On Thu, 2020-07-09 at 10:34 -0700, Jakub Kicinski wrote: > On Thu, 9 Jul 2020 15:12:41 +0200 Paolo Abeni wrote: > > exposes basic inet socket attribute, plus some MPTCP socket > > fields comprising PM status and MPTCP-level sequence numbers. > > > > Reviewed-by: Mat Martineau <mathew.j.martin...@linux.intel.com> > > Signed-off-by: Paolo Abeni <pab...@redhat.com> > > Any idea why sparse says this: > > include/net/sock.h:1612:31: warning: context imbalance in > 'mptcp_diag_get_info' - unexpected unlock > > ? 🤨
AFAICS, that is a sparse false positive, tied to unlock_sock_fast() usage. unlock_sock_fast() conditionally releases the socket lock, according to it's bool argument, and that fools sparse check: any unlock_sock_fast() user splats it. IMHO such warning should be addressed into [un]lock_sock_fast() function[s] - if possible at all. Outside the scope of this series. Matthieu suggested adding some comment to note the above, but I boldly opposed, as current unlock_sock_fast() users don't do that. Cheers, Paolo
