[patch] Add missing call to restore_child_signals_mask.

Doug Evans dje@google.com
Tue May 3 03:21:00 GMT 2011


Hi.

Perhaps there's an obscure reason why this patch is wrong,
but I can't think of one.
[The function calls block_child_signals at the start - all exit
paths should, presumably, call restore_child_signals_mask before returning.]

I will check it in in a few days if there are no objections.

2011-05-02  Doug Evans  <dje@google.com>

	* linux-nat.c (lin_lwp_attach_lwp): For !WIPSTOPPED case,
	adding missing call to restore_child_signals_mask.

Index: linux-nat.c
===================================================================
RCS file: /cvs/src/src/gdb/linux-nat.c,v
retrieving revision 1.202
diff -u -p -r1.202 linux-nat.c
--- linux-nat.c	29 Apr 2011 15:54:12 -0000	1.202
+++ linux-nat.c	3 May 2011 03:17:33 -0000
@@ -1476,7 +1476,10 @@ lin_lwp_attach_lwp (ptid_t ptid)
 
       status = linux_nat_post_attach_wait (ptid, 0, &cloned, &signalled);
       if (!WIFSTOPPED (status))
-	return -1;
+	{
+	  restore_child_signals_mask (&prev_mask);
+	  return -1;
+	}
 
       lp = add_lwp (ptid);
       lp->stopped = 1;



More information about the Gdb-patches mailing list