGTRAP.
If is_swbp == 0, we raced with uprobe_unregister(), we simply
restart this insn again.
The "difficult" case is is_swbp == -EFAULT, when we can't read
this memory. In this case I think we should restart too, and
this is more correct compared to the current code which sends
SIGTRAP.
Ignoring ENOMEM/etc from get_user_pages(), this can only happen
if another thread unmaps this memory before find_active_uprobe()
takes mmap_sem. It would be better to pretend it was unmapped
before this insn was executed, restart, and get SIGSEGV.
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Acked-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Anton Arapov <anton@redhat.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/20120529192947.GF8057@redhat.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
2012-05-30 03:29:47 +08:00
|
|
|
if (is_swbp > 0) {
|
|
|
|
/* No matching uprobe; signal SIGTRAP. */
|
|
|
|
send_sig(SIGTRAP, current, 0);
|
|
|
|
} else {
|
|
|
|
/*
|
|
|
|
* Either we raced with uprobe_unregister() or we can't
|
|
|
|
* access this memory. The latter is only possible if
|
|
|
|
* another thread plays with our ->mm. In both cases
|
|
|
|
* we can simply restart. If this vma was unmapped we
|
|
|
|
* can pretend this insn was not executed yet and get
|
|
|
|
* the (correct) SIGSEGV after restart.
|
|
|
|
*/
|
|
|
|
instruction_pointer_set(regs, bp_vaddr);
|
|
|
|
}
|
uprobes/core: Handle breakpoint and singlestep exceptions
Uprobes uses exception notifiers to get to know if a thread hit
a breakpoint or a singlestep exception.
When a thread hits a uprobe or is singlestepping post a uprobe
hit, the uprobe exception notifier sets its TIF_UPROBE bit,
which will then be checked on its return to userspace path
(do_notify_resume() ->uprobe_notify_resume()), where the
consumers handlers are run (in task context) based on the
defined filters.
Uprobe hits are thread specific and hence we need to maintain
information about if a task hit a uprobe, what uprobe was hit,
the slot where the original instruction was copied for xol so
that it can be singlestepped with appropriate fixups.
In some cases, special care is needed for instructions that are
executed out of line (xol). These are architecture specific
artefacts, such as handling RIP relative instructions on x86_64.
Since the instruction at which the uprobe was inserted is
executed out of line, architecture specific fixups are added so
that the thread continues normal execution in the presence of a
uprobe.
Postpone the signals until we execute the probed insn.
post_xol() path does a recalc_sigpending() before return to
user-mode, this ensures the signal can't be lost.
Uprobes relies on DIE_DEBUG notification to notify if a
singlestep is complete.
Adds x86 specific uprobe exception notifiers and appropriate
hooks needed to determine a uprobe hit and subsequent post
processing.
Add requisite x86 fixups for xol for uprobes. Specific cases
needing fixups include relative jumps (x86_64), calls, etc.
Where possible, we check and skip singlestepping the
breakpointed instructions. For now we skip single byte as well
as few multibyte nop instructions. However this can be extended
to other instructions too.
Credits to Oleg Nesterov for suggestions/patches related to
signal, breakpoint, singlestep handling code.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Jim Keniston <jkenisto@linux.vnet.ibm.com>
Cc: Linux-mm <linux-mm@kvack.org>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/20120313180011.29771.89027.sendpatchset@srdronam.in.ibm.com
[ Performed various cleanliness edits ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-03-14 02:00:11 +08:00
|
|
|
return;
|
|
|
|
}
|
2012-12-30 22:54:08 +08:00
|
|
|
|
|
|
|
/* change it in advance for ->handler() and restart */
|
|
|
|
instruction_pointer_set(regs, bp_vaddr);
|
|
|
|
|
2012-09-30 03:56:57 +08:00
|
|
|
/*
|
|
|
|
* TODO: move copy_insn/etc into _register and remove this hack.
|
|
|
|
* After we hit the bp, _unregister + _register can install the
|
|
|
|
* new and not-yet-analyzed uprobe at the same address, restart.
|
|
|
|
*/
|
2012-10-01 03:12:44 +08:00
|
|
|
if (unlikely(!test_bit(UPROBE_COPY_INSN, &uprobe->flags)))
|
2012-12-30 22:54:08 +08:00
|
|
|
goto out;
|
uprobes/core: Handle breakpoint and singlestep exceptions
Uprobes uses exception notifiers to get to know if a thread hit
a breakpoint or a singlestep exception.
When a thread hits a uprobe or is singlestepping post a uprobe
hit, the uprobe exception notifier sets its TIF_UPROBE bit,
which will then be checked on its return to userspace path
(do_notify_resume() ->uprobe_notify_resume()), where the
consumers handlers are run (in task context) based on the
defined filters.
Uprobe hits are thread specific and hence we need to maintain
information about if a task hit a uprobe, what uprobe was hit,
the slot where the original instruction was copied for xol so
that it can be singlestepped with appropriate fixups.
In some cases, special care is needed for instructions that are
executed out of line (xol). These are architecture specific
artefacts, such as handling RIP relative instructions on x86_64.
Since the instruction at which the uprobe was inserted is
executed out of line, architecture specific fixups are added so
that the thread continues normal execution in the presence of a
uprobe.
Postpone the signals until we execute the probed insn.
post_xol() path does a recalc_sigpending() before return to
user-mode, this ensures the signal can't be lost.
Uprobes relies on DIE_DEBUG notification to notify if a
singlestep is complete.
Adds x86 specific uprobe exception notifiers and appropriate
hooks needed to determine a uprobe hit and subsequent post
processing.
Add requisite x86 fixups for xol for uprobes. Specific cases
needing fixups include relative jumps (x86_64), calls, etc.
Where possible, we check and skip singlestepping the
breakpointed instructions. For now we skip single byte as well
as few multibyte nop instructions. However this can be extended
to other instructions too.
Credits to Oleg Nesterov for suggestions/patches related to
signal, breakpoint, singlestep handling code.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Jim Keniston <jkenisto@linux.vnet.ibm.com>
Cc: Linux-mm <linux-mm@kvack.org>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/20120313180011.29771.89027.sendpatchset@srdronam.in.ibm.com
[ Performed various cleanliness edits ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-03-14 02:00:11 +08:00
|
|
|
|
2018-11-23 00:10:31 +08:00
|
|
|
/*
|
|
|
|
* Pairs with the smp_wmb() in prepare_uprobe().
|
|
|
|
*
|
|
|
|
* Guarantees that if we see the UPROBE_COPY_INSN bit set, then
|
|
|
|
* we must also see the stores to &uprobe->arch performed by the
|
|
|
|
* prepare_uprobe() call.
|
|
|
|
*/
|
|
|
|
smp_rmb();
|
|
|
|
|
2013-11-26 08:35:25 +08:00
|
|
|
/* Tracing handlers use ->utask to communicate with fetch methods */
|
|
|
|
if (!get_utask())
|
|
|
|
goto out;
|
|
|
|
|
2014-02-04 03:25:49 +08:00
|
|
|
if (arch_uprobe_ignore(&uprobe->arch, regs))
|
|
|
|
goto out;
|
|
|
|
|
uprobes/core: Handle breakpoint and singlestep exceptions
Uprobes uses exception notifiers to get to know if a thread hit
a breakpoint or a singlestep exception.
When a thread hits a uprobe or is singlestepping post a uprobe
hit, the uprobe exception notifier sets its TIF_UPROBE bit,
which will then be checked on its return to userspace path
(do_notify_resume() ->uprobe_notify_resume()), where the
consumers handlers are run (in task context) based on the
defined filters.
Uprobe hits are thread specific and hence we need to maintain
information about if a task hit a uprobe, what uprobe was hit,
the slot where the original instruction was copied for xol so
that it can be singlestepped with appropriate fixups.
In some cases, special care is needed for instructions that are
executed out of line (xol). These are architecture specific
artefacts, such as handling RIP relative instructions on x86_64.
Since the instruction at which the uprobe was inserted is
executed out of line, architecture specific fixups are added so
that the thread continues normal execution in the presence of a
uprobe.
Postpone the signals until we execute the probed insn.
post_xol() path does a recalc_sigpending() before return to
user-mode, this ensures the signal can't be lost.
Uprobes relies on DIE_DEBUG notification to notify if a
singlestep is complete.
Adds x86 specific uprobe exception notifiers and appropriate
hooks needed to determine a uprobe hit and subsequent post
processing.
Add requisite x86 fixups for xol for uprobes. Specific cases
needing fixups include relative jumps (x86_64), calls, etc.
Where possible, we check and skip singlestepping the
breakpointed instructions. For now we skip single byte as well
as few multibyte nop instructions. However this can be extended
to other instructions too.
Credits to Oleg Nesterov for suggestions/patches related to
signal, breakpoint, singlestep handling code.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Jim Keniston <jkenisto@linux.vnet.ibm.com>
Cc: Linux-mm <linux-mm@kvack.org>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/20120313180011.29771.89027.sendpatchset@srdronam.in.ibm.com
[ Performed various cleanliness edits ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-03-14 02:00:11 +08:00
|
|
|
handler_chain(uprobe, regs);
|
2014-02-04 03:25:49 +08:00
|
|
|
|
uprobes: Kill UPROBE_SKIP_SSTEP and can_skip_sstep()
UPROBE_COPY_INSN, UPROBE_SKIP_SSTEP, and uprobe->flags must die. This
patch kills UPROBE_SKIP_SSTEP. I never understood why it was added;
not only it doesn't help, it harms.
It can only help to avoid arch_uprobe_skip_sstep() if it was already
called before and failed. But this is ugly, if we want to know whether
we can emulate this instruction or not we should do this analysis in
arch_uprobe_analyze_insn(), not when we hit this probe for the first
time.
And in fact this logic is simply wrong. arch_uprobe_skip_sstep() can
fail or not depending on the task/register state, if this insn can be
emulated but, say, put_user() fails we need to xol it this time, but
this doesn't mean we shouldn't try to emulate it when this or another
thread hits this bp next time.
And this is the actual reason for this change. We need to emulate the
"call" insn, but push(return-address) can obviously fail.
Per-arch notes:
x86: __skip_sstep() can only emulate "rep;nop". With this
change it will be called every time and most probably
for no reason.
This will be fixed by the next changes. We need to
change this suboptimal code anyway.
arm: Should not be affected. It has its own "bool simulate"
flag checked in arch_uprobe_skip_sstep().
ppc: Looks like, it can emulate almost everything. Does it
actually need to record the fact that emulate_step()
failed? Hopefully not. But if yes, it can add the ppc-
specific flag into arch_uprobe.
TODO: rename arch_uprobe_skip_sstep() to arch_uprobe_emulate_insn(),
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Reviewed-by: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Reviewed-by: David A. Long <dave.long@linaro.org>
Reviewed-by: Jim Keniston <jkenisto@us.ibm.com>
Acked-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
2014-03-31 00:56:22 +08:00
|
|
|
if (arch_uprobe_skip_sstep(&uprobe->arch, regs))
|
2012-09-15 00:31:23 +08:00
|
|
|
goto out;
|
uprobes/core: Handle breakpoint and singlestep exceptions
Uprobes uses exception notifiers to get to know if a thread hit
a breakpoint or a singlestep exception.
When a thread hits a uprobe or is singlestepping post a uprobe
hit, the uprobe exception notifier sets its TIF_UPROBE bit,
which will then be checked on its return to userspace path
(do_notify_resume() ->uprobe_notify_resume()), where the
consumers handlers are run (in task context) based on the
defined filters.
Uprobe hits are thread specific and hence we need to maintain
information about if a task hit a uprobe, what uprobe was hit,
the slot where the original instruction was copied for xol so
that it can be singlestepped with appropriate fixups.
In some cases, special care is needed for instructions that are
executed out of line (xol). These are architecture specific
artefacts, such as handling RIP relative instructions on x86_64.
Since the instruction at which the uprobe was inserted is
executed out of line, architecture specific fixups are added so
that the thread continues normal execution in the presence of a
uprobe.
Postpone the signals until we execute the probed insn.
post_xol() path does a recalc_sigpending() before return to
user-mode, this ensures the signal can't be lost.
Uprobes relies on DIE_DEBUG notification to notify if a
singlestep is complete.
Adds x86 specific uprobe exception notifiers and appropriate
hooks needed to determine a uprobe hit and subsequent post
processing.
Add requisite x86 fixups for xol for uprobes. Specific cases
needing fixups include relative jumps (x86_64), calls, etc.
Where possible, we check and skip singlestepping the
breakpointed instructions. For now we skip single byte as well
as few multibyte nop instructions. However this can be extended
to other instructions too.
Credits to Oleg Nesterov for suggestions/patches related to
signal, breakpoint, singlestep handling code.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Jim Keniston <jkenisto@linux.vnet.ibm.com>
Cc: Linux-mm <linux-mm@kvack.org>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/20120313180011.29771.89027.sendpatchset@srdronam.in.ibm.com
[ Performed various cleanliness edits ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-03-14 02:00:11 +08:00
|
|
|
|
2013-01-01 01:20:42 +08:00
|
|
|
if (!pre_ssout(uprobe, regs, bp_vaddr))
|
uprobes/core: Handle breakpoint and singlestep exceptions
Uprobes uses exception notifiers to get to know if a thread hit
a breakpoint or a singlestep exception.
When a thread hits a uprobe or is singlestepping post a uprobe
hit, the uprobe exception notifier sets its TIF_UPROBE bit,
which will then be checked on its return to userspace path
(do_notify_resume() ->uprobe_notify_resume()), where the
consumers handlers are run (in task context) based on the
defined filters.
Uprobe hits are thread specific and hence we need to maintain
information about if a task hit a uprobe, what uprobe was hit,
the slot where the original instruction was copied for xol so
that it can be singlestepped with appropriate fixups.
In some cases, special care is needed for instructions that are
executed out of line (xol). These are architecture specific
artefacts, such as handling RIP relative instructions on x86_64.
Since the instruction at which the uprobe was inserted is
executed out of line, architecture specific fixups are added so
that the thread continues normal execution in the presence of a
uprobe.
Postpone the signals until we execute the probed insn.
post_xol() path does a recalc_sigpending() before return to
user-mode, this ensures the signal can't be lost.
Uprobes relies on DIE_DEBUG notification to notify if a
singlestep is complete.
Adds x86 specific uprobe exception notifiers and appropriate
hooks needed to determine a uprobe hit and subsequent post
processing.
Add requisite x86 fixups for xol for uprobes. Specific cases
needing fixups include relative jumps (x86_64), calls, etc.
Where possible, we check and skip singlestepping the
breakpointed instructions. For now we skip single byte as well
as few multibyte nop instructions. However this can be extended
to other instructions too.
Credits to Oleg Nesterov for suggestions/patches related to
signal, breakpoint, singlestep handling code.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Jim Keniston <jkenisto@linux.vnet.ibm.com>
Cc: Linux-mm <linux-mm@kvack.org>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/20120313180011.29771.89027.sendpatchset@srdronam.in.ibm.com
[ Performed various cleanliness edits ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-03-14 02:00:11 +08:00
|
|
|
return;
|
|
|
|
|
uprobes: Kill UPROBE_SKIP_SSTEP and can_skip_sstep()
UPROBE_COPY_INSN, UPROBE_SKIP_SSTEP, and uprobe->flags must die. This
patch kills UPROBE_SKIP_SSTEP. I never understood why it was added;
not only it doesn't help, it harms.
It can only help to avoid arch_uprobe_skip_sstep() if it was already
called before and failed. But this is ugly, if we want to know whether
we can emulate this instruction or not we should do this analysis in
arch_uprobe_analyze_insn(), not when we hit this probe for the first
time.
And in fact this logic is simply wrong. arch_uprobe_skip_sstep() can
fail or not depending on the task/register state, if this insn can be
emulated but, say, put_user() fails we need to xol it this time, but
this doesn't mean we shouldn't try to emulate it when this or another
thread hits this bp next time.
And this is the actual reason for this change. We need to emulate the
"call" insn, but push(return-address) can obviously fail.
Per-arch notes:
x86: __skip_sstep() can only emulate "rep;nop". With this
change it will be called every time and most probably
for no reason.
This will be fixed by the next changes. We need to
change this suboptimal code anyway.
arm: Should not be affected. It has its own "bool simulate"
flag checked in arch_uprobe_skip_sstep().
ppc: Looks like, it can emulate almost everything. Does it
actually need to record the fact that emulate_step()
failed? Hopefully not. But if yes, it can add the ppc-
specific flag into arch_uprobe.
TODO: rename arch_uprobe_skip_sstep() to arch_uprobe_emulate_insn(),
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Reviewed-by: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Reviewed-by: David A. Long <dave.long@linaro.org>
Reviewed-by: Jim Keniston <jkenisto@us.ibm.com>
Acked-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
2014-03-31 00:56:22 +08:00
|
|
|
/* arch_uprobe_skip_sstep() succeeded, or restart if can't singlestep */
|
2012-09-15 00:31:23 +08:00
|
|
|
out:
|
2012-08-08 00:12:30 +08:00
|
|
|
put_uprobe(uprobe);
|
uprobes/core: Handle breakpoint and singlestep exceptions
Uprobes uses exception notifiers to get to know if a thread hit
a breakpoint or a singlestep exception.
When a thread hits a uprobe or is singlestepping post a uprobe
hit, the uprobe exception notifier sets its TIF_UPROBE bit,
which will then be checked on its return to userspace path
(do_notify_resume() ->uprobe_notify_resume()), where the
consumers handlers are run (in task context) based on the
defined filters.
Uprobe hits are thread specific and hence we need to maintain
information about if a task hit a uprobe, what uprobe was hit,
the slot where the original instruction was copied for xol so
that it can be singlestepped with appropriate fixups.
In some cases, special care is needed for instructions that are
executed out of line (xol). These are architecture specific
artefacts, such as handling RIP relative instructions on x86_64.
Since the instruction at which the uprobe was inserted is
executed out of line, architecture specific fixups are added so
that the thread continues normal execution in the presence of a
uprobe.
Postpone the signals until we execute the probed insn.
post_xol() path does a recalc_sigpending() before return to
user-mode, this ensures the signal can't be lost.
Uprobes relies on DIE_DEBUG notification to notify if a
singlestep is complete.
Adds x86 specific uprobe exception notifiers and appropriate
hooks needed to determine a uprobe hit and subsequent post
processing.
Add requisite x86 fixups for xol for uprobes. Specific cases
needing fixups include relative jumps (x86_64), calls, etc.
Where possible, we check and skip singlestepping the
breakpointed instructions. For now we skip single byte as well
as few multibyte nop instructions. However this can be extended
to other instructions too.
Credits to Oleg Nesterov for suggestions/patches related to
signal, breakpoint, singlestep handling code.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Jim Keniston <jkenisto@linux.vnet.ibm.com>
Cc: Linux-mm <linux-mm@kvack.org>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/20120313180011.29771.89027.sendpatchset@srdronam.in.ibm.com
[ Performed various cleanliness edits ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-03-14 02:00:11 +08:00
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Perform required fix-ups and disable singlestep.
|
|
|
|
* Allow pending signals to take effect.
|
|
|
|
*/
|
|
|
|
static void handle_singlestep(struct uprobe_task *utask, struct pt_regs *regs)
|
|
|
|
{
|
|
|
|
struct uprobe *uprobe;
|
2014-04-04 02:20:10 +08:00
|
|
|
int err = 0;
|
uprobes/core: Handle breakpoint and singlestep exceptions
Uprobes uses exception notifiers to get to know if a thread hit
a breakpoint or a singlestep exception.
When a thread hits a uprobe or is singlestepping post a uprobe
hit, the uprobe exception notifier sets its TIF_UPROBE bit,
which will then be checked on its return to userspace path
(do_notify_resume() ->uprobe_notify_resume()), where the
consumers handlers are run (in task context) based on the
defined filters.
Uprobe hits are thread specific and hence we need to maintain
information about if a task hit a uprobe, what uprobe was hit,
the slot where the original instruction was copied for xol so
that it can be singlestepped with appropriate fixups.
In some cases, special care is needed for instructions that are
executed out of line (xol). These are architecture specific
artefacts, such as handling RIP relative instructions on x86_64.
Since the instruction at which the uprobe was inserted is
executed out of line, architecture specific fixups are added so
that the thread continues normal execution in the presence of a
uprobe.
Postpone the signals until we execute the probed insn.
post_xol() path does a recalc_sigpending() before return to
user-mode, this ensures the signal can't be lost.
Uprobes relies on DIE_DEBUG notification to notify if a
singlestep is complete.
Adds x86 specific uprobe exception notifiers and appropriate
hooks needed to determine a uprobe hit and subsequent post
processing.
Add requisite x86 fixups for xol for uprobes. Specific cases
needing fixups include relative jumps (x86_64), calls, etc.
Where possible, we check and skip singlestepping the
breakpointed instructions. For now we skip single byte as well
as few multibyte nop instructions. However this can be extended
to other instructions too.
Credits to Oleg Nesterov for suggestions/patches related to
signal, breakpoint, singlestep handling code.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Jim Keniston <jkenisto@linux.vnet.ibm.com>
Cc: Linux-mm <linux-mm@kvack.org>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/20120313180011.29771.89027.sendpatchset@srdronam.in.ibm.com
[ Performed various cleanliness edits ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-03-14 02:00:11 +08:00
|
|
|
|
|
|
|
uprobe = utask->active_uprobe;
|
|
|
|
if (utask->state == UTASK_SSTEP_ACK)
|
2014-04-04 02:20:10 +08:00
|
|
|
err = arch_uprobe_post_xol(&uprobe->arch, regs);
|
uprobes/core: Handle breakpoint and singlestep exceptions
Uprobes uses exception notifiers to get to know if a thread hit
a breakpoint or a singlestep exception.
When a thread hits a uprobe or is singlestepping post a uprobe
hit, the uprobe exception notifier sets its TIF_UPROBE bit,
which will then be checked on its return to userspace path
(do_notify_resume() ->uprobe_notify_resume()), where the
consumers handlers are run (in task context) based on the
defined filters.
Uprobe hits are thread specific and hence we need to maintain
information about if a task hit a uprobe, what uprobe was hit,
the slot where the original instruction was copied for xol so
that it can be singlestepped with appropriate fixups.
In some cases, special care is needed for instructions that are
executed out of line (xol). These are architecture specific
artefacts, such as handling RIP relative instructions on x86_64.
Since the instruction at which the uprobe was inserted is
executed out of line, architecture specific fixups are added so
that the thread continues normal execution in the presence of a
uprobe.
Postpone the signals until we execute the probed insn.
post_xol() path does a recalc_sigpending() before return to
user-mode, this ensures the signal can't be lost.
Uprobes relies on DIE_DEBUG notification to notify if a
singlestep is complete.
Adds x86 specific uprobe exception notifiers and appropriate
hooks needed to determine a uprobe hit and subsequent post
processing.
Add requisite x86 fixups for xol for uprobes. Specific cases
needing fixups include relative jumps (x86_64), calls, etc.
Where possible, we check and skip singlestepping the
breakpointed instructions. For now we skip single byte as well
as few multibyte nop instructions. However this can be extended
to other instructions too.
Credits to Oleg Nesterov for suggestions/patches related to
signal, breakpoint, singlestep handling code.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Jim Keniston <jkenisto@linux.vnet.ibm.com>
Cc: Linux-mm <linux-mm@kvack.org>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/20120313180011.29771.89027.sendpatchset@srdronam.in.ibm.com
[ Performed various cleanliness edits ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-03-14 02:00:11 +08:00
|
|
|
else if (utask->state == UTASK_SSTEP_TRAPPED)
|
|
|
|
arch_uprobe_abort_xol(&uprobe->arch, regs);
|
|
|
|
else
|
|
|
|
WARN_ON_ONCE(1);
|
|
|
|
|
|
|
|
put_uprobe(uprobe);
|
|
|
|
utask->active_uprobe = NULL;
|
|
|
|
utask->state = UTASK_RUNNING;
|
2012-03-31 02:26:31 +08:00
|
|
|
xol_free_insn_slot(current);
|
uprobes/core: Handle breakpoint and singlestep exceptions
Uprobes uses exception notifiers to get to know if a thread hit
a breakpoint or a singlestep exception.
When a thread hits a uprobe or is singlestepping post a uprobe
hit, the uprobe exception notifier sets its TIF_UPROBE bit,
which will then be checked on its return to userspace path
(do_notify_resume() ->uprobe_notify_resume()), where the
consumers handlers are run (in task context) based on the
defined filters.
Uprobe hits are thread specific and hence we need to maintain
information about if a task hit a uprobe, what uprobe was hit,
the slot where the original instruction was copied for xol so
that it can be singlestepped with appropriate fixups.
In some cases, special care is needed for instructions that are
executed out of line (xol). These are architecture specific
artefacts, such as handling RIP relative instructions on x86_64.
Since the instruction at which the uprobe was inserted is
executed out of line, architecture specific fixups are added so
that the thread continues normal execution in the presence of a
uprobe.
Postpone the signals until we execute the probed insn.
post_xol() path does a recalc_sigpending() before return to
user-mode, this ensures the signal can't be lost.
Uprobes relies on DIE_DEBUG notification to notify if a
singlestep is complete.
Adds x86 specific uprobe exception notifiers and appropriate
hooks needed to determine a uprobe hit and subsequent post
processing.
Add requisite x86 fixups for xol for uprobes. Specific cases
needing fixups include relative jumps (x86_64), calls, etc.
Where possible, we check and skip singlestepping the
breakpointed instructions. For now we skip single byte as well
as few multibyte nop instructions. However this can be extended
to other instructions too.
Credits to Oleg Nesterov for suggestions/patches related to
signal, breakpoint, singlestep handling code.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Jim Keniston <jkenisto@linux.vnet.ibm.com>
Cc: Linux-mm <linux-mm@kvack.org>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/20120313180011.29771.89027.sendpatchset@srdronam.in.ibm.com
[ Performed various cleanliness edits ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-03-14 02:00:11 +08:00
|
|
|
|
|
|
|
spin_lock_irq(¤t->sighand->siglock);
|
|
|
|
recalc_sigpending(); /* see uprobe_deny_signal() */
|
|
|
|
spin_unlock_irq(¤t->sighand->siglock);
|
2014-04-04 02:20:10 +08:00
|
|
|
|
|
|
|
if (unlikely(err)) {
|
|
|
|
uprobe_warn(current, "execute the probed insn, sending SIGILL.");
|
2018-07-20 09:33:53 +08:00
|
|
|
force_sig(SIGILL, current);
|
2014-04-04 02:20:10 +08:00
|
|
|
}
|
uprobes/core: Handle breakpoint and singlestep exceptions
Uprobes uses exception notifiers to get to know if a thread hit
a breakpoint or a singlestep exception.
When a thread hits a uprobe or is singlestepping post a uprobe
hit, the uprobe exception notifier sets its TIF_UPROBE bit,
which will then be checked on its return to userspace path
(do_notify_resume() ->uprobe_notify_resume()), where the
consumers handlers are run (in task context) based on the
defined filters.
Uprobe hits are thread specific and hence we need to maintain
information about if a task hit a uprobe, what uprobe was hit,
the slot where the original instruction was copied for xol so
that it can be singlestepped with appropriate fixups.
In some cases, special care is needed for instructions that are
executed out of line (xol). These are architecture specific
artefacts, such as handling RIP relative instructions on x86_64.
Since the instruction at which the uprobe was inserted is
executed out of line, architecture specific fixups are added so
that the thread continues normal execution in the presence of a
uprobe.
Postpone the signals until we execute the probed insn.
post_xol() path does a recalc_sigpending() before return to
user-mode, this ensures the signal can't be lost.
Uprobes relies on DIE_DEBUG notification to notify if a
singlestep is complete.
Adds x86 specific uprobe exception notifiers and appropriate
hooks needed to determine a uprobe hit and subsequent post
processing.
Add requisite x86 fixups for xol for uprobes. Specific cases
needing fixups include relative jumps (x86_64), calls, etc.
Where possible, we check and skip singlestepping the
breakpointed instructions. For now we skip single byte as well
as few multibyte nop instructions. However this can be extended
to other instructions too.
Credits to Oleg Nesterov for suggestions/patches related to
signal, breakpoint, singlestep handling code.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Jim Keniston <jkenisto@linux.vnet.ibm.com>
Cc: Linux-mm <linux-mm@kvack.org>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/20120313180011.29771.89027.sendpatchset@srdronam.in.ibm.com
[ Performed various cleanliness edits ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-03-14 02:00:11 +08:00
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
2012-09-15 00:52:10 +08:00
|
|
|
* On breakpoint hit, breakpoint notifier sets the TIF_UPROBE flag and
|
|
|
|
* allows the thread to return from interrupt. After that handle_swbp()
|
|
|
|
* sets utask->active_uprobe.
|
uprobes/core: Handle breakpoint and singlestep exceptions
Uprobes uses exception notifiers to get to know if a thread hit
a breakpoint or a singlestep exception.
When a thread hits a uprobe or is singlestepping post a uprobe
hit, the uprobe exception notifier sets its TIF_UPROBE bit,
which will then be checked on its return to userspace path
(do_notify_resume() ->uprobe_notify_resume()), where the
consumers handlers are run (in task context) based on the
defined filters.
Uprobe hits are thread specific and hence we need to maintain
information about if a task hit a uprobe, what uprobe was hit,
the slot where the original instruction was copied for xol so
that it can be singlestepped with appropriate fixups.
In some cases, special care is needed for instructions that are
executed out of line (xol). These are architecture specific
artefacts, such as handling RIP relative instructions on x86_64.
Since the instruction at which the uprobe was inserted is
executed out of line, architecture specific fixups are added so
that the thread continues normal execution in the presence of a
uprobe.
Postpone the signals until we execute the probed insn.
post_xol() path does a recalc_sigpending() before return to
user-mode, this ensures the signal can't be lost.
Uprobes relies on DIE_DEBUG notification to notify if a
singlestep is complete.
Adds x86 specific uprobe exception notifiers and appropriate
hooks needed to determine a uprobe hit and subsequent post
processing.
Add requisite x86 fixups for xol for uprobes. Specific cases
needing fixups include relative jumps (x86_64), calls, etc.
Where possible, we check and skip singlestepping the
breakpointed instructions. For now we skip single byte as well
as few multibyte nop instructions. However this can be extended
to other instructions too.
Credits to Oleg Nesterov for suggestions/patches related to
signal, breakpoint, singlestep handling code.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Jim Keniston <jkenisto@linux.vnet.ibm.com>
Cc: Linux-mm <linux-mm@kvack.org>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/20120313180011.29771.89027.sendpatchset@srdronam.in.ibm.com
[ Performed various cleanliness edits ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-03-14 02:00:11 +08:00
|
|
|
*
|
2012-09-15 00:52:10 +08:00
|
|
|
* On singlestep exception, singlestep notifier sets the TIF_UPROBE flag
|
|
|
|
* and allows the thread to return from interrupt.
|
uprobes/core: Handle breakpoint and singlestep exceptions
Uprobes uses exception notifiers to get to know if a thread hit
a breakpoint or a singlestep exception.
When a thread hits a uprobe or is singlestepping post a uprobe
hit, the uprobe exception notifier sets its TIF_UPROBE bit,
which will then be checked on its return to userspace path
(do_notify_resume() ->uprobe_notify_resume()), where the
consumers handlers are run (in task context) based on the
defined filters.
Uprobe hits are thread specific and hence we need to maintain
information about if a task hit a uprobe, what uprobe was hit,
the slot where the original instruction was copied for xol so
that it can be singlestepped with appropriate fixups.
In some cases, special care is needed for instructions that are
executed out of line (xol). These are architecture specific
artefacts, such as handling RIP relative instructions on x86_64.
Since the instruction at which the uprobe was inserted is
executed out of line, architecture specific fixups are added so
that the thread continues normal execution in the presence of a
uprobe.
Postpone the signals until we execute the probed insn.
post_xol() path does a recalc_sigpending() before return to
user-mode, this ensures the signal can't be lost.
Uprobes relies on DIE_DEBUG notification to notify if a
singlestep is complete.
Adds x86 specific uprobe exception notifiers and appropriate
hooks needed to determine a uprobe hit and subsequent post
processing.
Add requisite x86 fixups for xol for uprobes. Specific cases
needing fixups include relative jumps (x86_64), calls, etc.
Where possible, we check and skip singlestepping the
breakpointed instructions. For now we skip single byte as well
as few multibyte nop instructions. However this can be extended
to other instructions too.
Credits to Oleg Nesterov for suggestions/patches related to
signal, breakpoint, singlestep handling code.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Jim Keniston <jkenisto@linux.vnet.ibm.com>
Cc: Linux-mm <linux-mm@kvack.org>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/20120313180011.29771.89027.sendpatchset@srdronam.in.ibm.com
[ Performed various cleanliness edits ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-03-14 02:00:11 +08:00
|
|
|
*
|
|
|
|
* While returning to userspace, thread notices the TIF_UPROBE flag and calls
|
|
|
|
* uprobe_notify_resume().
|
|
|
|
*/
|
|
|
|
void uprobe_notify_resume(struct pt_regs *regs)
|
|
|
|
{
|
|
|
|
struct uprobe_task *utask;
|
|
|
|
|
2012-09-15 01:05:46 +08:00
|
|
|
clear_thread_flag(TIF_UPROBE);
|
|
|
|
|
uprobes/core: Handle breakpoint and singlestep exceptions
Uprobes uses exception notifiers to get to know if a thread hit
a breakpoint or a singlestep exception.
When a thread hits a uprobe or is singlestepping post a uprobe
hit, the uprobe exception notifier sets its TIF_UPROBE bit,
which will then be checked on its return to userspace path
(do_notify_resume() ->uprobe_notify_resume()), where the
consumers handlers are run (in task context) based on the
defined filters.
Uprobe hits are thread specific and hence we need to maintain
information about if a task hit a uprobe, what uprobe was hit,
the slot where the original instruction was copied for xol so
that it can be singlestepped with appropriate fixups.
In some cases, special care is needed for instructions that are
executed out of line (xol). These are architecture specific
artefacts, such as handling RIP relative instructions on x86_64.
Since the instruction at which the uprobe was inserted is
executed out of line, architecture specific fixups are added so
that the thread continues normal execution in the presence of a
uprobe.
Postpone the signals until we execute the probed insn.
post_xol() path does a recalc_sigpending() before return to
user-mode, this ensures the signal can't be lost.
Uprobes relies on DIE_DEBUG notification to notify if a
singlestep is complete.
Adds x86 specific uprobe exception notifiers and appropriate
hooks needed to determine a uprobe hit and subsequent post
processing.
Add requisite x86 fixups for xol for uprobes. Specific cases
needing fixups include relative jumps (x86_64), calls, etc.
Where possible, we check and skip singlestepping the
breakpointed instructions. For now we skip single byte as well
as few multibyte nop instructions. However this can be extended
to other instructions too.
Credits to Oleg Nesterov for suggestions/patches related to
signal, breakpoint, singlestep handling code.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Jim Keniston <jkenisto@linux.vnet.ibm.com>
Cc: Linux-mm <linux-mm@kvack.org>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/20120313180011.29771.89027.sendpatchset@srdronam.in.ibm.com
[ Performed various cleanliness edits ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-03-14 02:00:11 +08:00
|
|
|
utask = current->utask;
|
2012-09-15 00:52:10 +08:00
|
|
|
if (utask && utask->active_uprobe)
|
uprobes/core: Handle breakpoint and singlestep exceptions
Uprobes uses exception notifiers to get to know if a thread hit
a breakpoint or a singlestep exception.
When a thread hits a uprobe or is singlestepping post a uprobe
hit, the uprobe exception notifier sets its TIF_UPROBE bit,
which will then be checked on its return to userspace path
(do_notify_resume() ->uprobe_notify_resume()), where the
consumers handlers are run (in task context) based on the
defined filters.
Uprobe hits are thread specific and hence we need to maintain
information about if a task hit a uprobe, what uprobe was hit,
the slot where the original instruction was copied for xol so
that it can be singlestepped with appropriate fixups.
In some cases, special care is needed for instructions that are
executed out of line (xol). These are architecture specific
artefacts, such as handling RIP relative instructions on x86_64.
Since the instruction at which the uprobe was inserted is
executed out of line, architecture specific fixups are added so
that the thread continues normal execution in the presence of a
uprobe.
Postpone the signals until we execute the probed insn.
post_xol() path does a recalc_sigpending() before return to
user-mode, this ensures the signal can't be lost.
Uprobes relies on DIE_DEBUG notification to notify if a
singlestep is complete.
Adds x86 specific uprobe exception notifiers and appropriate
hooks needed to determine a uprobe hit and subsequent post
processing.
Add requisite x86 fixups for xol for uprobes. Specific cases
needing fixups include relative jumps (x86_64), calls, etc.
Where possible, we check and skip singlestepping the
breakpointed instructions. For now we skip single byte as well
as few multibyte nop instructions. However this can be extended
to other instructions too.
Credits to Oleg Nesterov for suggestions/patches related to
signal, breakpoint, singlestep handling code.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Jim Keniston <jkenisto@linux.vnet.ibm.com>
Cc: Linux-mm <linux-mm@kvack.org>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/20120313180011.29771.89027.sendpatchset@srdronam.in.ibm.com
[ Performed various cleanliness edits ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-03-14 02:00:11 +08:00
|
|
|
handle_singlestep(utask, regs);
|
2012-09-15 00:52:10 +08:00
|
|
|
else
|
|
|
|
handle_swbp(regs);
|
uprobes/core: Handle breakpoint and singlestep exceptions
Uprobes uses exception notifiers to get to know if a thread hit
a breakpoint or a singlestep exception.
When a thread hits a uprobe or is singlestepping post a uprobe
hit, the uprobe exception notifier sets its TIF_UPROBE bit,
which will then be checked on its return to userspace path
(do_notify_resume() ->uprobe_notify_resume()), where the
consumers handlers are run (in task context) based on the
defined filters.
Uprobe hits are thread specific and hence we need to maintain
information about if a task hit a uprobe, what uprobe was hit,
the slot where the original instruction was copied for xol so
that it can be singlestepped with appropriate fixups.
In some cases, special care is needed for instructions that are
executed out of line (xol). These are architecture specific
artefacts, such as handling RIP relative instructions on x86_64.
Since the instruction at which the uprobe was inserted is
executed out of line, architecture specific fixups are added so
that the thread continues normal execution in the presence of a
uprobe.
Postpone the signals until we execute the probed insn.
post_xol() path does a recalc_sigpending() before return to
user-mode, this ensures the signal can't be lost.
Uprobes relies on DIE_DEBUG notification to notify if a
singlestep is complete.
Adds x86 specific uprobe exception notifiers and appropriate
hooks needed to determine a uprobe hit and subsequent post
processing.
Add requisite x86 fixups for xol for uprobes. Specific cases
needing fixups include relative jumps (x86_64), calls, etc.
Where possible, we check and skip singlestepping the
breakpointed instructions. For now we skip single byte as well
as few multibyte nop instructions. However this can be extended
to other instructions too.
Credits to Oleg Nesterov for suggestions/patches related to
signal, breakpoint, singlestep handling code.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Jim Keniston <jkenisto@linux.vnet.ibm.com>
Cc: Linux-mm <linux-mm@kvack.org>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/20120313180011.29771.89027.sendpatchset@srdronam.in.ibm.com
[ Performed various cleanliness edits ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-03-14 02:00:11 +08:00
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* uprobe_pre_sstep_notifier gets called from interrupt context as part of
|
|
|
|
* notifier mechanism. Set TIF_UPROBE flag and indicate breakpoint hit.
|
|
|
|
*/
|
|
|
|
int uprobe_pre_sstep_notifier(struct pt_regs *regs)
|
|
|
|
{
|
2013-04-04 00:00:35 +08:00
|
|
|
if (!current->mm)
|
|
|
|
return 0;
|
|
|
|
|
|
|
|
if (!test_bit(MMF_HAS_UPROBES, ¤t->mm->flags) &&
|
|
|
|
(!current->utask || !current->utask->return_instances))
|
uprobes/core: Handle breakpoint and singlestep exceptions
Uprobes uses exception notifiers to get to know if a thread hit
a breakpoint or a singlestep exception.
When a thread hits a uprobe or is singlestepping post a uprobe
hit, the uprobe exception notifier sets its TIF_UPROBE bit,
which will then be checked on its return to userspace path
(do_notify_resume() ->uprobe_notify_resume()), where the
consumers handlers are run (in task context) based on the
defined filters.
Uprobe hits are thread specific and hence we need to maintain
information about if a task hit a uprobe, what uprobe was hit,
the slot where the original instruction was copied for xol so
that it can be singlestepped with appropriate fixups.
In some cases, special care is needed for instructions that are
executed out of line (xol). These are architecture specific
artefacts, such as handling RIP relative instructions on x86_64.
Since the instruction at which the uprobe was inserted is
executed out of line, architecture specific fixups are added so
that the thread continues normal execution in the presence of a
uprobe.
Postpone the signals until we execute the probed insn.
post_xol() path does a recalc_sigpending() before return to
user-mode, this ensures the signal can't be lost.
Uprobes relies on DIE_DEBUG notification to notify if a
singlestep is complete.
Adds x86 specific uprobe exception notifiers and appropriate
hooks needed to determine a uprobe hit and subsequent post
processing.
Add requisite x86 fixups for xol for uprobes. Specific cases
needing fixups include relative jumps (x86_64), calls, etc.
Where possible, we check and skip singlestepping the
breakpointed instructions. For now we skip single byte as well
as few multibyte nop instructions. However this can be extended
to other instructions too.
Credits to Oleg Nesterov for suggestions/patches related to
signal, breakpoint, singlestep handling code.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Jim Keniston <jkenisto@linux.vnet.ibm.com>
Cc: Linux-mm <linux-mm@kvack.org>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/20120313180011.29771.89027.sendpatchset@srdronam.in.ibm.com
[ Performed various cleanliness edits ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-03-14 02:00:11 +08:00
|
|
|
return 0;
|
|
|
|
|
|
|
|
set_thread_flag(TIF_UPROBE);
|
|
|
|
return 1;
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* uprobe_post_sstep_notifier gets called in interrupt context as part of notifier
|
|
|
|
* mechanism. Set TIF_UPROBE flag and indicate completion of singlestep.
|
|
|
|
*/
|
|
|
|
int uprobe_post_sstep_notifier(struct pt_regs *regs)
|
|
|
|
{
|
|
|
|
struct uprobe_task *utask = current->utask;
|
|
|
|
|
|
|
|
if (!current->mm || !utask || !utask->active_uprobe)
|
|
|
|
/* task is currently not uprobed */
|
|
|
|
return 0;
|
|
|
|
|
|
|
|
utask->state = UTASK_SSTEP_ACK;
|
|
|
|
set_thread_flag(TIF_UPROBE);
|
|
|
|
return 1;
|
|
|
|
}
|
|
|
|
|
|
|
|
static struct notifier_block uprobe_exception_nb = {
|
|
|
|
.notifier_call = arch_uprobe_exception_notify,
|
|
|
|
.priority = INT_MAX-1, /* notified after kprobes, kgdb */
|
|
|
|
};
|
|
|
|
|
uprobes, mm, x86: Add the ability to install and remove uprobes breakpoints
Add uprobes support to the core kernel, with x86 support.
This commit adds the kernel facilities, the actual uprobes
user-space ABI and perf probe support comes in later commits.
General design:
Uprobes are maintained in an rb-tree indexed by inode and offset
(the offset here is from the start of the mapping). For a unique
(inode, offset) tuple, there can be at most one uprobe in the
rb-tree.
Since the (inode, offset) tuple identifies a unique uprobe, more
than one user may be interested in the same uprobe. This provides
the ability to connect multiple 'consumers' to the same uprobe.
Each consumer defines a handler and a filter (optional). The
'handler' is run every time the uprobe is hit, if it matches the
'filter' criteria.
The first consumer of a uprobe causes the breakpoint to be
inserted at the specified address and subsequent consumers are
appended to this list. On subsequent probes, the consumer gets
appended to the existing list of consumers. The breakpoint is
removed when the last consumer unregisters. For all other
unregisterations, the consumer is removed from the list of
consumers.
Given a inode, we get a list of the mms that have mapped the
inode. Do the actual registration if mm maps the page where a
probe needs to be inserted/removed.
We use a temporary list to walk through the vmas that map the
inode.
- The number of maps that map the inode, is not known before we
walk the rmap and keeps changing.
- extending vm_area_struct wasn't recommended, it's a
size-critical data structure.
- There can be more than one maps of the inode in the same mm.
We add callbacks to the mmap methods to keep an eye on text vmas
that are of interest to uprobes. When a vma of interest is mapped,
we insert the breakpoint at the right address.
Uprobe works by replacing the instruction at the address defined
by (inode, offset) with the arch specific breakpoint
instruction. We save a copy of the original instruction at the
uprobed address.
This is needed for:
a. executing the instruction out-of-line (xol).
b. instruction analysis for any subsequent fixups.
c. restoring the instruction back when the uprobe is unregistered.
We insert or delete a breakpoint instruction, and this
breakpoint instruction is assumed to be the smallest instruction
available on the platform. For fixed size instruction platforms
this is trivially true, for variable size instruction platforms
the breakpoint instruction is typically the smallest (often a
single byte).
Writing the instruction is done by COWing the page and changing
the instruction during the copy, this even though most platforms
allow atomic writes of the breakpoint instruction. This also
mirrors the behaviour of a ptrace() memory write to a PRIVATE
file map.
The core worker is derived from KSM's replace_page() logic.
In essence, similar to KSM:
a. allocate a new page and copy over contents of the page that
has the uprobed vaddr
b. modify the copy and insert the breakpoint at the required
address
c. switch the original page with the copy containing the
breakpoint
d. flush page tables.
replace_page() is being replicated here because of some minor
changes in the type of pages and also because Hugh Dickins had
plans to improve replace_page() for KSM specific work.
Instruction analysis on x86 is based on instruction decoder and
determines if an instruction can be probed and determines the
necessary fixups after singlestep. Instruction analysis is done
at probe insertion time so that we avoid having to repeat the
same analysis every time a probe is hit.
A lot of code here is due to the improvement/suggestions/inputs
from Peter Zijlstra.
Changelog:
(v10):
- Add code to clear REX.B prefix as suggested by Denys Vlasenko
and Masami Hiramatsu.
(v9):
- Use insn_offset_modrm as suggested by Masami Hiramatsu.
(v7):
Handle comments from Peter Zijlstra:
- Dont take reference to inode. (expect inode to uprobe_register to be sane).
- Use PTR_ERR to set the return value.
- No need to take reference to inode.
- use PTR_ERR to return error value.
- register and uprobe_unregister share code.
(v5):
- Modified del_consumer as per comments from Peter.
- Drop reference to inode before dropping reference to uprobe.
- Use i_size_read(inode) instead of inode->i_size.
- Ensure uprobe->consumers is NULL, before __uprobe_unregister() is called.
- Includes errno.h as recommended by Stephen Rothwell to fix a build issue
on sparc defconfig
- Remove restrictions while unregistering.
- Earlier code leaked inode references under some conditions while
registering/unregistering.
- Continue the vma-rmap walk even if the intermediate vma doesnt
meet the requirements.
- Validate the vma found by find_vma before inserting/removing the
breakpoint
- Call del_consumer under mutex_lock.
- Use hash locks.
- Handle mremap.
- Introduce find_least_offset_node() instead of close match logic in
find_uprobe
- Uprobes no more depends on MM_OWNER; No reference to task_structs
while inserting/removing a probe.
- Uses read_mapping_page instead of grab_cache_page so that the pages
have valid content.
- pass NULL to get_user_pages for the task parameter.
- call SetPageUptodate on the new page allocated in write_opcode.
- fix leaking a reference to the new page under certain conditions.
- Include Instruction Decoder if Uprobes gets defined.
- Remove const attributes for instruction prefix arrays.
- Uses mm_context to know if the application is 32 bit.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Also-written-by: Jim Keniston <jkenisto@us.ibm.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Roland McGrath <roland@hack.frob.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Anton Arapov <anton@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Denys Vlasenko <vda.linux@googlemail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linux-mm <linux-mm@kvack.org>
Link: http://lkml.kernel.org/r/20120209092642.GE16600@linux.vnet.ibm.com
[ Made various small edits to the commit log ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-02-09 17:26:42 +08:00
|
|
|
static int __init init_uprobes(void)
|
|
|
|
{
|
|
|
|
int i;
|
|
|
|
|
2012-11-26 05:48:37 +08:00
|
|
|
for (i = 0; i < UPROBES_HASH_SZ; i++)
|
uprobes, mm, x86: Add the ability to install and remove uprobes breakpoints
Add uprobes support to the core kernel, with x86 support.
This commit adds the kernel facilities, the actual uprobes
user-space ABI and perf probe support comes in later commits.
General design:
Uprobes are maintained in an rb-tree indexed by inode and offset
(the offset here is from the start of the mapping). For a unique
(inode, offset) tuple, there can be at most one uprobe in the
rb-tree.
Since the (inode, offset) tuple identifies a unique uprobe, more
than one user may be interested in the same uprobe. This provides
the ability to connect multiple 'consumers' to the same uprobe.
Each consumer defines a handler and a filter (optional). The
'handler' is run every time the uprobe is hit, if it matches the
'filter' criteria.
The first consumer of a uprobe causes the breakpoint to be
inserted at the specified address and subsequent consumers are
appended to this list. On subsequent probes, the consumer gets
appended to the existing list of consumers. The breakpoint is
removed when the last consumer unregisters. For all other
unregisterations, the consumer is removed from the list of
consumers.
Given a inode, we get a list of the mms that have mapped the
inode. Do the actual registration if mm maps the page where a
probe needs to be inserted/removed.
We use a temporary list to walk through the vmas that map the
inode.
- The number of maps that map the inode, is not known before we
walk the rmap and keeps changing.
- extending vm_area_struct wasn't recommended, it's a
size-critical data structure.
- There can be more than one maps of the inode in the same mm.
We add callbacks to the mmap methods to keep an eye on text vmas
that are of interest to uprobes. When a vma of interest is mapped,
we insert the breakpoint at the right address.
Uprobe works by replacing the instruction at the address defined
by (inode, offset) with the arch specific breakpoint
instruction. We save a copy of the original instruction at the
uprobed address.
This is needed for:
a. executing the instruction out-of-line (xol).
b. instruction analysis for any subsequent fixups.
c. restoring the instruction back when the uprobe is unregistered.
We insert or delete a breakpoint instruction, and this
breakpoint instruction is assumed to be the smallest instruction
available on the platform. For fixed size instruction platforms
this is trivially true, for variable size instruction platforms
the breakpoint instruction is typically the smallest (often a
single byte).
Writing the instruction is done by COWing the page and changing
the instruction during the copy, this even though most platforms
allow atomic writes of the breakpoint instruction. This also
mirrors the behaviour of a ptrace() memory write to a PRIVATE
file map.
The core worker is derived from KSM's replace_page() logic.
In essence, similar to KSM:
a. allocate a new page and copy over contents of the page that
has the uprobed vaddr
b. modify the copy and insert the breakpoint at the required
address
c. switch the original page with the copy containing the
breakpoint
d. flush page tables.
replace_page() is being replicated here because of some minor
changes in the type of pages and also because Hugh Dickins had
plans to improve replace_page() for KSM specific work.
Instruction analysis on x86 is based on instruction decoder and
determines if an instruction can be probed and determines the
necessary fixups after singlestep. Instruction analysis is done
at probe insertion time so that we avoid having to repeat the
same analysis every time a probe is hit.
A lot of code here is due to the improvement/suggestions/inputs
from Peter Zijlstra.
Changelog:
(v10):
- Add code to clear REX.B prefix as suggested by Denys Vlasenko
and Masami Hiramatsu.
(v9):
- Use insn_offset_modrm as suggested by Masami Hiramatsu.
(v7):
Handle comments from Peter Zijlstra:
- Dont take reference to inode. (expect inode to uprobe_register to be sane).
- Use PTR_ERR to set the return value.
- No need to take reference to inode.
- use PTR_ERR to return error value.
- register and uprobe_unregister share code.
(v5):
- Modified del_consumer as per comments from Peter.
- Drop reference to inode before dropping reference to uprobe.
- Use i_size_read(inode) instead of inode->i_size.
- Ensure uprobe->consumers is NULL, before __uprobe_unregister() is called.
- Includes errno.h as recommended by Stephen Rothwell to fix a build issue
on sparc defconfig
- Remove restrictions while unregistering.
- Earlier code leaked inode references under some conditions while
registering/unregistering.
- Continue the vma-rmap walk even if the intermediate vma doesnt
meet the requirements.
- Validate the vma found by find_vma before inserting/removing the
breakpoint
- Call del_consumer under mutex_lock.
- Use hash locks.
- Handle mremap.
- Introduce find_least_offset_node() instead of close match logic in
find_uprobe
- Uprobes no more depends on MM_OWNER; No reference to task_structs
while inserting/removing a probe.
- Uses read_mapping_page instead of grab_cache_page so that the pages
have valid content.
- pass NULL to get_user_pages for the task parameter.
- call SetPageUptodate on the new page allocated in write_opcode.
- fix leaking a reference to the new page under certain conditions.
- Include Instruction Decoder if Uprobes gets defined.
- Remove const attributes for instruction prefix arrays.
- Uses mm_context to know if the application is 32 bit.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Also-written-by: Jim Keniston <jkenisto@us.ibm.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Roland McGrath <roland@hack.frob.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Anton Arapov <anton@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Denys Vlasenko <vda.linux@googlemail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linux-mm <linux-mm@kvack.org>
Link: http://lkml.kernel.org/r/20120209092642.GE16600@linux.vnet.ibm.com
[ Made various small edits to the commit log ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-02-09 17:26:42 +08:00
|
|
|
mutex_init(&uprobes_mmap_mutex[i]);
|
uprobes/core: Handle breakpoint and singlestep exceptions
Uprobes uses exception notifiers to get to know if a thread hit
a breakpoint or a singlestep exception.
When a thread hits a uprobe or is singlestepping post a uprobe
hit, the uprobe exception notifier sets its TIF_UPROBE bit,
which will then be checked on its return to userspace path
(do_notify_resume() ->uprobe_notify_resume()), where the
consumers handlers are run (in task context) based on the
defined filters.
Uprobe hits are thread specific and hence we need to maintain
information about if a task hit a uprobe, what uprobe was hit,
the slot where the original instruction was copied for xol so
that it can be singlestepped with appropriate fixups.
In some cases, special care is needed for instructions that are
executed out of line (xol). These are architecture specific
artefacts, such as handling RIP relative instructions on x86_64.
Since the instruction at which the uprobe was inserted is
executed out of line, architecture specific fixups are added so
that the thread continues normal execution in the presence of a
uprobe.
Postpone the signals until we execute the probed insn.
post_xol() path does a recalc_sigpending() before return to
user-mode, this ensures the signal can't be lost.
Uprobes relies on DIE_DEBUG notification to notify if a
singlestep is complete.
Adds x86 specific uprobe exception notifiers and appropriate
hooks needed to determine a uprobe hit and subsequent post
processing.
Add requisite x86 fixups for xol for uprobes. Specific cases
needing fixups include relative jumps (x86_64), calls, etc.
Where possible, we check and skip singlestepping the
breakpointed instructions. For now we skip single byte as well
as few multibyte nop instructions. However this can be extended
to other instructions too.
Credits to Oleg Nesterov for suggestions/patches related to
signal, breakpoint, singlestep handling code.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Jim Keniston <jkenisto@linux.vnet.ibm.com>
Cc: Linux-mm <linux-mm@kvack.org>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/20120313180011.29771.89027.sendpatchset@srdronam.in.ibm.com
[ Performed various cleanliness edits ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-03-14 02:00:11 +08:00
|
|
|
|
2012-11-15 02:03:42 +08:00
|
|
|
if (percpu_init_rwsem(&dup_mmap_sem))
|
|
|
|
return -ENOMEM;
|
|
|
|
|
uprobes/core: Handle breakpoint and singlestep exceptions
Uprobes uses exception notifiers to get to know if a thread hit
a breakpoint or a singlestep exception.
When a thread hits a uprobe or is singlestepping post a uprobe
hit, the uprobe exception notifier sets its TIF_UPROBE bit,
which will then be checked on its return to userspace path
(do_notify_resume() ->uprobe_notify_resume()), where the
consumers handlers are run (in task context) based on the
defined filters.
Uprobe hits are thread specific and hence we need to maintain
information about if a task hit a uprobe, what uprobe was hit,
the slot where the original instruction was copied for xol so
that it can be singlestepped with appropriate fixups.
In some cases, special care is needed for instructions that are
executed out of line (xol). These are architecture specific
artefacts, such as handling RIP relative instructions on x86_64.
Since the instruction at which the uprobe was inserted is
executed out of line, architecture specific fixups are added so
that the thread continues normal execution in the presence of a
uprobe.
Postpone the signals until we execute the probed insn.
post_xol() path does a recalc_sigpending() before return to
user-mode, this ensures the signal can't be lost.
Uprobes relies on DIE_DEBUG notification to notify if a
singlestep is complete.
Adds x86 specific uprobe exception notifiers and appropriate
hooks needed to determine a uprobe hit and subsequent post
processing.
Add requisite x86 fixups for xol for uprobes. Specific cases
needing fixups include relative jumps (x86_64), calls, etc.
Where possible, we check and skip singlestepping the
breakpointed instructions. For now we skip single byte as well
as few multibyte nop instructions. However this can be extended
to other instructions too.
Credits to Oleg Nesterov for suggestions/patches related to
signal, breakpoint, singlestep handling code.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Jim Keniston <jkenisto@linux.vnet.ibm.com>
Cc: Linux-mm <linux-mm@kvack.org>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/20120313180011.29771.89027.sendpatchset@srdronam.in.ibm.com
[ Performed various cleanliness edits ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-03-14 02:00:11 +08:00
|
|
|
return register_die_notifier(&uprobe_exception_nb);
|
uprobes, mm, x86: Add the ability to install and remove uprobes breakpoints
Add uprobes support to the core kernel, with x86 support.
This commit adds the kernel facilities, the actual uprobes
user-space ABI and perf probe support comes in later commits.
General design:
Uprobes are maintained in an rb-tree indexed by inode and offset
(the offset here is from the start of the mapping). For a unique
(inode, offset) tuple, there can be at most one uprobe in the
rb-tree.
Since the (inode, offset) tuple identifies a unique uprobe, more
than one user may be interested in the same uprobe. This provides
the ability to connect multiple 'consumers' to the same uprobe.
Each consumer defines a handler and a filter (optional). The
'handler' is run every time the uprobe is hit, if it matches the
'filter' criteria.
The first consumer of a uprobe causes the breakpoint to be
inserted at the specified address and subsequent consumers are
appended to this list. On subsequent probes, the consumer gets
appended to the existing list of consumers. The breakpoint is
removed when the last consumer unregisters. For all other
unregisterations, the consumer is removed from the list of
consumers.
Given a inode, we get a list of the mms that have mapped the
inode. Do the actual registration if mm maps the page where a
probe needs to be inserted/removed.
We use a temporary list to walk through the vmas that map the
inode.
- The number of maps that map the inode, is not known before we
walk the rmap and keeps changing.
- extending vm_area_struct wasn't recommended, it's a
size-critical data structure.
- There can be more than one maps of the inode in the same mm.
We add callbacks to the mmap methods to keep an eye on text vmas
that are of interest to uprobes. When a vma of interest is mapped,
we insert the breakpoint at the right address.
Uprobe works by replacing the instruction at the address defined
by (inode, offset) with the arch specific breakpoint
instruction. We save a copy of the original instruction at the
uprobed address.
This is needed for:
a. executing the instruction out-of-line (xol).
b. instruction analysis for any subsequent fixups.
c. restoring the instruction back when the uprobe is unregistered.
We insert or delete a breakpoint instruction, and this
breakpoint instruction is assumed to be the smallest instruction
available on the platform. For fixed size instruction platforms
this is trivially true, for variable size instruction platforms
the breakpoint instruction is typically the smallest (often a
single byte).
Writing the instruction is done by COWing the page and changing
the instruction during the copy, this even though most platforms
allow atomic writes of the breakpoint instruction. This also
mirrors the behaviour of a ptrace() memory write to a PRIVATE
file map.
The core worker is derived from KSM's replace_page() logic.
In essence, similar to KSM:
a. allocate a new page and copy over contents of the page that
has the uprobed vaddr
b. modify the copy and insert the breakpoint at the required
address
c. switch the original page with the copy containing the
breakpoint
d. flush page tables.
replace_page() is being replicated here because of some minor
changes in the type of pages and also because Hugh Dickins had
plans to improve replace_page() for KSM specific work.
Instruction analysis on x86 is based on instruction decoder and
determines if an instruction can be probed and determines the
necessary fixups after singlestep. Instruction analysis is done
at probe insertion time so that we avoid having to repeat the
same analysis every time a probe is hit.
A lot of code here is due to the improvement/suggestions/inputs
from Peter Zijlstra.
Changelog:
(v10):
- Add code to clear REX.B prefix as suggested by Denys Vlasenko
and Masami Hiramatsu.
(v9):
- Use insn_offset_modrm as suggested by Masami Hiramatsu.
(v7):
Handle comments from Peter Zijlstra:
- Dont take reference to inode. (expect inode to uprobe_register to be sane).
- Use PTR_ERR to set the return value.
- No need to take reference to inode.
- use PTR_ERR to return error value.
- register and uprobe_unregister share code.
(v5):
- Modified del_consumer as per comments from Peter.
- Drop reference to inode before dropping reference to uprobe.
- Use i_size_read(inode) instead of inode->i_size.
- Ensure uprobe->consumers is NULL, before __uprobe_unregister() is called.
- Includes errno.h as recommended by Stephen Rothwell to fix a build issue
on sparc defconfig
- Remove restrictions while unregistering.
- Earlier code leaked inode references under some conditions while
registering/unregistering.
- Continue the vma-rmap walk even if the intermediate vma doesnt
meet the requirements.
- Validate the vma found by find_vma before inserting/removing the
breakpoint
- Call del_consumer under mutex_lock.
- Use hash locks.
- Handle mremap.
- Introduce find_least_offset_node() instead of close match logic in
find_uprobe
- Uprobes no more depends on MM_OWNER; No reference to task_structs
while inserting/removing a probe.
- Uses read_mapping_page instead of grab_cache_page so that the pages
have valid content.
- pass NULL to get_user_pages for the task parameter.
- call SetPageUptodate on the new page allocated in write_opcode.
- fix leaking a reference to the new page under certain conditions.
- Include Instruction Decoder if Uprobes gets defined.
- Remove const attributes for instruction prefix arrays.
- Uses mm_context to know if the application is 32 bit.
Signed-off-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Also-written-by: Jim Keniston <jkenisto@us.ibm.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Roland McGrath <roland@hack.frob.com>
Cc: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Arnaldo Carvalho de Melo <acme@infradead.org>
Cc: Anton Arapov <anton@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Denys Vlasenko <vda.linux@googlemail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linux-mm <linux-mm@kvack.org>
Link: http://lkml.kernel.org/r/20120209092642.GE16600@linux.vnet.ibm.com
[ Made various small edits to the commit log ]
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2012-02-09 17:26:42 +08:00
|
|
|
}
|
2013-11-01 02:28:22 +08:00
|
|
|
__initcall(init_uprobes);
|