On 11/24/23 9:29 AM, Song Liu wrote:

On Wed, Nov 8, 2023 at 10:22 AM Junxiao Bi [off-list ref] wrote:

quoted

This reverts commit 5e2cf333b7bd5d3e62595a44d598a254c697cd74.

That commit introduced the following race and can cause system hung.

  md_write_start:             raid5d:
  // mddev->in_sync == 1
  set "MD_SB_CHANGE_PENDING"
                             // running before md_write_start wakeup it
                              waiting "MD_SB_CHANGE_PENDING" cleared
                              >>>>>>>>> hung
  wakeup mddev->thread
  ...
  waiting "MD_SB_CHANGE_PENDING" cleared
  >>>> hung, raid5d should clear this flag
  but get hung by same flag.

The issue reverted commit fixing is fixed by last patch in a new way.

Fixes: 5e2cf333b7bd ("md/raid5: Wait for MD_SB_CHANGE_PENDING in raid5d")
Signed-off-by: Junxiao Bi <redacted>

The set looks good to me. Thanks!

Thanks for the review.

Quick question: from the earlier thread, the issue was observed in
production. Have you reproduced the issue and thus verified the fix
works as expected?

I didn't try reproducing this since the system hung on the code where 
the bad commit added, after revert it, this issue will not reproduce any 
more.

Thanks,

Junxiao.

Thanks,
Song