Thread (26 messages) flat view 26 messages, 9 authors, 2024-08-27

Re: [PATCH net] net: drop bad gso csum_start and offset in virtio_net_hdr

From: Adrian Vladu <hidden>
Date: 2024-08-14 10:10:28
Also in: regressions, stable

Hello,

The 6.6.y branch has the patch already in the stable queue -> https://git.kernel.org/pub/scm/linux/kernel/git/stable/stable-queue.git/commit/?id=3e713b73c01fac163a5c8cb0953d1e300407a773, and it should be available in the 6.6.46 upcoming minor.

Thanks, Adrian.
________________________________________
From: Christian Heusel
Sent: Wednesday, August 14, 2024 1:05 PM
To: Michael S. Tsirkin
Cc: Greg KH; Adrian Vladu; willemdebruijn.kernel@gmail.com; alexander.duyck@gmail.com; arefev@swemel.ru; davem@davemloft.net; edumazet@google.com; jasowang@redhat.com; kuba@kernel.org; netdev@vger.kernel.org; pabeni@redhat.com; stable@vger.kernel.org; willemb@google.com; regressions@lists.linux.dev
Subject: Re: [PATCH net] net: drop bad gso csum_start and offset in virtio_net_hdr


On 24/08/14 05:54AM, Michael S. Tsirkin wrote:
On Wed, Aug 14, 2024 at 11:46:30AM +0200, Christian Heusel wrote:
quoted
On 24/08/08 11:52AM, Christian Heusel wrote:
quoted
quoted
On 24/08/08 08:38AM, Greg KH wrote:
quoted
quoted
quoted
On Wed, Aug 07, 2024 at 08:34:48PM +0200, Christian Heusel wrote:
quoted
quoted
quoted
quoted
On 24/08/07 04:12PM, Greg KH wrote:
quoted
quoted
quoted
quoted
quoted
On Mon, Aug 05, 2024 at 09:28:29PM +0000, avladu@cloudbasesolutions.com wrote:
quoted
quoted
quoted
quoted
quoted
quoted
Hello,
quoted
quoted
quoted
quoted
quoted
quoted
quoted
quoted
quoted
quoted
quoted
quoted
This patch needs to be backported to the stable 6.1.x and 6.64.x branches, as the initial patch https://github.com/torvalds/linux/commit/e269d79c7d35aa3808b1f3c1737d63dab504ddc8 was backported a few days ago: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/include/linux/virtio_net.h?h=3Dv6.1.103&id=3D5b1997487a3f3373b0f580c8a20b56c1b64b0775
quoted
quoted
quoted
quoted
quoted
quoted
https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/include/linux/virtio_net.h?h=3Dv6.6.44&id=3D90d41ebe0cd4635f6410471efc1dd71b33e894cf
quoted
quoted
quoted
quoted
quoted
quoted
quoted
quoted
quoted
quoted
Please provide a working backport, the change does not properly
quoted
quoted
quoted
quoted
quoted
cherry-pick.
quoted
quoted
quoted
quoted
quoted
quoted
quoted
quoted
quoted
quoted
greg k-h
quoted
quoted
quoted
quoted
quoted
quoted
quoted
quoted
Hey Greg, hey Sasha,
quoted
quoted
quoted
quoted
quoted
quoted
quoted
quoted
this patch also needs backporting to the 6.6.y and 6.10.y series as the
quoted
quoted
quoted
quoted
buggy commit was backported to to all three series.
quoted
quoted
quoted
quoted
quoted
quoted
What buggy commit?
quoted
quoted
quoted
quoted
The issue is that commit e269d79c7d35 ("net: missing check virtio")
quoted
quoted
introduces a bug which is fixed by 89add40066f9 ("net: drop bad gso
quoted
quoted
csum_start and offset in virtio_net_hdr") which it also carries a
quoted
quoted
"Fixes:" tag for.
quoted
quoted
quoted
quoted
Therefore it would be good to also get 89add40066f9 backported.
quoted
quoted
quoted
quoted
quoted
And how was this tested, it does not apply cleanly to the trees for me
quoted
quoted
quoted
at all.
quoted
quoted
quoted
quoted
I have tested this with the procedure as described in [0]:
quoted
quoted
quoted
quoted
    $ git switch linux-6.10.y
quoted
quoted
    $ git cherry-pick -x 89add40066f9ed9abe5f7f886fe5789ff7e0c50e
quoted
quoted
    Auto-merging net/ipv4/udp_offload.c
quoted
quoted
    [linux-6.10.y fbc0d2bea065] net: drop bad gso csum_start and offset in virtio_net_hdr
quoted
quoted
     Author: Willem de Bruijn [off-list ref]
quoted
quoted
     Date: Mon Jul 29 16:10:12 2024 -0400
quoted
quoted
     3 files changed, 12 insertions(+), 11 deletions(-)
quoted
quoted
quoted
quoted
This also works for linux-6.6.y, but not for linux-6.1.y, as it fails
quoted
quoted
with a merge error there.
quoted
quoted
quoted
quoted
The relevant commit is confirmed to fix the issue in the relevant Githu
quoted
quoted
issue here[1]:
quoted
quoted
quoted
quoted
    @marek22k commented
quoted
quoted
    > They both fix the problem for me.
quoted
quoted
quoted
quoted
quoted
confused,
quoted
quoted
quoted
quoted
Sorry for the confusion! I hope the above clears things up a little :)
quoted
quoted
quoted
quoted
quoted
greg k-h
quoted
quoted
quoted
quoted
Cheers,
quoted
quoted
Christian
quoted
quoted
quoted
quoted
[0]: https://lore.kernel.org/all/2024060624-platinum-ladies-9214@gregkh/ (local)
quoted
quoted
[1]: https://github.com/tailscale/tailscale/issues/13041#issuecomment-2272326491
quoted
quoted
Since I didn't hear from anybody so far about the above issue it's a bit
quoted
unclear on how to proceed here. I still think that I would make sense to
quoted
go with my above suggestion about patching at least 2 out of the 3
quoted
stable series where the patch applies cleanly.
quoted
quoted
     ~ Chris
Do what Greg said:
       Please provide a working backport, the change does not properly
       cherry-pick.
that means, post backported patches to stable, copy list.


Alright, will do!
Keyboard shortcuts
hback out one level
jnext message in thread
kprevious message in thread
ldrill in
Escclose help / fold thread tree
?toggle this help