Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::3:1 as permitted sender) client-ip=2620:137:e000::3:1;
Date:   Fri, 8 Dec 2023 09:35:14 +0100
From:   Steffen Klassert <steffen.klassert@secunet.com>
To:     Eyal Birger <eyal.birger@gmail.com>
CC:     Daniel Xu <dxu@dxuuu.xyz>,
        Herbert Xu <herbert@gondor.apana.org.au>,
        <netdev@vger.kernel.org>, <linux-kernel@vger.kernel.org>,
        <alexei.starovoitov@gmail.com>, <devel@linux-ipsec.org>,
        <eddyz87@gmail.com>, <edumazet@google.com>,
        Eyal Birger <eyal@metanetworks.com>, <yonghong.song@linux.dev>,
        <kuba@kernel.org>, <bpf@vger.kernel.org>, <pabeni@redhat.com>,
        <davem@davemloft.net>
Subject: Re: [devel-ipsec] [PATCH bpf-next v4 01/10] xfrm: bpf: Move
 xfrm_interface_bpf.c to xfrm_bpf.c
Message-ID: <ZXLVQuqSG7TSjQxD@gauss3.secunet.de>
References: <cover.1701722991.git.dxu@dxuuu.xyz>
 <a385991bb4f36133e15d6eacb72ed22a3c02da16.1701722991.git.dxu@dxuuu.xyz>
 <ZXGx7H/Spv634xgX@gauss3.secunet.de>
 <CAHsH6GtmhP=hZcf2Qv=21dAOSb5dD4GDa+QYdLFz9_FsCZq6tA@mail.gmail.com>
MIME-Version: 1.0
Content-Type: text/plain; charset="utf-8"
Content-Disposition: inline
Content-Transfer-Encoding: 8bit
In-Reply-To: <CAHsH6GtmhP=hZcf2Qv=21dAOSb5dD4GDa+QYdLFz9_FsCZq6tA@mail.gmail.com>
Precedence: bulk

On Thu, Dec 07, 2023 at 01:08:08PM -0800, Eyal Birger wrote:
> Hi Daniel,
> 
> On Thu, Dec 7, 2023 at 3:52 AM Steffen Klassert via Devel
> <devel@linux-ipsec.org> wrote:
> >
> > On Mon, Dec 04, 2023 at 01:56:21PM -0700, Daniel Xu wrote:
> > > This commit moves the contents of xfrm_interface_bpf.c into a new file,
> > > xfrm_bpf.c This is in preparation for adding more xfrm kfuncs. We'd like
> > > to keep all the bpf integrations in a single file.
> 
> This takes away the nice ability to reload the xfrm interface
> related kfuncs when reloading the xfrm interface.
> 
> I also find it a little strange that the kfuncs would be available
> when the xfrm interface isn't loaded.
> 
> So imho it makes sense that these kfuncs would be built
> as part of the module and not as part of the core.

I proposed to merge all the bpf extensions into one file.
With that I wanted to avoid to have 'many' files under
/net/xfrm that fall basically under bpf maintainance
scope. But if there are practical reasons to have these
spilted, I'm OK with that too.