mbox series

[for,7.2-rc3,v3,0/7] fix vhost-user issues with CI

Message ID 20221128164105.1191058-1-alex.bennee@linaro.org
Headers show
Series fix vhost-user issues with CI | expand

Message

Alex Bennée Nov. 28, 2022, 4:40 p.m. UTC 
Hi,

Hopefully this is the final iteration to fix the vhost-user issues
that are currently plaguing the release. I've prevented the circular
closing for the vhost_dev structure by generalising the solution used
by virtio-user-blk which punts the close off to an aio instance.

The memory leak from:

  gpio->vhost_dev.vqs = g_new0(struct vhost_virtqueue, gpio->vhost_dev.nvqs);

still occurs which is because we never call vu_gpio_device_unrealize()
in the test. However its unclear why this is the case. We don't seem
to unrealize the vhost-user-network tests either and clang doesn't
complain about that. However as its not triggered by CI I guess we can
live with it for now.

I've included Stefano's:

  vhost: enable vrings in vhost_dev_start() for vhost-user devices

in this series as it makes sense and improves the vring state errors.
However it's up to you if you want to include it in the eventual PR.

I have had at least one green run on CI now:

  https://gitlab.com/stsquad/qemu/-/pipelines/707015814

Please review.

Alex Bennée (6):
  include/hw: attempt to document VirtIO feature variables
  include/hw: VM state takes precedence in virtio_device_should_start
  tests/qtests: override "force-legacy" for gpio virtio-mmio tests
  hw/virtio: ensure a valid host_feature set for virtio-user-gpio
  hw/virtio: add started_vu status field to vhost-user-gpio
  hw/virtio: generalise CHR_EVENT_CLOSED handling

Stefano Garzarella (1):
  vhost: enable vrings in vhost_dev_start() for vhost-user devices

 include/hw/virtio/vhost-user-gpio.h | 10 ++++
 include/hw/virtio/vhost-user.h      | 18 ++++++++
 include/hw/virtio/vhost.h           | 31 +++++++++++--
 include/hw/virtio/virtio.h          | 42 ++++++++++++++---
 backends/cryptodev-vhost.c          |  4 +-
 backends/vhost-user.c               |  4 +-
 hw/block/vhost-user-blk.c           | 45 +++---------------
 hw/net/vhost_net.c                  |  8 ++--
 hw/scsi/vhost-scsi-common.c         |  4 +-
 hw/virtio/vhost-user-fs.c           |  4 +-
 hw/virtio/vhost-user-gpio.c         | 32 +++++++++----
 hw/virtio/vhost-user-i2c.c          |  4 +-
 hw/virtio/vhost-user-rng.c          |  4 +-
 hw/virtio/vhost-user.c              | 71 +++++++++++++++++++++++++++++
 hw/virtio/vhost-vsock-common.c      |  4 +-
 hw/virtio/vhost.c                   | 44 ++++++++++++++++--
 tests/qtest/libqos/virtio-gpio.c    |  3 +-
 hw/virtio/trace-events              |  4 +-
 18 files changed, 251 insertions(+), 85 deletions(-)

- 
2.34.1

Comments

Michael S. Tsirkin Nov. 30, 2022, 6:57 a.m. UTC | #1 
Patch 1 is good but inappropriate for 7.2
Patch 2 should be last in series.
Patch 4 we are dropping.
I thought hard about it, I think we should patch vhost user net too
because of the risk introduced by patch 2 (which affects everyone).
Can be a patch on top though.

Besides this, for series:

Reviewed-by: Michael S. Tsirkin <mst@redhat.com>