[commit] Fix inf-ttrace.c

[commit] Fix inf-ttrace.c

[Mark Kettenis]

I'm not sure what happened here, but while checking out some other
HP-UX related changes, I discovered that gdb didn't work properly on
HP-UX anymore.  Turns out GDB doesn't really like
TARGET_WAITKIND_EXECD events.  Either this broke somewhere along the
line, or I accidentally comitted a inf-ttrace.c that didn't work.
Since the exec following stuff needs some major surgery anyway, I'll
commit this now, such that we have at least working HP-UX support in
the next release.

Mark


Index: ChangeLog
from  Mark Kettenis  <kettenis@gnu.org>

	* inf-ttrace.c (inf_ttrace_wait): Comment out TARGET_WAITKIND_EXEC
	code; return TARGET_WAITKIND_STOPPED instead.

Index: inf-ttrace.c
===================================================================
RCS file: /cvs/src/src/gdb/inf-ttrace.c,v
retrieving revision 1.15
diff -u -p -r1.15 inf-ttrace.c
--- inf-ttrace.c 28 Oct 2005 18:20:35 -0000 1.15
+++ inf-ttrace.c 29 Oct 2005 21:18:31 -0000
@@ -894,6 +894,10 @@ inf_ttrace_wait (ptid_t ptid, struct tar
 #endif
 
     case TTEVT_EXEC:
+      /* FIXME: kettenis/20051029: GDB doesn't really know how to deal
+	 with TARGET_WAITKIND_EXECD events yet.  So we make it look
+	 like a SIGTRAP instead.  */
+#if 0
       ourstatus->kind = TARGET_WAITKIND_EXECD;
       ourstatus->value.execd_pathname =
 	xmalloc (tts.tts_u.tts_exec.tts_pathlen + 1);
@@ -902,6 +906,10 @@ inf_ttrace_wait (ptid_t ptid, struct tar
 		  tts.tts_u.tts_exec.tts_pathlen, 0) == -1)
 	perror_with_name (("ttrace"));
       ourstatus->value.execd_pathname[tts.tts_u.tts_exec.tts_pathlen] = 0;
+#else
+      ourstatus->kind = TARGET_WAITKIND_STOPPED;
+      ourstatus->value.sig = TARGET_SIGNAL_TRAP;
+#endif
       break;
 
     case TTEVT_EXIT:

Re: [commit] Fix inf-ttrace.c

[Daniel Jacobowitz]

On Sat, Oct 29, 2005 at 11:22:44PM +0200, Mark Kettenis wrote:
> I'm not sure what happened here, but while checking out some other
> HP-UX related changes, I discovered that gdb didn't work properly on
> HP-UX anymore.  Turns out GDB doesn't really like
> TARGET_WAITKIND_EXECD events.  Either this broke somewhere along the
> line, or I accidentally comitted a inf-ttrace.c that didn't work.
> Since the exec following stuff needs some major surgery anyway, I'll
> commit this now, such that we have at least working HP-UX support in
> the next release.

It definitely isn't something that broke along the line - I didn't
contribute exec support for GNU/Linux because of how badly it works in
GDB.

We just don't have the user interface for it, either.  I've been
thinking about this on and off, but no inspiration yet...

-- 
Daniel Jacobowitz
CodeSourcery, LLC