linux-kernel - [PATCH v3 0/5] virtio: drop sizing vqs during init

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-next>] [day] [month] [year] [list]

Message-ID: <20220815215938.154999-1-mst@redhat.com>
Date:   Mon, 15 Aug 2022 18:00:21 -0400
From:   "Michael S. Tsirkin" <mst@...hat.com>
To:     linux-kernel@...r.kernel.org
Cc:     Xuan Zhuo <xuanzhuo@...ux.alibaba.com>,
        Jason Wang <jasowang@...hat.com>,
        "David S. Miller" <davem@...emloft.net>,
        Eric Dumazet <edumazet@...gle.com>,
        Jakub Kicinski <kuba@...nel.org>,
        Paolo Abeni <pabeni@...hat.com>,
        virtualization@...ts.linux-foundation.org, netdev@...r.kernel.org,
        Linus Torvalds <torvalds@...ux-foundation.org>,
        Jens Axboe <axboe@...nel.dk>,
        James Bottomley <James.Bottomley@...senpartnership.com>,
        "Martin K. Petersen" <martin.petersen@...cle.com>,
        Guenter Roeck <linux@...ck-us.net>,
        Greg KH <gregkh@...uxfoundation.org>
Subject: [PATCH v3 0/5] virtio: drop sizing vqs during init

Reporting after I botched up v2 posting. Sorry about the noise.

Supplying size during init does not work for all transports.
In fact for legacy pci doing that causes a memory
corruption which was reported on Google Cloud.

We might get away with changing size to size_hint so it's
safe to ignore and then fixing legacy to ignore the hint.

But the benefit is unclear in any case, so let's revert for now.
Any new version will have to come with
- documentation of performance gains
- performance testing showing existing workflows
  are not harmed materially. especially ones with
  bursty traffic
- report of testing on legacy devices


Huge shout out to Andres Freund for the effort spent reproducing and
debugging!  Thanks to Guenter Roeck for help with testing!


changes from v2
	drop unrelated patches
changes from v1
	revert the ring size api, it's unused now

Michael S. Tsirkin (5):
  virtio_net: Revert "virtio_net: set the default max ring size by
    find_vqs()"
  virtio: Revert "virtio: add helper virtio_find_vqs_ctx_size()"
  virtio-mmio: Revert "virtio_mmio: support the arg sizes of find_vqs()"
  virtio_pci: Revert "virtio_pci: support the arg sizes of find_vqs()"
  virtio: Revert "virtio: find_vqs() add arg sizes"

 arch/um/drivers/virtio_uml.c             |  2 +-
 drivers/net/virtio_net.c                 | 42 +++---------------------
 drivers/platform/mellanox/mlxbf-tmfifo.c |  1 -
 drivers/remoteproc/remoteproc_virtio.c   |  1 -
 drivers/s390/virtio/virtio_ccw.c         |  1 -
 drivers/virtio/virtio_mmio.c             |  9 ++---
 drivers/virtio/virtio_pci_common.c       | 20 +++++------
 drivers/virtio/virtio_pci_common.h       |  3 +-
 drivers/virtio/virtio_pci_legacy.c       |  6 +---
 drivers/virtio/virtio_pci_modern.c       | 17 +++-------
 drivers/virtio/virtio_vdpa.c             |  1 -
 include/linux/virtio_config.h            | 26 +++------------
 12 files changed, 28 insertions(+), 101 deletions(-)

-- 
MST