netdev - Re: [PATCH net-next v2] xsk: skip validating skb list in xmit path

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: <CAL+tcoA1LMjxKgQb4WZZ8LeipbGU038is21M_y+kc93eoUpBCA@mail.gmail.com>
Date: Thu, 17 Jul 2025 08:06:48 +0800
From: Jason Xing <kerneljasonxing@...il.com>
To: Jakub Kicinski <kuba@...nel.org>
Cc: davem@...emloft.net, edumazet@...gle.com, pabeni@...hat.com, 
	bjorn@...nel.org, magnus.karlsson@...el.com, maciej.fijalkowski@...el.com, 
	jonathan.lemon@...il.com, sdf@...ichev.me, ast@...nel.org, 
	daniel@...earbox.net, hawk@...nel.org, john.fastabend@...il.com, joe@...a.to, 
	willemdebruijn.kernel@...il.com, bpf@...r.kernel.org, netdev@...r.kernel.org, 
	Jason Xing <kernelxing@...cent.com>
Subject: Re: [PATCH net-next v2] xsk: skip validating skb list in xmit path

On Thu, Jul 17, 2025 at 7:43 AM Jakub Kicinski <kuba@...nel.org> wrote:
>
> On Thu, 17 Jul 2025 07:37:42 +0800 Jason Xing wrote:
> > On Thu, Jul 17, 2025 at 5:56 AM Jakub Kicinski <kuba@...nel.org> wrote:
> > > On Wed, 16 Jul 2025 20:27:25 +0800 Jason Xing wrote:
> > > > This patch only does one thing that removes validate_xmit_skb_list()
> > > > for xsk.
> > >
> > > Please no, I understand that it's fun to optimize the fallback paths
> > > but it increases the complexity of the stack.
> >
> > Are you suggesting to remove this description? And I see you marked it
> > as 'rejected', so it seems that I should use the V1 patch which
> > doesn't increase the complexity.
>
> No, I'm questioning optimizing the copy mode AF_XDP in the first place.

Yesterday, Stan asked me the same question, but the zerocopy for me is
obviously not qualified for deployment right now. Let me rephrase a
bit:

1) There are still many VMs that don't support zerocopy in the real world.
2) Some old drivers have various unknown problems. I posted one at
https://lore.kernel.org/all/CAL+tcoCTHTptwmok9vhp7GEwQgMhNsBJxT3PStJDeVOLR_-Q3g@mail.gmail.com/

To be honest, this patch really only does one thing as the commit
says. It might look very complex, but if readers take a deep look they
will find only one removal of that validation for xsk in the hot path.
Nothing more and nothing less. So IMHO, it doesn't bring more complex
codes here.

And removal of one validation indeed contributes to the transmission.
I believe there remain a number of applications using copy mode
currently. And maintainers of xsk don't regard copy mode as orphaned,
right?

Jakub, if you can, please don't mark it as rejected. Thanks.

Thanks,
Jason