Re: [PATCH] mm/slab: initialize slab->stride early to avoid memory ordering issues

[Harry Yoo <harry.yoo@oracle.com>]

On Wed, Feb 25, 2026 at 02:44:24PM +0530, Venkat Rao Bagalkote wrote:
> > > Thanks for the patch. I did ran the complete test suite, and unfortunately
> > > issue is reproducing.
>
> > Oops, thanks for confirming that it's still reproduced!
> > That's really helpful.
> > 
> > Perhaps I should start considering cases where it's not a memory
> > ordering issue, but let's check one more thing before moving on.
> > could you please test if it still reproduces with the following patch?
> > 
> > If it's still reproducible, it should not be due to the memory ordering
> > issue between obj_exts and stride.
> > 
> > ---8<---
> > From: Harry Yoo <harry.yoo@oracle.com>
> > Date: Mon, 23 Feb 2026 16:58:09 +0900
> > Subject: mm/slab: enforce slab->stride -> slab->obj_exts ordering
> > 
> > I tried to avoid unnecessary memory barriers for efficiency,
> > but the original bug is still reproducible.
> > 
> > Probably I missed a case where an object is allocated on a CPU
> > and then freed on a different CPU without involving spinlock.
> > 
> > I'm not sure if I did not cover edge cases or if it's caused by
> > something other than memory ordering issue.
> > 
> > Anyway, let's find out by introducing heavy memory barriers!
> > 
> > Always ensure that updates to stride is visible before obj_exts.
> > 
> > ---

[...]

> With this patch, issue is not reproduced. So looks good.

Thanks a lot, Venkat! That's really helpful.

I think that's enough signal to assume that memory ordering is playing
a role here, unless it happens to be masking another issue.
Even so, it's important to enforce the ordering anyway.

But having smp_load_acquire() on every alloc/free fastpath doesn't
sound great to me. Let me think a bit about it and come up with
a reasonable solution (this time, hopefully no hole in the ordering).

Since it's a bug I'm working on it with high priority.

Again, thanks a lot for testing!

-- 
Cheers,
Harry / Hyeonggon