git: 7b5053ce2296 - main - tcp_input: remove comments and assertions about tcpbinfo locking
Gleb Smirnoff
glebius at FreeBSD.org
Tue Apr 20 17:06:35 UTC 2021
The branch main has been updated by glebius:
URL: https://cgit.FreeBSD.org/src/commit/?id=7b5053ce2296e0fd96f162e308beb344b4ee654c
commit 7b5053ce2296e0fd96f162e308beb344b4ee654c
Author: Gleb Smirnoff <glebius at FreeBSD.org>
AuthorDate: 2021-04-16 21:39:19 +0000
Commit: Gleb Smirnoff <glebius at FreeBSD.org>
CommitDate: 2021-04-20 17:02:20 +0000
tcp_input: remove comments and assertions about tcpbinfo locking
They aren't valid since d40c0d47cd2.
---
sys/netinet/tcp_input.c | 12 ------------
1 file changed, 12 deletions(-)
diff --git a/sys/netinet/tcp_input.c b/sys/netinet/tcp_input.c
index 8592f3313725..cac67024705e 100644
--- a/sys/netinet/tcp_input.c
+++ b/sys/netinet/tcp_input.c
@@ -984,16 +984,6 @@ findpcb:
* or duplicate segments arriving late. If this segment was a
* legitimate new connection attempt, the old INPCB gets removed and
* we can try again to find a listening socket.
- *
- * At this point, due to earlier optimism, we may hold only an inpcb
- * lock, and not the inpcbinfo write lock. If so, we need to try to
- * acquire it, or if that fails, acquire a reference on the inpcb,
- * drop all locks, acquire a global write lock, and then re-acquire
- * the inpcb lock. We may at that point discover that another thread
- * has tried to free the inpcb, in which case we need to loop back
- * and try to find a new inpcb to deliver to.
- *
- * XXXRW: It may be time to rethink timewait locking.
*/
if (inp->inp_flags & INP_TIMEWAIT) {
tcp_dooptions(&to, optp, optlen,
@@ -1361,7 +1351,6 @@ tfo_socket_result:
* Entry added to syncache and mbuf consumed.
* Only the listen socket is unlocked by syncache_add().
*/
- INP_INFO_WUNLOCK_ASSERT(&V_tcbinfo);
return (IPPROTO_DONE);
} else if (tp->t_state == TCPS_LISTEN) {
/*
@@ -1414,7 +1403,6 @@ dropunlock:
INP_UNLOCK(inp);
drop:
- INP_INFO_WUNLOCK_ASSERT(&V_tcbinfo);
if (s != NULL)
free(s, M_TCPLOG);
if (m != NULL)
More information about the dev-commits-src-main
mailing list