From: Laurent Vivier <lvivier@redhat.com>
To: David Gibson <david@gibson.dropbear.id.au>
Cc: passt-dev@passt.top
Subject: Re: [PATCH 12/12] vhost-user,udp: Use 2 iovec entries per element
Date: Wed, 11 Mar 2026 14:05:58 +0100 [thread overview]
Message-ID: <afbd29b2-4bbf-4851-b203-b7a7f8c0ad34@redhat.com> (raw)
In-Reply-To: <aaTh3JQBYifmV9vs@zatzit>
On 3/2/26 02:03, David Gibson wrote:
> On Fri, Feb 27, 2026 at 03:03:30PM +0100, Laurent Vivier wrote:
>> iPXE places the vnet header in one virtqueue descriptor and the payload
>> in another. When passt maps these descriptors, it needs two iovecs per
>> virtqueue element to handle this layout.
>>
>> Without this, passt crashes with:
>>
>> ASSERTION FAILED in virtqueue_map_desc (virtio.c:403): num_sg < max_num_sg
>>
>> Signed-off-by: Laurent Vivier <lvivier@redhat.com>
>
> Reviewed-by: David Gibson <david@gibson.dropbear.id.au>
>
> It looks like a reasonable short term fix for the specific iPXE
> problem.
>
> Longer term, having a fixed 2 iovecs per element still seems fragile,
> if we encounter some other occasionally-used guest side virtio driver
> which splits its Tx buffers up in an even weirder way. It would be
> nicer if we can actually take our cue from what the guest actually
> gives us. I'm not sure how feasible that is, though.
I agree with that. I'm going to rework this part to allocate a variable number of iovec
per element. It will be cleaner and more correct.
Thanks,
Laurent
prev parent reply other threads:[~2026-03-11 13:06 UTC|newest]
Thread overview: 26+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-02-27 14:03 [PATCH 00/12] vhost-user,udp: Handle multiple iovec entries per virtqueue element Laurent Vivier
2026-02-27 14:03 ` [PATCH 01/12] iov: Add iov_tail_truncate() and iov_tail_zero_end() Laurent Vivier
2026-03-01 23:47 ` David Gibson
2026-02-27 14:03 ` [PATCH 02/12] vhost-user: Use ARRAY_SIZE(elem) instead of VIRTQUEUE_MAX_SIZE Laurent Vivier
2026-03-01 23:48 ` David Gibson
2026-02-27 14:03 ` [PATCH 03/12] udp_vu: Use iov_tail to manage virtqueue buffers Laurent Vivier
2026-03-02 0:03 ` David Gibson
2026-02-27 14:03 ` [PATCH 04/12] udp_vu: Move virtqueue management from udp_vu_sock_recv() to its caller Laurent Vivier
2026-03-02 0:05 ` David Gibson
2026-02-27 14:03 ` [PATCH 05/12] iov: Add IOV_PUT_HEADER() to write header data back to iov_tail Laurent Vivier
2026-03-02 0:08 ` David Gibson
2026-02-27 14:03 ` [PATCH 06/12] udp: Pass iov_tail to udp_update_hdr4()/udp_update_hdr6() Laurent Vivier
2026-03-02 0:13 ` David Gibson
2026-02-27 14:03 ` [PATCH 07/12] udp_vu: Use iov_tail in udp_vu_prepare() Laurent Vivier
2026-03-02 0:24 ` David Gibson
2026-02-27 14:03 ` [PATCH 08/12] vu_common: Pass iov_tail to vu_set_vnethdr() Laurent Vivier
2026-03-02 0:51 ` David Gibson
2026-02-27 14:03 ` [PATCH 09/12] vu_common: Accept explicit iovec counts in vu_set_element() Laurent Vivier
2026-03-02 0:54 ` David Gibson
2026-02-27 14:03 ` [PATCH 10/12] vu_common: Accept explicit iovec count per element in vu_init_elem() Laurent Vivier
2026-03-02 0:55 ` David Gibson
2026-02-27 14:03 ` [PATCH 11/12] vu_common: Prepare to use multibuffer with guest RX Laurent Vivier
2026-03-02 0:59 ` David Gibson
2026-02-27 14:03 ` [PATCH 12/12] vhost-user,udp: Use 2 iovec entries per element Laurent Vivier
2026-03-02 1:03 ` David Gibson
2026-03-11 13:05 ` Laurent Vivier [this message]
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=afbd29b2-4bbf-4851-b203-b7a7f8c0ad34@redhat.com \
--to=lvivier@redhat.com \
--cc=david@gibson.dropbear.id.au \
--cc=passt-dev@passt.top \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
Code repositories for project(s) associated with this public inbox
https://passt.top/passt
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for IMAP folder(s).