[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <ZjOJLXIQS23nkaW8@gauss3.secunet.de>
Date: Thu, 2 May 2024 14:38:05 +0200
From: Steffen Klassert <steffen.klassert@...unet.com>
To: Antony Antony <antony@...nome.org>
CC: <netdev@...r.kernel.org>, <devel@...ux-ipsec.org>, Paul Wouters
<paul@...ats.ca>, Antony Antony <antony.antony@...unet.com>, Tobias Brunner
<tobias@...ongswan.org>, Daniel Xu <dxu@...uu.xyz>
Subject: Re: [PATCH ipsec-next 0/3] Add support for per cpu xfrm states.
On Thu, May 02, 2024 at 02:20:25PM +0200, Antony Antony wrote:
>
> On Fri, Apr 12, 2024 at 08:05:50AM +0200, Steffen Klassert via Devel wrote:
> > Add support for per cpu xfrm states.
> >
> > This patchset implements the xfrm part of per cpu SAs as specified in:
> >
> > https://datatracker.ietf.org/doc/draft-ietf-ipsecme-multi-sa-performance/
> >
> > Patch 1 adds the cpu as a lookup key and config option to to generate
> > acquire messages for each cpu.
> >
> > Patch 2 caches outbound states at the policy.
> >
> > Patch 3 caches inbound states on a new percpu state cache.
> >
> > Please review and test.
>
> Hi Steffen,
>
> I tried xfrm-pcpu-v8 branch, and get these kernel splats. I think it happens
> of the pervious version too. This kernel build has KASAN enabled.
I've introduced this in v8 when I replaced get_cpu by smp_processor_id.
I'll fix this when I rebase next time.
Powered by blists - more mailing lists