linux-kernel - Re: [PATCH v4 13/22] slab: remove the do_slab

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <CAADnVQLEupQB3O1RzFAPoVvg-frojmjFQhisAUcxTJOmkz02nw@mail.gmail.com>
Date: Fri, 23 Jan 2026 10:15:35 -0800
From: Alexei Starovoitov <alexei.starovoitov@...il.com>
To: Vlastimil Babka <vbabka@...e.cz>
Cc: Harry Yoo <harry.yoo@...cle.com>, Petr Tesarik <ptesarik@...e.com>, 
	Christoph Lameter <cl@...two.org>, David Rientjes <rientjes@...gle.com>, 
	Roman Gushchin <roman.gushchin@...ux.dev>, Hao Li <hao.li@...ux.dev>, 
	Andrew Morton <akpm@...ux-foundation.org>, Uladzislau Rezki <urezki@...il.com>, 
	"Liam R. Howlett" <Liam.Howlett@...cle.com>, Suren Baghdasaryan <surenb@...gle.com>, 
	Sebastian Andrzej Siewior <bigeasy@...utronix.de>, Alexei Starovoitov <ast@...nel.org>, linux-mm <linux-mm@...ck.org>, 
	LKML <linux-kernel@...r.kernel.org>, 
	"open list:Real-time Linux (PREEMPT_RT):Keyword:PREEMPT_RT" <linux-rt-devel@...ts.linux.dev>, bpf <bpf@...r.kernel.org>, 
	kasan-dev <kasan-dev@...glegroups.com>
Subject: Re: [PATCH v4 13/22] slab: remove the do_slab_free() fastpath

On Thu, Jan 22, 2026 at 10:53 PM Vlastimil Babka <vbabka@...e.cz> wrote:
>
> We have removed cpu slab usage from allocation paths. Now remove
> do_slab_free() which was freeing objects to the cpu slab when
> the object belonged to it. Instead call __slab_free() directly,
> which was previously the fallback.
>
> This simplifies kfree_nolock() - when freeing to percpu sheaf
> fails, we can call defer_free() directly.
>
> Also remove functions that became unused.
>
> Reviewed-by: Harry Yoo <harry.yoo@...cle.com>
> Reviewed-by: Hao Li <hao.li@...ux.dev>
> Reviewed-by: Suren Baghdasaryan <surenb@...gle.com>
> Signed-off-by: Vlastimil Babka <vbabka@...e.cz>

...

> @@ -6557,7 +6429,12 @@ void kfree_nolock(const void *object)
>                         return;
>         }
>
> -       do_slab_free(s, slab, x, x, 0, _RET_IP_);
> +       /*
> +        * __slab_free() can locklessly cmpxchg16 into a slab, but then it might
> +        * need to take spin_lock for further processing.
> +        * Avoid the complexity and simply add to a deferred list.
> +        */
> +       defer_free(s, x);

We'll re-benchmark this when it lands.
The current defer_free() path due to slab != c->slab is causing
spikes in some of the workloads.
I think with sheaves
slab_nid(slab) == numa_mem_id() -> free_to_pcs()
path should be exercised a lot more often and fall back to defer_free()
should be rare.
Just leaving a mental note.

Acked-by: Alexei Starovoitov <ast@...nel.org>