Re: [Cluster-devel] [PATCH] gfs2: fix race between gfs2_freeze_func and unmount

2019-04-30 Thread Abhijith Das
NACK. Andreas mentioned that the description could be more descriptive and that we should be using clear_bit_unlock() instead of clear_bit(). I'll post a v2 shortly with these changes. Cheers! --Abhi On Tue, Apr 30, 2019 at 12:48 PM Abhi Das wrote: > gfs2_unfreee() doesn't wait for

[Cluster-devel] [PATCH] gfs2: fix race between gfs2_freeze_func and unmount

2019-04-30 Thread Abhi Das
gfs2_unfreee() doesn't wait for gfs2_freeze_func() to complete. If a umount is issued right after unfreeze, it could result in an inconsistent filesystem because some journal data (statfs update) wasn't written out. This patch causes gfs2_unfreeze() to wait for gfs2_freeze_func() to complete