svn commit: r346360 - head/sys/netinet

[John Baldwin]

Author: jhb
Date: Thu Apr 18 23:21:26 2019
New Revision: 346360
URL: https://svnweb.freebsd.org/changeset/base/346360

Log:
  Push down INP_WLOCK slightly in tcp_ctloutput.
  
  The inp lock is not needed for testing the V6 flag as that flag is set
  once when the inp is created and never changes.  For non-TCP socket
  options the lock is immediately dropped after checking that flag.
  This just pushes the lock down to only be acquired for TCP socket
  options.
  
  This isn't a hot-path, more a cosmetic cleanup I noticed while reading
  the code.
  
  Reviewed by:	bz
  MFC after:	1 month
  Sponsored by:	Netflix
  Differential Revision:	https://reviews.freebsd.org/D19740

Modified:
  head/sys/netinet/tcp_usrreq.c

Modified: head/sys/netinet/tcp_usrreq.c
==============================================================================
--- head/sys/netinet/tcp_usrreq.c	Thu Apr 18 22:52:12 2019	(r346359)
+++ head/sys/netinet/tcp_usrreq.c	Thu Apr 18 23:21:26 2019	(r346360)
@@ -1578,11 +1578,9 @@ tcp_ctloutput(struct socket *so, struct sockopt *sopt)
 	error = 0;
 	inp = sotoinpcb(so);
 	KASSERT(inp != NULL, ("tcp_ctloutput: inp == NULL"));
-	INP_WLOCK(inp);
 	if (sopt->sopt_level != IPPROTO_TCP) {
 #ifdef INET6
 		if (inp->inp_vflag & INP_IPV6PROTO) {
-			INP_WUNLOCK(inp);
 			error = ip6_ctloutput(so, sopt);
 			/*
 			 * In case of the IPV6_USE_MIN_MTU socket option,
@@ -1627,12 +1625,12 @@ tcp_ctloutput(struct socket *so, struct sockopt *sopt)
 #endif
 #ifdef INET
 		{
-			INP_WUNLOCK(inp);
 			error = ip_ctloutput(so, sopt);
 		}
 #endif
 		return (error);
 	}
+	INP_WLOCK(inp);
 	if (inp->inp_flags & (INP_TIMEWAIT | INP_DROPPED)) {
 		INP_WUNLOCK(inp);
 		return (ECONNRESET);