Re: [syzbot] [cgroups?] [mm?] WARNING in memcg1_swapout

[Johannes Weiner]

On Wed, Jan 21, 2026 at 08:35:26AM -0800, Andrew Morton wrote:
> On Wed, 21 Jan 2026 09:58:42 -0500 Johannes Weiner <hannes@xxxxxxxxxxx> wrote:
> 
> > > > I tested with the latest linux-next in sysbot. It is working fine
> > > 
> > > Great, thanks.  But we still don't have review for this one.
> > 
> > IOW, this is not necessary anymore. Kairui's (cc'd) fix which you
> > picked up fixed the syzbot reported problem.
> > 
> > > For some reason I don't have cc:stable on this - could people
> > > make a recommendation?
> > 
> > So this:
> > 
> > > From: Deepanshu Kartikey <kartikey406@xxxxxxxxx>
> > > Subject: mm/swap_cgroup: fix kernel BUG in swap_cgroup_record
> > 
> > can be dropped.
> > 
> 
> OK, thanks, I'll drop Deepanshu's "mm/swap_cgroup: fix kernel BUG in
> swap_cgroup_record".  I understand that Kairui's
> mm-swap-use-swap-cache-as-the-swap-in-synchronize-layer-fix
> (https://lkml.kernel.org/r/CAMgjq7CGUnzOVG7uSaYjzw9wD7w2dSKOHprJfaEp4CcGLgE3iw@xxxxxxxxxxxxxx)
> has addressed this mm-unstable issue.

Yes.

> However Deepanshu's "mm/swap_cgroup: fix kernel BUG in
> swap_cgroup_record" has
> 
> 	Fixes: 1a4e58cce84e ("mm: introduce MADV_PAGEOUT")
> 
> Which I assume was mistaken?

Yes, this appeared to be a misdiagnosis of the bug sequence.
MADV_PAGEOUT was in the stacktrace, but not the culprit.