[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: SOMAXCONN
From: |
Mark Kettenis |
Subject: |
Re: SOMAXCONN |
Date: |
Wed, 25 Jul 2001 23:52:00 +0200 |
From: Roland McGrath <roland@gnu.org>
Date: Wed, 25 Jul 2001 17:08:04 -0400 (EDT)
> Yup. The only thing that we hadn't worked out yet was the actual
> definition of SOMAXCONN. If you agree that 128 is a sensible value,
> I'll put the new file in sysdeps/unix/bsd4.4/bits.
Yes, please do.
Done. This means we have to fiddle the pfinet glue-include stuff a
bit though, otherwise it won't compile anymore. How about the
attached patch? The put_cmsg() duplication is a bit odd, but we
should probably change the second one to do something useful.
Index: pfinet/ChangeLog
from Mark Kettenis <kettenis@gnu.org>
* glue-include/linux/socket.h (SOMAXCONN): Only define if it
wasn't already.
(msg_control, msg_controllen, struct cmsghdr, cmsg_len, cmsg_type,
cmsg_level, CMSG_FIRSTHDR, CMSG_NXTHDR, CMSG_DATA, CMSG_LEN): Only
define if CMSG_DATA wasn't already defined.
Index: pfinet/glue-include/linux/socket.h
===================================================================
RCS file: /cvsroot/hurd/hurd/pfinet/glue-include/linux/socket.h,v
retrieving revision 1.2
diff -u -p -r1.2 socket.h
--- pfinet/glue-include/linux/socket.h 2000/02/07 02:25:46 1.2
+++ pfinet/glue-include/linux/socket.h 2001/07/25 21:48:30
@@ -48,9 +48,11 @@
#define SO_BSDCOMPAT 192
/* Maximum queue length specifiable by listen. */
+#ifndef SOMAXCONN
#define SOMAXCONN 128
+#endif
-/* XXX */
+#ifndef CMSG_DATA
#define msg_control msg_accrights
#define msg_controllen msg_accrightslen
struct cmsghdr { int cmsg_garbage; };
@@ -65,7 +67,11 @@ put_cmsg(struct msghdr *msg, int level,
#define CMSG_DATA(cmsg) (0)
#define CMSG_ALIGN(size) (0)
#define CMSG_LEN(size) (0)
-
+#else
+static inline int
+put_cmsg(struct msghdr *msg, int level, int type, int len, void *data)
+{ return 0; }
+#endif
#define MSG_NOSIGNAL 0
#define MSG_ERRQUEUE 0
- SOMAXCONN, Known Human Nick Rusnov, 2001/07/24