These new controls toggle experimental support for a new CPU feature,
the straightforward extension of largepages from the pmd level to the
pud level, which allows 1GB (kernel) TLBs instead of 2MB TLBs.
Turn it off by default, as this code has not been tested well enough yet.
Use the CONFIG_DIRECT_GBPAGES=y .config option or gbpages on the
boot line can be used to enable it. If enabled in the .config then
nogbpages boot option disables it.
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
recently the 64-bit allyesconfig bzImage kernel started spontaneously
rebooting during early bootup.
after a few fun hours spent with early init debugging, it turns out
that we've got this rather annoying limit on the size of the kernel
image:
#define KERNEL_TEXT_SIZE (40*1024*1024)
which limit my vmlinux just happened to pass:
text data bss dec hex filename
29703744 4222751 8646224 42572719 2899baf vmlinux
40 MB is 42572719 bytes, so my vmlinux was just 1.5% above this limit :-/
So it happily crashed right in head_64.S, which - as we all know - is
the most debuggable code in the whole architecture ;-)
So increase the limit to allow an up to 128MB kernel image to be mapped.
(should anyone be that crazy or lazy)
We have a full 4K of pagetable (level2_kernel_pgt) allocated for these
mappings already, so there's no RAM overhead and the limit was rather
pointless and arbitrary.
Signed-off-by: Ingo Molnar <mingo@elte.hu>
The early boot code maps KERNEL_TEXT_SIZE (currently 40MB) starting
from __START_KERNEL_map. The kernel itself only needs _text to _end
mapped in the high alias. On relocatible kernels the ASM setup code
adjusts the compile time created high mappings to the relocation. This
creates invalid pmd entries for negative offsets:
0xffffffff80000000 -> pmd entry: ffffffffff2001e3
It points outside of the physical address space and is marked present.
This starts at the virtual address __START_KERNEL_map and goes up to
the point where the first valid physical address (0x0) is mapped.
Zap the mappings before _text and after _end right away in early
boot. This removes also the invalid entries.
Furthermore it simplifies the range check for high aliases.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Acked-by: H. Peter Anvin <hpa@zytor.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
extern should not appear in C files. Also, the definitions
do not match the prototype currently, not sure what way you
want to go with this, I've switched the prototype to return
int, but I can see going to the void return as well.
Signed-off-by: Harvey Harrison <harvey.harrison@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
DEBUG_PAGEALLOC was not possible on 64-bit due to its early-bootup
hardcoded reliance on PSE pages, and the unrobustness of the runtime
splitup of large pages. The splitup ended in recursive calls to
alloc_pages() when a page for a pte split was requested.
Avoid the recursion with a preallocated page pool, which is used to
split up large mappings and gets refilled in the return path of
kernel_map_pages after the split has been done. The size of the page
pool is adjusted to the available memory.
This part just implements the page pool and the initialization w/o
using it yet.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
This patchset adds a flags variable to reserve_bootmem() and uses the
BOOTMEM_EXCLUSIVE flag in crashkernel reservation code to detect collisions
between crashkernel area and already used memory.
This patch:
Change the reserve_bootmem() function to accept a new flag BOOTMEM_EXCLUSIVE.
If that flag is set, the function returns with -EBUSY if the memory already
has been reserved in the past. This is to avoid conflicts.
Because that code runs before SMP initialisation, there's no race condition
inside reserve_bootmem_core().
[akpm@linux-foundation.org: coding-style fixes]
[akpm@linux-foundation.org: fix powerpc build]
Signed-off-by: Bernhard Walle <bwalle@suse.de>
Cc: <linux-arch@vger.kernel.org>
Cc: "Eric W. Biederman" <ebiederm@xmission.com>
Cc: Vivek Goyal <vgoyal@in.ibm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
The .rodata section shouldn't just be read-only,
but also non-executable. This is free since we've broken
up the 2MB page already anyway.
also update test_nx to check for this.
Signed-off-by: Arjan van de Ven <arjan@linux.intel.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Use set_pte() for setting up the 2MB pages in the direct mapping.
Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
one early crash on one 8 node 256g machine:
Command line: console=uart8250,io,0x3f8,115200n8 initrd=kernel.org/mydisk11_x86_64.gz rw root=/dev/ram0 debug initcall_debug apic=debug acpi.debug_level=0x0000000f pci=routeirq ip=dhcp load_ramdisk=1 ramdisk_size=131072 BOOT_IMAGE=kernel.org/bzImage_2.6.25_k8.1
BIOS-provided physical RAM map:
BIOS-e820: 0000000000000000 - 000000000009bc00 (usable)
BIOS-e820: 000000000009bc00 - 00000000000a0000 (reserved)
BIOS-e820: 00000000000e6000 - 0000000000100000 (reserved)
BIOS-e820: 0000000000100000 - 00000000dffe0000 (usable)
BIOS-e820: 00000000dffe0000 - 00000000dffee000 (ACPI data)
BIOS-e820: 00000000dffee000 - 00000000dffff050 (ACPI NVS)
BIOS-e820: 00000000dffff050 - 00000000e0000000 (reserved)
BIOS-e820: 00000000fec00000 - 00000000fec01000 (reserved)
BIOS-e820: 00000000fee00000 - 00000000fee01000 (reserved)
BIOS-e820: 00000000ff700000 - 0000000100000000 (reserved)
BIOS-e820: 0000000100000000 - 0000004020000000 (usable)
Early serial console at I/O port 0x3f8 (options '115200n8')
console [uart0] enabled
end_pfn_map = 67239936
Kernel panic - not syncing: Duplicated early reservation d40000-e42000
Pid: 0, comm: swapper Not tainted 2.6.24-smp-g5a514e21-dirty #3
Call Trace:
[<ffffffff80221545>] lapic_get_maxlvt+0x0/0x10
[<ffffffff80221657>] clear_local_APIC+0x5/0xcf
[<ffffffff80221726>] disable_local_APIC+0x5/0x17
[<ffffffff8021fe16>] smp_send_stop+0x46/0x4c
[<ffffffff80235293>] panic+0x94/0x13e
[<ffffffff80bc3b03>] sctp_eps_proc_init+0x12/0x34
[<ffffffff80b9f1c5>] reserve_early+0x30/0x6c
[<ffffffff80803925>] init_memory_mapping+0x2cd/0x2dc
[<ffffffff80b9dc01>] setup_arch+0x21f/0x44e
[<ffffffff80b978be>] start_kernel+0x6f/0x2c7
[<ffffffff80b971cc>] _sinittext+0x1cc/0x1d3
it turns out there is overlap between pgtable and bss...
in System.map we have
ffffffff80d40420 b rsi_table
ffffffff80d40620 B krb5_seq_lock
ffffffff80d40628 b i.20437
ffffffff80d40630 b xprt_rdma_inline_write_padding
ffffffff80d40638 b sunrpc_table_header
ffffffff80d40640 b zero
ffffffff80d40644 b min_memreg
ffffffff80d40648 b rpcrdma_tk_lock_g
ffffffff80d40650 B sctp_assocs_id_lock
ffffffff80d40658 B proc_net_sctp
ffffffff80d40660 B sctp_assocs_id
ffffffff80d40680 B sysctl_sctp_mem
ffffffff80d40690 B sysctl_sctp_rmem
ffffffff80d406a0 B sysctl_sctp_wmem
ffffffff80d406b0 b sctp_ctl_socket
ffffffff80d406b8 b sctp_pf_inet6_specific
ffffffff80d406c0 b sctp_pf_inet_specific
ffffffff80d406c8 b sctp_af_v4_specific
ffffffff80d406d0 b sctp_af_v6_specific
ffffffff80d406d8 b sctp_rand.33270
ffffffff80d406dc b sctp_memory_pressure
ffffffff80d406e0 b sctp_sockets_allocated
ffffffff80d406e4 b sctp_memory_allocated
ffffffff80d406e8 b sctp_sysctl_header
ffffffff80d406f0 b zero
ffffffff80d406f4 A __bss_stop
ffffffff80d406f4 A _end
need to round up table_start to PAGE_SIZE.
also make the panic more informative.
Signed-off-by: Yinghai Lu <yinghai.lu@sun.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
debug incorrect/late access to init memory, by permanently unmapping
the init memory ranges. Depends on CONFIG_DEBUG_PAGEALLOC=y.
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
It looks like a mismerge put the rodata self-check in the wrong spot; move
it to the right place after marking the .rodata section read only.
Signed-off-by: Arjan van de Ven <arjan@linux.intel.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Latest update; I now have 4 NX tests, but 2 fail so they're #if 0'd.
I also cleaned up the NX test code quite a bit, and got rid of the ugly
exception table sorting stuff.
From: Arjan van de Ven <arjan@linux.intel.com>
This patch adds testcases for the CONFIG_DEBUG_RODATA configuration option
as well as the NX CPU feature/mappings. Both testcases can move to tests/
once that patch gets merged into mainline.
(I'm half considering moving the rodata test into mm/init.c but I'll
wait with that until init.c is unified)
As part of this I had to fix a not-quite-right alignment in the vmlinux.lds.h
for the RODATA sections, which lead to 1 page less being marked read only.
Signed-off-by: Arjan van de Ven <arjan@linux.intel.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
When we free initmem, various rodata and CPA checks may have left
memory read only.. this patch ensures that the memory is writable
before we free it.
Signed-off-by: Arjan van de Ven <arjan@linux.intel.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
The set_memory_* and set_pages_* family of API's currently requires the
callers to do a global tlb flush after the function call; forgetting this is
a very nasty deathtrap. This patch moves the global tlb flush into
each of the callers
Signed-off-by: Arjan van de Ven <arjan@linux.intel.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
This patch converts various users of change_page_attr() to the new,
more intent driven set_page_*/set_memory_* API set.
Signed-off-by: Arjan van de Ven <arjan@linux.intel.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Fix early_ioremap() on x86-64
I had ACPI failures on several machines since a few days. Symptom
was NUMA nodes not getting detected or worse cores not getting detected.
They all came from ACPI not being able to read various of its tables. I finally
bisected it down to Jeremy's "put _PAGE_GLOBAL into PAGE_KERNEL" change.
With that the fix was fairly obvious. The problem was that early_ioremap()
didn't use a "_all" flush that would affect the global PTEs too. So
with global bits getting used everywhere now an early_ioremap would
not actually flush a mapping if something else was mapped previously
on that slot (which can happen with early_iounmap inbetween)
This patch changes all flushes in init_64.c to be __flush_tlb_all()
and fixes the problem here.
Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
When CONFIG_DEBUG_RODATA is enabled undo the ro mapping and redo it again.
This gives some simple testing for change_page_attr().
Signed-off-by: Andi Kleen <ak@suse.de>
Acked-by: Jan Beulich <jbeulich@novell.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
On x86-64 there are several memory allocations before bootmem. To avoid
them stomping on each other they used to be all hard coded in bad_area().
Replace this with an array that is filled as needed.
This cleans up the code considerably and allows to expand its use.
Cc: peterz@infradead.org
Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
empty_zero_page is in .bss section, and it is cleared in clear_bss by
x86_64_start_kernel(). So don't clear that again in mem_init
Signed-off-by: Yinghai Lu <yinghai.lu@sun.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
It only has a single use, which can be trivially replaced.
Signed-off-by: Jeremy Fitzhardinge <Jeremy.Fitzhardinge@citrix.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
This patch does some whitespace cleanups in the paging code to fix some
checkpatch.pl warnings of my formerly merged cleanup patches.
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
This minor cleanup replaces _KERNPG_TABLE with the __PAGE_KERNEL* for 2MB PTEs
in the 64-bit memory initialization code. The __PAGE_KERNEL* defines are more
appropriate for PTEs.
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Use sparsemem as the only memory model for UP, SMP and NUMA. Measurements
indicate that DISCONTIGMEM has a higher overhead than sparsemem. And
FLATMEMs benefits are minimal. So I think its best to simply standardize
on sparsemem.
Results of page allocator tests (test can be had via git from slab git
tree branch tests)
Measurements in cycle counts. 1000 allocations were performed and then the
average cycle count was calculated.
Order FlatMem Discontig SparseMem
0 639 665 641
1 567 647 593
2 679 774 692
3 763 967 781
4 961 1501 962
5 1356 2344 1392
6 2224 3982 2336
7 4869 7225 5074
8 12500 14048 12732
9 27926 28223 28165
10 58578 58714 58682
(Note that FlatMem is an SMP config and the rest NUMA configurations)
Memory use:
SMP Sparsemem
-------------
Kernel size:
text data bss dec hex filename
3849268 397739 1264856 5511863 541ab7 vmlinux
total used free shared buffers cached
Mem: 8242252 41164 8201088 0 352 11512
-/+ buffers/cache: 29300 8212952
Swap: 9775512 0 9775512
SMP Flatmem
-----------
Kernel size:
text data bss dec hex filename
3844612 397739 1264536 5506887 540747 vmlinux
So 4.5k growth in text size vs. FLATMEM.
total used free shared buffers cached
Mem: 8244052 40544 8203508 0 352 11484
-/+ buffers/cache: 28708 8215344
2k growth in overall memory use after boot.
NUMA discontig:
text data bss dec hex filename
3888124 470659 1276504 5635287 55fcd7 vmlinux
total used free shared buffers cached
Mem: 8256256 56908 8199348 0 352 11496
-/+ buffers/cache: 45060 8211196
Swap: 9775512 0 9775512
NUMA sparse:
text data bss dec hex filename
3896428 470659 1276824 5643911 561e87 vmlinux
8k text growth. Given that we fully inline virt_to_page and friends now
that is rather good.
total used free shared buffers cached
Mem: 8264720 57240 8207480 0 352 11516
-/+ buffers/cache: 45372 8219348
Swap: 9775512 0 9775512
The total available memory is increased by 8k.
This patch makes sparsemem the default and removes discontig and
flatmem support from x86.
[ akpm@linux-foundation.org: allnoconfig build fix ]
Acked-by: Andi Kleen <ak@suse.de>
Signed-off-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
This reverts commit 2e1c49db4c.
First off, testing in Fedora has shown it to cause boot failures,
bisected down by Martin Ebourne, and reported by Dave Jobes. So the
commit will likely be reverted in the 2.6.23 stable kernels.
Secondly, in the 2.6.24 model, x86-64 has now grown support for
SPARSEMEM_VMEMMAP, which disables the relevant code anyway, so while the
bug is not visible any more, it's become invisible due to the code just
being irrelevant and no longer enabled on the only architecture that
this ever affected.
Reported-by: Dave Jones <davej@redhat.com>
Tested-by: Martin Ebourne <fedora@ebourne.me.uk>
Cc: Zou Nan hai <nanhai.zou@intel.com>
Cc: Suresh Siddha <suresh.b.siddha@intel.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Acked-by: Andy Whitcroft <apw@shadowen.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Now, arch dependent code around CONFIG_MEMORY_HOTREMOVE is a mess.
This patch cleans up them. This is against 2.6.23-rc6-mm1.
- fix compile failure on ia64/ CONFIG_MEMORY_HOTPLUG && !CONFIG_MEMORY_HOTREMOVE case.
- For !CONFIG_MEMORY_HOTREMOVE, add generic no-op remove_memory(),
which returns -EINVAL.
- removed remove_pages() only used in powerpc.
- removed no-op remove_memory() in i386, sh, sparc64, x86_64.
- only powerpc returns -ENOSYS at memory hot remove(no-op). changes it
to return -EINVAL.
Note:
Currently, only ia64 supports CONFIG_MEMORY_HOTREMOVE. I welcome other
archs if there are requirements and testers.
Signed-off-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
x86_64 uses 2M page table entries to map its 1-1 kernel space. We also
implement the virtual memmap using 2M page table entries. So there is no
additional runtime overhead over FLATMEM, initialisation is slightly more
complex. As FLATMEM still references memory to obtain the mem_map pointer and
SPARSEMEM_VMEMMAP uses a compile time constant, SPARSEMEM_VMEMMAP should be
superior.
With this SPARSEMEM becomes the most efficient way of handling virt_to_page,
pfn_to_page and friends for UP, SMP and NUMA on x86_64.
[apw@shadowen.org: code resplit, style fixups]
[apw@shadowen.org: vmemmap x86_64: ensure end of section memmap is initialised]
Signed-off-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Andy Whitcroft <apw@shadowen.org>
Acked-by: Mel Gorman <mel@csn.ul.ie>
Cc: Andi Kleen <ak@suse.de>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Ingo Molnar