On Fri, Nov 22, 2024 at 10:46 AM Yu Kuai <yukuai1@xxxxxxxxxxxxxxx> wrote: > > Hi, > > 在 2024/11/22 10:06, Xiao Ni 写道: > > On Mon, Nov 18, 2024 at 7:44 PM Yu Kuai <yukuai1@xxxxxxxxxxxxxxx> wrote: > >> > >> From: Yu Kuai <yukuai3@xxxxxxxxxx> > >> > >> There are two BUG reports that raid5 will hang at > >> bitmap_startwrite([1],[2]), root cause is that bitmap start write and end > >> write is unbalanced, and while reviewing raid5 code, it's found that > > > > Hi Kuai > > > > It's better to describe more about "unbalanced" in the patch. For > > raid5, bitmap is set and cleared based on stripe->dev[] now. It looks > > like the set operation matches the clear operation already. > > Ok, one place that I found is that raid5 can do extra end write while > stripe->dev[].towrite is NULL, the null checking is missing. I'll > mention that in the next version. Does this can cause the deadlock? Regards Xiao > > ... > > > > > This patch looks good to me. > > > > Reviewd-by: Xiao Ni <xni@xxxxxxxxxx> > > > > Thanks for the review. I'll also remove the unused STRIPE_BITMAP_PENDING > in v2. > > Kuai > > > > > . > > >
