On 5/17/23 11:08, Eelco Chaudron wrote:
This series adds an operation callback which gets called every time the
library wants to call eventfd_write(). This eventfd_write() call could
result in a system call, which could potentially block the PMD thread.

The callback function can decide whether it's ok to handle the
eventfd_write() now or have the newly introduced function,
rte_vhost_notify_guest(), called at a later time.

This can be used by 3rd party applications, like OVS, to avoid system
calls being called as part of the PMD threads.

v3:
     - Changed ABI compatibility code to no longer use a boolean
       to avoid having to disable specific GCC warnings.
     - Moved the fd check fix to a separate patch (patch 3/4).
     - Fixed some coding style issues.

v2: - Used vhost_virtqueue->index to find index for operation.
     - Aligned function name to VDUSE RFC patchset.
     - Added error and offload statistics counter.
     - Mark new API as experimental.
     - Change the virtual queue spin lock to read/write spin lock.
     - Made shared counters atomic.
     - Add versioned rte_vhost_driver_callback_register() for
       ABI compliance.

Eelco Chaudron (4):
       vhost: change vhost_virtqueue access lock to a read/write one
       vhost: make the guest_notifications statistic counter atomic
       vhost: fix invalid call FD handling
       vhost: add device op to offload the interrupt kick


  lib/eal/include/generic/rte_rwlock.h | 17 +++++
  lib/vhost/meson.build                |  2 +
  lib/vhost/rte_vhost.h                | 23 ++++++-
  lib/vhost/socket.c                   | 63 +++++++++++++++++--
  lib/vhost/version.map                |  9 +++
  lib/vhost/vhost.c                    | 92 +++++++++++++++++++++-------
  lib/vhost/vhost.h                    | 69 ++++++++++++++-------
  lib/vhost/vhost_user.c               | 14 ++---
  lib/vhost/virtio_net.c               | 90 +++++++++++++--------------
  9 files changed, 278 insertions(+), 101 deletions(-)



Applied to dpdk-next-virtio/main.

Thanks,
Maxime

Reply via email to