108d3943c0
367 Commits
| Author | SHA1 | Message | Date | |
|---|---|---|---|---|
Frederic Weisbecker
|
108d3943c0 |
reiserfs: Relax the lock before truncating pages
While truncating a file, reiserfs_setattr() calls inode_setattr() that will truncate the mapping for the given inode, but for that it needs the pages locks. In order to release these, the owners need the reiserfs lock to complete their jobs. But they can't, as we don't release it before calling inode_setattr(). We need to do that to fix the following softlockups: INFO: task flush-8:0:2149 blocked for more than 120 seconds. "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message. flush-8:0 D f51af998 0 2149 2 0x00000000 f51af9ac 00000092 00000002 f51af998 c2803304 00000000 c1894ad0 010f3000 f51af9cc c1462604 c189ef80 f51af974 c1710304 f715b450 f715b5ec c2807c40 00000000 0005bb00 c2803320 c102c55b c1710304 c2807c50 c2803304 00000246 Call Trace: [<c1462604>] ? schedule+0x434/0xb20 [<c102c55b>] ? resched_task+0x4b/0x70 [<c106fa22>] ? mark_held_locks+0x62/0x80 [<c146414d>] ? mutex_lock_nested+0x1fd/0x350 [<c14640b9>] mutex_lock_nested+0x169/0x350 [<c1178cde>] ? reiserfs_write_lock+0x2e/0x40 [<c1178cde>] reiserfs_write_lock+0x2e/0x40 [<c11719a2>] do_journal_end+0xc2/0xe70 [<c1172912>] journal_end+0xb2/0x120 [<c11686b3>] ? pathrelse+0x33/0xb0 [<c11729e4>] reiserfs_end_persistent_transaction+0x64/0x70 [<c1153caa>] reiserfs_get_block+0x12ba/0x15f0 [<c106fa22>] ? mark_held_locks+0x62/0x80 [<c1154b24>] reiserfs_writepage+0xa74/0xe80 [<c1465a27>] ? _raw_spin_unlock_irq+0x27/0x50 [<c11f3d25>] ? radix_tree_gang_lookup_tag_slot+0x95/0xc0 [<c10b5377>] ? find_get_pages_tag+0x127/0x1a0 [<c106fa22>] ? mark_held_locks+0x62/0x80 [<c106fcd4>] ? trace_hardirqs_on_caller+0x124/0x170 [<c10bc1e0>] __writepage+0x10/0x40 [<c10bc9ab>] write_cache_pages+0x16b/0x320 [<c10bc1d0>] ? __writepage+0x0/0x40 [<c10bcb88>] generic_writepages+0x28/0x40 [<c10bcbd5>] do_writepages+0x35/0x40 [<c11059f7>] writeback_single_inode+0xc7/0x330 [<c11067b2>] writeback_inodes_wb+0x2c2/0x490 [<c1106a86>] wb_writeback+0x106/0x1b0 [<c1106cf6>] wb_do_writeback+0x106/0x1e0 [<c1106c18>] ? wb_do_writeback+0x28/0x1e0 [<c1106e0a>] bdi_writeback_task+0x3a/0xb0 [<c10cbb13>] bdi_start_fn+0x63/0xc0 [<c10cbab0>] ? bdi_start_fn+0x0/0xc0 [<c105d1f4>] kthread+0x74/0x80 [<c105d180>] ? kthread+0x0/0x80 [<c100327a>] kernel_thread_helper+0x6/0x10 3 locks held by flush-8:0/2149: #0: (&type->s_umount_key#30){+++++.}, at: [<c110676f>] writeback_inodes_wb+0x27f/0x490 #1: (&journal->j_mutex){+.+...}, at: [<c117199a>] do_journal_end+0xba/0xe70 #2: (&REISERFS_SB(s)->lock){+.+.+.}, at: [<c1178cde>] reiserfs_write_lock+0x2e/0x40 INFO: task fstest:3813 blocked for more than 120 seconds. "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message. fstest D 00000002 0 3813 3812 0x00000000 f5103c94 00000082 f5103c40 00000002 f5ad5450 00000007 f5103c28 011f3000 00000006 f5ad5450 c10bb005 00000480 c1710304 f5ad5450 f5ad55ec c2907c40 00000001 f5ad5450 f5103c74 00000046 00000002 f5ad5450 00000007 f5103c6c Call Trace: [<c10bb005>] ? free_hot_cold_page+0x1d5/0x280 [<c1462d64>] io_schedule+0x74/0xc0 [<c10b5a45>] sync_page+0x35/0x60 [<c146325a>] __wait_on_bit_lock+0x4a/0x90 [<c10b5a10>] ? sync_page+0x0/0x60 [<c10b59e5>] __lock_page+0x85/0x90 [<c105d660>] ? wake_bit_function+0x0/0x60 [<c10bf654>] truncate_inode_pages_range+0x1e4/0x2d0 [<c10bf75f>] truncate_inode_pages+0x1f/0x30 [<c10bf7cf>] truncate_pagecache+0x5f/0xa0 [<c10bf86a>] vmtruncate+0x5a/0x70 [<c10fdb7d>] inode_setattr+0x5d/0x190 [<c1150117>] reiserfs_setattr+0x1f7/0x2f0 [<c1464569>] ? down_write+0x49/0x70 [<c10fde01>] notify_change+0x151/0x330 [<c10e6f3d>] do_truncate+0x6d/0xa0 [<c10f4ce2>] do_filp_open+0x9a2/0xcf0 [<c1465aec>] ? _raw_spin_unlock+0x2c/0x50 [<c10fec50>] ? alloc_fd+0xe0/0x100 [<c10e602d>] do_sys_open+0x6d/0x130 [<c1002cfb>] ? sysenter_exit+0xf/0x16 [<c10e615e>] sys_open+0x2e/0x40 [<c1002ccc>] sysenter_do_call+0x12/0x32 3 locks held by fstest/3813: #0: (&sb->s_type->i_mutex_key#4){+.+.+.}, at: [<c10e6f33>] do_truncate+0x63/0xa0 #1: (&sb->s_type->i_alloc_sem_key#3){+.+.+.}, at: [<c10fdf07>] notify_change+0x257/0x330 #2: (&REISERFS_SB(s)->lock){+.+.+.}, at: [<c1178c8e>] reiserfs_write_lock_once+0x2e/0x50 Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com> Cc: Christian Kujau <lists@nerdbynature.de> Cc: Alexander Beregalov <a.beregalov@gmail.com> Cc: Chris Mason <chris.mason@oracle.com> Cc: Ingo Molnar <mingo@elte.hu> |
||
|
Frederic Weisbecker
|
5fe1533fda |
reiserfs: Fix recursive lock on lchown
On chown, reiserfs will call reiserfs_setattr() to change the owner of the given inode, but it may also recursively call reiserfs_setattr() to propagate the owner change to the private xattr files for this inode. Hence, the reiserfs lock may be acquired twice which is not wanted as reiserfs_setattr() calls journal_begin() that is going to try to relax the lock in order to safely acquire the journal mutex. Using reiserfs_write_lock_once() from reiserfs_setattr() solves the problem. This fixes the following warning, that precedes a lockdep report. WARNING: at fs/reiserfs/lock.c:95 reiserfs_lock_check_recursive+0x3f/0x50() Hardware name: MS-7418 Unwanted recursive reiserfs lock! Pid: 4189, comm: fsstress Not tainted 2.6.33-rc2-tip-atom+ #195 Call Trace: [<c1178bff>] ? reiserfs_lock_check_recursive+0x3f/0x50 [<c1178bff>] ? reiserfs_lock_check_recursive+0x3f/0x50 [<c103f7ac>] warn_slowpath_common+0x6c/0xc0 [<c1178bff>] ? reiserfs_lock_check_recursive+0x3f/0x50 [<c103f84b>] warn_slowpath_fmt+0x2b/0x30 [<c1178bff>] reiserfs_lock_check_recursive+0x3f/0x50 [<c1172ae3>] do_journal_begin_r+0x83/0x350 [<c1172f2d>] journal_begin+0x7d/0x140 [<c106509a>] ? in_group_p+0x2a/0x30 [<c10fda71>] ? inode_change_ok+0x91/0x140 [<c115007d>] reiserfs_setattr+0x15d/0x2e0 [<c10f9bf3>] ? dput+0xe3/0x140 [<c1465adc>] ? _raw_spin_unlock+0x2c/0x50 [<c117831d>] chown_one_xattr+0xd/0x10 [<c11780a3>] reiserfs_for_each_xattr+0x113/0x2c0 [<c1178310>] ? chown_one_xattr+0x0/0x10 [<c14641e9>] ? mutex_lock_nested+0x2a9/0x350 [<c117826f>] reiserfs_chown_xattrs+0x1f/0x60 [<c106509a>] ? in_group_p+0x2a/0x30 [<c10fda71>] ? inode_change_ok+0x91/0x140 [<c1150046>] reiserfs_setattr+0x126/0x2e0 [<c1177c20>] ? reiserfs_getxattr+0x0/0x90 [<c11b0d57>] ? cap_inode_need_killpriv+0x37/0x50 [<c10fde01>] notify_change+0x151/0x330 [<c10e659f>] chown_common+0x6f/0x90 [<c10e67bd>] sys_lchown+0x6d/0x80 [<c1002ccc>] sysenter_do_call+0x12/0x32 ---[ end trace 7c2b77224c1442fc ]--- Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com> Cc: Christian Kujau <lists@nerdbynature.de> Cc: Alexander Beregalov <a.beregalov@gmail.com> Cc: Chris Mason <chris.mason@oracle.com> Cc: Ingo Molnar <mingo@elte.hu> |
||
|
Frederic Weisbecker
|
f3e22f48f3 |
reiserfs: Fix mistake in down_write() conversion
Fix a mistake in commit
|
||
|
Frederic Weisbecker
|
835d5247d9 |
reiserfs: Safely acquire i_mutex from xattr_rmdir
Relax the reiserfs lock before taking the inode mutex from xattr_rmdir() to avoid the usual reiserfs lock <-> inode mutex bad dependency. Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com> Tested-by: Christian Kujau <lists@nerdbynature.de> Cc: Alexander Beregalov <a.beregalov@gmail.com> Cc: Chris Mason <chris.mason@oracle.com> Cc: Ingo Molnar <mingo@elte.hu> |
||
|
Frederic Weisbecker
|
8b513f56d4 |
reiserfs: Safely acquire i_mutex from reiserfs_for_each_xattr
Relax the reiserfs lock before taking the inode mutex from reiserfs_for_each_xattr() to avoid the usual bad dependencies: ======================================================= [ INFO: possible circular locking dependency detected ] 2.6.32-atom #179 ------------------------------------------------------- rm/3242 is trying to acquire lock: (&sb->s_type->i_mutex_key#4/3){+.+.+.}, at: [<c11428ef>] reiserfs_for_each_xattr+0x23f/0x290 but task is already holding lock: (&REISERFS_SB(s)->lock){+.+.+.}, at: [<c1143389>] reiserfs_write_lock+0x29/0x40 which lock already depends on the new lock. the existing dependency chain (in reverse order) is: -> #1 (&REISERFS_SB(s)->lock){+.+.+.}: [<c105ea7f>] __lock_acquire+0x11ff/0x19e0 [<c105f2c8>] lock_acquire+0x68/0x90 [<c1401aab>] mutex_lock_nested+0x5b/0x340 [<c1143339>] reiserfs_write_lock_once+0x29/0x50 [<c1117022>] reiserfs_lookup+0x62/0x140 [<c10bd85f>] __lookup_hash+0xef/0x110 [<c10bf21d>] lookup_one_len+0x8d/0xc0 [<c1141e3a>] open_xa_dir+0xea/0x1b0 [<c1142720>] reiserfs_for_each_xattr+0x70/0x290 [<c11429ba>] reiserfs_delete_xattrs+0x1a/0x60 [<c111ea2f>] reiserfs_delete_inode+0x9f/0x150 [<c10c9c32>] generic_delete_inode+0xa2/0x170 [<c10c9d4f>] generic_drop_inode+0x4f/0x70 [<c10c8b07>] iput+0x47/0x50 [<c10c0965>] do_unlinkat+0xd5/0x160 [<c10c0b13>] sys_unlinkat+0x23/0x40 [<c1002ec4>] sysenter_do_call+0x12/0x32 -> #0 (&sb->s_type->i_mutex_key#4/3){+.+.+.}: [<c105f176>] __lock_acquire+0x18f6/0x19e0 [<c105f2c8>] lock_acquire+0x68/0x90 [<c1401aab>] mutex_lock_nested+0x5b/0x340 [<c11428ef>] reiserfs_for_each_xattr+0x23f/0x290 [<c11429ba>] reiserfs_delete_xattrs+0x1a/0x60 [<c111ea2f>] reiserfs_delete_inode+0x9f/0x150 [<c10c9c32>] generic_delete_inode+0xa2/0x170 [<c10c9d4f>] generic_drop_inode+0x4f/0x70 [<c10c8b07>] iput+0x47/0x50 [<c10c0965>] do_unlinkat+0xd5/0x160 [<c10c0b13>] sys_unlinkat+0x23/0x40 [<c1002ec4>] sysenter_do_call+0x12/0x32 other info that might help us debug this: 1 lock held by rm/3242: #0: (&REISERFS_SB(s)->lock){+.+.+.}, at: [<c1143389>] reiserfs_write_lock+0x29/0x40 stack backtrace: Pid: 3242, comm: rm Not tainted 2.6.32-atom #179 Call Trace: [<c13ffa13>] ? printk+0x18/0x1a [<c105d33a>] print_circular_bug+0xca/0xd0 [<c105f176>] __lock_acquire+0x18f6/0x19e0 [<c105c932>] ? mark_held_locks+0x62/0x80 [<c105cc3b>] ? trace_hardirqs_on+0xb/0x10 [<c1401098>] ? mutex_unlock+0x8/0x10 [<c105f2c8>] lock_acquire+0x68/0x90 [<c11428ef>] ? reiserfs_for_each_xattr+0x23f/0x290 [<c11428ef>] ? reiserfs_for_each_xattr+0x23f/0x290 [<c1401aab>] mutex_lock_nested+0x5b/0x340 [<c11428ef>] ? reiserfs_for_each_xattr+0x23f/0x290 [<c11428ef>] reiserfs_for_each_xattr+0x23f/0x290 [<c1143180>] ? delete_one_xattr+0x0/0x100 [<c11429ba>] reiserfs_delete_xattrs+0x1a/0x60 [<c1143339>] ? reiserfs_write_lock_once+0x29/0x50 [<c111ea2f>] reiserfs_delete_inode+0x9f/0x150 [<c11b0d4f>] ? _atomic_dec_and_lock+0x4f/0x70 [<c111e990>] ? reiserfs_delete_inode+0x0/0x150 [<c10c9c32>] generic_delete_inode+0xa2/0x170 [<c10c9d4f>] generic_drop_inode+0x4f/0x70 [<c10c8b07>] iput+0x47/0x50 [<c10c0965>] do_unlinkat+0xd5/0x160 [<c1401098>] ? mutex_unlock+0x8/0x10 [<c10c3e0d>] ? vfs_readdir+0x7d/0xb0 [<c10c3af0>] ? filldir64+0x0/0xf0 [<c1002ef3>] ? sysenter_exit+0xf/0x16 [<c105cbe4>] ? trace_hardirqs_on_caller+0x124/0x170 [<c10c0b13>] sys_unlinkat+0x23/0x40 [<c1002ec4>] sysenter_do_call+0x12/0x32 Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com> Tested-by: Christian Kujau <lists@nerdbynature.de> Cc: Alexander Beregalov <a.beregalov@gmail.com> Cc: Chris Mason <chris.mason@oracle.com> Cc: Ingo Molnar <mingo@elte.hu> |
||
|
Frederic Weisbecker
|
4dd859697f |
reiserfs: Fix journal mutex <-> inode mutex lock inversion
We need to relax the reiserfs lock before locking the inode mutex from xattr_unlink(), otherwise we'll face the usual bad dependencies: ======================================================= [ INFO: possible circular locking dependency detected ] 2.6.32-atom #178 ------------------------------------------------------- rm/3202 is trying to acquire lock: (&journal->j_mutex){+.+...}, at: [<c113c234>] do_journal_begin_r+0x94/0x360 but task is already holding lock: (&sb->s_type->i_mutex_key#4/2){+.+...}, at: [<c1142a67>] xattr_unlink+0x57/0xb0 which lock already depends on the new lock. the existing dependency chain (in reverse order) is: -> #2 (&sb->s_type->i_mutex_key#4/2){+.+...}: [<c105ea7f>] __lock_acquire+0x11ff/0x19e0 [<c105f2c8>] lock_acquire+0x68/0x90 [<c1401a7b>] mutex_lock_nested+0x5b/0x340 [<c1142a67>] xattr_unlink+0x57/0xb0 [<c1143179>] delete_one_xattr+0x29/0x100 [<c11427bb>] reiserfs_for_each_xattr+0x10b/0x290 [<c11429ba>] reiserfs_delete_xattrs+0x1a/0x60 [<c111ea2f>] reiserfs_delete_inode+0x9f/0x150 [<c10c9c32>] generic_delete_inode+0xa2/0x170 [<c10c9d4f>] generic_drop_inode+0x4f/0x70 [<c10c8b07>] iput+0x47/0x50 [<c10c0965>] do_unlinkat+0xd5/0x160 [<c10c0b13>] sys_unlinkat+0x23/0x40 [<c1002ec4>] sysenter_do_call+0x12/0x32 -> #1 (&REISERFS_SB(s)->lock){+.+.+.}: [<c105ea7f>] __lock_acquire+0x11ff/0x19e0 [<c105f2c8>] lock_acquire+0x68/0x90 [<c1401a7b>] mutex_lock_nested+0x5b/0x340 [<c1143359>] reiserfs_write_lock+0x29/0x40 [<c113c23c>] do_journal_begin_r+0x9c/0x360 [<c113c680>] journal_begin+0x80/0x130 [<c1127363>] reiserfs_remount+0x223/0x4e0 [<c10b6dd6>] do_remount_sb+0xa6/0x140 [<c10ce6a0>] do_mount+0x560/0x750 [<c10ce914>] sys_mount+0x84/0xb0 [<c1002ec4>] sysenter_do_call+0x12/0x32 -> #0 (&journal->j_mutex){+.+...}: [<c105f176>] __lock_acquire+0x18f6/0x19e0 [<c105f2c8>] lock_acquire+0x68/0x90 [<c1401a7b>] mutex_lock_nested+0x5b/0x340 [<c113c234>] do_journal_begin_r+0x94/0x360 [<c113c680>] journal_begin+0x80/0x130 [<c1116d63>] reiserfs_unlink+0x83/0x2e0 [<c1142a74>] xattr_unlink+0x64/0xb0 [<c1143179>] delete_one_xattr+0x29/0x100 [<c11427bb>] reiserfs_for_each_xattr+0x10b/0x290 [<c11429ba>] reiserfs_delete_xattrs+0x1a/0x60 [<c111ea2f>] reiserfs_delete_inode+0x9f/0x150 [<c10c9c32>] generic_delete_inode+0xa2/0x170 [<c10c9d4f>] generic_drop_inode+0x4f/0x70 [<c10c8b07>] iput+0x47/0x50 [<c10c0965>] do_unlinkat+0xd5/0x160 [<c10c0b13>] sys_unlinkat+0x23/0x40 [<c1002ec4>] sysenter_do_call+0x12/0x32 other info that might help us debug this: 2 locks held by rm/3202: #0: (&sb->s_type->i_mutex_key#4/3){+.+.+.}, at: [<c114274b>] reiserfs_for_each_xattr+0x9b/0x290 #1: (&sb->s_type->i_mutex_key#4/2){+.+...}, at: [<c1142a67>] xattr_unlink+0x57/0xb0 stack backtrace: Pid: 3202, comm: rm Not tainted 2.6.32-atom #178 Call Trace: [<c13ff9e3>] ? printk+0x18/0x1a [<c105d33a>] print_circular_bug+0xca/0xd0 [<c105f176>] __lock_acquire+0x18f6/0x19e0 [<c1142a67>] ? xattr_unlink+0x57/0xb0 [<c105f2c8>] lock_acquire+0x68/0x90 [<c113c234>] ? do_journal_begin_r+0x94/0x360 [<c113c234>] ? do_journal_begin_r+0x94/0x360 [<c1401a7b>] mutex_lock_nested+0x5b/0x340 [<c113c234>] ? do_journal_begin_r+0x94/0x360 [<c113c234>] do_journal_begin_r+0x94/0x360 [<c10411b6>] ? run_timer_softirq+0x1a6/0x220 [<c103cb00>] ? __do_softirq+0x50/0x140 [<c113c680>] journal_begin+0x80/0x130 [<c103cba2>] ? __do_softirq+0xf2/0x140 [<c104f72f>] ? hrtimer_interrupt+0xdf/0x220 [<c1116d63>] reiserfs_unlink+0x83/0x2e0 [<c105c932>] ? mark_held_locks+0x62/0x80 [<c11b8d08>] ? trace_hardirqs_on_thunk+0xc/0x10 [<c1002fd8>] ? restore_all_notrace+0x0/0x18 [<c1142a67>] ? xattr_unlink+0x57/0xb0 [<c1142a74>] xattr_unlink+0x64/0xb0 [<c1143179>] delete_one_xattr+0x29/0x100 [<c11427bb>] reiserfs_for_each_xattr+0x10b/0x290 [<c1143150>] ? delete_one_xattr+0x0/0x100 [<c1401cb9>] ? mutex_lock_nested+0x299/0x340 [<c11429ba>] reiserfs_delete_xattrs+0x1a/0x60 [<c1143309>] ? reiserfs_write_lock_once+0x29/0x50 [<c111ea2f>] reiserfs_delete_inode+0x9f/0x150 [<c11b0d1f>] ? _atomic_dec_and_lock+0x4f/0x70 [<c111e990>] ? reiserfs_delete_inode+0x0/0x150 [<c10c9c32>] generic_delete_inode+0xa2/0x170 [<c10c9d4f>] generic_drop_inode+0x4f/0x70 [<c10c8b07>] iput+0x47/0x50 [<c10c0965>] do_unlinkat+0xd5/0x160 [<c1401068>] ? mutex_unlock+0x8/0x10 [<c10c3e0d>] ? vfs_readdir+0x7d/0xb0 [<c10c3af0>] ? filldir64+0x0/0xf0 [<c1002ef3>] ? sysenter_exit+0xf/0x16 [<c105cbe4>] ? trace_hardirqs_on_caller+0x124/0x170 [<c10c0b13>] sys_unlinkat+0x23/0x40 [<c1002ec4>] sysenter_do_call+0x12/0x32 Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com> Tested-by: Christian Kujau <lists@nerdbynature.de> Cc: Alexander Beregalov <a.beregalov@gmail.com> Cc: Chris Mason <chris.mason@oracle.com> Cc: Ingo Molnar <mingo@elte.hu> |
||
|
Frederic Weisbecker
|
c674905ca7 |
reiserfs: Fix unwanted recursive reiserfs lock in reiserfs_unlink()
reiserfs_unlink() may or may not be called under the reiserfs lock. But it also takes the reiserfs lock and can then acquire it recursively which leads to do_journal_begin_r() that fails to relax the reiserfs lock before grabbing the journal mutex, creating an unexpected lock inversion. We need to ensure reiserfs_unlink() won't get the reiserfs lock recursively using reiserfs_write_lock_once(). This fixes the following warning that precedes a lock inversion report (reiserfs lock <-> journal mutex). ------------[ cut here ]------------ WARNING: at fs/reiserfs/lock.c:95 reiserfs_lock_check_recursive+0x3a/0x50() Hardware name: MS-7418 Unwanted recursive reiserfs lock! Pid: 3208, comm: dbench Not tainted 2.6.32-atom #177 Call Trace: [<c114327a>] ? reiserfs_lock_check_recursive+0x3a/0x50 [<c114327a>] ? reiserfs_lock_check_recursive+0x3a/0x50 [<c10373a7>] warn_slowpath_common+0x67/0xc0 [<c114327a>] ? reiserfs_lock_check_recursive+0x3a/0x50 [<c1037446>] warn_slowpath_fmt+0x26/0x30 [<c114327a>] reiserfs_lock_check_recursive+0x3a/0x50 [<c113c213>] do_journal_begin_r+0x83/0x360 [<c105eb16>] ? __lock_acquire+0x1296/0x19e0 [<c1142a57>] ? xattr_unlink+0x57/0xb0 [<c113c670>] journal_begin+0x80/0x130 [<c1116d5d>] reiserfs_unlink+0x7d/0x2d0 [<c1142a57>] ? xattr_unlink+0x57/0xb0 [<c1142a57>] ? xattr_unlink+0x57/0xb0 [<c1142a57>] ? xattr_unlink+0x57/0xb0 [<c1142a64>] xattr_unlink+0x64/0xb0 [<c1143169>] delete_one_xattr+0x29/0x100 [<c11427ab>] reiserfs_for_each_xattr+0x10b/0x290 [<c1143140>] ? delete_one_xattr+0x0/0x100 [<c1401ca9>] ? mutex_lock_nested+0x299/0x340 [<c11429aa>] reiserfs_delete_xattrs+0x1a/0x60 [<c11432f9>] ? reiserfs_write_lock_once+0x29/0x50 [<c111ea1f>] reiserfs_delete_inode+0x9f/0x150 [<c11b0d0f>] ? _atomic_dec_and_lock+0x4f/0x70 [<c111e980>] ? reiserfs_delete_inode+0x0/0x150 [<c10c9c32>] generic_delete_inode+0xa2/0x170 [<c10c9d4f>] generic_drop_inode+0x4f/0x70 [<c10c8b07>] iput+0x47/0x50 [<c10c0965>] do_unlinkat+0xd5/0x160 [<c10505c6>] ? up_read+0x16/0x30 [<c1022ab7>] ? do_page_fault+0x187/0x330 [<c1002fd8>] ? restore_all_notrace+0x0/0x18 [<c1022930>] ? do_page_fault+0x0/0x330 [<c105cbe4>] ? trace_hardirqs_on_caller+0x124/0x170 [<c10c0a00>] sys_unlink+0x10/0x20 [<c1002ec4>] sysenter_do_call+0x12/0x32 ---[ end trace 2e35d71a6cc69d0c ]--- Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com> Tested-by: Christian Kujau <lists@nerdbynature.de> Cc: Alexander Beregalov <a.beregalov@gmail.com> Cc: Chris Mason <chris.mason@oracle.com> Cc: Ingo Molnar <mingo@elte.hu> |
||
|
Frederic Weisbecker
|
3f14fea6bb |
reiserfs: Relax lock before open xattr dir in reiserfs_xattr_set_handle()
We call xattr_lookup() from reiserfs_xattr_get(). We then hold the reiserfs lock when we grab the i_mutex. But later, we may relax the reiserfs lock, creating dependency inversion between both locks. The lookups and creation jobs ar already protected by the inode mutex, so we can safely relax the reiserfs lock, dropping the unwanted reiserfs lock -> i_mutex dependency, as shown in the following lockdep report: ======================================================= [ INFO: possible circular locking dependency detected ] 2.6.32-atom #173 ------------------------------------------------------- cp/3204 is trying to acquire lock: (&REISERFS_SB(s)->lock){+.+.+.}, at: [<c11432b9>] reiserfs_write_lock_once+0x29/0x50 but task is already holding lock: (&sb->s_type->i_mutex_key#4/3){+.+.+.}, at: [<c1141e18>] open_xa_dir+0xd8/0x1b0 which lock already depends on the new lock. the existing dependency chain (in reverse order) is: -> #1 (&sb->s_type->i_mutex_key#4/3){+.+.+.}: [<c105ea7f>] __lock_acquire+0x11ff/0x19e0 [<c105f2c8>] lock_acquire+0x68/0x90 [<c1401a2b>] mutex_lock_nested+0x5b/0x340 [<c1141d83>] open_xa_dir+0x43/0x1b0 [<c1142722>] reiserfs_for_each_xattr+0x62/0x260 [<c114299a>] reiserfs_delete_xattrs+0x1a/0x60 [<c111ea1f>] reiserfs_delete_inode+0x9f/0x150 [<c10c9c32>] generic_delete_inode+0xa2/0x170 [<c10c9d4f>] generic_drop_inode+0x4f/0x70 [<c10c8b07>] iput+0x47/0x50 [<c10c0965>] do_unlinkat+0xd5/0x160 [<c10c0a00>] sys_unlink+0x10/0x20 [<c1002ec4>] sysenter_do_call+0x12/0x32 -> #0 (&REISERFS_SB(s)->lock){+.+.+.}: [<c105f176>] __lock_acquire+0x18f6/0x19e0 [<c105f2c8>] lock_acquire+0x68/0x90 [<c1401a2b>] mutex_lock_nested+0x5b/0x340 [<c11432b9>] reiserfs_write_lock_once+0x29/0x50 [<c1117012>] reiserfs_lookup+0x62/0x140 [<c10bd85f>] __lookup_hash+0xef/0x110 [<c10bf21d>] lookup_one_len+0x8d/0xc0 [<c1141e2a>] open_xa_dir+0xea/0x1b0 [<c1141fe5>] xattr_lookup+0x15/0x160 [<c1142476>] reiserfs_xattr_get+0x56/0x2a0 [<c1144042>] reiserfs_get_acl+0xa2/0x360 [<c114461a>] reiserfs_cache_default_acl+0x3a/0x160 [<c111789c>] reiserfs_mkdir+0x6c/0x2c0 [<c10bea96>] vfs_mkdir+0xd6/0x180 [<c10c0c10>] sys_mkdirat+0xc0/0xd0 [<c10c0c40>] sys_mkdir+0x20/0x30 [<c1002ec4>] sysenter_do_call+0x12/0x32 other info that might help us debug this: 2 locks held by cp/3204: #0: (&sb->s_type->i_mutex_key#4/1){+.+.+.}, at: [<c10bd8d6>] lookup_create+0x26/0xa0 #1: (&sb->s_type->i_mutex_key#4/3){+.+.+.}, at: [<c1141e18>] open_xa_dir+0xd8/0x1b0 stack backtrace: Pid: 3204, comm: cp Not tainted 2.6.32-atom #173 Call Trace: [<c13ff993>] ? printk+0x18/0x1a [<c105d33a>] print_circular_bug+0xca/0xd0 [<c105f176>] __lock_acquire+0x18f6/0x19e0 [<c105d3aa>] ? check_usage+0x6a/0x460 [<c105f2c8>] lock_acquire+0x68/0x90 [<c11432b9>] ? reiserfs_write_lock_once+0x29/0x50 [<c11432b9>] ? reiserfs_write_lock_once+0x29/0x50 [<c1401a2b>] mutex_lock_nested+0x5b/0x340 [<c11432b9>] ? reiserfs_write_lock_once+0x29/0x50 [<c11432b9>] reiserfs_write_lock_once+0x29/0x50 [<c1117012>] reiserfs_lookup+0x62/0x140 [<c105ccca>] ? debug_check_no_locks_freed+0x8a/0x140 [<c105cbe4>] ? trace_hardirqs_on_caller+0x124/0x170 [<c10bd85f>] __lookup_hash+0xef/0x110 [<c10bf21d>] lookup_one_len+0x8d/0xc0 [<c1141e2a>] open_xa_dir+0xea/0x1b0 [<c1141fe5>] xattr_lookup+0x15/0x160 [<c1142476>] reiserfs_xattr_get+0x56/0x2a0 [<c1144042>] reiserfs_get_acl+0xa2/0x360 [<c10ca2e7>] ? new_inode+0x27/0xa0 [<c114461a>] reiserfs_cache_default_acl+0x3a/0x160 [<c1402eb7>] ? _spin_unlock+0x27/0x40 [<c111789c>] reiserfs_mkdir+0x6c/0x2c0 [<c10c7cb8>] ? __d_lookup+0x108/0x190 [<c105c932>] ? mark_held_locks+0x62/0x80 [<c1401c8d>] ? mutex_lock_nested+0x2bd/0x340 [<c10bd17a>] ? generic_permission+0x1a/0xa0 [<c11788fe>] ? security_inode_permission+0x1e/0x20 [<c10bea96>] vfs_mkdir+0xd6/0x180 [<c10c0c10>] sys_mkdirat+0xc0/0xd0 [<c10505c6>] ? up_read+0x16/0x30 [<c1002fd8>] ? restore_all_notrace+0x0/0x18 [<c10c0c40>] sys_mkdir+0x20/0x30 [<c1002ec4>] sysenter_do_call+0x12/0x32 Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com> Tested-by: Christian Kujau <lists@nerdbynature.de> Cc: Alexander Beregalov <a.beregalov@gmail.com> Cc: Chris Mason <chris.mason@oracle.com> Cc: Ingo Molnar <mingo@elte.hu> |
||
|
Frederic Weisbecker
|
0523676d3f |
reiserfs: Relax reiserfs lock while freeing the journal
Keeping the reiserfs lock while freeing the journal on umount path triggers a lock inversion between bdev->bd_mutex and the reiserfs lock. We don't need the reiserfs lock at this stage. The filesystem is not usable anymore, and there are no more pending commits, everything got flushed (even this operation was done in parallel and didn't required the reiserfs lock from the current process). This fixes the following lockdep report: ======================================================= [ INFO: possible circular locking dependency detected ] 2.6.32-atom #172 ------------------------------------------------------- umount/3904 is trying to acquire lock: (&bdev->bd_mutex){+.+.+.}, at: [<c10de2c2>] __blkdev_put+0x22/0x160 but task is already holding lock: (&REISERFS_SB(s)->lock){+.+.+.}, at: [<c1143279>] reiserfs_write_lock+0x29/0x40 which lock already depends on the new lock. the existing dependency chain (in reverse order) is: -> #3 (&REISERFS_SB(s)->lock){+.+.+.}: [<c105ea7f>] __lock_acquire+0x11ff/0x19e0 [<c105f2c8>] lock_acquire+0x68/0x90 [<c140199b>] mutex_lock_nested+0x5b/0x340 [<c1143229>] reiserfs_write_lock_once+0x29/0x50 [<c111c485>] reiserfs_get_block+0x85/0x1620 [<c10e1040>] do_mpage_readpage+0x1f0/0x6d0 [<c10e1640>] mpage_readpages+0xc0/0x100 [<c1119b89>] reiserfs_readpages+0x19/0x20 [<c108f1ec>] __do_page_cache_readahead+0x1bc/0x260 [<c108f2b8>] ra_submit+0x28/0x40 [<c1087e3e>] filemap_fault+0x40e/0x420 [<c109b5fd>] __do_fault+0x3d/0x430 [<c109d47e>] handle_mm_fault+0x12e/0x790 [<c1022a65>] do_page_fault+0x135/0x330 [<c1403663>] error_code+0x6b/0x70 [<c10ef9ca>] load_elf_binary+0x82a/0x1a10 [<c10ba130>] search_binary_handler+0x90/0x1d0 [<c10bb70f>] do_execve+0x1df/0x250 [<c1001746>] sys_execve+0x46/0x70 [<c1002fa5>] syscall_call+0x7/0xb -> #2 (&mm->mmap_sem){++++++}: [<c105ea7f>] __lock_acquire+0x11ff/0x19e0 [<c105f2c8>] lock_acquire+0x68/0x90 [<c109b1ab>] might_fault+0x8b/0xb0 [<c11b8f52>] copy_to_user+0x32/0x70 [<c10c3b94>] filldir64+0xa4/0xf0 [<c1109116>] sysfs_readdir+0x116/0x210 [<c10c3e1d>] vfs_readdir+0x8d/0xb0 [<c10c3ea9>] sys_getdents64+0x69/0xb0 [<c1002ec4>] sysenter_do_call+0x12/0x32 -> #1 (sysfs_mutex){+.+.+.}: [<c105ea7f>] __lock_acquire+0x11ff/0x19e0 [<c105f2c8>] lock_acquire+0x68/0x90 [<c140199b>] mutex_lock_nested+0x5b/0x340 [<c110951c>] sysfs_addrm_start+0x2c/0xb0 [<c1109aa0>] create_dir+0x40/0x90 [<c1109b1b>] sysfs_create_dir+0x2b/0x50 [<c11b2352>] kobject_add_internal+0xc2/0x1b0 [<c11b2531>] kobject_add_varg+0x31/0x50 [<c11b25ac>] kobject_add+0x2c/0x60 [<c1258294>] device_add+0x94/0x560 [<c11036ea>] add_partition+0x18a/0x2a0 [<c110418a>] rescan_partitions+0x33a/0x450 [<c10de5bf>] __blkdev_get+0x12f/0x2d0 [<c10de76a>] blkdev_get+0xa/0x10 [<c11034b8>] register_disk+0x108/0x130 [<c11a87a9>] add_disk+0xd9/0x130 [<c12998e5>] sd_probe_async+0x105/0x1d0 [<c10528af>] async_thread+0xcf/0x230 [<c104bfd4>] kthread+0x74/0x80 [<c1003aab>] kernel_thread_helper+0x7/0x3c -> #0 (&bdev->bd_mutex){+.+.+.}: [<c105f176>] __lock_acquire+0x18f6/0x19e0 [<c105f2c8>] lock_acquire+0x68/0x90 [<c140199b>] mutex_lock_nested+0x5b/0x340 [<c10de2c2>] __blkdev_put+0x22/0x160 [<c10de40a>] blkdev_put+0xa/0x10 [<c113ce22>] free_journal_ram+0xd2/0x130 [<c113ea18>] do_journal_release+0x98/0x190 [<c113eb2a>] journal_release+0xa/0x10 [<c1128eb6>] reiserfs_put_super+0x36/0x130 [<c10b776f>] generic_shutdown_super+0x4f/0xe0 [<c10b7825>] kill_block_super+0x25/0x40 [<c11255df>] reiserfs_kill_sb+0x7f/0x90 [<c10b7f4a>] deactivate_super+0x7a/0x90 [<c10cccd8>] mntput_no_expire+0x98/0xd0 [<c10ccfcc>] sys_umount+0x4c/0x310 [<c10cd2a9>] sys_oldumount+0x19/0x20 [<c1002ec4>] sysenter_do_call+0x12/0x32 other info that might help us debug this: 2 locks held by umount/3904: #0: (&type->s_umount_key#30){+++++.}, at: [<c10b7f45>] deactivate_super+0x75/0x90 #1: (&REISERFS_SB(s)->lock){+.+.+.}, at: [<c1143279>] reiserfs_write_lock+0x29/0x40 stack backtrace: Pid: 3904, comm: umount Not tainted 2.6.32-atom #172 Call Trace: [<c13ff903>] ? printk+0x18/0x1a [<c105d33a>] print_circular_bug+0xca/0xd0 [<c105f176>] __lock_acquire+0x18f6/0x19e0 [<c108b66f>] ? free_pcppages_bulk+0x1f/0x250 [<c105f2c8>] lock_acquire+0x68/0x90 [<c10de2c2>] ? __blkdev_put+0x22/0x160 [<c10de2c2>] ? __blkdev_put+0x22/0x160 [<c140199b>] mutex_lock_nested+0x5b/0x340 [<c10de2c2>] ? __blkdev_put+0x22/0x160 [<c105c932>] ? mark_held_locks+0x62/0x80 [<c10afe12>] ? kfree+0x92/0xd0 [<c10de2c2>] __blkdev_put+0x22/0x160 [<c105cc3b>] ? trace_hardirqs_on+0xb/0x10 [<c10de40a>] blkdev_put+0xa/0x10 [<c113ce22>] free_journal_ram+0xd2/0x130 [<c113ea18>] do_journal_release+0x98/0x190 [<c113eb2a>] journal_release+0xa/0x10 [<c1128eb6>] reiserfs_put_super+0x36/0x130 [<c1050596>] ? up_write+0x16/0x30 [<c10b776f>] generic_shutdown_super+0x4f/0xe0 [<c10b7825>] kill_block_super+0x25/0x40 [<c10f41e0>] ? vfs_quota_off+0x0/0x20 [<c11255df>] reiserfs_kill_sb+0x7f/0x90 [<c10b7f4a>] deactivate_super+0x7a/0x90 [<c10cccd8>] mntput_no_expire+0x98/0xd0 [<c10ccfcc>] sys_umount+0x4c/0x310 [<c10cd2a9>] sys_oldumount+0x19/0x20 [<c1002ec4>] sysenter_do_call+0x12/0x32 Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com> Cc: Alexander Beregalov <a.beregalov@gmail.com> Cc: Chris Mason <chris.mason@oracle.com> Cc: Ingo Molnar <mingo@elte.hu> |
||
|
Frederic Weisbecker
|
27026a05bb |
reiserfs: Fix reiserfs lock <-> i_mutex dependency inversion on xattr
While deleting the xattrs of an inode, we hold the reiserfs lock and grab the inode->i_mutex of the targeted inode and the root private xattr directory. Later on, we may relax the reiserfs lock for various reasons, this creates inverted dependencies. We can remove the reiserfs lock -> i_mutex dependency by relaxing the former before calling open_xa_dir(). This is fine because the lookup and creation of xattr private directories done in open_xa_dir() are covered by the targeted inode mutexes. And deeper operations in the tree are still done under the write lock. This fixes the following lockdep report: ======================================================= [ INFO: possible circular locking dependency detected ] 2.6.32-atom #173 ------------------------------------------------------- cp/3204 is trying to acquire lock: (&REISERFS_SB(s)->lock){+.+.+.}, at: [<c11432b9>] reiserfs_write_lock_once+0x29/0x50 but task is already holding lock: (&sb->s_type->i_mutex_key#4/3){+.+.+.}, at: [<c1141e18>] open_xa_dir+0xd8/0x1b0 which lock already depends on the new lock. the existing dependency chain (in reverse order) is: -> #1 (&sb->s_type->i_mutex_key#4/3){+.+.+.}: [<c105ea7f>] __lock_acquire+0x11ff/0x19e0 [<c105f2c8>] lock_acquire+0x68/0x90 [<c1401a2b>] mutex_lock_nested+0x5b/0x340 [<c1141d83>] open_xa_dir+0x43/0x1b0 [<c1142722>] reiserfs_for_each_xattr+0x62/0x260 [<c114299a>] reiserfs_delete_xattrs+0x1a/0x60 [<c111ea1f>] reiserfs_delete_inode+0x9f/0x150 [<c10c9c32>] generic_delete_inode+0xa2/0x170 [<c10c9d4f>] generic_drop_inode+0x4f/0x70 [<c10c8b07>] iput+0x47/0x50 [<c10c0965>] do_unlinkat+0xd5/0x160 [<c10c0a00>] sys_unlink+0x10/0x20 [<c1002ec4>] sysenter_do_call+0x12/0x32 -> #0 (&REISERFS_SB(s)->lock){+.+.+.}: [<c105f176>] __lock_acquire+0x18f6/0x19e0 [<c105f2c8>] lock_acquire+0x68/0x90 [<c1401a2b>] mutex_lock_nested+0x5b/0x340 [<c11432b9>] reiserfs_write_lock_once+0x29/0x50 [<c1117012>] reiserfs_lookup+0x62/0x140 [<c10bd85f>] __lookup_hash+0xef/0x110 [<c10bf21d>] lookup_one_len+0x8d/0xc0 [<c1141e2a>] open_xa_dir+0xea/0x1b0 [<c1141fe5>] xattr_lookup+0x15/0x160 [<c1142476>] reiserfs_xattr_get+0x56/0x2a0 [<c1144042>] reiserfs_get_acl+0xa2/0x360 [<c114461a>] reiserfs_cache_default_acl+0x3a/0x160 [<c111789c>] reiserfs_mkdir+0x6c/0x2c0 [<c10bea96>] vfs_mkdir+0xd6/0x180 [<c10c0c10>] sys_mkdirat+0xc0/0xd0 [<c10c0c40>] sys_mkdir+0x20/0x30 [<c1002ec4>] sysenter_do_call+0x12/0x32 other info that might help us debug this: 2 locks held by cp/3204: #0: (&sb->s_type->i_mutex_key#4/1){+.+.+.}, at: [<c10bd8d6>] lookup_create+0x26/0xa0 #1: (&sb->s_type->i_mutex_key#4/3){+.+.+.}, at: [<c1141e18>] open_xa_dir+0xd8/0x1b0 stack backtrace: Pid: 3204, comm: cp Not tainted 2.6.32-atom #173 Call Trace: [<c13ff993>] ? printk+0x18/0x1a [<c105d33a>] print_circular_bug+0xca/0xd0 [<c105f176>] __lock_acquire+0x18f6/0x19e0 [<c105d3aa>] ? check_usage+0x6a/0x460 [<c105f2c8>] lock_acquire+0x68/0x90 [<c11432b9>] ? reiserfs_write_lock_once+0x29/0x50 [<c11432b9>] ? reiserfs_write_lock_once+0x29/0x50 [<c1401a2b>] mutex_lock_nested+0x5b/0x340 [<c11432b9>] ? reiserfs_write_lock_once+0x29/0x50 [<c11432b9>] reiserfs_write_lock_once+0x29/0x50 [<c1117012>] reiserfs_lookup+0x62/0x140 [<c105ccca>] ? debug_check_no_locks_freed+0x8a/0x140 [<c105cbe4>] ? trace_hardirqs_on_caller+0x124/0x170 [<c10bd85f>] __lookup_hash+0xef/0x110 [<c10bf21d>] lookup_one_len+0x8d/0xc0 [<c1141e2a>] open_xa_dir+0xea/0x1b0 [<c1141fe5>] xattr_lookup+0x15/0x160 [<c1142476>] reiserfs_xattr_get+0x56/0x2a0 [<c1144042>] reiserfs_get_acl+0xa2/0x360 [<c10ca2e7>] ? new_inode+0x27/0xa0 [<c114461a>] reiserfs_cache_default_acl+0x3a/0x160 [<c1402eb7>] ? _spin_unlock+0x27/0x40 [<c111789c>] reiserfs_mkdir+0x6c/0x2c0 [<c10c7cb8>] ? __d_lookup+0x108/0x190 [<c105c932>] ? mark_held_locks+0x62/0x80 [<c1401c8d>] ? mutex_lock_nested+0x2bd/0x340 [<c10bd17a>] ? generic_permission+0x1a/0xa0 [<c11788fe>] ? security_inode_permission+0x1e/0x20 [<c10bea96>] vfs_mkdir+0xd6/0x180 [<c10c0c10>] sys_mkdirat+0xc0/0xd0 [<c10505c6>] ? up_read+0x16/0x30 [<c1002fd8>] ? restore_all_notrace+0x0/0x18 [<c10c0c40>] sys_mkdir+0x20/0x30 [<c1002ec4>] sysenter_do_call+0x12/0x32 v2: Don't drop reiserfs_mutex_lock_nested_safe() as we'll still need it later Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com> Tested-by: Christian Kujau <lists@nerdbynature.de> Cc: Alexander Beregalov <a.beregalov@gmail.com> Cc: Chris Mason <chris.mason@oracle.com> Cc: Ingo Molnar <mingo@elte.hu> |
||
|
Frederic Weisbecker
|
c4a62ca362 |
reiserfs: Warn on lock relax if taken recursively
When we relax the reiserfs lock to avoid creating unwanted dependencies against others locks while grabbing these, we want to ensure it has not been taken recursively, otherwise the lock won't be really relaxed. Only its depth will be decreased. The unwanted dependency would then actually happen. To prevent from that, add a reiserfs_lock_check_recursive() call in the places that need it. Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com> Cc: Alexander Beregalov <a.beregalov@gmail.com> Cc: Chris Mason <chris.mason@oracle.com> Cc: Ingo Molnar <mingo@elte.hu> |
||
|
Frederic Weisbecker
|
0719d34347 |
reiserfs: Fix reiserfs lock <-> i_xattr_sem dependency inversion
i_xattr_sem depends on the reiserfs lock. But after we grab i_xattr_sem, we may relax/relock the reiserfs lock while waiting on a freezed filesystem, creating a dependency inversion between the two locks. In order to avoid the i_xattr_sem -> reiserfs lock dependency, let's create a reiserfs_down_read_safe() that acts like reiserfs_mutex_lock_safe(): relax the reiserfs lock while grabbing another lock to avoid undesired dependencies induced by the heivyweight reiserfs lock. This fixes the following warning: [ 990.005931] ======================================================= [ 990.012373] [ INFO: possible circular locking dependency detected ] [ 990.013233] 2.6.33-rc1 #1 [ 990.013233] ------------------------------------------------------- [ 990.013233] dbench/1891 is trying to acquire lock: [ 990.013233] (&REISERFS_SB(s)->lock){+.+.+.}, at: [<ffffffff81159505>] reiserfs_write_lock+0x35/0x50 [ 990.013233] [ 990.013233] but task is already holding lock: [ 990.013233] (&REISERFS_I(inode)->i_xattr_sem){+.+.+.}, at: [<ffffffff8115899a>] reiserfs_xattr_set_handle+0x8a/0x470 [ 990.013233] [ 990.013233] which lock already depends on the new lock. [ 990.013233] [ 990.013233] [ 990.013233] the existing dependency chain (in reverse order) is: [ 990.013233] [ 990.013233] -> #1 (&REISERFS_I(inode)->i_xattr_sem){+.+.+.}: [ 990.013233] [<ffffffff81063afc>] __lock_acquire+0xf9c/0x1560 [ 990.013233] [<ffffffff8106414f>] lock_acquire+0x8f/0xb0 [ 990.013233] [<ffffffff814ac194>] down_write+0x44/0x80 [ 990.013233] [<ffffffff8115899a>] reiserfs_xattr_set_handle+0x8a/0x470 [ 990.013233] [<ffffffff81158e30>] reiserfs_xattr_set+0xb0/0x150 [ 990.013233] [<ffffffff8115a6aa>] user_set+0x8a/0x90 [ 990.013233] [<ffffffff8115901a>] reiserfs_setxattr+0xaa/0xb0 [ 990.013233] [<ffffffff810e2596>] __vfs_setxattr_noperm+0x36/0xa0 [ 990.013233] [<ffffffff810e26bc>] vfs_setxattr+0xbc/0xc0 [ 990.013233] [<ffffffff810e2780>] setxattr+0xc0/0x150 [ 990.013233] [<ffffffff810e289d>] sys_fsetxattr+0x8d/0xa0 [ 990.013233] [<ffffffff81002dab>] system_call_fastpath+0x16/0x1b [ 990.013233] [ 990.013233] -> #0 (&REISERFS_SB(s)->lock){+.+.+.}: [ 990.013233] [<ffffffff81063e30>] __lock_acquire+0x12d0/0x1560 [ 990.013233] [<ffffffff8106414f>] lock_acquire+0x8f/0xb0 [ 990.013233] [<ffffffff814aba77>] __mutex_lock_common+0x47/0x3b0 [ 990.013233] [<ffffffff814abebe>] mutex_lock_nested+0x3e/0x50 [ 990.013233] [<ffffffff81159505>] reiserfs_write_lock+0x35/0x50 [ 990.013233] [<ffffffff811340e5>] reiserfs_prepare_write+0x45/0x180 [ 990.013233] [<ffffffff81158bb6>] reiserfs_xattr_set_handle+0x2a6/0x470 [ 990.013233] [<ffffffff81158e30>] reiserfs_xattr_set+0xb0/0x150 [ 990.013233] [<ffffffff8115a6aa>] user_set+0x8a/0x90 [ 990.013233] [<ffffffff8115901a>] reiserfs_setxattr+0xaa/0xb0 [ 990.013233] [<ffffffff810e2596>] __vfs_setxattr_noperm+0x36/0xa0 [ 990.013233] [<ffffffff810e26bc>] vfs_setxattr+0xbc/0xc0 [ 990.013233] [<ffffffff810e2780>] setxattr+0xc0/0x150 [ 990.013233] [<ffffffff810e289d>] sys_fsetxattr+0x8d/0xa0 [ 990.013233] [<ffffffff81002dab>] system_call_fastpath+0x16/0x1b [ 990.013233] [ 990.013233] other info that might help us debug this: [ 990.013233] [ 990.013233] 2 locks held by dbench/1891: [ 990.013233] #0: (&sb->s_type->i_mutex_key#12){+.+.+.}, at: [<ffffffff810e2678>] vfs_setxattr+0x78/0xc0 [ 990.013233] #1: (&REISERFS_I(inode)->i_xattr_sem){+.+.+.}, at: [<ffffffff8115899a>] reiserfs_xattr_set_handle+0x8a/0x470 [ 990.013233] [ 990.013233] stack backtrace: [ 990.013233] Pid: 1891, comm: dbench Not tainted 2.6.33-rc1 #1 [ 990.013233] Call Trace: [ 990.013233] [<ffffffff81061639>] print_circular_bug+0xe9/0xf0 [ 990.013233] [<ffffffff81063e30>] __lock_acquire+0x12d0/0x1560 [ 990.013233] [<ffffffff8115899a>] ? reiserfs_xattr_set_handle+0x8a/0x470 [ 990.013233] [<ffffffff8106414f>] lock_acquire+0x8f/0xb0 [ 990.013233] [<ffffffff81159505>] ? reiserfs_write_lock+0x35/0x50 [ 990.013233] [<ffffffff8115899a>] ? reiserfs_xattr_set_handle+0x8a/0x470 [ 990.013233] [<ffffffff814aba77>] __mutex_lock_common+0x47/0x3b0 [ 990.013233] [<ffffffff81159505>] ? reiserfs_write_lock+0x35/0x50 [ 990.013233] [<ffffffff81159505>] ? reiserfs_write_lock+0x35/0x50 [ 990.013233] [<ffffffff81062592>] ? mark_held_locks+0x72/0xa0 [ 990.013233] [<ffffffff814ab81d>] ? __mutex_unlock_slowpath+0xbd/0x140 [ 990.013233] [<ffffffff810628ad>] ? trace_hardirqs_on_caller+0x14d/0x1a0 [ 990.013233] [<ffffffff814abebe>] mutex_lock_nested+0x3e/0x50 [ 990.013233] [<ffffffff81159505>] reiserfs_write_lock+0x35/0x50 [ 990.013233] [<ffffffff811340e5>] reiserfs_prepare_write+0x45/0x180 [ 990.013233] [<ffffffff81158bb6>] reiserfs_xattr_set_handle+0x2a6/0x470 [ 990.013233] [<ffffffff81158e30>] reiserfs_xattr_set+0xb0/0x150 [ 990.013233] [<ffffffff814abcb4>] ? __mutex_lock_common+0x284/0x3b0 [ 990.013233] [<ffffffff8115a6aa>] user_set+0x8a/0x90 [ 990.013233] [<ffffffff8115901a>] reiserfs_setxattr+0xaa/0xb0 [ 990.013233] [<ffffffff810e2596>] __vfs_setxattr_noperm+0x36/0xa0 [ 990.013233] [<ffffffff810e26bc>] vfs_setxattr+0xbc/0xc0 [ 990.013233] [<ffffffff810e2780>] setxattr+0xc0/0x150 [ 990.013233] [<ffffffff81056018>] ? sched_clock_cpu+0xb8/0x100 [ 990.013233] [<ffffffff8105eded>] ? trace_hardirqs_off+0xd/0x10 [ 990.013233] [<ffffffff810560a3>] ? cpu_clock+0x43/0x50 [ 990.013233] [<ffffffff810c6820>] ? fget+0xb0/0x110 [ 990.013233] [<ffffffff810c6770>] ? fget+0x0/0x110 [ 990.013233] [<ffffffff81002ddc>] ? sysret_check+0x27/0x62 [ 990.013233] [<ffffffff810e289d>] sys_fsetxattr+0x8d/0xa0 [ 990.013233] [<ffffffff81002dab>] system_call_fastpath+0x16/0x1b Reported-and-tested-by: Christian Kujau <lists@nerdbynature.de> Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com> Cc: Alexander Beregalov <a.beregalov@gmail.com> Cc: Chris Mason <chris.mason@oracle.com> Cc: Ingo Molnar <mingo@elte.hu> |
||
|
Frederic Weisbecker
|
98ea3f50bc |
reiserfs: Fix remaining in-reclaim-fs <-> reclaim-fs-on locking inversion
Commit
|
||
|
Frederic Weisbecker
|
47376ceba5 |
reiserfs: Fix reiserfs lock <-> inode mutex dependency inversion
The reiserfs lock -> inode mutex dependency gets inverted when we relax the lock while walking to the tree. To fix this, use a specialized version of reiserfs_mutex_lock_safe that takes care of mutex subclasses. Then we can grab the inode mutex with I_MUTEX_XATTR subclass without any reiserfs lock dependency. This fixes the following report: [ INFO: possible circular locking dependency detected ] 2.6.32-06793-gf405425-dirty #2 ------------------------------------------------------- mv/18566 is trying to acquire lock: (&REISERFS_SB(s)->lock){+.+.+.}, at: [<c1110708>] reiserfs_write_lock+0x28= /0x40 but task is already holding lock: (&sb->s_type->i_mutex_key#5/3){+.+.+.}, at: [<c111033c>] reiserfs_for_each_xattr+0x10c/0x380 which lock already depends on the new lock. the existing dependency chain (in reverse order) is: -> #1 (&sb->s_type->i_mutex_key#5/3){+.+.+.}: [<c104f723>] validate_chain+0xa23/0xf70 [<c1050155>] __lock_acquire+0x4e5/0xa70 [<c105075a>] lock_acquire+0x7a/0xa0 [<c134c76f>] mutex_lock_nested+0x5f/0x2b0 [<c11102b4>] reiserfs_for_each_xattr+0x84/0x380 [<c1110615>] reiserfs_delete_xattrs+0x15/0x50 [<c10ef57f>] reiserfs_delete_inode+0x8f/0x140 [<c10a565c>] generic_delete_inode+0x9c/0x150 [<c10a574d>] generic_drop_inode+0x3d/0x60 [<c10a4667>] iput+0x47/0x50 [<c109cc0b>] do_unlinkat+0xdb/0x160 [<c109cca0>] sys_unlink+0x10/0x20 [<c1002c50>] sysenter_do_call+0x12/0x36 -> #0 (&REISERFS_SB(s)->lock){+.+.+.}: [<c104fc68>] validate_chain+0xf68/0xf70 [<c1050155>] __lock_acquire+0x4e5/0xa70 [<c105075a>] lock_acquire+0x7a/0xa0 [<c134c76f>] mutex_lock_nested+0x5f/0x2b0 [<c1110708>] reiserfs_write_lock+0x28/0x40 [<c1103d6b>] search_by_key+0x1f7b/0x21b0 [<c10e73ef>] search_by_entry_key+0x1f/0x3b0 [<c10e77f7>] reiserfs_find_entry+0x77/0x400 [<c10e81e5>] reiserfs_lookup+0x85/0x130 [<c109a144>] __lookup_hash+0xb4/0x110 [<c109b763>] lookup_one_len+0xb3/0x100 [<c1110350>] reiserfs_for_each_xattr+0x120/0x380 [<c1110615>] reiserfs_delete_xattrs+0x15/0x50 [<c10ef57f>] reiserfs_delete_inode+0x8f/0x140 [<c10a565c>] generic_delete_inode+0x9c/0x150 [<c10a574d>] generic_drop_inode+0x3d/0x60 [<c10a4667>] iput+0x47/0x50 [<c10a1c4f>] dentry_iput+0x6f/0xf0 [<c10a1d74>] d_kill+0x24/0x50 [<c10a396b>] dput+0x5b/0x120 [<c109ca89>] sys_renameat+0x1b9/0x230 [<c109cb28>] sys_rename+0x28/0x30 [<c1002c50>] sysenter_do_call+0x12/0x36 other info that might help us debug this: 2 locks held by mv/18566: #0: (&sb->s_type->i_mutex_key#5/1){+.+.+.}, at: [<c109b6ac>] lock_rename+0xcc/0xd0 #1: (&sb->s_type->i_mutex_key#5/3){+.+.+.}, at: [<c111033c>] reiserfs_for_each_xattr+0x10c/0x380 stack backtrace: Pid: 18566, comm: mv Tainted: G C 2.6.32-06793-gf405425-dirty #2 Call Trace: [<c134b252>] ? printk+0x18/0x1e [<c104e790>] print_circular_bug+0xc0/0xd0 [<c104fc68>] validate_chain+0xf68/0xf70 [<c104c8cb>] ? trace_hardirqs_off+0xb/0x10 [<c1050155>] __lock_acquire+0x4e5/0xa70 [<c105075a>] lock_acquire+0x7a/0xa0 [<c1110708>] ? reiserfs_write_lock+0x28/0x40 [<c134c76f>] mutex_lock_nested+0x5f/0x2b0 [<c1110708>] ? reiserfs_write_lock+0x28/0x40 [<c1110708>] ? reiserfs_write_lock+0x28/0x40 [<c134b60a>] ? schedule+0x27a/0x440 [<c1110708>] reiserfs_write_lock+0x28/0x40 [<c1103d6b>] search_by_key+0x1f7b/0x21b0 [<c1050176>] ? __lock_acquire+0x506/0xa70 [<c1051267>] ? lock_release_non_nested+0x1e7/0x340 [<c1110708>] ? reiserfs_write_lock+0x28/0x40 [<c104e354>] ? trace_hardirqs_on_caller+0x124/0x170 [<c104e3ab>] ? trace_hardirqs_on+0xb/0x10 [<c1042a55>] ? T.316+0x15/0x1a0 [<c1042d2d>] ? sched_clock_cpu+0x9d/0x100 [<c10e73ef>] search_by_entry_key+0x1f/0x3b0 [<c134bf2a>] ? __mutex_unlock_slowpath+0x9a/0x120 [<c104e354>] ? trace_hardirqs_on_caller+0x124/0x170 [<c10e77f7>] reiserfs_find_entry+0x77/0x400 [<c10e81e5>] reiserfs_lookup+0x85/0x130 [<c1042d2d>] ? sched_clock_cpu+0x9d/0x100 [<c109a144>] __lookup_hash+0xb4/0x110 [<c109b763>] lookup_one_len+0xb3/0x100 [<c1110350>] reiserfs_for_each_xattr+0x120/0x380 [<c110ffe0>] ? delete_one_xattr+0x0/0x1c0 [<c1003342>] ? math_error+0x22/0x150 [<c1110708>] ? reiserfs_write_lock+0x28/0x40 [<c1110615>] reiserfs_delete_xattrs+0x15/0x50 [<c1110708>] ? reiserfs_write_lock+0x28/0x40 [<c10ef57f>] reiserfs_delete_inode+0x8f/0x140 [<c10a561f>] ? generic_delete_inode+0x5f/0x150 [<c10ef4f0>] ? reiserfs_delete_inode+0x0/0x140 [<c10a565c>] generic_delete_inode+0x9c/0x150 [<c10a574d>] generic_drop_inode+0x3d/0x60 [<c10a4667>] iput+0x47/0x50 [<c10a1c4f>] dentry_iput+0x6f/0xf0 [<c10a1d74>] d_kill+0x24/0x50 [<c10a396b>] dput+0x5b/0x120 [<c109ca89>] sys_renameat+0x1b9/0x230 [<c1042d2d>] ? sched_clock_cpu+0x9d/0x100 [<c104c8cb>] ? trace_hardirqs_off+0xb/0x10 [<c1042dde>] ? cpu_clock+0x4e/0x60 [<c1350825>] ? do_page_fault+0x155/0x370 [<c1041816>] ? up_read+0x16/0x30 [<c1350825>] ? do_page_fault+0x155/0x370 [<c109cb28>] sys_rename+0x28/0x30 [<c1002c50>] sysenter_do_call+0x12/0x36 Reported-by: Alexander Beregalov <a.beregalov@gmail.com> Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com> Cc: Chris Mason <chris.mason@oracle.com> Cc: Ingo Molnar <mingo@elte.hu> Cc: Thomas Gleixner <tglx@linutronix.de> |
||
|
Frederic Weisbecker
|
cb1c2e51c5 |
reiserfs: Fix reiserfs lock and journal lock inversion dependency
When we were using the bkl, we didn't care about dependencies against other locks, but the mutex conversion created new ones, which is why we have reiserfs_mutex_lock_safe(), which unlocks the reiserfs lock before acquiring another mutex. But this trick actually fails if we have acquired the reiserfs lock recursively, as we try to unlock it to acquire the new mutex without inverted dependency, but we eventually only decrease its depth. This happens in the case of a nested inode creation/deletion. Say we have no space left on the device, we create an inode and tak the lock but fail to create its entry, then we release the inode using iput(), which calls reiserfs_delete_inode() that takes the reiserfs lock recursively. The path eventually ends up in journal_begin() where we try to take the journal safely but we fail because of the reiserfs lock recursion: [ INFO: possible circular locking dependency detected ] 2.6.32-06486-g053fe57 #2 ------------------------------------------------------- vi/23454 is trying to acquire lock: (&journal->j_mutex){+.+...}, at: [<c110dac4>] do_journal_begin_r+0x64/0x2f0 but task is already holding lock: (&REISERFS_SB(s)->lock){+.+.+.}, at: [<c11106a8>] reiserfs_write_lock+0x28/0x40 which lock already depends on the new lock. the existing dependency chain (in reverse order) is: -> #1 (&REISERFS_SB(s)->lock){+.+.+.}: [<c104f8f3>] validate_chain+0xa23/0xf70 [<c1050325>] __lock_acquire+0x4e5/0xa70 [<c105092a>] lock_acquire+0x7a/0xa0 [<c134c78f>] mutex_lock_nested+0x5f/0x2b0 [<c11106a8>] reiserfs_write_lock+0x28/0x40 [<c110dacb>] do_journal_begin_r+0x6b/0x2f0 [<c110ddcf>] journal_begin+0x7f/0x120 [<c10f76c2>] reiserfs_remount+0x212/0x4d0 [<c1093997>] do_remount_sb+0x67/0x140 [<c10a9ca6>] do_mount+0x436/0x6b0 [<c10a9f86>] sys_mount+0x66/0xa0 [<c1002c50>] sysenter_do_call+0x12/0x36 -> #0 (&journal->j_mutex){+.+...}: [<c104fe38>] validate_chain+0xf68/0xf70 [<c1050325>] __lock_acquire+0x4e5/0xa70 [<c105092a>] lock_acquire+0x7a/0xa0 [<c134c78f>] mutex_lock_nested+0x5f/0x2b0 [<c110dac4>] do_journal_begin_r+0x64/0x2f0 [<c110ddcf>] journal_begin+0x7f/0x120 [<c10ef52f>] reiserfs_delete_inode+0x9f/0x140 [<c10a55fc>] generic_delete_inode+0x9c/0x150 [<c10a56ed>] generic_drop_inode+0x3d/0x60 [<c10a4607>] iput+0x47/0x50 [<c10e915c>] reiserfs_create+0x16c/0x1c0 [<c109a9c1>] vfs_create+0xc1/0x130 [<c109dbec>] do_filp_open+0x81c/0x920 [<c109004f>] do_sys_open+0x4f/0x110 [<c1090179>] sys_open+0x29/0x40 [<c1002c50>] sysenter_do_call+0x12/0x36 other info that might help us debug this: 2 locks held by vi/23454: #0: (&sb->s_type->i_mutex_key#5){+.+.+.}, at: [<c109d64e>] do_filp_open+0x27e/0x920 #1: (&REISERFS_SB(s)->lock){+.+.+.}, at: [<c11106a8>] reiserfs_write_lock+0x28/0x40 stack backtrace: Pid: 23454, comm: vi Not tainted 2.6.32-06486-g053fe57 #2 Call Trace: [<c134b202>] ? printk+0x18/0x1e [<c104e960>] print_circular_bug+0xc0/0xd0 [<c104fe38>] validate_chain+0xf68/0xf70 [<c104ca9b>] ? trace_hardirqs_off+0xb/0x10 [<c1050325>] __lock_acquire+0x4e5/0xa70 [<c105092a>] lock_acquire+0x7a/0xa0 [<c110dac4>] ? do_journal_begin_r+0x64/0x2f0 [<c134c78f>] mutex_lock_nested+0x5f/0x2b0 [<c110dac4>] ? do_journal_begin_r+0x64/0x2f0 [<c110dac4>] ? do_journal_begin_r+0x64/0x2f0 [<c110ff80>] ? delete_one_xattr+0x0/0x1c0 [<c110dac4>] do_journal_begin_r+0x64/0x2f0 [<c110ddcf>] journal_begin+0x7f/0x120 [<c11105b5>] ? reiserfs_delete_xattrs+0x15/0x50 [<c10ef52f>] reiserfs_delete_inode+0x9f/0x140 [<c10a55bf>] ? generic_delete_inode+0x5f/0x150 [<c10ef490>] ? reiserfs_delete_inode+0x0/0x140 [<c10a55fc>] generic_delete_inode+0x9c/0x150 [<c10a56ed>] generic_drop_inode+0x3d/0x60 [<c10a4607>] iput+0x47/0x50 [<c10e915c>] reiserfs_create+0x16c/0x1c0 [<c1099a5d>] ? inode_permission+0x7d/0xa0 [<c109a9c1>] vfs_create+0xc1/0x130 [<c10e8ff0>] ? reiserfs_create+0x0/0x1c0 [<c109dbec>] do_filp_open+0x81c/0x920 [<c104ca9b>] ? trace_hardirqs_off+0xb/0x10 [<c134dc0d>] ? _spin_unlock+0x1d/0x20 [<c10a6eea>] ? alloc_fd+0xba/0xf0 [<c109004f>] do_sys_open+0x4f/0x110 [<c1090179>] sys_open+0x29/0x40 [<c1002c50>] sysenter_do_call+0x12/0x36 To fix this, use reiserfs_lock_once() from reiserfs_delete_inode() which prevents from adding reiserfs lock recursion. Reported-by: Alexander Beregalov <a.beregalov@gmail.com> Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com> Cc: Chris Mason <chris.mason@oracle.com> Cc: Ingo Molnar <mingo@elte.hu> Cc: Thomas Gleixner <tglx@linutronix.de> |
||
|
Frederic Weisbecker
|
500f5a0bf5 |
reiserfs: Fix possible recursive lock
While allocating the bitmap using vmalloc, we hold the reiserfs lock,
which makes lockdep later reporting a possible deadlock as we may
swap out pages to allocate memory and then take the reiserfs lock
recursively:
inconsistent {RECLAIM_FS-ON-W} -> {IN-RECLAIM_FS-W} usage.
kswapd0/312 [HC0[0]:SC0[0]:HE1:SE1] takes:
(&REISERFS_SB(s)->lock){+.+.?.}, at: [<c11108a8>] reiserfs_write_lock+0x28/0x40
{RECLAIM_FS-ON-W} state was registered at:
[<c104e1c2>] mark_held_locks+0x62/0x90
[<c104e28a>] lockdep_trace_alloc+0x9a/0xc0
[<c108e396>] kmem_cache_alloc+0x26/0xf0
[<c10850ec>] __get_vm_area_node+0x6c/0xf0
[<c10857de>] __vmalloc_node+0x7e/0xa0
[<c108597b>] vmalloc+0x2b/0x30
[<c10e00b9>] reiserfs_init_bitmap_cache+0x39/0x70
[<c10f8178>] reiserfs_fill_super+0x2e8/0xb90
[<c1094345>] get_sb_bdev+0x145/0x180
[<c10f5a11>] get_super_block+0x21/0x30
[<c10931f0>] vfs_kern_mount+0x40/0xd0
[<c10932d9>] do_kern_mount+0x39/0xd0
[<c10a9857>] do_mount+0x2c7/0x6b0
[<c10a9ca6>] sys_mount+0x66/0xa0
[<c161589b>] mount_block_root+0xc4/0x245
[<c1615a75>] mount_root+0x59/0x5f
[<c1615b8c>] prepare_namespace+0x111/0x14b
[<c1615269>] kernel_init+0xcf/0xdb
[<c10031fb>] kernel_thread_helper+0x7/0x1c
This is actually fine for two reasons: we call vmalloc at mount time
then it's not in the swapping out path. Also the reiserfs lock can be
acquired recursively, but since its implementation depends on a mutex,
it's hard and not necessary worth it to teach that to lockdep.
The lock is useless at mount time anyway, at least until we replay the
journal. But let's remove it from this path later as this needs
more thinking and is a sensible change.
For now we can just relax the lock around vmalloc,
Reported-by: Alexander Beregalov <a.beregalov@gmail.com>
Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Chris Mason <chris.mason@oracle.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Thomas Gleixner <tglx@linutronix.de>
|
||
|
Frederic Weisbecker
|
6548698f92 |
Merge commit 'v2.6.32' into reiserfs/kill-bkl
Merge-reason: The tree was based 2.6.31. It's better to be up to date with 2.6.32. Although no conflicting changes were made in between, it gives benchmarking results closer to the lastest kernel behaviour. |
||
|
Frederic Weisbecker
|
1d2c6cfd40 |
kill-the-bkl/reiserfs: turn GFP_ATOMIC flag to GFP_NOFS in reiserfs_get_block()
GFP_ATOMIC was used in reiserfs_get_block to not lose the Bkl so that nobody can modify the tree in the middle of its work. Now that we kicked out the bkl, we can use a more friendly flag. We use GFP_NOFS here because we already hold the reiserfs lock. Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com> Cc: Jeff Mahoney <jeffm@suse.com> Cc: Chris Mason <chris.mason@oracle.com> Cc: Ingo Molnar <mingo@elte.hu> Cc: Alexander Beregalov <a.beregalov@gmail.com> Cc: Laurent Riffard <laurent.riffard@free.fr> Cc: Thomas Gleixner <tglx@linutronix.de> |
||
|
Frederic Weisbecker
|
27b3a5c51b |
kill-the-bkl/reiserfs: drop the fs race watchdog from _get_block_create_0()
We had a watchdog in _get_block_create_0() that jumped to a fixup retry path in case the bkl got relaxed while calling kmap(). This is not necessary anymore since we now have a reiserfs lock that is not implicitly relaxed while sleeping. Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com> Cc: Jeff Mahoney <jeffm@suse.com> Cc: Chris Mason <chris.mason@oracle.com> Cc: Ingo Molnar <mingo@elte.hu> Cc: Alexander Beregalov <a.beregalov@gmail.com> Cc: Laurent Riffard <laurent.riffard@free.fr> Cc: Thomas Gleixner <tglx@linutronix.de> |
||
|
Frederic Weisbecker
|
205cb37b89 |
kill-the-bkl/reiserfs: definitely drop the bkl from reiserfs_ioctl()
The reiserfs ioctl path doesn't need the big kernel lock anymore , now that the filesystem synchronizes through its own lock. We can then turn reiserfs_ioctl() into an unlocked_ioctl callback. Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com> Cc: Jeff Mahoney <jeffm@suse.com> Cc: Chris Mason <chris.mason@oracle.com> Cc: Ingo Molnar <mingo@elte.hu> Cc: Alexander Beregalov <a.beregalov@gmail.com> Cc: Laurent Riffard <laurent.riffard@free.fr> Cc: Thomas Gleixner <tglx@linutronix.de> |
||
|
Frederic Weisbecker
|
ac78a07893 |
kill-the-bkl/reiserfs: always lock the ioctl path
Reiserfs uses the ioctl callback for its file operations, which means that its ioctl path is still locked by the bkl, this was synchronizing with the rest of the filsystem operations. We have changed that by locking it with the new reiserfs lock but we do that only from the compat_ioctl callback. Fix that by locking reiserfs_ioctl() everytime. Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com> Cc: Jeff Mahoney <jeffm@suse.com> Cc: Chris Mason <chris.mason@oracle.com> Cc: Ingo Molnar <mingo@elte.hu> Cc: Alexander Beregalov <a.beregalov@gmail.com> Cc: Laurent Riffard <laurent.riffard@free.fr> Cc: Thomas Gleixner <tglx@linutronix.de> |
||
|
Frederic Weisbecker
|
48f6ba5e69 |
kill-the-bkl/reiserfs: fix reiserfs lock to cpu_add_remove_lock dependency
While creating the reiserfs workqueue during the journal initialization, we are holding the reiserfs lock, but create_workqueue() also holds the cpu_add_remove_lock, creating then the following dependency: - reiserfs lock -> cpu_add_remove_lock But we also have the following existing dependencies: - mm->mmap_sem -> reiserfs lock - cpu_add_remove_lock -> cpu_hotplug.lock -> slub_lock -> sysfs_mutex The merged dependency chain then becomes: - mm->mmap_sem -> reiserfs lock -> cpu_add_remove_lock -> cpu_hotplug.lock -> slub_lock -> sysfs_mutex But when we fill a dir entry in sysfs_readir(), we are holding the sysfs_mutex and we also might fault while copying the directory entry to the user, leading to the following dependency: - sysfs_mutex -> mm->mmap_sem The end result is then a lock inversion between sysfs_mutex and mm->mmap_sem, as reported in the following lockdep warning: [ INFO: possible circular locking dependency detected ] 2.6.31-07095-g25a3912 #4 ------------------------------------------------------- udevadm/790 is trying to acquire lock: (&mm->mmap_sem){++++++}, at: [<c1098942>] might_fault+0x72/0xc0 but task is already holding lock: (sysfs_mutex){+.+.+.}, at: [<c110813c>] sysfs_readdir+0x7c/0x260 which lock already depends on the new lock. the existing dependency chain (in reverse order) is: -> #5 (sysfs_mutex){+.+.+.}: [...] -> #4 (slub_lock){+++++.}: [...] -> #3 (cpu_hotplug.lock){+.+.+.}: [...] -> #2 (cpu_add_remove_lock){+.+.+.}: [...] -> #1 (&REISERFS_SB(s)->lock){+.+.+.}: [...] -> #0 (&mm->mmap_sem){++++++}: [...] This can be fixed by relaxing the reiserfs lock while creating the workqueue. This is fine to relax the lock here, we just keep it around to pass through reiserfs lock checks and for paranoid reasons. Reported-by: Alexander Beregalov <a.beregalov@gmail.com> Tested-by: Alexander Beregalov <a.beregalov@gmail.com> Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com> Cc: Jeff Mahoney <jeffm@suse.com> Cc: Chris Mason <chris.mason@oracle.com> Cc: Ingo Molnar <mingo@elte.hu> Cc: Alexander Beregalov <a.beregalov@gmail.com> Cc: Laurent Riffard <laurent.riffard@free.fr> |
||
Alexey Dobriyan
|
0d54b217a2 |
const: make struct super_block::s_qcop const
Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org> |
||
|
Alexey Dobriyan
|
61e225dc34 |
const: make struct super_block::dq_op const
Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org> |
||
|
Frederic Weisbecker
|
193be0ee17 |
kill-the-bkl/reiserfs: Fix induced mm->mmap_sem to sysfs_mutex dependency
Alexander Beregalov reported the following warning: ======================================================= [ INFO: possible circular locking dependency detected ] 2.6.31-03149-gdcc030a #1 ------------------------------------------------------- udevadm/716 is trying to acquire lock: (&mm->mmap_sem){++++++}, at: [<c107249a>] might_fault+0x4a/0xa0 but task is already holding lock: (sysfs_mutex){+.+.+.}, at: [<c10cb9aa>] sysfs_readdir+0x5a/0x200 which lock already depends on the new lock. the existing dependency chain (in reverse order) is: -> #3 (sysfs_mutex){+.+.+.}: [...] -> #2 (&bdev->bd_mutex){+.+.+.}: [...] -> #1 (&REISERFS_SB(s)->lock){+.+.+.}: [...] -> #0 (&mm->mmap_sem){++++++}: [...] On reiserfs mount path, we take the reiserfs lock and while initializing the journal, we open the device, taking the bdev->bd_mutex. Then rescan_partition() may signal the change to sysfs. We have then the following dependency: reiserfs_lock -> bd_mutex -> sysfs_mutex Later, while entering reiserfs_readpage() after a pagefault in an mmaped reiserfs file, we are holding the mm->mmap_sem, and we are going to take the reiserfs lock too. We have then the following dependency: mm->mmap_sem -> reiserfs_lock which, expanded with the previous dependency gives us: mm->mmap_sem -> reiserfs_lock -> bd_mutex -> sysfs_mutex Now while entering the sysfs readdir path, we are holding the sysfs_mutex. And when we copy a directory entry to the user buffer, we might fault and then take the mm->mmap_sem lock. Which leads to the circular locking dependency reported. We can fix that by relaxing the reiserfs lock during the call to journal_init_dev(), which is the place where we open the mounted device. This is fine to relax the lock here because we are in the begining of the reiserfs mount path and there is nothing to protect at this time, the journal is not intialized. We just keep this lock around for paranoid reasons. Reported-by: Alexander Beregalov <a.beregalov@gmail.com> Tested-by: Alexander Beregalov <a.beregalov@gmail.com> Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com> Cc: Jeff Mahoney <jeffm@suse.com> Cc: Chris Mason <chris.mason@oracle.com> Cc: Ingo Molnar <mingo@elte.hu> Cc: Alexander Beregalov <a.beregalov@gmail.com> Cc: Laurent Riffard <laurent.riffard@free.fr> |
||
|
Frederic Weisbecker
|
8050318598 |
kill-the-bkl/reiserfs: panic in case of lock imbalance
Until now, trying to unlock the reiserfs write lock whereas the current task doesn't hold it lead to a simple warning. We should actually warn and panic in this case to avoid the user datas to reach an unstable state. Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com> Cc: Jeff Mahoney <jeffm@suse.com> Cc: Chris Mason <chris.mason@oracle.com> Cc: Ingo Molnar <mingo@elte.hu> Cc: Alexander Beregalov <a.beregalov@gmail.com> Cc: Laurent Riffard <laurent.riffard@free.fr> |
||
|
Frederic Weisbecker
|
7e94277050 |
kill-the-bkl/reiserfs: fix recursive reiserfs write lock in reiserfs_commit_write()
reiserfs_commit_write() is always called with the write lock held. Thus the current calls to reiserfs_write_lock() in this function are acquiring the lock recursively. We can safely drop them. This also solves further assumptions for this lock to be really released while calling reiserfs_write_unlock(). Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com> Cc: Jeff Mahoney <jeffm@suse.com> Cc: Chris Mason <chris.mason@oracle.com> Cc: Ingo Molnar <mingo@elte.hu> Cc: Alexander Beregalov <a.beregalov@gmail.com> Cc: Laurent Riffard <laurent.riffard@free.fr> |
||
|
Frederic Weisbecker
|
b10ab4c337 |
kill-the-bkl/reiserfs: fix recursive reiserfs lock in reiserfs_mkdir()
reiserfs_mkdir() acquires the reiserfs lock, assuming it has been called from the dir inodes callbacks, without the lock held. But it can also be called from other internal sites such as reiserfs_xattr_init() which already holds the lock. This recursive locking leads to further wrong assumptions. For example, later calls to reiserfs_mutex_lock_safe() won't actually unlock the reiserfs lock the time we acquire a given mutex, creating unexpected lock inversions. Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com> Cc: Jeff Mahoney <jeffm@suse.com> Cc: Chris Mason <chris.mason@oracle.com> Cc: Ingo Molnar <mingo@elte.hu> Cc: Alexander Beregalov <a.beregalov@gmail.com> Cc: Laurent Riffard <laurent.riffard@free.fr> |
||
|
Frederic Weisbecker
|
ae635c0bbd |
kill-the-bkl/reiserfs: fix "reiserfs lock" / "inode mutex" lock inversion dependency
reiserfs_xattr_init is called with the reiserfs write lock held, but if the ".reiserfs_priv" entry is not created, we take the superblock root directory inode mutex until .reiserfs_priv is created. This creates a lock dependency inversion against other sites such as reiserfs_file_release() which takes an inode mutex and the reiserfs lock after. Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com> Cc: Jeff Mahoney <jeffm@suse.com> Cc: Chris Mason <chris.mason@oracle.com> Cc: Ingo Molnar <mingo@elte.hu> Cc: Alexander Beregalov <a.beregalov@gmail.com> Cc: Laurent Riffard <laurent.riffard@free.fr> |
||
|
Frederic Weisbecker
|
08f14fc896 |
kill-the-bkl/reiserfs: move the concurrent tree accesses checks per superblock
When do_balance() balances the tree, a trick is performed to provide the ability for other tree writers/readers to check whether do_balance() is executing concurrently (requires CONFIG_REISERFS_CHECK). This is done to protect concurrent accesses to the tree. The trick is the following: When do_balance is called, a unique global variable called cur_tb takes a pointer to the current tree to be rebalanced. Once do_balance finishes its work, cur_tb takes the NULL value. Then, concurrent tree readers/writers just have to check the value of cur_tb to ensure do_balance isn't executing concurrently. If it is, then it proves that schedule() occured on do_balance(), which then relaxed the bkl that protected the tree. Now that the bkl has be turned into a mutex, this check is still fine even though do_balance() becomes preemptible: the write lock will not be automatically released on schedule(), so the tree is still protected. But this is only fine if we have a single reiserfs mountpoint. Indeed, because the bkl is a global lock, it didn't allowed concurrent executions between a tree reader/writer in a mount point and a do_balance() on another tree from another mountpoint. So assuming all these readers/writers weren't supposed to be reentrant, the current check now sometimes detect false positives with the current per-superblock mutex which allows this reentrancy. This patch keeps the concurrent tree accesses check but moves it per superblock, so that only trees from a same mount point are checked to be not accessed concurrently. [ Impact: fix spurious panic while running several reiserfs mount-points ] Cc: Jeff Mahoney <jeffm@suse.com> Cc: Chris Mason <chris.mason@oracle.com> Cc: Ingo Molnar <mingo@elte.hu> Cc: Alexander Beregalov <a.beregalov@gmail.com> Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com> |
||
|
Frederic Weisbecker
|
c72e05756b |
kill-the-bkl/reiserfs: acquire the inode mutex safely
While searching a pathname, an inode mutex can be acquired in do_lookup() which calls reiserfs_lookup() which in turn acquires the write lock. On the other side reiserfs_fill_super() can acquire the write_lock and then call reiserfs_lookup_privroot() which can acquire an inode mutex (the root of the mount point). So we theoretically risk an AB - BA lock inversion that could lead to a deadlock. As for other lock dependencies found since the bkl to mutex conversion, the fix is to use reiserfs_mutex_lock_safe() which drops the lock dependency to the write lock. [ Impact: fix a possible deadlock with reiserfs ] Cc: Jeff Mahoney <jeffm@suse.com> Cc: Chris Mason <chris.mason@oracle.com> Cc: Ingo Molnar <mingo@elte.hu> Cc: Alexander Beregalov <a.beregalov@gmail.com> Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com> |
||
|
Frederic Weisbecker
|
2ac626955e |
kill-the-bkl/reiserfs: unlock only when needed in search_by_key
search_by_key() is the site which most requires the lock. This is mostly because it is a very central function and also because it releases/reaqcuires the write lock at least once each time it is called. Such release/reacquire creates a lot of contention in this place and also opens more the window which let another thread changing the tree. When it happens, the current path searching over the tree must be retried from the beggining (the root) which is a wasteful and time consuming recovery. This patch factorizes two release/reacquire sequences: - reading leaf nodes blocks - reading current block The latter immediately follows the former. The whole sequence is safe as a single unlocked section because we check just after if the tree has changed during these operations. Cc: Jeff Mahoney <jeffm@suse.com> Cc: Chris Mason <chris.mason@oracle.com> Cc: Ingo Molnar <mingo@elte.hu> Cc: Alexander Beregalov <a.beregalov@gmail.com> Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com> |
||
|
Frederic Weisbecker
|
c63e3c0b24 |
kill-the-bkl/reiserfs: use mutex_lock in reiserfs_mutex_lock_safe
reiserfs_mutex_lock_safe() is a hack to avoid any dependency between
an internal reiserfs mutex and the write lock, it has been proposed
to follow the old bkl logic.
The code does the following:
while (!mutex_trylock(m)) {
reiserfs_write_unlock(s);
schedule();
reiserfs_write_lock(s);
}
It then imitate the implicit behaviour of the lock when it was
a Bkl and hadn't such dependency:
mutex_lock(m) {
if (fastpath)
let's go
else {
wait_for_mutex() {
schedule() {
unlock_kernel()
reacquire_lock_kernel()
}
}
}
}
The problem is that by using such explicit schedule(), we don't
benefit of the adaptive mutex spinning on owner.
The logic in use now is:
reiserfs_write_unlock(s);
mutex_lock(m); // -> possible adaptive spinning
reiserfs_write_lock(s);
[ Impact: restore the use of adaptive spinning mutexes in reiserfs ]
Cc: Jeff Mahoney <jeffm@suse.com>
Cc: Chris Mason <chris.mason@oracle.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Alexander Beregalov <a.beregalov@gmail.com>
Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
|
||
|
Frederic Weisbecker
|
d6f5b0aa08 |
kill-the-bkl/reiserfs: factorize the locking in reiserfs_write_end()
reiserfs_write_end() is a hot path in reiserfs. We have two wasteful write lock lock/release inside that can be gathered without changing the code logic. This patch factorizes them out in a single protected section, reducing the number of contentions inside. [ Impact: reduce lock contention in a reiserfs hotpath ] Cc: Jeff Mahoney <jeffm@suse.com> Cc: Chris Mason <chris.mason@oracle.com> Cc: Ingo Molnar <mingo@elte.hu> Cc: Alexander Beregalov <a.beregalov@gmail.com> Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com> |
||
|
Frederic Weisbecker
|
09eb47a7c5 |
kill-the-bkl/reiserfs: reduce number of contentions in search_by_key()
search_by_key() is a central function in reiserfs which searches the patch in the fs tree from the root to a node given its key. It is the function that is most requesting the write lock because it's a path very often used. Also we forget to release the lock while reading the next tree node, making us holding the lock in a wasteful way. Then we release the lock while reading the current node and its childs, all-in-one. It should be safe because we have a reference to these blocks and even if we read a block that will be concurrently changed, we have an fs_changed check later that will make us retry the path from the root. [ Impact: release the write lock while unused in a hot path ] Cc: Jeff Mahoney <jeffm@suse.com> Cc: Chris Mason <chris.mason@oracle.com> Cc: Ingo Molnar <mingo@elte.hu> Cc: Alexander Beregalov <a.beregalov@gmail.com> Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com> |
||
|
Frederic Weisbecker
|
b1c839bb2d |
kill-the-bkl/reiserfs: don't hold the write recursively in reiserfs_lookup()
The write lock can be acquired recursively in reiserfs_lookup(). But we may want to *really* release the lock before possible rescheduling from a reiserfs_lookup() callee. Hence we want to only acquire the lock once (ie: not recursively). [ Impact: prevent from possible false unreleased write lock on sleeping ] Cc: Jeff Mahoney <jeffm@suse.com> Cc: Chris Mason <chris.mason@oracle.com> Cc: Ingo Molnar <mingo@elte.hu> Cc: Alexander Beregalov <a.beregalov@gmail.com> Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com> |
||
|
Frederic Weisbecker
|
26931309a4 |
kill-the-bkl/reiserfs: lock only once on reiserfs_get_block()
reiserfs_get_block() is one of these sites where the write lock might be acquired recursively. It's a particular problem because this function is called very often. It's a hot spot which needs to reschedule() periodically while converting direct items to indirect ones because it can take some time. Then if we are applying the write lock release/reacquire pattern on schedule() here, it may not produce the desired effect since we may have locked in more than one depth. The solution is to use reiserfs_write_lock_once() which won't try to reacquire the lock recursively. Then the lock will be *really* released before schedule(). Also, we only release the lock if TIF_NEED_RESCHED is set to not create wasteful numerous contentions. [ Impact: fix a too long holded lock case in reiserfs_get_block() ] Cc: Jeff Mahoney <jeffm@suse.com> Cc: Chris Mason <chris.mason@oracle.com> Cc: Ingo Molnar <mingo@elte.hu> Cc: Alexander Beregalov <a.beregalov@gmail.com> Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com> |
||
|
Frederic Weisbecker
|
6e3647acb4 |
kill-the-BKL/reiserfs: release the write lock on flush_commit_list()
flush_commit_list() uses ll_rw_block() to commit the pending log blocks. ll_rw_block() might sleep, and the bkl was released at this point. Then we can also relax the write lock at this point. [ Impact: release the reiserfs write lock when it is not needed ] Cc: Jeff Mahoney <jeffm@suse.com> Cc: Chris Mason <chris.mason@oracle.com> Cc: Alexander Beregalov <a.beregalov@gmail.com> Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com> |
||
|
Frederic Weisbecker
|
4c5eface5d |
kill-the-BKL/reiserfs: release the write lock inside reiserfs_read_bitmap_block()
reiserfs_read_bitmap_block() uses sb_bread() to read the bitmap block. This helper might sleep. Then, when the bkl was used, it was released at this point. We can then relax the write lock too here. [ Impact: release the reiserfs write lock when it is not needed ] Cc: Jeff Mahoney <jeffm@suse.com> Cc: Chris Mason <chris.mason@oracle.com> Cc: Alexander Beregalov <a.beregalov@gmail.com> Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com> |
||
|
Frederic Weisbecker
|
148d3504c1 |
kill-the-BKL/reiserfs: release the write lock inside get_neighbors()
get_neighbors() is used to get the left and/or right blocks against a given one in order to balance a tree. sb_bread() is used to read the buffer of these neighors blocks and while it waits for this operation, it might sleep. The bkl was released at this point, and then we can also release the write lock before calling sb_bread(). This is safe because if the filesystem is changed after this lock release, the function returns REPEAT_SEARCH (aka SCHEDULE_OCCURRED in the function header comments) in order to repeat the neighbhor research. [ Impact: release the reiserfs write lock when it is not needed ] Cc: Jeff Mahoney <jeffm@suse.com> Cc: Chris Mason <chris.mason@oracle.com> Cc: Alexander Beregalov <a.beregalov@gmail.com> Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com> |
||
|
Frederic Weisbecker
|
5e69e3a449 |
kill-the-BKL/reiserfs: release write lock while rescheduling on prepare_for_delete_or_cut()
prepare_for_delete_or_cut() can process several types of items, including indirect items, ie: items which contain no file data but pointers to unformatted nodes scattering the datas of a file. In this case it has to zero out these pointers to block numbers of unformatted nodes and release the bitmap from these block numbers. It can take some time, so a rescheduling() is performed between each block processed. We can safely release the write lock while rescheduling(), like the bkl did, because the code checks just after if the item has moved after sleeping. [ Impact: release the reiserfs write lock when it is not needed ] Cc: Jeff Mahoney <jeffm@suse.com> Cc: Chris Mason <chris.mason@oracle.com> Cc: Alexander Beregalov <a.beregalov@gmail.com> Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com> |
||
|
Frederic Weisbecker
|
e6950a4da3 |
kill-the-BKL/reiserfs: release the write lock before rescheduling on do_journal_end()
When do_journal_end() copies data to the journal blocks buffers in memory, it reschedules if needed between each block copied and dirtyfied. We can also release the write lock at this rescheduling stage, like did the bkl implicitly. [ Impact: release the reiserfs write lock when it is not needed ] Cc: Jeff Mahoney <jeffm@suse.com> Cc: Chris Mason <chris.mason@oracle.com> Cc: Alexander Beregalov <a.beregalov@gmail.com> Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com> |
||
|
Frederic Weisbecker
|
dc8f6d8936 |
kill-the-BKL/reiserfs: only acquire the write lock once in reiserfs_dirty_inode
Impact: fix a deadlock
reiserfs_dirty_inode() is the super_operations::dirty_inode() callback
of reiserfs. It can be called from different contexts where the write
lock can be already held.
But this function also grab the write lock (possibly recursively).
Subsequent release of the lock before sleep will actually not release
the lock if the caller of mark_inode_dirty() (which in turn calls
reiserfs_dirty_inode()) already owns the lock.
A typical case:
reiserfs_write_end() {
acquire_write_lock()
mark_inode_dirty() {
reiserfs_dirty_inode() {
reacquire_write_lock() {
journal_begin() {
do_journal_begin_r() {
/*
* fail to release, still
* one depth of lock
*/
release_write_lock()
reiserfs_wait_on_write_block() {
wait_event()
The event is usually provided by something which needs the write lock but
it hasn't been released.
We use reiserfs_write_lock_once() here to ensure we only grab the
write lock in one level.
Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Alessio Igor Bogani <abogani@texware.it>
Cc: Jeff Mahoney <jeffm@suse.com>
Cc: Chris Mason <chris.mason@oracle.com>
LKML-Reference: <1239680065-25013-4-git-send-email-fweisbec@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
|
||
|
Frederic Weisbecker
|
22c963addc |
kill-the-BKL/reiserfs: lock only once in reiserfs_truncate_file
Impact: fix a deadlock
reiserfs_truncate_file() can be called from multiple context where
the write lock can be already hold or not.
This function also acquire (possibly recursively) the write
lock. Subsequent releases before sleeping will not actually release
the lock because we may be in more than one lock depth degree.
A typical case is:
reiserfs_file_release {
acquire_the_lock()
reiserfs_truncate_file()
reacquire_the_lock()
journal_begin() {
do_journal_begin_r() {
reiserfs_wait_on_write_block() {
/*
* Not released because still one
* depth owned
*/
release_lock()
wait_for_event()
At this stage the event never happen because the one which provides
it needs the write lock.
We use reiserfs_write_lock_once() here to ensure that we don't acquire the
write lock recursively.
Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Alessio Igor Bogani <abogani@texware.it>
Cc: Jeff Mahoney <jeffm@suse.com>
Cc: Alexander Beregalov <a.beregalov@gmail.com>
Cc: Chris Mason <chris.mason@oracle.com>
LKML-Reference: <1239680065-25013-3-git-send-email-fweisbec@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
|
||
|
Frederic Weisbecker
|
daf88c8983 |
kill-the-BKL/reiserfs: provide a tool to lock only once the write lock
Sometimes we don't want to recursively hold the per superblock write lock because we want to be sure it is actually released when we come to sleep. This patch introduces the necessary tools for that. reiserfs_write_lock_once() does the same job than reiserfs_write_lock() except that it won't try to acquire recursively the lock if the current task already owns it. Also the lock_depth before the call of this function is returned. reiserfs_write_unlock_once() unlock only if reiserfs_write_lock_once() returned a depth equal to -1, ie: only if it actually locked. Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com> Cc: Alessio Igor Bogani <abogani@texware.it> Cc: Jeff Mahoney <jeffm@suse.com> Cc: Alexander Beregalov <a.beregalov@gmail.com> Cc: Chris Mason <chris.mason@oracle.com> LKML-Reference: <1239680065-25013-2-git-send-email-fweisbec@gmail.com> Signed-off-by: Ingo Molnar <mingo@elte.hu> |
||
|
Frederic Weisbecker
|
a412f9efdd |
reiserfs, kill-the-BKL: fix unsafe j_flush_mutex lock
Impact: fix a deadlock The j_flush_mutex is acquired safely in journal.c: if we can't take it, we free the reiserfs per superblock lock and wait a bit. But we have a remaining place in kupdate_transactions() where j_flush_mutex is still acquired traditionnaly. Thus the following scenario (warned by lockdep) can happen: A B mutex_lock(&write_lock) mutex_lock(&write_lock) mutex_lock(&j_flush_mutex) mutex_lock(&j_flush_mutex) //block mutex_unlock(&write_lock) sleep... mutex_lock(&write_lock) //deadlock Fix this by using reiserfs_mutex_lock_safe() in kupdate_transactions(). Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com> Cc: Alessio Igor Bogani <abogani@texware.it> Cc: Jeff Mahoney <jeffm@suse.com> LKML-Reference: <1239660635-12940-1-git-send-email-fweisbec@gmail.com> Signed-off-by: Ingo Molnar <mingo@elte.hu> |
||
|
Frederic Weisbecker
|
8ebc423238 |
reiserfs: kill-the-BKL
This patch is an attempt to remove the Bkl based locking scheme from reiserfs and is intended. It is a bit inspired from an old attempt by Peter Zijlstra: http://lkml.indiana.edu/hypermail/linux/kernel/0704.2/2174.html The bkl is heavily used in this filesystem to prevent from concurrent write accesses on the filesystem. Reiserfs makes a deep use of the specific properties of the Bkl: - It can be acqquired recursively by a same task - It is released on the schedule() calls and reacquired when schedule() returns The two properties above are a roadmap for the reiserfs write locking so it's very hard to simply replace it with a common mutex. - We need a recursive-able locking unless we want to restructure several blocks of the code. - We need to identify the sites where the bkl was implictly relaxed (schedule, wait, sync, etc...) so that we can in turn release and reacquire our new lock explicitly. Such implicit releases of the lock are often required to let other resources producer/consumer do their job or we can suffer unexpected starvations or deadlocks. So the new lock that replaces the bkl here is a per superblock mutex with a specific property: it can be acquired recursively by a same task, like the bkl. For such purpose, we integrate a lock owner and a lock depth field on the superblock information structure. The first axis on this patch is to turn reiserfs_write_(un)lock() function into a wrapper to manage this mutex. Also some explicit calls to lock_kernel() have been converted to reiserfs_write_lock() helpers. The second axis is to find the important blocking sites (schedule...(), wait_on_buffer(), sync_dirty_buffer(), etc...) and then apply an explicit release of the write lock on these locations before blocking. Then we can safely wait for those who can give us resources or those who need some. Typically this is a fight between the current writer, the reiserfs workqueue (aka the async commiter) and the pdflush threads. The third axis is a consequence of the second. The write lock is usually on top of a lock dependency chain which can include the journal lock, the flush lock or the commit lock. So it's dangerous to release and trying to reacquire the write lock while we still hold other locks. This is fine with the bkl: T1 T2 lock_kernel() mutex_lock(A) unlock_kernel() // do something lock_kernel() mutex_lock(A) -> already locked by T1 schedule() (and then unlock_kernel()) lock_kernel() mutex_unlock(A) .... This is not fine with a mutex: T1 T2 mutex_lock(write) mutex_lock(A) mutex_unlock(write) // do something mutex_lock(write) mutex_lock(A) -> already locked by T1 schedule() mutex_lock(write) -> already locked by T2 deadlock The solution in this patch is to provide a helper which releases the write lock and sleep a bit if we can't lock a mutex that depend on it. It's another simulation of the bkl behaviour. The last axis is to locate the fs callbacks that are called with the bkl held, according to Documentation/filesystem/Locking. Those are: - reiserfs_remount - reiserfs_fill_super - reiserfs_put_super Reiserfs didn't need to explicitly lock because of the context of these callbacks. But now we must take care of that with the new locking. After this patch, reiserfs suffers from a slight performance regression (for now). On UP, a high volume write with dd reports an average of 27 MB/s instead of 30 MB/s without the patch applied. Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com> Reviewed-by: Ingo Molnar <mingo@elte.hu> Cc: Jeff Mahoney <jeffm@suse.com> Cc: Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: Bron Gondwana <brong@fastmail.fm> Cc: Andrew Morton <akpm@linux-foundation.org> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Alexander Viro <viro@zeniv.linux.org.uk> LKML-Reference: <1239070789-13354-1-git-send-email-fweisbec@gmail.com> Signed-off-by: Ingo Molnar <mingo@elte.hu> |
||
|
Alexey Dobriyan
|
405f55712d |
headers: smp_lock.h redux
* Remove smp_lock.h from files which don't need it (including some headers!) * Add smp_lock.h to files which do need it * Make smp_lock.h include conditional in hardirq.h It's needed only for one kernel_locked() usage which is under CONFIG_PREEMPT This will make hardirq.h inclusion cheaper for every PREEMPT=n config (which includes allmodconfig/allyesconfig, BTW) Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org> |
||
Jens Axboe
|
8aa7e847d8 |
Fix congestion_wait() sync/async vs read/write confusion
Commit
|
||
|
Alexey Dobriyan
|
b43f3cbd21 |
headers: mnt_namespace.h redux
Fix various silly problems wrt mnt_namespace.h: - exit_mnt_ns() isn't used, remove it - done that, sched.h and nsproxy.h inclusions aren't needed - mount.h inclusion was need for vfsmount_lock, but no longer - remove mnt_namespace.h inclusion from files which don't use anything from mnt_namespace.h Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org> |