linux/fs/fscache/object-list.c

412 lines
11 KiB
C
Raw Normal View History

/* Global fscache object list maintainer and viewer
*
* Copyright (C) 2009 Red Hat, Inc. All Rights Reserved.
* Written by David Howells (dhowells@redhat.com)
*
* This program is free software; you can redistribute it and/or
* modify it under the terms of the GNU General Public Licence
* as published by the Free Software Foundation; either version
* 2 of the Licence, or (at your option) any later version.
*/
#define FSCACHE_DEBUG_LEVEL COOKIE
#include <linux/module.h>
#include <linux/seq_file.h>
include cleanup: Update gfp.h and slab.h includes to prepare for breaking implicit slab.h inclusion from percpu.h percpu.h is included by sched.h and module.h and thus ends up being included when building most .c files. percpu.h includes slab.h which in turn includes gfp.h making everything defined by the two files universally available and complicating inclusion dependencies. percpu.h -> slab.h dependency is about to be removed. Prepare for this change by updating users of gfp and slab facilities include those headers directly instead of assuming availability. As this conversion needs to touch large number of source files, the following script is used as the basis of conversion. http://userweb.kernel.org/~tj/misc/slabh-sweep.py The script does the followings. * Scan files for gfp and slab usages and update includes such that only the necessary includes are there. ie. if only gfp is used, gfp.h, if slab is used, slab.h. * When the script inserts a new include, it looks at the include blocks and try to put the new include such that its order conforms to its surrounding. It's put in the include block which contains core kernel includes, in the same order that the rest are ordered - alphabetical, Christmas tree, rev-Xmas-tree or at the end if there doesn't seem to be any matching order. * If the script can't find a place to put a new include (mostly because the file doesn't have fitting include block), it prints out an error message indicating which .h file needs to be added to the file. The conversion was done in the following steps. 1. The initial automatic conversion of all .c files updated slightly over 4000 files, deleting around 700 includes and adding ~480 gfp.h and ~3000 slab.h inclusions. The script emitted errors for ~400 files. 2. Each error was manually checked. Some didn't need the inclusion, some needed manual addition while adding it to implementation .h or embedding .c file was more appropriate for others. This step added inclusions to around 150 files. 3. The script was run again and the output was compared to the edits from #2 to make sure no file was left behind. 4. Several build tests were done and a couple of problems were fixed. e.g. lib/decompress_*.c used malloc/free() wrappers around slab APIs requiring slab.h to be added manually. 5. The script was run on all .h files but without automatically editing them as sprinkling gfp.h and slab.h inclusions around .h files could easily lead to inclusion dependency hell. Most gfp.h inclusion directives were ignored as stuff from gfp.h was usually wildly available and often used in preprocessor macros. Each slab.h inclusion directive was examined and added manually as necessary. 6. percpu.h was updated not to include slab.h. 7. Build test were done on the following configurations and failures were fixed. CONFIG_GCOV_KERNEL was turned off for all tests (as my distributed build env didn't work with gcov compiles) and a few more options had to be turned off depending on archs to make things build (like ipr on powerpc/64 which failed due to missing writeq). * x86 and x86_64 UP and SMP allmodconfig and a custom test config. * powerpc and powerpc64 SMP allmodconfig * sparc and sparc64 SMP allmodconfig * ia64 SMP allmodconfig * s390 SMP allmodconfig * alpha SMP allmodconfig * um on x86_64 SMP allmodconfig 8. percpu.h modifications were reverted so that it could be applied as a separate patch and serve as bisection point. Given the fact that I had only a couple of failures from tests on step 6, I'm fairly confident about the coverage of this conversion patch. If there is a breakage, it's likely to be something in one of the arch headers which should be easily discoverable easily on most builds of the specific arch. Signed-off-by: Tejun Heo <tj@kernel.org> Guess-its-ok-by: Christoph Lameter <cl@linux-foundation.org> Cc: Ingo Molnar <mingo@redhat.com> Cc: Lee Schermerhorn <Lee.Schermerhorn@hp.com>
2010-03-24 16:04:11 +08:00
#include <linux/slab.h>
#include <linux/key.h>
#include <keys/user-type.h>
#include "internal.h"
static struct rb_root fscache_object_list;
static DEFINE_RWLOCK(fscache_object_list_lock);
struct fscache_objlist_data {
unsigned long config; /* display configuration */
#define FSCACHE_OBJLIST_CONFIG_KEY 0x00000001 /* show object keys */
#define FSCACHE_OBJLIST_CONFIG_AUX 0x00000002 /* show object auxdata */
#define FSCACHE_OBJLIST_CONFIG_COOKIE 0x00000004 /* show objects with cookies */
#define FSCACHE_OBJLIST_CONFIG_NOCOOKIE 0x00000008 /* show objects without cookies */
#define FSCACHE_OBJLIST_CONFIG_BUSY 0x00000010 /* show busy objects */
#define FSCACHE_OBJLIST_CONFIG_IDLE 0x00000020 /* show idle objects */
#define FSCACHE_OBJLIST_CONFIG_PENDWR 0x00000040 /* show objects with pending writes */
#define FSCACHE_OBJLIST_CONFIG_NOPENDWR 0x00000080 /* show objects without pending writes */
#define FSCACHE_OBJLIST_CONFIG_READS 0x00000100 /* show objects with active reads */
#define FSCACHE_OBJLIST_CONFIG_NOREADS 0x00000200 /* show objects without active reads */
#define FSCACHE_OBJLIST_CONFIG_EVENTS 0x00000400 /* show objects with events */
#define FSCACHE_OBJLIST_CONFIG_NOEVENTS 0x00000800 /* show objects without no events */
#define FSCACHE_OBJLIST_CONFIG_WORK 0x00001000 /* show objects with work */
#define FSCACHE_OBJLIST_CONFIG_NOWORK 0x00002000 /* show objects without work */
u8 buf[512]; /* key and aux data buffer */
};
/*
* Add an object to the object list
* - we use the address of the fscache_object structure as the key into the
* tree
*/
void fscache_objlist_add(struct fscache_object *obj)
{
struct fscache_object *xobj;
struct rb_node **p = &fscache_object_list.rb_node, *parent = NULL;
ASSERT(RB_EMPTY_NODE(&obj->objlist_link));
write_lock(&fscache_object_list_lock);
while (*p) {
parent = *p;
xobj = rb_entry(parent, struct fscache_object, objlist_link);
if (obj < xobj)
p = &(*p)->rb_left;
else if (obj > xobj)
p = &(*p)->rb_right;
else
BUG();
}
rb_link_node(&obj->objlist_link, parent, p);
rb_insert_color(&obj->objlist_link, &fscache_object_list);
write_unlock(&fscache_object_list_lock);
}
FS-Cache: Simplify cookie retention for fscache_objects, fixing oops Simplify the way fscache cache objects retain their cookie. The way I implemented the cookie storage handling made synchronisation a pain (ie. the object state machine can't rely on the cookie actually still being there). Instead of the the object being detached from the cookie and the cookie being freed in __fscache_relinquish_cookie(), we defer both operations: (*) The detachment of the object from the list in the cookie now takes place in fscache_drop_object() and is thus governed by the object state machine (fscache_detach_from_cookie() has been removed). (*) The release of the cookie is now in fscache_object_destroy() - which is called by the cache backend just before it frees the object. This means that the fscache_cookie struct is now available to the cache all the way through from ->alloc_object() to ->drop_object() and ->put_object() - meaning that it's no longer necessary to take object->lock to guarantee access. However, __fscache_relinquish_cookie() doesn't wait for the object to go all the way through to destruction before letting the netfs proceed. That would massively slow down the netfs. Since __fscache_relinquish_cookie() leaves the cookie around, in must therefore break all attachments to the netfs - which includes ->def, ->netfs_data and any outstanding page read/writes. To handle this, struct fscache_cookie now has an n_active counter: (1) This starts off initialised to 1. (2) Any time the cache needs to get at the netfs data, it calls fscache_use_cookie() to increment it - if it is not zero. If it was zero, then access is not permitted. (3) When the cache has finished with the data, it calls fscache_unuse_cookie() to decrement it. This does a wake-up on it if it reaches 0. (4) __fscache_relinquish_cookie() decrements n_active and then waits for it to reach 0. The initialisation to 1 in step (1) ensures that we only get wake ups when we're trying to get rid of the cookie. This leaves __fscache_relinquish_cookie() a lot simpler. *** This fixes a problem in the current code whereby if fscache_invalidate() is followed sufficiently quickly by fscache_relinquish_cookie() then it is possible for __fscache_relinquish_cookie() to have detached the cookie from the object and cleared the pointer before a thread is dispatched to process the invalidation state in the object state machine. Since the pending write clearance was deferred to the invalidation state to make it asynchronous, we need to either wait in relinquishment for the stores tree to be cleared in the invalidation state or we need to handle the clearance in relinquishment. Further, if the relinquishment code does clear the tree, then the invalidation state need to make the clearance contingent on still having the cookie to hand (since that's where the tree is rooted) and we have to prevent the cookie from disappearing for the duration. This can lead to an oops like the following: BUG: unable to handle kernel NULL pointer dereference at 000000000000000c ... RIP: 0010:[<ffffffff8151023e>] _spin_lock+0xe/0x30 ... CR2: 000000000000000c ... ... Process kslowd002 (...) .... Call Trace: [<ffffffffa01c3278>] fscache_invalidate_writes+0x38/0xd0 [fscache] [<ffffffff810096f0>] ? __switch_to+0xd0/0x320 [<ffffffff8105e759>] ? find_busiest_queue+0x69/0x150 [<ffffffff8110ddd4>] ? slow_work_enqueue+0x104/0x180 [<ffffffffa01c1303>] fscache_object_slow_work_execute+0x5e3/0x9d0 [fscache] [<ffffffff81096b67>] ? bit_waitqueue+0x17/0xd0 [<ffffffff8110e233>] slow_work_execute+0x233/0x310 [<ffffffff8110e515>] slow_work_thread+0x205/0x360 [<ffffffff81096ca0>] ? autoremove_wake_function+0x0/0x40 [<ffffffff8110e310>] ? slow_work_thread+0x0/0x360 [<ffffffff81096936>] kthread+0x96/0xa0 [<ffffffff8100c0ca>] child_rip+0xa/0x20 [<ffffffff810968a0>] ? kthread+0x0/0xa0 [<ffffffff8100c0c0>] ? child_rip+0x0/0x20 The parameter to fscache_invalidate_writes() was object->cookie which is NULL. Signed-off-by: David Howells <dhowells@redhat.com> Tested-By: Milosz Tanski <milosz@adfin.com> Acked-by: Jeff Layton <jlayton@redhat.com>
2013-05-11 02:50:26 +08:00
/*
* Remove an object from the object list.
*/
FS-Cache: Simplify cookie retention for fscache_objects, fixing oops Simplify the way fscache cache objects retain their cookie. The way I implemented the cookie storage handling made synchronisation a pain (ie. the object state machine can't rely on the cookie actually still being there). Instead of the the object being detached from the cookie and the cookie being freed in __fscache_relinquish_cookie(), we defer both operations: (*) The detachment of the object from the list in the cookie now takes place in fscache_drop_object() and is thus governed by the object state machine (fscache_detach_from_cookie() has been removed). (*) The release of the cookie is now in fscache_object_destroy() - which is called by the cache backend just before it frees the object. This means that the fscache_cookie struct is now available to the cache all the way through from ->alloc_object() to ->drop_object() and ->put_object() - meaning that it's no longer necessary to take object->lock to guarantee access. However, __fscache_relinquish_cookie() doesn't wait for the object to go all the way through to destruction before letting the netfs proceed. That would massively slow down the netfs. Since __fscache_relinquish_cookie() leaves the cookie around, in must therefore break all attachments to the netfs - which includes ->def, ->netfs_data and any outstanding page read/writes. To handle this, struct fscache_cookie now has an n_active counter: (1) This starts off initialised to 1. (2) Any time the cache needs to get at the netfs data, it calls fscache_use_cookie() to increment it - if it is not zero. If it was zero, then access is not permitted. (3) When the cache has finished with the data, it calls fscache_unuse_cookie() to decrement it. This does a wake-up on it if it reaches 0. (4) __fscache_relinquish_cookie() decrements n_active and then waits for it to reach 0. The initialisation to 1 in step (1) ensures that we only get wake ups when we're trying to get rid of the cookie. This leaves __fscache_relinquish_cookie() a lot simpler. *** This fixes a problem in the current code whereby if fscache_invalidate() is followed sufficiently quickly by fscache_relinquish_cookie() then it is possible for __fscache_relinquish_cookie() to have detached the cookie from the object and cleared the pointer before a thread is dispatched to process the invalidation state in the object state machine. Since the pending write clearance was deferred to the invalidation state to make it asynchronous, we need to either wait in relinquishment for the stores tree to be cleared in the invalidation state or we need to handle the clearance in relinquishment. Further, if the relinquishment code does clear the tree, then the invalidation state need to make the clearance contingent on still having the cookie to hand (since that's where the tree is rooted) and we have to prevent the cookie from disappearing for the duration. This can lead to an oops like the following: BUG: unable to handle kernel NULL pointer dereference at 000000000000000c ... RIP: 0010:[<ffffffff8151023e>] _spin_lock+0xe/0x30 ... CR2: 000000000000000c ... ... Process kslowd002 (...) .... Call Trace: [<ffffffffa01c3278>] fscache_invalidate_writes+0x38/0xd0 [fscache] [<ffffffff810096f0>] ? __switch_to+0xd0/0x320 [<ffffffff8105e759>] ? find_busiest_queue+0x69/0x150 [<ffffffff8110ddd4>] ? slow_work_enqueue+0x104/0x180 [<ffffffffa01c1303>] fscache_object_slow_work_execute+0x5e3/0x9d0 [fscache] [<ffffffff81096b67>] ? bit_waitqueue+0x17/0xd0 [<ffffffff8110e233>] slow_work_execute+0x233/0x310 [<ffffffff8110e515>] slow_work_thread+0x205/0x360 [<ffffffff81096ca0>] ? autoremove_wake_function+0x0/0x40 [<ffffffff8110e310>] ? slow_work_thread+0x0/0x360 [<ffffffff81096936>] kthread+0x96/0xa0 [<ffffffff8100c0ca>] child_rip+0xa/0x20 [<ffffffff810968a0>] ? kthread+0x0/0xa0 [<ffffffff8100c0c0>] ? child_rip+0x0/0x20 The parameter to fscache_invalidate_writes() was object->cookie which is NULL. Signed-off-by: David Howells <dhowells@redhat.com> Tested-By: Milosz Tanski <milosz@adfin.com> Acked-by: Jeff Layton <jlayton@redhat.com>
2013-05-11 02:50:26 +08:00
void fscache_objlist_remove(struct fscache_object *obj)
{
if (RB_EMPTY_NODE(&obj->objlist_link))
return;
write_lock(&fscache_object_list_lock);
BUG_ON(RB_EMPTY_ROOT(&fscache_object_list));
rb_erase(&obj->objlist_link, &fscache_object_list);
write_unlock(&fscache_object_list_lock);
}
/*
* find the object in the tree on or after the specified index
*/
static struct fscache_object *fscache_objlist_lookup(loff_t *_pos)
{
struct fscache_object *pobj, *obj = NULL, *minobj = NULL;
struct rb_node *p;
unsigned long pos;
if (*_pos >= (unsigned long) ERR_PTR(-ENOENT))
return NULL;
pos = *_pos;
/* banners (can't represent line 0 by pos 0 as that would involve
* returning a NULL pointer) */
if (pos == 0)
return (struct fscache_object *)(long)++(*_pos);
if (pos < 3)
return (struct fscache_object *)pos;
pobj = (struct fscache_object *)pos;
p = fscache_object_list.rb_node;
while (p) {
obj = rb_entry(p, struct fscache_object, objlist_link);
if (pobj < obj) {
if (!minobj || minobj > obj)
minobj = obj;
p = p->rb_left;
} else if (pobj > obj) {
p = p->rb_right;
} else {
minobj = obj;
break;
}
obj = NULL;
}
if (!minobj)
*_pos = (unsigned long) ERR_PTR(-ENOENT);
else if (minobj != obj)
*_pos = (unsigned long) minobj;
return minobj;
}
/*
* set up the iterator to start reading from the first line
*/
static void *fscache_objlist_start(struct seq_file *m, loff_t *_pos)
__acquires(&fscache_object_list_lock)
{
read_lock(&fscache_object_list_lock);
return fscache_objlist_lookup(_pos);
}
/*
* move to the next line
*/
static void *fscache_objlist_next(struct seq_file *m, void *v, loff_t *_pos)
{
(*_pos)++;
return fscache_objlist_lookup(_pos);
}
/*
* clean up after reading
*/
static void fscache_objlist_stop(struct seq_file *m, void *v)
__releases(&fscache_object_list_lock)
{
read_unlock(&fscache_object_list_lock);
}
/*
* display an object
*/
static int fscache_objlist_show(struct seq_file *m, void *v)
{
struct fscache_objlist_data *data = m->private;
struct fscache_object *obj = v;
FS-Cache: Simplify cookie retention for fscache_objects, fixing oops Simplify the way fscache cache objects retain their cookie. The way I implemented the cookie storage handling made synchronisation a pain (ie. the object state machine can't rely on the cookie actually still being there). Instead of the the object being detached from the cookie and the cookie being freed in __fscache_relinquish_cookie(), we defer both operations: (*) The detachment of the object from the list in the cookie now takes place in fscache_drop_object() and is thus governed by the object state machine (fscache_detach_from_cookie() has been removed). (*) The release of the cookie is now in fscache_object_destroy() - which is called by the cache backend just before it frees the object. This means that the fscache_cookie struct is now available to the cache all the way through from ->alloc_object() to ->drop_object() and ->put_object() - meaning that it's no longer necessary to take object->lock to guarantee access. However, __fscache_relinquish_cookie() doesn't wait for the object to go all the way through to destruction before letting the netfs proceed. That would massively slow down the netfs. Since __fscache_relinquish_cookie() leaves the cookie around, in must therefore break all attachments to the netfs - which includes ->def, ->netfs_data and any outstanding page read/writes. To handle this, struct fscache_cookie now has an n_active counter: (1) This starts off initialised to 1. (2) Any time the cache needs to get at the netfs data, it calls fscache_use_cookie() to increment it - if it is not zero. If it was zero, then access is not permitted. (3) When the cache has finished with the data, it calls fscache_unuse_cookie() to decrement it. This does a wake-up on it if it reaches 0. (4) __fscache_relinquish_cookie() decrements n_active and then waits for it to reach 0. The initialisation to 1 in step (1) ensures that we only get wake ups when we're trying to get rid of the cookie. This leaves __fscache_relinquish_cookie() a lot simpler. *** This fixes a problem in the current code whereby if fscache_invalidate() is followed sufficiently quickly by fscache_relinquish_cookie() then it is possible for __fscache_relinquish_cookie() to have detached the cookie from the object and cleared the pointer before a thread is dispatched to process the invalidation state in the object state machine. Since the pending write clearance was deferred to the invalidation state to make it asynchronous, we need to either wait in relinquishment for the stores tree to be cleared in the invalidation state or we need to handle the clearance in relinquishment. Further, if the relinquishment code does clear the tree, then the invalidation state need to make the clearance contingent on still having the cookie to hand (since that's where the tree is rooted) and we have to prevent the cookie from disappearing for the duration. This can lead to an oops like the following: BUG: unable to handle kernel NULL pointer dereference at 000000000000000c ... RIP: 0010:[<ffffffff8151023e>] _spin_lock+0xe/0x30 ... CR2: 000000000000000c ... ... Process kslowd002 (...) .... Call Trace: [<ffffffffa01c3278>] fscache_invalidate_writes+0x38/0xd0 [fscache] [<ffffffff810096f0>] ? __switch_to+0xd0/0x320 [<ffffffff8105e759>] ? find_busiest_queue+0x69/0x150 [<ffffffff8110ddd4>] ? slow_work_enqueue+0x104/0x180 [<ffffffffa01c1303>] fscache_object_slow_work_execute+0x5e3/0x9d0 [fscache] [<ffffffff81096b67>] ? bit_waitqueue+0x17/0xd0 [<ffffffff8110e233>] slow_work_execute+0x233/0x310 [<ffffffff8110e515>] slow_work_thread+0x205/0x360 [<ffffffff81096ca0>] ? autoremove_wake_function+0x0/0x40 [<ffffffff8110e310>] ? slow_work_thread+0x0/0x360 [<ffffffff81096936>] kthread+0x96/0xa0 [<ffffffff8100c0ca>] child_rip+0xa/0x20 [<ffffffff810968a0>] ? kthread+0x0/0xa0 [<ffffffff8100c0c0>] ? child_rip+0x0/0x20 The parameter to fscache_invalidate_writes() was object->cookie which is NULL. Signed-off-by: David Howells <dhowells@redhat.com> Tested-By: Milosz Tanski <milosz@adfin.com> Acked-by: Jeff Layton <jlayton@redhat.com>
2013-05-11 02:50:26 +08:00
struct fscache_cookie *cookie;
unsigned long config = data->config;
char _type[3], *type;
u8 *buf = data->buf, *p;
if ((unsigned long) v == 1) {
seq_puts(m, "OBJECT PARENT STAT CHLDN OPS OOP IPR EX READS"
FS-Cache: Fix object state machine to have separate work and wait states Fix object state machine to have separate work and wait states as that makes it easier to envision. There are now three kinds of state: (1) Work state. This is an execution state. No event processing is performed by a work state. The function attached to a work state returns a pointer indicating the next state to which the OSM should transition. Returning NO_TRANSIT repeats the current state, but goes back to the scheduler first. (2) Wait state. This is an event processing state. No execution is performed by a wait state. Wait states are just tables of "if event X occurs, clear it and transition to state Y". The dispatcher returns to the scheduler if none of the events in which the wait state has an interest are currently pending. (3) Out-of-band state. This is a special work state. Transitions to normal states can be overridden when an unexpected event occurs (eg. I/O error). Instead the dispatcher disables and clears the OOB event and transits to the specified work state. This then acts as an ordinary work state, though object->state points to the overridden destination. Returning NO_TRANSIT resumes the overridden transition. In addition, the states have names in their definitions, so there's no need for tables of state names. Further, the EV_REQUEUE event is no longer necessary as that is automatic for work states. Since the states are now separate structs rather than values in an enum, it's not possible to use comparisons other than (non-)equality between them, so use some object->flags to indicate what phase an object is in. The EV_RELEASE, EV_RETIRE and EV_WITHDRAW events have been squished into one (EV_KILL). An object flag now carries the information about retirement. Similarly, the RELEASING, RECYCLING and WITHDRAWING states have been merged into an KILL_OBJECT state and additional states have been added for handling waiting dependent objects (JUMPSTART_DEPS and KILL_DEPENDENTS). A state has also been added for synchronising with parent object initialisation (WAIT_FOR_PARENT) and another for initiating look up (PARENT_READY). Signed-off-by: David Howells <dhowells@redhat.com> Tested-By: Milosz Tanski <milosz@adfin.com> Acked-by: Jeff Layton <jlayton@redhat.com>
2013-05-11 02:50:26 +08:00
" EM EV FL S"
" | NETFS_COOKIE_DEF TY FL NETFS_DATA");
if (config & (FSCACHE_OBJLIST_CONFIG_KEY |
FSCACHE_OBJLIST_CONFIG_AUX))
seq_puts(m, " ");
if (config & FSCACHE_OBJLIST_CONFIG_KEY)
seq_puts(m, "OBJECT_KEY");
if ((config & (FSCACHE_OBJLIST_CONFIG_KEY |
FSCACHE_OBJLIST_CONFIG_AUX)) ==
(FSCACHE_OBJLIST_CONFIG_KEY | FSCACHE_OBJLIST_CONFIG_AUX))
seq_puts(m, ", ");
if (config & FSCACHE_OBJLIST_CONFIG_AUX)
seq_puts(m, "AUX_DATA");
seq_puts(m, "\n");
return 0;
}
if ((unsigned long) v == 2) {
seq_puts(m, "======== ======== ==== ===== === === === == ====="
FS-Cache: Fix object state machine to have separate work and wait states Fix object state machine to have separate work and wait states as that makes it easier to envision. There are now three kinds of state: (1) Work state. This is an execution state. No event processing is performed by a work state. The function attached to a work state returns a pointer indicating the next state to which the OSM should transition. Returning NO_TRANSIT repeats the current state, but goes back to the scheduler first. (2) Wait state. This is an event processing state. No execution is performed by a wait state. Wait states are just tables of "if event X occurs, clear it and transition to state Y". The dispatcher returns to the scheduler if none of the events in which the wait state has an interest are currently pending. (3) Out-of-band state. This is a special work state. Transitions to normal states can be overridden when an unexpected event occurs (eg. I/O error). Instead the dispatcher disables and clears the OOB event and transits to the specified work state. This then acts as an ordinary work state, though object->state points to the overridden destination. Returning NO_TRANSIT resumes the overridden transition. In addition, the states have names in their definitions, so there's no need for tables of state names. Further, the EV_REQUEUE event is no longer necessary as that is automatic for work states. Since the states are now separate structs rather than values in an enum, it's not possible to use comparisons other than (non-)equality between them, so use some object->flags to indicate what phase an object is in. The EV_RELEASE, EV_RETIRE and EV_WITHDRAW events have been squished into one (EV_KILL). An object flag now carries the information about retirement. Similarly, the RELEASING, RECYCLING and WITHDRAWING states have been merged into an KILL_OBJECT state and additional states have been added for handling waiting dependent objects (JUMPSTART_DEPS and KILL_DEPENDENTS). A state has also been added for synchronising with parent object initialisation (WAIT_FOR_PARENT) and another for initiating look up (PARENT_READY). Signed-off-by: David Howells <dhowells@redhat.com> Tested-By: Milosz Tanski <milosz@adfin.com> Acked-by: Jeff Layton <jlayton@redhat.com>
2013-05-11 02:50:26 +08:00
" == == == ="
" | ================ == == ================");
if (config & (FSCACHE_OBJLIST_CONFIG_KEY |
FSCACHE_OBJLIST_CONFIG_AUX))
seq_puts(m, " ================");
seq_puts(m, "\n");
return 0;
}
/* filter out any unwanted objects */
#define FILTER(criterion, _yes, _no) \
do { \
unsigned long yes = FSCACHE_OBJLIST_CONFIG_##_yes; \
unsigned long no = FSCACHE_OBJLIST_CONFIG_##_no; \
if (criterion) { \
if (!(config & yes)) \
return 0; \
} else { \
if (!(config & no)) \
return 0; \
} \
} while(0)
FS-Cache: Simplify cookie retention for fscache_objects, fixing oops Simplify the way fscache cache objects retain their cookie. The way I implemented the cookie storage handling made synchronisation a pain (ie. the object state machine can't rely on the cookie actually still being there). Instead of the the object being detached from the cookie and the cookie being freed in __fscache_relinquish_cookie(), we defer both operations: (*) The detachment of the object from the list in the cookie now takes place in fscache_drop_object() and is thus governed by the object state machine (fscache_detach_from_cookie() has been removed). (*) The release of the cookie is now in fscache_object_destroy() - which is called by the cache backend just before it frees the object. This means that the fscache_cookie struct is now available to the cache all the way through from ->alloc_object() to ->drop_object() and ->put_object() - meaning that it's no longer necessary to take object->lock to guarantee access. However, __fscache_relinquish_cookie() doesn't wait for the object to go all the way through to destruction before letting the netfs proceed. That would massively slow down the netfs. Since __fscache_relinquish_cookie() leaves the cookie around, in must therefore break all attachments to the netfs - which includes ->def, ->netfs_data and any outstanding page read/writes. To handle this, struct fscache_cookie now has an n_active counter: (1) This starts off initialised to 1. (2) Any time the cache needs to get at the netfs data, it calls fscache_use_cookie() to increment it - if it is not zero. If it was zero, then access is not permitted. (3) When the cache has finished with the data, it calls fscache_unuse_cookie() to decrement it. This does a wake-up on it if it reaches 0. (4) __fscache_relinquish_cookie() decrements n_active and then waits for it to reach 0. The initialisation to 1 in step (1) ensures that we only get wake ups when we're trying to get rid of the cookie. This leaves __fscache_relinquish_cookie() a lot simpler. *** This fixes a problem in the current code whereby if fscache_invalidate() is followed sufficiently quickly by fscache_relinquish_cookie() then it is possible for __fscache_relinquish_cookie() to have detached the cookie from the object and cleared the pointer before a thread is dispatched to process the invalidation state in the object state machine. Since the pending write clearance was deferred to the invalidation state to make it asynchronous, we need to either wait in relinquishment for the stores tree to be cleared in the invalidation state or we need to handle the clearance in relinquishment. Further, if the relinquishment code does clear the tree, then the invalidation state need to make the clearance contingent on still having the cookie to hand (since that's where the tree is rooted) and we have to prevent the cookie from disappearing for the duration. This can lead to an oops like the following: BUG: unable to handle kernel NULL pointer dereference at 000000000000000c ... RIP: 0010:[<ffffffff8151023e>] _spin_lock+0xe/0x30 ... CR2: 000000000000000c ... ... Process kslowd002 (...) .... Call Trace: [<ffffffffa01c3278>] fscache_invalidate_writes+0x38/0xd0 [fscache] [<ffffffff810096f0>] ? __switch_to+0xd0/0x320 [<ffffffff8105e759>] ? find_busiest_queue+0x69/0x150 [<ffffffff8110ddd4>] ? slow_work_enqueue+0x104/0x180 [<ffffffffa01c1303>] fscache_object_slow_work_execute+0x5e3/0x9d0 [fscache] [<ffffffff81096b67>] ? bit_waitqueue+0x17/0xd0 [<ffffffff8110e233>] slow_work_execute+0x233/0x310 [<ffffffff8110e515>] slow_work_thread+0x205/0x360 [<ffffffff81096ca0>] ? autoremove_wake_function+0x0/0x40 [<ffffffff8110e310>] ? slow_work_thread+0x0/0x360 [<ffffffff81096936>] kthread+0x96/0xa0 [<ffffffff8100c0ca>] child_rip+0xa/0x20 [<ffffffff810968a0>] ? kthread+0x0/0xa0 [<ffffffff8100c0c0>] ? child_rip+0x0/0x20 The parameter to fscache_invalidate_writes() was object->cookie which is NULL. Signed-off-by: David Howells <dhowells@redhat.com> Tested-By: Milosz Tanski <milosz@adfin.com> Acked-by: Jeff Layton <jlayton@redhat.com>
2013-05-11 02:50:26 +08:00
cookie = obj->cookie;
if (~config) {
FS-Cache: Simplify cookie retention for fscache_objects, fixing oops Simplify the way fscache cache objects retain their cookie. The way I implemented the cookie storage handling made synchronisation a pain (ie. the object state machine can't rely on the cookie actually still being there). Instead of the the object being detached from the cookie and the cookie being freed in __fscache_relinquish_cookie(), we defer both operations: (*) The detachment of the object from the list in the cookie now takes place in fscache_drop_object() and is thus governed by the object state machine (fscache_detach_from_cookie() has been removed). (*) The release of the cookie is now in fscache_object_destroy() - which is called by the cache backend just before it frees the object. This means that the fscache_cookie struct is now available to the cache all the way through from ->alloc_object() to ->drop_object() and ->put_object() - meaning that it's no longer necessary to take object->lock to guarantee access. However, __fscache_relinquish_cookie() doesn't wait for the object to go all the way through to destruction before letting the netfs proceed. That would massively slow down the netfs. Since __fscache_relinquish_cookie() leaves the cookie around, in must therefore break all attachments to the netfs - which includes ->def, ->netfs_data and any outstanding page read/writes. To handle this, struct fscache_cookie now has an n_active counter: (1) This starts off initialised to 1. (2) Any time the cache needs to get at the netfs data, it calls fscache_use_cookie() to increment it - if it is not zero. If it was zero, then access is not permitted. (3) When the cache has finished with the data, it calls fscache_unuse_cookie() to decrement it. This does a wake-up on it if it reaches 0. (4) __fscache_relinquish_cookie() decrements n_active and then waits for it to reach 0. The initialisation to 1 in step (1) ensures that we only get wake ups when we're trying to get rid of the cookie. This leaves __fscache_relinquish_cookie() a lot simpler. *** This fixes a problem in the current code whereby if fscache_invalidate() is followed sufficiently quickly by fscache_relinquish_cookie() then it is possible for __fscache_relinquish_cookie() to have detached the cookie from the object and cleared the pointer before a thread is dispatched to process the invalidation state in the object state machine. Since the pending write clearance was deferred to the invalidation state to make it asynchronous, we need to either wait in relinquishment for the stores tree to be cleared in the invalidation state or we need to handle the clearance in relinquishment. Further, if the relinquishment code does clear the tree, then the invalidation state need to make the clearance contingent on still having the cookie to hand (since that's where the tree is rooted) and we have to prevent the cookie from disappearing for the duration. This can lead to an oops like the following: BUG: unable to handle kernel NULL pointer dereference at 000000000000000c ... RIP: 0010:[<ffffffff8151023e>] _spin_lock+0xe/0x30 ... CR2: 000000000000000c ... ... Process kslowd002 (...) .... Call Trace: [<ffffffffa01c3278>] fscache_invalidate_writes+0x38/0xd0 [fscache] [<ffffffff810096f0>] ? __switch_to+0xd0/0x320 [<ffffffff8105e759>] ? find_busiest_queue+0x69/0x150 [<ffffffff8110ddd4>] ? slow_work_enqueue+0x104/0x180 [<ffffffffa01c1303>] fscache_object_slow_work_execute+0x5e3/0x9d0 [fscache] [<ffffffff81096b67>] ? bit_waitqueue+0x17/0xd0 [<ffffffff8110e233>] slow_work_execute+0x233/0x310 [<ffffffff8110e515>] slow_work_thread+0x205/0x360 [<ffffffff81096ca0>] ? autoremove_wake_function+0x0/0x40 [<ffffffff8110e310>] ? slow_work_thread+0x0/0x360 [<ffffffff81096936>] kthread+0x96/0xa0 [<ffffffff8100c0ca>] child_rip+0xa/0x20 [<ffffffff810968a0>] ? kthread+0x0/0xa0 [<ffffffff8100c0c0>] ? child_rip+0x0/0x20 The parameter to fscache_invalidate_writes() was object->cookie which is NULL. Signed-off-by: David Howells <dhowells@redhat.com> Tested-By: Milosz Tanski <milosz@adfin.com> Acked-by: Jeff Layton <jlayton@redhat.com>
2013-05-11 02:50:26 +08:00
FILTER(cookie->def,
COOKIE, NOCOOKIE);
FS-Cache: Fix object state machine to have separate work and wait states Fix object state machine to have separate work and wait states as that makes it easier to envision. There are now three kinds of state: (1) Work state. This is an execution state. No event processing is performed by a work state. The function attached to a work state returns a pointer indicating the next state to which the OSM should transition. Returning NO_TRANSIT repeats the current state, but goes back to the scheduler first. (2) Wait state. This is an event processing state. No execution is performed by a wait state. Wait states are just tables of "if event X occurs, clear it and transition to state Y". The dispatcher returns to the scheduler if none of the events in which the wait state has an interest are currently pending. (3) Out-of-band state. This is a special work state. Transitions to normal states can be overridden when an unexpected event occurs (eg. I/O error). Instead the dispatcher disables and clears the OOB event and transits to the specified work state. This then acts as an ordinary work state, though object->state points to the overridden destination. Returning NO_TRANSIT resumes the overridden transition. In addition, the states have names in their definitions, so there's no need for tables of state names. Further, the EV_REQUEUE event is no longer necessary as that is automatic for work states. Since the states are now separate structs rather than values in an enum, it's not possible to use comparisons other than (non-)equality between them, so use some object->flags to indicate what phase an object is in. The EV_RELEASE, EV_RETIRE and EV_WITHDRAW events have been squished into one (EV_KILL). An object flag now carries the information about retirement. Similarly, the RELEASING, RECYCLING and WITHDRAWING states have been merged into an KILL_OBJECT state and additional states have been added for handling waiting dependent objects (JUMPSTART_DEPS and KILL_DEPENDENTS). A state has also been added for synchronising with parent object initialisation (WAIT_FOR_PARENT) and another for initiating look up (PARENT_READY). Signed-off-by: David Howells <dhowells@redhat.com> Tested-By: Milosz Tanski <milosz@adfin.com> Acked-by: Jeff Layton <jlayton@redhat.com>
2013-05-11 02:50:26 +08:00
FILTER(fscache_object_is_active(obj) ||
obj->n_ops != 0 ||
obj->n_obj_ops != 0 ||
obj->flags ||
!list_empty(&obj->dependents),
BUSY, IDLE);
FILTER(test_bit(FSCACHE_OBJECT_PENDING_WRITE, &obj->flags),
PENDWR, NOPENDWR);
FILTER(atomic_read(&obj->n_reads),
READS, NOREADS);
FILTER(obj->events & obj->event_mask,
EVENTS, NOEVENTS);
FILTER(work_busy(&obj->work), WORK, NOWORK);
}
seq_printf(m,
FS-Cache: Fix object state machine to have separate work and wait states Fix object state machine to have separate work and wait states as that makes it easier to envision. There are now three kinds of state: (1) Work state. This is an execution state. No event processing is performed by a work state. The function attached to a work state returns a pointer indicating the next state to which the OSM should transition. Returning NO_TRANSIT repeats the current state, but goes back to the scheduler first. (2) Wait state. This is an event processing state. No execution is performed by a wait state. Wait states are just tables of "if event X occurs, clear it and transition to state Y". The dispatcher returns to the scheduler if none of the events in which the wait state has an interest are currently pending. (3) Out-of-band state. This is a special work state. Transitions to normal states can be overridden when an unexpected event occurs (eg. I/O error). Instead the dispatcher disables and clears the OOB event and transits to the specified work state. This then acts as an ordinary work state, though object->state points to the overridden destination. Returning NO_TRANSIT resumes the overridden transition. In addition, the states have names in their definitions, so there's no need for tables of state names. Further, the EV_REQUEUE event is no longer necessary as that is automatic for work states. Since the states are now separate structs rather than values in an enum, it's not possible to use comparisons other than (non-)equality between them, so use some object->flags to indicate what phase an object is in. The EV_RELEASE, EV_RETIRE and EV_WITHDRAW events have been squished into one (EV_KILL). An object flag now carries the information about retirement. Similarly, the RELEASING, RECYCLING and WITHDRAWING states have been merged into an KILL_OBJECT state and additional states have been added for handling waiting dependent objects (JUMPSTART_DEPS and KILL_DEPENDENTS). A state has also been added for synchronising with parent object initialisation (WAIT_FOR_PARENT) and another for initiating look up (PARENT_READY). Signed-off-by: David Howells <dhowells@redhat.com> Tested-By: Milosz Tanski <milosz@adfin.com> Acked-by: Jeff Layton <jlayton@redhat.com>
2013-05-11 02:50:26 +08:00
"%8x %8x %s %5u %3u %3u %3u %2u %5u %2lx %2lx %2lx %1x | ",
obj->debug_id,
obj->parent ? obj->parent->debug_id : -1,
FS-Cache: Fix object state machine to have separate work and wait states Fix object state machine to have separate work and wait states as that makes it easier to envision. There are now three kinds of state: (1) Work state. This is an execution state. No event processing is performed by a work state. The function attached to a work state returns a pointer indicating the next state to which the OSM should transition. Returning NO_TRANSIT repeats the current state, but goes back to the scheduler first. (2) Wait state. This is an event processing state. No execution is performed by a wait state. Wait states are just tables of "if event X occurs, clear it and transition to state Y". The dispatcher returns to the scheduler if none of the events in which the wait state has an interest are currently pending. (3) Out-of-band state. This is a special work state. Transitions to normal states can be overridden when an unexpected event occurs (eg. I/O error). Instead the dispatcher disables and clears the OOB event and transits to the specified work state. This then acts as an ordinary work state, though object->state points to the overridden destination. Returning NO_TRANSIT resumes the overridden transition. In addition, the states have names in their definitions, so there's no need for tables of state names. Further, the EV_REQUEUE event is no longer necessary as that is automatic for work states. Since the states are now separate structs rather than values in an enum, it's not possible to use comparisons other than (non-)equality between them, so use some object->flags to indicate what phase an object is in. The EV_RELEASE, EV_RETIRE and EV_WITHDRAW events have been squished into one (EV_KILL). An object flag now carries the information about retirement. Similarly, the RELEASING, RECYCLING and WITHDRAWING states have been merged into an KILL_OBJECT state and additional states have been added for handling waiting dependent objects (JUMPSTART_DEPS and KILL_DEPENDENTS). A state has also been added for synchronising with parent object initialisation (WAIT_FOR_PARENT) and another for initiating look up (PARENT_READY). Signed-off-by: David Howells <dhowells@redhat.com> Tested-By: Milosz Tanski <milosz@adfin.com> Acked-by: Jeff Layton <jlayton@redhat.com>
2013-05-11 02:50:26 +08:00
obj->state->short_name,
obj->n_children,
obj->n_ops,
obj->n_obj_ops,
obj->n_in_progress,
obj->n_exclusive,
atomic_read(&obj->n_reads),
obj->event_mask,
obj->events,
obj->flags,
work_busy(&obj->work));
FS-Cache: Simplify cookie retention for fscache_objects, fixing oops Simplify the way fscache cache objects retain their cookie. The way I implemented the cookie storage handling made synchronisation a pain (ie. the object state machine can't rely on the cookie actually still being there). Instead of the the object being detached from the cookie and the cookie being freed in __fscache_relinquish_cookie(), we defer both operations: (*) The detachment of the object from the list in the cookie now takes place in fscache_drop_object() and is thus governed by the object state machine (fscache_detach_from_cookie() has been removed). (*) The release of the cookie is now in fscache_object_destroy() - which is called by the cache backend just before it frees the object. This means that the fscache_cookie struct is now available to the cache all the way through from ->alloc_object() to ->drop_object() and ->put_object() - meaning that it's no longer necessary to take object->lock to guarantee access. However, __fscache_relinquish_cookie() doesn't wait for the object to go all the way through to destruction before letting the netfs proceed. That would massively slow down the netfs. Since __fscache_relinquish_cookie() leaves the cookie around, in must therefore break all attachments to the netfs - which includes ->def, ->netfs_data and any outstanding page read/writes. To handle this, struct fscache_cookie now has an n_active counter: (1) This starts off initialised to 1. (2) Any time the cache needs to get at the netfs data, it calls fscache_use_cookie() to increment it - if it is not zero. If it was zero, then access is not permitted. (3) When the cache has finished with the data, it calls fscache_unuse_cookie() to decrement it. This does a wake-up on it if it reaches 0. (4) __fscache_relinquish_cookie() decrements n_active and then waits for it to reach 0. The initialisation to 1 in step (1) ensures that we only get wake ups when we're trying to get rid of the cookie. This leaves __fscache_relinquish_cookie() a lot simpler. *** This fixes a problem in the current code whereby if fscache_invalidate() is followed sufficiently quickly by fscache_relinquish_cookie() then it is possible for __fscache_relinquish_cookie() to have detached the cookie from the object and cleared the pointer before a thread is dispatched to process the invalidation state in the object state machine. Since the pending write clearance was deferred to the invalidation state to make it asynchronous, we need to either wait in relinquishment for the stores tree to be cleared in the invalidation state or we need to handle the clearance in relinquishment. Further, if the relinquishment code does clear the tree, then the invalidation state need to make the clearance contingent on still having the cookie to hand (since that's where the tree is rooted) and we have to prevent the cookie from disappearing for the duration. This can lead to an oops like the following: BUG: unable to handle kernel NULL pointer dereference at 000000000000000c ... RIP: 0010:[<ffffffff8151023e>] _spin_lock+0xe/0x30 ... CR2: 000000000000000c ... ... Process kslowd002 (...) .... Call Trace: [<ffffffffa01c3278>] fscache_invalidate_writes+0x38/0xd0 [fscache] [<ffffffff810096f0>] ? __switch_to+0xd0/0x320 [<ffffffff8105e759>] ? find_busiest_queue+0x69/0x150 [<ffffffff8110ddd4>] ? slow_work_enqueue+0x104/0x180 [<ffffffffa01c1303>] fscache_object_slow_work_execute+0x5e3/0x9d0 [fscache] [<ffffffff81096b67>] ? bit_waitqueue+0x17/0xd0 [<ffffffff8110e233>] slow_work_execute+0x233/0x310 [<ffffffff8110e515>] slow_work_thread+0x205/0x360 [<ffffffff81096ca0>] ? autoremove_wake_function+0x0/0x40 [<ffffffff8110e310>] ? slow_work_thread+0x0/0x360 [<ffffffff81096936>] kthread+0x96/0xa0 [<ffffffff8100c0ca>] child_rip+0xa/0x20 [<ffffffff810968a0>] ? kthread+0x0/0xa0 [<ffffffff8100c0c0>] ? child_rip+0x0/0x20 The parameter to fscache_invalidate_writes() was object->cookie which is NULL. Signed-off-by: David Howells <dhowells@redhat.com> Tested-By: Milosz Tanski <milosz@adfin.com> Acked-by: Jeff Layton <jlayton@redhat.com>
2013-05-11 02:50:26 +08:00
if (fscache_use_cookie(obj)) {
uint16_t keylen = 0, auxlen = 0;
FS-Cache: Simplify cookie retention for fscache_objects, fixing oops Simplify the way fscache cache objects retain their cookie. The way I implemented the cookie storage handling made synchronisation a pain (ie. the object state machine can't rely on the cookie actually still being there). Instead of the the object being detached from the cookie and the cookie being freed in __fscache_relinquish_cookie(), we defer both operations: (*) The detachment of the object from the list in the cookie now takes place in fscache_drop_object() and is thus governed by the object state machine (fscache_detach_from_cookie() has been removed). (*) The release of the cookie is now in fscache_object_destroy() - which is called by the cache backend just before it frees the object. This means that the fscache_cookie struct is now available to the cache all the way through from ->alloc_object() to ->drop_object() and ->put_object() - meaning that it's no longer necessary to take object->lock to guarantee access. However, __fscache_relinquish_cookie() doesn't wait for the object to go all the way through to destruction before letting the netfs proceed. That would massively slow down the netfs. Since __fscache_relinquish_cookie() leaves the cookie around, in must therefore break all attachments to the netfs - which includes ->def, ->netfs_data and any outstanding page read/writes. To handle this, struct fscache_cookie now has an n_active counter: (1) This starts off initialised to 1. (2) Any time the cache needs to get at the netfs data, it calls fscache_use_cookie() to increment it - if it is not zero. If it was zero, then access is not permitted. (3) When the cache has finished with the data, it calls fscache_unuse_cookie() to decrement it. This does a wake-up on it if it reaches 0. (4) __fscache_relinquish_cookie() decrements n_active and then waits for it to reach 0. The initialisation to 1 in step (1) ensures that we only get wake ups when we're trying to get rid of the cookie. This leaves __fscache_relinquish_cookie() a lot simpler. *** This fixes a problem in the current code whereby if fscache_invalidate() is followed sufficiently quickly by fscache_relinquish_cookie() then it is possible for __fscache_relinquish_cookie() to have detached the cookie from the object and cleared the pointer before a thread is dispatched to process the invalidation state in the object state machine. Since the pending write clearance was deferred to the invalidation state to make it asynchronous, we need to either wait in relinquishment for the stores tree to be cleared in the invalidation state or we need to handle the clearance in relinquishment. Further, if the relinquishment code does clear the tree, then the invalidation state need to make the clearance contingent on still having the cookie to hand (since that's where the tree is rooted) and we have to prevent the cookie from disappearing for the duration. This can lead to an oops like the following: BUG: unable to handle kernel NULL pointer dereference at 000000000000000c ... RIP: 0010:[<ffffffff8151023e>] _spin_lock+0xe/0x30 ... CR2: 000000000000000c ... ... Process kslowd002 (...) .... Call Trace: [<ffffffffa01c3278>] fscache_invalidate_writes+0x38/0xd0 [fscache] [<ffffffff810096f0>] ? __switch_to+0xd0/0x320 [<ffffffff8105e759>] ? find_busiest_queue+0x69/0x150 [<ffffffff8110ddd4>] ? slow_work_enqueue+0x104/0x180 [<ffffffffa01c1303>] fscache_object_slow_work_execute+0x5e3/0x9d0 [fscache] [<ffffffff81096b67>] ? bit_waitqueue+0x17/0xd0 [<ffffffff8110e233>] slow_work_execute+0x233/0x310 [<ffffffff8110e515>] slow_work_thread+0x205/0x360 [<ffffffff81096ca0>] ? autoremove_wake_function+0x0/0x40 [<ffffffff8110e310>] ? slow_work_thread+0x0/0x360 [<ffffffff81096936>] kthread+0x96/0xa0 [<ffffffff8100c0ca>] child_rip+0xa/0x20 [<ffffffff810968a0>] ? kthread+0x0/0xa0 [<ffffffff8100c0c0>] ? child_rip+0x0/0x20 The parameter to fscache_invalidate_writes() was object->cookie which is NULL. Signed-off-by: David Howells <dhowells@redhat.com> Tested-By: Milosz Tanski <milosz@adfin.com> Acked-by: Jeff Layton <jlayton@redhat.com>
2013-05-11 02:50:26 +08:00
switch (cookie->def->type) {
case 0:
type = "IX";
break;
case 1:
type = "DT";
break;
default:
sprintf(_type, "%02u", cookie->def->type);
type = _type;
break;
}
FS-Cache: Simplify cookie retention for fscache_objects, fixing oops Simplify the way fscache cache objects retain their cookie. The way I implemented the cookie storage handling made synchronisation a pain (ie. the object state machine can't rely on the cookie actually still being there). Instead of the the object being detached from the cookie and the cookie being freed in __fscache_relinquish_cookie(), we defer both operations: (*) The detachment of the object from the list in the cookie now takes place in fscache_drop_object() and is thus governed by the object state machine (fscache_detach_from_cookie() has been removed). (*) The release of the cookie is now in fscache_object_destroy() - which is called by the cache backend just before it frees the object. This means that the fscache_cookie struct is now available to the cache all the way through from ->alloc_object() to ->drop_object() and ->put_object() - meaning that it's no longer necessary to take object->lock to guarantee access. However, __fscache_relinquish_cookie() doesn't wait for the object to go all the way through to destruction before letting the netfs proceed. That would massively slow down the netfs. Since __fscache_relinquish_cookie() leaves the cookie around, in must therefore break all attachments to the netfs - which includes ->def, ->netfs_data and any outstanding page read/writes. To handle this, struct fscache_cookie now has an n_active counter: (1) This starts off initialised to 1. (2) Any time the cache needs to get at the netfs data, it calls fscache_use_cookie() to increment it - if it is not zero. If it was zero, then access is not permitted. (3) When the cache has finished with the data, it calls fscache_unuse_cookie() to decrement it. This does a wake-up on it if it reaches 0. (4) __fscache_relinquish_cookie() decrements n_active and then waits for it to reach 0. The initialisation to 1 in step (1) ensures that we only get wake ups when we're trying to get rid of the cookie. This leaves __fscache_relinquish_cookie() a lot simpler. *** This fixes a problem in the current code whereby if fscache_invalidate() is followed sufficiently quickly by fscache_relinquish_cookie() then it is possible for __fscache_relinquish_cookie() to have detached the cookie from the object and cleared the pointer before a thread is dispatched to process the invalidation state in the object state machine. Since the pending write clearance was deferred to the invalidation state to make it asynchronous, we need to either wait in relinquishment for the stores tree to be cleared in the invalidation state or we need to handle the clearance in relinquishment. Further, if the relinquishment code does clear the tree, then the invalidation state need to make the clearance contingent on still having the cookie to hand (since that's where the tree is rooted) and we have to prevent the cookie from disappearing for the duration. This can lead to an oops like the following: BUG: unable to handle kernel NULL pointer dereference at 000000000000000c ... RIP: 0010:[<ffffffff8151023e>] _spin_lock+0xe/0x30 ... CR2: 000000000000000c ... ... Process kslowd002 (...) .... Call Trace: [<ffffffffa01c3278>] fscache_invalidate_writes+0x38/0xd0 [fscache] [<ffffffff810096f0>] ? __switch_to+0xd0/0x320 [<ffffffff8105e759>] ? find_busiest_queue+0x69/0x150 [<ffffffff8110ddd4>] ? slow_work_enqueue+0x104/0x180 [<ffffffffa01c1303>] fscache_object_slow_work_execute+0x5e3/0x9d0 [fscache] [<ffffffff81096b67>] ? bit_waitqueue+0x17/0xd0 [<ffffffff8110e233>] slow_work_execute+0x233/0x310 [<ffffffff8110e515>] slow_work_thread+0x205/0x360 [<ffffffff81096ca0>] ? autoremove_wake_function+0x0/0x40 [<ffffffff8110e310>] ? slow_work_thread+0x0/0x360 [<ffffffff81096936>] kthread+0x96/0xa0 [<ffffffff8100c0ca>] child_rip+0xa/0x20 [<ffffffff810968a0>] ? kthread+0x0/0xa0 [<ffffffff8100c0c0>] ? child_rip+0x0/0x20 The parameter to fscache_invalidate_writes() was object->cookie which is NULL. Signed-off-by: David Howells <dhowells@redhat.com> Tested-By: Milosz Tanski <milosz@adfin.com> Acked-by: Jeff Layton <jlayton@redhat.com>
2013-05-11 02:50:26 +08:00
seq_printf(m, "%-16s %s %2lx %16p",
cookie->def->name,
type,
cookie->flags,
cookie->netfs_data);
if (cookie->def->get_key &&
config & FSCACHE_OBJLIST_CONFIG_KEY)
keylen = cookie->def->get_key(cookie->netfs_data,
buf, 400);
if (cookie->def->get_aux &&
config & FSCACHE_OBJLIST_CONFIG_AUX)
auxlen = cookie->def->get_aux(cookie->netfs_data,
buf + keylen, 512 - keylen);
fscache_unuse_cookie(obj);
if (keylen > 0 || auxlen > 0) {
seq_puts(m, " ");
for (p = buf; keylen > 0; keylen--)
seq_printf(m, "%02x", *p++);
if (auxlen > 0) {
if (config & FSCACHE_OBJLIST_CONFIG_KEY)
seq_puts(m, ", ");
for (; auxlen > 0; auxlen--)
seq_printf(m, "%02x", *p++);
}
}
seq_puts(m, "\n");
FS-Cache: Simplify cookie retention for fscache_objects, fixing oops Simplify the way fscache cache objects retain their cookie. The way I implemented the cookie storage handling made synchronisation a pain (ie. the object state machine can't rely on the cookie actually still being there). Instead of the the object being detached from the cookie and the cookie being freed in __fscache_relinquish_cookie(), we defer both operations: (*) The detachment of the object from the list in the cookie now takes place in fscache_drop_object() and is thus governed by the object state machine (fscache_detach_from_cookie() has been removed). (*) The release of the cookie is now in fscache_object_destroy() - which is called by the cache backend just before it frees the object. This means that the fscache_cookie struct is now available to the cache all the way through from ->alloc_object() to ->drop_object() and ->put_object() - meaning that it's no longer necessary to take object->lock to guarantee access. However, __fscache_relinquish_cookie() doesn't wait for the object to go all the way through to destruction before letting the netfs proceed. That would massively slow down the netfs. Since __fscache_relinquish_cookie() leaves the cookie around, in must therefore break all attachments to the netfs - which includes ->def, ->netfs_data and any outstanding page read/writes. To handle this, struct fscache_cookie now has an n_active counter: (1) This starts off initialised to 1. (2) Any time the cache needs to get at the netfs data, it calls fscache_use_cookie() to increment it - if it is not zero. If it was zero, then access is not permitted. (3) When the cache has finished with the data, it calls fscache_unuse_cookie() to decrement it. This does a wake-up on it if it reaches 0. (4) __fscache_relinquish_cookie() decrements n_active and then waits for it to reach 0. The initialisation to 1 in step (1) ensures that we only get wake ups when we're trying to get rid of the cookie. This leaves __fscache_relinquish_cookie() a lot simpler. *** This fixes a problem in the current code whereby if fscache_invalidate() is followed sufficiently quickly by fscache_relinquish_cookie() then it is possible for __fscache_relinquish_cookie() to have detached the cookie from the object and cleared the pointer before a thread is dispatched to process the invalidation state in the object state machine. Since the pending write clearance was deferred to the invalidation state to make it asynchronous, we need to either wait in relinquishment for the stores tree to be cleared in the invalidation state or we need to handle the clearance in relinquishment. Further, if the relinquishment code does clear the tree, then the invalidation state need to make the clearance contingent on still having the cookie to hand (since that's where the tree is rooted) and we have to prevent the cookie from disappearing for the duration. This can lead to an oops like the following: BUG: unable to handle kernel NULL pointer dereference at 000000000000000c ... RIP: 0010:[<ffffffff8151023e>] _spin_lock+0xe/0x30 ... CR2: 000000000000000c ... ... Process kslowd002 (...) .... Call Trace: [<ffffffffa01c3278>] fscache_invalidate_writes+0x38/0xd0 [fscache] [<ffffffff810096f0>] ? __switch_to+0xd0/0x320 [<ffffffff8105e759>] ? find_busiest_queue+0x69/0x150 [<ffffffff8110ddd4>] ? slow_work_enqueue+0x104/0x180 [<ffffffffa01c1303>] fscache_object_slow_work_execute+0x5e3/0x9d0 [fscache] [<ffffffff81096b67>] ? bit_waitqueue+0x17/0xd0 [<ffffffff8110e233>] slow_work_execute+0x233/0x310 [<ffffffff8110e515>] slow_work_thread+0x205/0x360 [<ffffffff81096ca0>] ? autoremove_wake_function+0x0/0x40 [<ffffffff8110e310>] ? slow_work_thread+0x0/0x360 [<ffffffff81096936>] kthread+0x96/0xa0 [<ffffffff8100c0ca>] child_rip+0xa/0x20 [<ffffffff810968a0>] ? kthread+0x0/0xa0 [<ffffffff8100c0c0>] ? child_rip+0x0/0x20 The parameter to fscache_invalidate_writes() was object->cookie which is NULL. Signed-off-by: David Howells <dhowells@redhat.com> Tested-By: Milosz Tanski <milosz@adfin.com> Acked-by: Jeff Layton <jlayton@redhat.com>
2013-05-11 02:50:26 +08:00
} else {
seq_puts(m, "<no_netfs>\n");
FS-Cache: Simplify cookie retention for fscache_objects, fixing oops Simplify the way fscache cache objects retain their cookie. The way I implemented the cookie storage handling made synchronisation a pain (ie. the object state machine can't rely on the cookie actually still being there). Instead of the the object being detached from the cookie and the cookie being freed in __fscache_relinquish_cookie(), we defer both operations: (*) The detachment of the object from the list in the cookie now takes place in fscache_drop_object() and is thus governed by the object state machine (fscache_detach_from_cookie() has been removed). (*) The release of the cookie is now in fscache_object_destroy() - which is called by the cache backend just before it frees the object. This means that the fscache_cookie struct is now available to the cache all the way through from ->alloc_object() to ->drop_object() and ->put_object() - meaning that it's no longer necessary to take object->lock to guarantee access. However, __fscache_relinquish_cookie() doesn't wait for the object to go all the way through to destruction before letting the netfs proceed. That would massively slow down the netfs. Since __fscache_relinquish_cookie() leaves the cookie around, in must therefore break all attachments to the netfs - which includes ->def, ->netfs_data and any outstanding page read/writes. To handle this, struct fscache_cookie now has an n_active counter: (1) This starts off initialised to 1. (2) Any time the cache needs to get at the netfs data, it calls fscache_use_cookie() to increment it - if it is not zero. If it was zero, then access is not permitted. (3) When the cache has finished with the data, it calls fscache_unuse_cookie() to decrement it. This does a wake-up on it if it reaches 0. (4) __fscache_relinquish_cookie() decrements n_active and then waits for it to reach 0. The initialisation to 1 in step (1) ensures that we only get wake ups when we're trying to get rid of the cookie. This leaves __fscache_relinquish_cookie() a lot simpler. *** This fixes a problem in the current code whereby if fscache_invalidate() is followed sufficiently quickly by fscache_relinquish_cookie() then it is possible for __fscache_relinquish_cookie() to have detached the cookie from the object and cleared the pointer before a thread is dispatched to process the invalidation state in the object state machine. Since the pending write clearance was deferred to the invalidation state to make it asynchronous, we need to either wait in relinquishment for the stores tree to be cleared in the invalidation state or we need to handle the clearance in relinquishment. Further, if the relinquishment code does clear the tree, then the invalidation state need to make the clearance contingent on still having the cookie to hand (since that's where the tree is rooted) and we have to prevent the cookie from disappearing for the duration. This can lead to an oops like the following: BUG: unable to handle kernel NULL pointer dereference at 000000000000000c ... RIP: 0010:[<ffffffff8151023e>] _spin_lock+0xe/0x30 ... CR2: 000000000000000c ... ... Process kslowd002 (...) .... Call Trace: [<ffffffffa01c3278>] fscache_invalidate_writes+0x38/0xd0 [fscache] [<ffffffff810096f0>] ? __switch_to+0xd0/0x320 [<ffffffff8105e759>] ? find_busiest_queue+0x69/0x150 [<ffffffff8110ddd4>] ? slow_work_enqueue+0x104/0x180 [<ffffffffa01c1303>] fscache_object_slow_work_execute+0x5e3/0x9d0 [fscache] [<ffffffff81096b67>] ? bit_waitqueue+0x17/0xd0 [<ffffffff8110e233>] slow_work_execute+0x233/0x310 [<ffffffff8110e515>] slow_work_thread+0x205/0x360 [<ffffffff81096ca0>] ? autoremove_wake_function+0x0/0x40 [<ffffffff8110e310>] ? slow_work_thread+0x0/0x360 [<ffffffff81096936>] kthread+0x96/0xa0 [<ffffffff8100c0ca>] child_rip+0xa/0x20 [<ffffffff810968a0>] ? kthread+0x0/0xa0 [<ffffffff8100c0c0>] ? child_rip+0x0/0x20 The parameter to fscache_invalidate_writes() was object->cookie which is NULL. Signed-off-by: David Howells <dhowells@redhat.com> Tested-By: Milosz Tanski <milosz@adfin.com> Acked-by: Jeff Layton <jlayton@redhat.com>
2013-05-11 02:50:26 +08:00
}
return 0;
}
static const struct seq_operations fscache_objlist_ops = {
.start = fscache_objlist_start,
.stop = fscache_objlist_stop,
.next = fscache_objlist_next,
.show = fscache_objlist_show,
};
/*
* get the configuration for filtering the list
*/
static void fscache_objlist_config(struct fscache_objlist_data *data)
{
#ifdef CONFIG_KEYS
const struct user_key_payload *confkey;
unsigned long config;
struct key *key;
const char *buf;
int len;
key = request_key(&key_type_user, "fscache:objlist", NULL);
if (IS_ERR(key))
goto no_config;
config = 0;
rcu_read_lock();
KEYS: Differentiate uses of rcu_dereference_key() and user_key_payload() rcu_dereference_key() and user_key_payload() are currently being used in two different, incompatible ways: (1) As a wrapper to rcu_dereference() - when only the RCU read lock used to protect the key. (2) As a wrapper to rcu_dereference_protected() - when the key semaphor is used to protect the key and the may be being modified. Fix this by splitting both of the key wrappers to produce: (1) RCU accessors for keys when caller has the key semaphore locked: dereference_key_locked() user_key_payload_locked() (2) RCU accessors for keys when caller holds the RCU read lock: dereference_key_rcu() user_key_payload_rcu() This should fix following warning in the NFS idmapper =============================== [ INFO: suspicious RCU usage. ] 4.10.0 #1 Tainted: G W ------------------------------- ./include/keys/user-type.h:53 suspicious rcu_dereference_protected() usage! other info that might help us debug this: rcu_scheduler_active = 2, debug_locks = 0 1 lock held by mount.nfs/5987: #0: (rcu_read_lock){......}, at: [<d000000002527abc>] nfs_idmap_get_key+0x15c/0x420 [nfsv4] stack backtrace: CPU: 1 PID: 5987 Comm: mount.nfs Tainted: G W 4.10.0 #1 Call Trace: dump_stack+0xe8/0x154 (unreliable) lockdep_rcu_suspicious+0x140/0x190 nfs_idmap_get_key+0x380/0x420 [nfsv4] nfs_map_name_to_uid+0x2a0/0x3b0 [nfsv4] decode_getfattr_attrs+0xfac/0x16b0 [nfsv4] decode_getfattr_generic.constprop.106+0xbc/0x150 [nfsv4] nfs4_xdr_dec_lookup_root+0xac/0xb0 [nfsv4] rpcauth_unwrap_resp+0xe8/0x140 [sunrpc] call_decode+0x29c/0x910 [sunrpc] __rpc_execute+0x140/0x8f0 [sunrpc] rpc_run_task+0x170/0x200 [sunrpc] nfs4_call_sync_sequence+0x68/0xa0 [nfsv4] _nfs4_lookup_root.isra.44+0xd0/0xf0 [nfsv4] nfs4_lookup_root+0xe0/0x350 [nfsv4] nfs4_lookup_root_sec+0x70/0xa0 [nfsv4] nfs4_find_root_sec+0xc4/0x100 [nfsv4] nfs4_proc_get_rootfh+0x5c/0xf0 [nfsv4] nfs4_get_rootfh+0x6c/0x190 [nfsv4] nfs4_server_common_setup+0xc4/0x260 [nfsv4] nfs4_create_server+0x278/0x3c0 [nfsv4] nfs4_remote_mount+0x50/0xb0 [nfsv4] mount_fs+0x74/0x210 vfs_kern_mount+0x78/0x220 nfs_do_root_mount+0xb0/0x140 [nfsv4] nfs4_try_mount+0x60/0x100 [nfsv4] nfs_fs_mount+0x5ec/0xda0 [nfs] mount_fs+0x74/0x210 vfs_kern_mount+0x78/0x220 do_mount+0x254/0xf70 SyS_mount+0x94/0x100 system_call+0x38/0xe0 Reported-by: Jan Stancek <jstancek@redhat.com> Signed-off-by: David Howells <dhowells@redhat.com> Tested-by: Jan Stancek <jstancek@redhat.com> Signed-off-by: James Morris <james.l.morris@oracle.com>
2017-03-01 23:11:23 +08:00
confkey = user_key_payload_rcu(key);
buf = confkey->data;
for (len = confkey->datalen - 1; len >= 0; len--) {
switch (buf[len]) {
case 'K': config |= FSCACHE_OBJLIST_CONFIG_KEY; break;
case 'A': config |= FSCACHE_OBJLIST_CONFIG_AUX; break;
case 'C': config |= FSCACHE_OBJLIST_CONFIG_COOKIE; break;
case 'c': config |= FSCACHE_OBJLIST_CONFIG_NOCOOKIE; break;
case 'B': config |= FSCACHE_OBJLIST_CONFIG_BUSY; break;
case 'b': config |= FSCACHE_OBJLIST_CONFIG_IDLE; break;
case 'W': config |= FSCACHE_OBJLIST_CONFIG_PENDWR; break;
case 'w': config |= FSCACHE_OBJLIST_CONFIG_NOPENDWR; break;
case 'R': config |= FSCACHE_OBJLIST_CONFIG_READS; break;
case 'r': config |= FSCACHE_OBJLIST_CONFIG_NOREADS; break;
case 'S': config |= FSCACHE_OBJLIST_CONFIG_WORK; break;
case 's': config |= FSCACHE_OBJLIST_CONFIG_NOWORK; break;
}
}
rcu_read_unlock();
key_put(key);
if (!(config & (FSCACHE_OBJLIST_CONFIG_COOKIE | FSCACHE_OBJLIST_CONFIG_NOCOOKIE)))
config |= FSCACHE_OBJLIST_CONFIG_COOKIE | FSCACHE_OBJLIST_CONFIG_NOCOOKIE;
if (!(config & (FSCACHE_OBJLIST_CONFIG_BUSY | FSCACHE_OBJLIST_CONFIG_IDLE)))
config |= FSCACHE_OBJLIST_CONFIG_BUSY | FSCACHE_OBJLIST_CONFIG_IDLE;
if (!(config & (FSCACHE_OBJLIST_CONFIG_PENDWR | FSCACHE_OBJLIST_CONFIG_NOPENDWR)))
config |= FSCACHE_OBJLIST_CONFIG_PENDWR | FSCACHE_OBJLIST_CONFIG_NOPENDWR;
if (!(config & (FSCACHE_OBJLIST_CONFIG_READS | FSCACHE_OBJLIST_CONFIG_NOREADS)))
config |= FSCACHE_OBJLIST_CONFIG_READS | FSCACHE_OBJLIST_CONFIG_NOREADS;
if (!(config & (FSCACHE_OBJLIST_CONFIG_EVENTS | FSCACHE_OBJLIST_CONFIG_NOEVENTS)))
config |= FSCACHE_OBJLIST_CONFIG_EVENTS | FSCACHE_OBJLIST_CONFIG_NOEVENTS;
if (!(config & (FSCACHE_OBJLIST_CONFIG_WORK | FSCACHE_OBJLIST_CONFIG_NOWORK)))
config |= FSCACHE_OBJLIST_CONFIG_WORK | FSCACHE_OBJLIST_CONFIG_NOWORK;
data->config = config;
return;
no_config:
#endif
data->config = ULONG_MAX;
}
/*
* open "/proc/fs/fscache/objects" to provide a list of active objects
* - can be configured by a user-defined key added to the caller's keyrings
*/
static int fscache_objlist_open(struct inode *inode, struct file *file)
{
struct fscache_objlist_data *data;
data = __seq_open_private(file, &fscache_objlist_ops, sizeof(*data));
if (!data)
return -ENOMEM;
/* get the configuration key */
fscache_objlist_config(data);
return 0;
}
/*
* clean up on close
*/
static int fscache_objlist_release(struct inode *inode, struct file *file)
{
struct seq_file *m = file->private_data;
kfree(m->private);
m->private = NULL;
return seq_release(inode, file);
}
const struct file_operations fscache_objlist_fops = {
.open = fscache_objlist_open,
.read = seq_read,
.llseek = seq_lseek,
.release = fscache_objlist_release,
};