diff options
| author | Tejun Heo <tj@kernel.org> | 2021-07-28 01:12:20 +0200 |
|---|---|---|
| committer | Tejun Heo <tj@kernel.org> | 2021-07-28 01:12:20 +0200 |
| commit | c3df5fb57fe8756d67fd56ed29da65cdfde839f9 (patch) | |
| tree | 58e8de70658fb85ae8e451194bd9aa742618acec /kernel/cgroup/rstat.c | |
| parent | cgroup1: fix leaked context root causing sporadic NULL deref in LTP (diff) | |
| download | linux-c3df5fb57fe8756d67fd56ed29da65cdfde839f9.tar.xz linux-c3df5fb57fe8756d67fd56ed29da65cdfde839f9.zip | |
cgroup: rstat: fix A-A deadlock on 32bit around u64_stats_sync
0fa294fb1985 ("cgroup: Replace cgroup_rstat_mutex with a spinlock") added
cgroup_rstat_flush_irqsafe() allowing flushing to happen from the irq
context. However, rstat paths use u64_stats_sync to synchronize access to
64bit stat counters on 32bit machines. u64_stats_sync is implemented using
seq_lock and trying to read from an irq context can lead to A-A deadlock if
the irq happens to interrupt the stat update.
Fix it by using the irqsafe variants - u64_stats_update_begin_irqsave() and
u64_stats_update_end_irqrestore() - in the update paths. Note that none of
this matters on 64bit machines. All these are just for 32bit SMP setups.
Note that the interface was introduced way back, its first and currently
only use was recently added by 2d146aa3aa84 ("mm: memcontrol: switch to
rstat"). Stable tagging targets this commit.
Signed-off-by: Tejun Heo <tj@kernel.org>
Reported-by: Rik van Riel <riel@surriel.com>
Fixes: 2d146aa3aa84 ("mm: memcontrol: switch to rstat")
Cc: stable@vger.kernel.org # v5.13+
Diffstat (limited to 'kernel/cgroup/rstat.c')
| -rw-r--r-- | kernel/cgroup/rstat.c | 19 |
1 files changed, 11 insertions, 8 deletions
diff --git a/kernel/cgroup/rstat.c b/kernel/cgroup/rstat.c index 7f0e58917432..b264ab5652ba 100644 --- a/kernel/cgroup/rstat.c +++ b/kernel/cgroup/rstat.c @@ -347,19 +347,20 @@ static void cgroup_base_stat_flush(struct cgroup *cgrp, int cpu) } static struct cgroup_rstat_cpu * -cgroup_base_stat_cputime_account_begin(struct cgroup *cgrp) +cgroup_base_stat_cputime_account_begin(struct cgroup *cgrp, unsigned long *flags) { struct cgroup_rstat_cpu *rstatc; rstatc = get_cpu_ptr(cgrp->rstat_cpu); - u64_stats_update_begin(&rstatc->bsync); + *flags = u64_stats_update_begin_irqsave(&rstatc->bsync); return rstatc; } static void cgroup_base_stat_cputime_account_end(struct cgroup *cgrp, - struct cgroup_rstat_cpu *rstatc) + struct cgroup_rstat_cpu *rstatc, + unsigned long flags) { - u64_stats_update_end(&rstatc->bsync); + u64_stats_update_end_irqrestore(&rstatc->bsync, flags); cgroup_rstat_updated(cgrp, smp_processor_id()); put_cpu_ptr(rstatc); } @@ -367,18 +368,20 @@ static void cgroup_base_stat_cputime_account_end(struct cgroup *cgrp, void __cgroup_account_cputime(struct cgroup *cgrp, u64 delta_exec) { struct cgroup_rstat_cpu *rstatc; + unsigned long flags; - rstatc = cgroup_base_stat_cputime_account_begin(cgrp); + rstatc = cgroup_base_stat_cputime_account_begin(cgrp, &flags); rstatc->bstat.cputime.sum_exec_runtime += delta_exec; - cgroup_base_stat_cputime_account_end(cgrp, rstatc); + cgroup_base_stat_cputime_account_end(cgrp, rstatc, flags); } void __cgroup_account_cputime_field(struct cgroup *cgrp, enum cpu_usage_stat index, u64 delta_exec) { struct cgroup_rstat_cpu *rstatc; + unsigned long flags; - rstatc = cgroup_base_stat_cputime_account_begin(cgrp); + rstatc = cgroup_base_stat_cputime_account_begin(cgrp, &flags); switch (index) { case CPUTIME_USER: @@ -394,7 +397,7 @@ void __cgroup_account_cputime_field(struct cgroup *cgrp, break; } - cgroup_base_stat_cputime_account_end(cgrp, rstatc); + cgroup_base_stat_cputime_account_end(cgrp, rstatc, flags); } /* |