Re: 5.14.9 aarch64 OOPS Workqueue: btrfs-delalloc btrfs_work_helper

[Nikolay Borisov <nborisov@suse.com>]

On 25.10.21 г. 22:40, Chris Murphy wrote:
> Got another sysrq+t here, while dnf is completely hung while 'dnf
> install kernel-debuginfo' packages, for a long time without any call
> traces or indication why it's stuck. ps aux shows it's running, but
> consuming no meaningful cpu; top shows very high ~25% wa, the rest is
> idle. Essentially no user or system process consumption.
> 
> https://bugzilla.redhat.com/attachment.cgi?id=1836995
> 

<snip>


I think I identified a race that could cause the crash, can you apply the 
following diff and re-run the tests and leave them for a couple of days. 
Preferably apply it on 5.4.10 so that there is the highest chance to reproduce: 

diff --git a/fs/btrfs/async-thread.c b/fs/btrfs/async-thread.c
index 309516e6a968..a3d788dcbd34 100644
--- a/fs/btrfs/async-thread.c
+++ b/fs/btrfs/async-thread.c
@@ -234,6 +234,11 @@ static void run_ordered_work(struct __btrfs_workqueue *wq,
                                  ordered_list);
                if (!test_bit(WORK_DONE_BIT, &work->flags))
                        break;
+               /*
+                * Orders all subsequent loads after WORK_DONE_BIT, paired with
+                * the smp_mb__before_atomic in btrfs_work_helper
+                */
+               smp_rmb();
 
                /*
                 * we are going to call the ordered done function, but
@@ -317,6 +322,12 @@ static void btrfs_work_helper(struct work_struct *normal_work)
        thresh_exec_hook(wq);
        work->func(work);
        if (need_order) {
+               /*
+                * Ensures all вритес done in ->func are ordered before
+                * setting the WORK_DONE_BIT making them visible to ordered
+                * func
+                */
+               smp_mb__before_atomic();
                set_bit(WORK_DONE_BIT, &work->flags);
                run_ordered_work(wq, work);
        } else {