[PATCH bpf] bpf, sockmap: use READ_ONCE/WRITE_ONCE for saved_data_ready in the verdict path · Dudu Lu <hidden> · 2026-10-21
Re: [PATCH bpf] bpf, sockmap: use READ_ONCE/WRITE_ONCE for saved_data_ready in the verdict path · bot+bpf-ci@kernel.org · 2026-10-21
Re: [PATCH bpf] bpf, sockmap: use READ_ONCE/WRITE_ONCE for saved_data_ready in the verdict path · Alexei Starovoitov <hidden> · 2026-10-21
Re: [PATCH bpf] bpf, sockmap: use READ_ONCE/WRITE_ONCE for saved_data_ready in the verdict path · Jakub Sitnicki <jakub@cloudflare.com> · 2026-10-21
[PATCH bpf v2] bpf, sockmap: annotate the psock teardown data races · Dudu Lu <hidden> · 2026-10-22
Re: [PATCH bpf v2] bpf, sockmap: annotate the psock teardown data races · Alexei Starovoitov <hidden> · 2026-10-22
Re: [PATCH bpf v2] bpf, sockmap: annotate the psock teardown data races · Jakub Sitnicki <jakub@cloudflare.com> · 2026-10-23
Re: [PATCH bpf v2] bpf, sockmap: annotate the psock teardown data races · Alexei Starovoitov <hidden> · 2026-10-23

Re: [PATCH bpf] bpf, sockmap: use READ_ONCE/WRITE_ONCE for saved_data_ready in the verdict path

From: Alexei Starovoitov <hidden>
Date: 2026-10-21 16:45:12
Also in: bpf

On Wed, Oct 21, 2026 at 08:56 AM Dudu Lu [off-list ref] wrote:

sk_psock_stop_strp() uses WRITE_ONCE for the same field.  sock_map_link()
/ sock_map_unref() touch the same psock from another CPU without the
lock, so KCSAN reports the race:

  BUG: KCSAN: data-race in sock_map_link / sock_map_unref
  BUG: KCSAN: data-race in sk_psock_start_verdict / sock_map_unref

sashiko is right. KCSAN will print both with this patch applied.

pw-bot: cr

`h`	back out one level
`j`	next message in thread
`k`	previous message in thread
`l`	drill in
`Esc`	close help / fold thread tree
`?`	toggle this help