[net-next,v2,0/2] vsock: avoid queuing on workqueue if possible

Message

Luigi Leonardi via B4 Relay July 1, 2024, 2:28 p.m. UTC

This series introduces an optimization for vsock/virtio to reduce latency:
When the guest sends a packet to the host, and the workqueue is empty,
if there is enough space, the packet is put directly in the virtqueue.

In this v2 I replaced a mutex_lock with a mutex_trylock because it was inside
a RCU critical section. I also added a check on tx_run, so if the
module is being removed the packet is not queued. I'd like to thank Stefano
for reporting the tx_run issue.

v1->v2
Applied all Stefano's suggestions:
    - Minor code style changes
    - Minor commit text rewrite
Performed more experiments:
     - Check if all the packets go directly to the vq (Matias' suggestion)
     - Used iperf3 to see if there is any improvement in overall throughput
      from guest to host
     - Pinned the vhost process to a pCPU.
     - Run fio using 512B payload
Rebased on latest net-next

Signed-off-by: Luigi Leonardi <[e-mail]>
---
Marco Pinna (2):
      vsock/virtio: refactor virtio_transport_send_pkt_work
      vsock/virtio: avoid enqueue packets when work queue is empty

 net/vmw_vsock/virtio_transport.c | 171 +++++++++++++++++++++++++--------------
 1 file changed, 109 insertions(+), 62 deletions(-)
---
base-commit: 2e7b471121b09e7fa8ffb437bfa0e59d13f96053
change-id: 20240701-pinna-611e8b6cdda0

Best regards,