Discussion:
sctp bug
(too old to reply)
Maxime Villard
2019-03-07 12:34:51 UTC
Permalink
Looks like there is a missing sbsync here [1], in addition to mbuf_removed
that is not set correctly, this means we wreck pointers.

[1] https://nxr.netbsd.org/xref/src/sys/kern/uipc_socket.c?r=1.269#1362

--
Posted automagically by a mail2news gateway at muc.de e.V.
Please direct questions, flames, donations, etc. to news-***@muc.de
Robert Swindells
2019-03-07 18:50:35 UTC
Permalink
Post by Maxime Villard
Looks like there is a missing sbsync here [1], in addition to
mbuf_removed
that is not set correctly, this means we wreck pointers.
[1] https://nxr.netbsd.org/xref/src/sys/kern/uipc_socket.c?r=1.269#1362
How is this connected to SCTP ?

--
Posted automagically by a mail2news gateway at muc.de e.V.
Please direct questions, flames, donations, etc. to news-***@muc.de
Maxime Villard
2019-03-07 22:00:43 UTC
Permalink
Post by Robert Swindells
Post by Maxime Villard
Looks like there is a missing sbsync here [1], in addition to mbuf_removed
that is not set correctly, this means we wreck pointers.
[1] https://nxr.netbsd.org/xref/src/sys/kern/uipc_socket.c?r=1.269#1362
How is this connected to SCTP ?
PR_ADDR_OPT is used by only SCTP, and the code itself was introduced as part
of SCTP. I don't see any external use, so it is 100% connected to SCTP. (And
besides, there is a comment that says just that.)

--
Posted automagically by a mail2news gateway at muc.de e.V.
Please direct questions, flames, donations, etc. to news-***@muc.de
Loading...