[PATCH net-next v2 0/2] vsock: receive queue hint for io_uring

AF_VSOCK stream receives never fill msghdr.msg_inq. io_uring initialises the
hint to -1 and treats that as "unknown", so it cannot set
IORING_CQE_F_SOCK_NONEMPTY, and it speculatively retries a multishot receive
after every completion, including when the queue is already empty.

Patch 1 fills the hint at the common exit of __vsock_connectible_recvmsg(),
reusing vsock_stream_has_data(), the transport callback that SIOCINQ already
uses. Following tcp_inq_hint(), it reports a terminal hint of 1 once the
connection is finished, so the caller still performs the receive that
observes EOF. Patch 2 adds selftest coverage.

Numbers come from a ping-pong over vsock loopback using io_uring multishot
receive, 4 KiB messages, the queue drained between messages, with kernel
receives counted by a kprobe. The baseline is the same tree with patch 1
reverted:

  entries per delivered message   1.97 -> 1.00       (5000 messages)
  receiver CPU time, median       1.640s -> 1.587s   (-3.2%)
  receiver wall time, median      3.260s -> 3.183s   (-2.4%)

Timing is 25 runs of 50000 messages per kernel; CPU gives Mann-Whitney
p=0.006, wall time does not separate from noise (p=0.12), as expected for a
latency-bound workload.

The hint is gated to SOCK_STREAM; seqpacket is untouched. SO_INQ and SCM_INQ
are deliberately left out: they need their own per-socket state and
ancillary-message semantics, and a byte-count contract would have to account
for transports reporting less than the whole queue. That can follow.

Testing: built and run under virtme-ng on loopback, where the new cases pass
and vsock_test still passes 41 of 41 with a clean dmesg. Other transports
were reviewed but not tested. Hyper-V reports the current packet rather than
the whole queue and can return -EIO; an undercount only understates what is
readable, and a negative result reports -1, the same "unknown" value io_uring
presets.

Changes in v2:
- vsock: rename the helper to vsock_stream_inq_hint(), fold the NULL
  transport check into the finished check, drop the int cast, assign
  msg_inq directly, and document why only SOCK_STREAM reports a hint
  (Stefano)
- vsock/test: move expect_res() to util.c, add uring to the SOCK_NONEMPTY
  helper name, skip instead of failing when SIOCINQ is not supported, and
  drop the redundant final barrier of the EOF test (Stefano)
- vsock/test: queue two chunks before arming the multishot receive and
  check SOCK_NONEMPTY is set on the first completion and clear on the
  second, so the test fails without patch 1 (Bobby)
- v1: https://lore.kernel.org/netdev/20260920194710.1114748-1-devnexen@gmail.com/ (local)

David Carlier (2):
  vsock: report pending receive data to io_uring
  vsock/test: cover receive queue hints

 net/vmw_vsock/af_vsock.c               |  35 +++
 tools/testing/vsock/util.c             |   9 +
 tools/testing/vsock/util.h             |   1 +
 tools/testing/vsock/vsock_uring_test.c | 391 +++++++++++++++++++++++++
 4 files changed, 436 insertions(+)


base-commit: cea851523034d83dc61d882b1a6a3ad273480f7b
-- 
2.55.0