From 118ddafeebc0fdea668f848c9aaf8b845cdb9176 Mon Sep 17 00:00:00 2001 From: Roland McGrath Date: Thu, 3 Feb 2000 01:22:34 +0000 Subject: [PATCH] . --- ChangeLog | 14 ++++++++++++++ 1 file changed, 14 insertions(+) diff --git a/ChangeLog b/ChangeLog index ed23a77ad69..eb967c15eb6 100644 --- a/ChangeLog +++ b/ChangeLog @@ -1,5 +1,19 @@ 2000-02-02 Roland McGrath + * sysdeps/mach/hurd/readlink.c (__readlink): Don't clobber original + LEN value with io_read result; save it and compare them just in case + the server bogusly returns more than we asked for. + * sysdeps/mach/hurd/recv.c (recv): Likewise for socket_recv. + * sysdeps/mach/hurd/recvfrom.c (recvfrom): Likewise. + * hurd/fd-read.c (_hurd_fd_read): Likewise for io_read. + Initialize NREAD before the RPC. I think this must only have failed + to bomb before because all the servers always return ool data(?). + + * sysdeps/mach/hurd/getsockopt.c (getsockopt): Fix inverted sense of + test: *OPTLEN gets BUFLEN iff BUFLEN is lesser. Move that test out of + BUF != OPTVAL condition, since it is a separate issue. Short-circuit + the memcpy and dealloc for BUFLEN==0. + * hurd/hurdlookup.c (CHECK_ELOOP): New macro does SYMLOOP_MAX check; deallocate *RESULT before returning ELOOP. Don't conditionalize on SYMLOOP_MAX, we always better have one. -- 2.47.2