[poky] [PATCH 16/24] xtrans: update its patch's Upstream-Status

Dexuan Cui dexuan.cui at intel.com
Tue May 10 23:27:05 PDT 2011


From: Dexuan Cui <dexuan.cui at intel.com>

remove abstract_socket_fix.patch as it's not used at all for long.

Signed-off-by: Dexuan Cui <dexuan.cui at intel.com>
---
 .../xorg-lib/xtrans/abstract_socket_fix.patch      |   58 --------------------
 1 files changed, 0 insertions(+), 58 deletions(-)
 delete mode 100644 meta/recipes-graphics/xorg-lib/xtrans/abstract_socket_fix.patch

diff --git a/meta/recipes-graphics/xorg-lib/xtrans/abstract_socket_fix.patch b/meta/recipes-graphics/xorg-lib/xtrans/abstract_socket_fix.patch
deleted file mode 100644
index 197da07..0000000
--- a/meta/recipes-graphics/xorg-lib/xtrans/abstract_socket_fix.patch
+++ /dev/null
@@ -1,58 +0,0 @@
-From 556a351de83fc6f401b02213dae95731553c216d Mon Sep 17 00:00:00 2001
-From: Loïc Minier <lool at dooz.org>
-Date: Mon, 24  Mar  2008  19:38:27  +0000
-Subject: Bug #10489: Don't retry unix socket connect()s on ENOENT.
-
-If the socket isn't there, it's not gonna magically appear.  Either it's
-a server socket and you should have just waited for the SIGUSR1 from the
-server, or it's a stale reference to an ICE socket.
-
-However, do retry once, so fallback from abstract to filesystem namespace
-works.
-
-Originally Debian bug #385976.
----
-diff --git a/Xtranssock.c b/Xtranssock.c
-index 8a7d2ec..da4afe9 100644
---- a/Xtranssock.c
-+++ b/Xtranssock.c
-@@ -2146,8 +2146,13 @@ TRANS(SocketUNIXConnect) (XtransConnInfo ciptr, char *host, char *port)
- 	    errno = olderrno;
- 	    
- 	    /*
--	     * If the error was ENOENT, the server may be starting up
--	     * and we should try again.
-+	     * If the error was ENOENT, the server may be starting up; we used
-+	     * to suggest to try again in this case with
-+	     * TRANS_TRY_CONNECT_AGAIN, but this introduced problems for
-+	     * processes still referencing stale sockets in their environment.
-+	     * Hence, we now return a hard error, TRANS_CONNECT_FAILED, and it
-+	     * is suggested that higher level stacks handle retries on their
-+	     * level when they face a slow starting server.
- 	     *
- 	     * If the error was EWOULDBLOCK or EINPROGRESS then the socket
- 	     * was non-blocking and we should poll using select
-@@ -2161,14 +2166,14 @@ TRANS(SocketUNIXConnect) (XtransConnInfo ciptr, char *host, char *port)
- 	    else if (olderrno == EINTR)
- 		return TRANS_TRY_CONNECT_AGAIN;
- 	    else if (olderrno == ENOENT) {
--		/*
--		* If opening as abstract socket failed, try again "normally"
--		*/
--		if (abstract)
-+		/* If opening as abstract socket failed, try again normally */
-+		if (abstract) {
- 		    ciptr->transptr->flags &= ~(TRANS_ABSTRACT);
--		return TRANS_TRY_CONNECT_AGAIN;
--	    }
--	    else {
-+		    return TRANS_TRY_CONNECT_AGAIN;
-+		} else {
-+		    return TRANS_CONNECT_FAILED;
-+		}
-+	    } else {
- 		PRMSG (2,"SocketUNIXConnect: Can't connect: errno = %d\n",
- 		       EGET(),0, 0);
- 
---
-cgit v0.7.2-37-g538c
-- 
1.7.2




More information about the poky mailing list