Re: [PATCH v2 2/2] mm: zswap: Tie per-CPU acomp_ctx lifetime to the pool.

[Yosry Ahmed]

On Fri, Mar 27, 2026 at 12:30 PM Kanchana P. Sridhar
<kanchanapsridhar2026@xxxxxxxxx> wrote:
>
> On Thu, Mar 26, 2026 at 7:23 PM Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx> wrote:
> >
> > On Mon, 16 Mar 2026 18:48:02 -0700 "Kanchana P. Sridhar" <kanchanapsridhar2026@xxxxxxxxx> wrote:
> >
> > > Currently, per-CPU acomp_ctx are allocated on pool creation and/or CPU
> > > hotplug, and destroyed on pool destruction or CPU hotunplug. This
> > > complicates the lifetime management to save memory while a CPU is
> > > offlined, which is not very common.
> > >
> > > Simplify lifetime management by allocating per-CPU acomp_ctx once on
> > > pool creation (or CPU hotplug for CPUs onlined later), and keeping them
> > > allocated until the pool is destroyed.
> > >
> > > ...
> > >
> >
> > This is a tricky-looking patch and I haven't yet recorded any reviews,
> > so could someone please dig in?
>
> Hi Andrew, Yosry,
>
> To provide some background, this patch [1] in my v14 patch-series was
> Acked-by Yosry, with a minor change requested [2] that is addressed by
> the current patch.
>
> Since Sashiko flagged the same issue as [2], I did not automatically
> carry forward the Acked-by.
> Yosry, can you please confirm if your Acked-by and comments in [2] are
> addressed in this current patch?

The patch looks good to me, feel free to add:

Acked-by: Yosry Ahmed <yosry@xxxxxxxxxx>

That being said, it might be a good idea to resend the patches (new
version for rebase + Ack?) to get Sashiko to take a look, hopefully it
will be able to apply/review the patches this time.