This is the mail archive of the gcc-patches@gcc.gnu.org mailing list for the GCC project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

[incremental] Patch: FYI: minor fileno cleanup


I'm checking this in on the incremental-compiler branch.

I was using hard-coded file descriptor numbers instead of using fileno.

Tom

ChangeLog:
2008-03-27  Tom Tromey  <tromey@redhat.com>

	* server.c (server_start): Use fileno.
	* toplev.c (server_callback): Use fileno.

Index: toplev.c
===================================================================
--- toplev.c	(revision 133491)
+++ toplev.c	(working copy)
@@ -2550,8 +2550,8 @@
 
   /* The diagnostic machinery doesn't need this, but pex does.  Also
      GCC itself seems to write to stderr a lot ... */
-  saved_stderr = dup (2);
-  dup2 (fd, 2);
+  saved_stderr = dup (fileno (stderr));
+  dup2 (fd, fileno (stderr));
 
   /* FIXME: make a new global_dc.  Arrange to unlink assembler output
      file on error.  */
@@ -2588,7 +2588,7 @@
   /* Make sure to close dup'd stderr (by overwriting it), so that
      client will terminate properly.  The server loop will take care
      of the fd we were passed.  */
-  dup2 (saved_stderr, 2);
+  dup2 (saved_stderr, fileno (stderr));
   close (saved_stderr);
 
   return server_back_end_status && ! (errorcount || sorrycount);
Index: server.c
===================================================================
--- server.c	(revision 132956)
+++ server.c	(working copy)
@@ -164,8 +164,8 @@
 	close (fds[0]);
 	/* Don't need these any more -- but keep stderr while
 	   debugging.  */
-	close (0);
-	close (1);
+	close (fileno (stdin));
+	close (fileno (stdout));
 	execv (args[0], args);
 	error ("exec of server failed: %s", xstrerror (errno));
 	exit (FATAL_EXIT_CODE);


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]