bdi: make sure congestion states are clear on free
Subsystems (5)
PATCH 1/2 bdi: make sure congestion states are clear on free
FUSE has a bug where it fails to clear congestion states if a connection gets aborted while congested, which can leave nr_wb_congested[] stuck until reboot causing wait_iff_congested() to wait spuriously.
While the bdi owner, FUSE, is primarily responsible for clearing congestion states before destroying bdi_writebacks, bdi layer can ensure that congestion states are not leaked beyond bdi_writeback lifecycle.
--- include/linux/backing-dev.h | 14 +++++++++++++- mm/backing-dev.c | 2 +- 2 files changed, 14 insertions(+), 2 deletions(-)
include/linux/backing-dev.h +13 -1
@@ -220,6 +220,18 @@ static inline int bdi_sched_wait(void *wreturn 0 ; } +static inline void __wb_congested_free (struct bdi_writeback_congested *congested) +{ +/* +* Make sure congestion states are cleared before freeing to avoid +* nr_wb_congested() corruption which can lead to misbehaving +* wait_iff_congested(). +*/ +clear_wb_congested (congested,BLK_RW_SYNC ); +clear_wb_congested (congested,BLK_RW_ASYNC ); +kfree (congested); +} +#ifdef CONFIG_CGROUP_WRITEBACK struct bdi_writeback_congested *
@@ -409,7 +421,7 @@ wb_congested_get_create(struct backing_dstatic inline void wb_congested_put (struct bdi_writeback_congested *congested) {if (atomic_dec_and_test (&congested->refcnt)) -kfree (congested); +__wb_congested_free (congested); }static inline struct bdi_writeback *wb_find_current (struct backing_dev_info *bdi)
mm/backing-dev.c +1 -1
@@ -509,7 +509,7 @@ void wb_congested_put(struct bdi_writeba }spin_unlock_irqrestore (&cgwb_lock, flags); -kfree (congested); +__wb_congested_free (congested); }static void cgwb_release_workfn (struct work_struct *work)
PATCH 2/2 FUSE: fix congested state leak on aborted connections
If a connection gets aborted while congested, FUSE can leave nr_wb_congested[] stuck until reboot causing wait_iff_congested() to wait spuriously which can lead to severe performance degradation.
The leak is caused by gating congestion state clearing with fc->connected test in request_end(). This was added way back in 2009 by 26c3679101db ("fuse: destroy bdi on umount"). While the commit description doesn't explain why the test was added, it most likely was to avoid dereferencing bdi after it got destroyed.
Since then, bdi lifetime rules have changed many times and now we're always guaranteed to have access to the bdi while the superblock is alive (fc->sb).
Drop fc->connected conditional to avoid leaking congestion states.
--- fs/fuse/dev.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-)
fs/fuse/dev.c +1 -2
@@ -381,8 +381,7 @@ static void request_end(struct fuse_connif (!fc->blocked &&waitqueue_active (&fc->blocked_waitq))wake_up (&fc->blocked_waitq); -if (fc->num_background == fc->congestion_threshold && - fc->connected && fc->sb) { +if (fc->num_background == fc->congestion_threshold && fc->sb) {clear_bdi_congested (fc->sb->s_bdi,BLK_RW_SYNC );clear_bdi_congested (fc->sb->s_bdi,BLK_RW_ASYNC ); }
Quoted 10 lines · Tejun Heo
FUSE has a bug where it fails to clear congestion states if a connection gets aborted while congested, which can leave nr_wb_congested[] stuck until reboot causing
wait_iff_congested()to wait spuriously.While the bdi owner, FUSE, is primarily responsible for clearing congestion states before destroying
bdi_writebacks, bdi layer can ensure that congestion states are not leaked beyondbdi_writebacklifecycle.
Quoted 5 lines
Signed-off-by Tejun Heo <[email protected]>Reported-by Joshua Miller <[email protected]>Cc Johannes Weiner <[email protected]>Cc Jan Kara <[email protected]>
Quoted 10 lines · Tejun Heo
FUSE has a bug where it fails to clear congestion states if a connection gets aborted while congested, which can leave nr_wb_congested[] stuck until reboot causing
wait_iff_congested()to wait spuriously.While the bdi owner, FUSE, is primarily responsible for clearing congestion states before destroying
bdi_writebacks, bdi layer can ensure that congestion states are not leaked beyondbdi_writebacklifecycle.
Quoted 5 lines
Signed-off-by Tejun Heo <[email protected]>Reported-by Joshua Miller <[email protected]>Cc Johannes Weiner <[email protected]>Cc Jan Kara <[email protected]>
Looks good. You can add:
Honza
Quoted 46 lines
--- include/linux/backing-dev.h | 14 +++++++++++++- mm/backing-dev.c | 2 +- 2 files changed, 14 insertions(+), 2 deletions(-)include/linux/backing-dev.h +13
-1@@ -220,6 +220,18 @@ static inline int bdi_sched_wait(void *wreturn 0 ; } +static inline void __wb_congested_free (struct bdi_writeback_congested *congested) +{ +/* +* Make sure congestion states are cleared before freeing to avoid +* nr_wb_congested() corruption which can lead to misbehaving +* wait_iff_congested(). +*/ +clear_wb_congested (congested,BLK_RW_SYNC ); +clear_wb_congested (congested,BLK_RW_ASYNC ); +kfree (congested); +} +#ifdef CONFIG_CGROUP_WRITEBACK struct bdi_writeback_congested *@@ -409,7 +421,7 @@ wb_congested_get_create(struct backing_dstatic inline void wb_congested_put (struct bdi_writeback_congested *congested) {if (atomic_dec_and_test (&congested->refcnt)) -kfree (congested); +__wb_congested_free (congested); }static inline struct bdi_writeback *wb_find_current (struct backing_dev_info *bdi)mm/backing-dev.c +1
-1@@ -509,7 +509,7 @@ void wb_congested_put(struct bdi_writeba }spin_unlock_irqrestore (&cgwb_lock, flags); -kfree (congested); +__wb_congested_free (congested); }static void cgwb_release_workfn (struct work_struct *work)
Signature · 2 lines
Jan Kara <[email protected]> SUSE Labs, CR
Re: [PATCH 2/2] FUSE: fix congested state leak on aborted connections
In reply to Tejun HeoQuoted 14 lines · Tejun Heo
If a connection gets aborted while congested, FUSE can leave nr_wb_congested[] stuck until reboot causing
wait_iff_congested()to wait spuriously which can lead to severe performance degradation.The leak is caused by gating congestion state clearing with
fc->connectedtest inrequest_end(). This was added way back in 2009 by26c3679101db("fuse: destroy bdi on umount"). While the commit description doesn't explain why the test was added, it most likely was to avoid dereferencing bdi after it got destroyed.Since then, bdi lifetime rules have changed many times and now we're always guaranteed to have access to the bdi while the superblock is alive (
fc->sb).
Quoted 8 lines
Drop
fc->connectedconditional to avoid leaking congestion states.Signed-off-by Tejun Heo <[email protected]>Reported-by Joshua Miller <[email protected]>Cc Johannes Weiner <[email protected]>Cc Miklos Szeredi <[email protected]>Cc Jan Kara <[email protected]>
Yeah, this should be fine AFAICT but my knowledge of FUSE is very cursory. Anyway:
Honza
Quoted 16 lines
--- fs/fuse/dev.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-)fs/fuse/dev.c +1
-2@@ -381,8 +381,7 @@ static void request_end(struct fuse_connif (!fc->blocked &&waitqueue_active (&fc->blocked_waitq))wake_up (&fc->blocked_waitq); -if (fc->num_background == fc->congestion_threshold && - fc->connected && fc->sb) { +if (fc->num_background == fc->congestion_threshold && fc->sb) {clear_bdi_congested (fc->sb->s_bdi,BLK_RW_SYNC );clear_bdi_congested (fc->sb->s_bdi,BLK_RW_ASYNC ); }
Signature · 2 lines
Jan Kara <[email protected]> SUSE Labs, CR
Miklos Szeredi
maintainerRe: [PATCH 2/2] FUSE: fix congested state leak on aborted connections
Quoted 24 lines · Jan Kara
Quoted 14 lines · Tejun Heo
If a connection gets aborted while congested, FUSE can leave nr_wb_congested[] stuck until reboot causing
wait_iff_congested()to wait spuriously which can lead to severe performance degradation.The leak is caused by gating congestion state clearing with
fc->connectedtest inrequest_end(). This was added way back in 2009 by26c3679101db("fuse: destroy bdi on umount"). While the commit description doesn't explain why the test was added, it most likely was to avoid dereferencing bdi after it got destroyed.Since then, bdi lifetime rules have changed many times and now we're always guaranteed to have access to the bdi while the superblock is alive (
fc->sb).Quoted 8 lines
Drop
fc->connectedconditional to avoid leaking congestion states.Signed-off-by Tejun Heo <[email protected]>Reported-by Joshua Miller <[email protected]>Cc Johannes Weiner <[email protected]>Cc Miklos Szeredi <[email protected]>Cc Jan Kara <[email protected]>
Quoted 4 lines
Yeah, this should be fine AFAICT but my knowledge of FUSE is very cursory. Anyway:
Acked-by Jan Kara <[email protected]>
Can't say I fully understand how the global "is any bdi congested" state is used in direct reclaim, but the patch is an obvious improvement, so applied.
Thanks,
Miklos