Re: [PATCH RFC 0/3] md/raid1: data corruption with serialization

On Thu, Feb 5, 2026 at 9:44 AM Yu Kuai <[email protected]> wrote:
>
> 在 2026/2/5 8:34, Xiao Ni 写道:
> > I know the reason why it needs serialization for writemostly case. At
> > first, I asked the same question if we need to handle overlap if a
> > filesystem is on raid. I got the answer from AI, filesystem can submit
> > overlap bios simultaneously, maybe it's the reason why
> > serialize_policy was introduced? And database is another case which
> > can write to raid1 directly without a filesystem. So I started this
> > job. If we don't need to consider overlap, why do we need
> > serialize_policy? cc guoqing who is the author of serialize_policy
> > also.
>
> Please don't just trust answer from AI, filesystem can't submit overlap

Thanks for pointing out this.

> bios concurrently. And I already said serialize_policy is for writemostly
> case.

de31ee949739aba9ce7dbb8b10e72c6fce0e76c7 (md: reorgnize
mddev_create/destroy_serial_pool). serialize_policy is not for
writemostly.

>
> And I also said we don't need to consider user manage data on raw disk
> and write to same area concurrently, database will not do this.
>
> The reason is not complicated, if you write to the same LBA concurrent,
> it's pointless because you don't know what you'll read after that.

Thanks for the explanation.

>
> --
> Thansk,
> Kuai
>