Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
Date:   Tue, 30 Aug 2022 20:59:32 +0200
From:   Sebastian Andrzej Siewior <bigeasy@linutronix.de>
To:     Maurizio Lombardi <mlombard@redhat.com>
Cc:     Vlastimil Babka <vbabka@suse.cz>, linux-mm <linux-mm@kvack.org>,
        David Rientjes <rientjes@google.com>,
        Pekka Enberg <penberg@kernel.org>,
        Joonsoo Kim <iamjoonsoo.kim@lge.com>,
        Andrew Morton <akpm@linux-foundation.org>,
        LKML <linux-kernel@vger.kernel.org>, tglx@linutronix.de
Subject: Re: [PATCH RFC] mm: slub: fix flush_cpu_slab()/__free_slab()
 invocations in task context.
Message-ID: <Yw5eFGRreMA0I95S@linutronix.de>
References: <20220829154805.1205507-1-mlombard@redhat.com>
 <Yw3lXaDTfvJcEM4Q@linutronix.de>
 <CAFL455=4SBhJ4LpiPQr+PmL2ShuTpR=bAo8JVfe-2x3cg85cbQ@mail.gmail.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Disposition: inline
In-Reply-To: <CAFL455=4SBhJ4LpiPQr+PmL2ShuTpR=bAo8JVfe-2x3cg85cbQ@mail.gmail.com>
Precedence: bulk

On 2022-08-30 17:48:06 [+0200], Maurizio Lombardi wrote:
> Hmm, this is not good indeed. I guess I should have used for_each_online_cpu()
> instead of on_each_cpu_cond().

But you must not invoked that function cross-CPU since the code expects
to access the per-CPU variables always from the local-CPU.

> >
> > Couldn't we instead use a workqueue with that WQ_MEM_RECLAIM bit? It may
> > reclaim memory after all ;)
> 
> That should also fix it, do you think it would be ok to allocate a workqueue in
> in kmem_cache_init() ?

I guess so. There is probably none that is already available.

> Thanks,
> Maurizio

Sebastian