[PATCH bpf 0/2] sockmap fix for test_map failure

From: John Fastabend <john.fastabend@gmail.com>
To: alexei.starovoitov@gmail.com, daniel@iogearbox.net, andrii@kernel.org
Cc: john.fastabend@gmail.com, bpf@vger.kernel.org, netdev@vger.kernel.org
Subject: [PATCH bpf 0/2] sockmap fix for test_map failure
Date: Fri, 19 Nov 2021 10:14:16 -0800	[thread overview]
Message-ID: <20211119181418.353932-1-john.fastabend@gmail.com> (raw)

CI test_map runs started failing because of a regression in the
sockmap tests. The case, caught by test_maps is that progs attached
to sockets are not detatched currently when sockets are removed
from a map. We resolve this in two patches. The first patch
fixes a subtle issue found from code review and the second
patch addresses the reported CI issue. This was recently introduced
by a race fix, see patches for details.

Sorry for the hassle here, seems we missed ./test_maps run before
pushing the offending patch or maybe we just got lucky on the
run we did locally. Either way should be resolved now.

Thanks,
John

John Fastabend (2):
  bpf, sockmap: Attach map progs to psock early for feature probes
  bpf, sockmap: Re-evaluate proto ops when psock is removed from sockmap

 net/core/skmsg.c    |  5 +++++
 net/core/sock_map.c | 15 ++++++++++-----
 2 files changed, 15 insertions(+), 5 deletions(-)

-- 
2.33.0