Linux kernel mirror (for testing)
git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git
kernel
os
linux
Merge tag 'wq-for-7.0-rc6-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/tj/wq
Pull workqueue fix from Tejun Heo:
- Fix false positive stall reports on weakly ordered architectures
where the lockless worklist/timestamp check in the watchdog can
observe stale values due to memory reordering.
Recheck under pool->lock to confirm.
* tag 'wq-for-7.0-rc6-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/tj/wq:
workqueue: Better describe stall check
workqueue: Fix false positive stall reports
+22
-3
+22
-3
kernel/workqueue.c
+22
-3
kernel/workqueue.c
···
7699
7699
else
7700
7700
ts = touched;
7701
7701
7702
-
/* did we stall? */
7702
+
/*
7703
+
* Did we stall?
7704
+
*
7705
+
* Do a lockless check first to do not disturb the system.
7706
+
*
7707
+
* Prevent false positives by double checking the timestamp
7708
+
* under pool->lock. The lock makes sure that the check reads
7709
+
* an updated pool->last_progress_ts when this CPU saw
7710
+
* an already updated pool->worklist above. It seems better
7711
+
* than adding another barrier into __queue_work() which
7712
+
* is a hotter path.
7713
+
*/
7703
7714
if (time_after(now, ts + thresh)) {
7715
+
scoped_guard(raw_spinlock_irqsave, &pool->lock) {
7716
+
pool_ts = pool->last_progress_ts;
7717
+
if (time_after(pool_ts, touched))
7718
+
ts = pool_ts;
7719
+
else
7720
+
ts = touched;
7721
+
}
7722
+
if (!time_after(now, ts + thresh))
7723
+
continue;
7724
+
7704
7725
lockup_detected = true;
7705
7726
stall_time = jiffies_to_msecs(now - pool_ts) / 1000;
7706
7727
max_stall_time = max(max_stall_time, stall_time);
···
7733
7712
pr_cont_pool_info(pool);
7734
7713
pr_cont(" stuck for %us!\n", stall_time);
7735
7714
}
7736
-
7737
-
7738
7715
}
7739
7716
7740
7717
if (lockup_detected)