* testsuite fixes for bleeding edge gcc
@ 2004-10-14 7:53 Zack Weinberg
2004-10-14 12:04 ` Michael Chastain
0 siblings, 1 reply; 6+ messages in thread
From: Zack Weinberg @ 2004-10-14 7:53 UTC (permalink / raw)
To: gdb-patches
Current development versions of GCC are substantially pickier about
"built-in" library functions being properly declared, which causes a
large number of GDB testsuite failures.
The appended patch adds all the missing declarations to the 6.2-branch
version of the testsuite. I haven't tried mainline. Thoughts?
zw
* gdb.base/auxv.c: Declare abort.
* gdb.base/charset.c: Include stdlib.h.
* gdb.base/complex.c: Declare abort and exit.
* gdb.base/freebpcmd.c: Include stdio.h.
* gdb.base/gcore.c: Include string.h.
* gdb.base/sigaltstack.c: Include stdlib.h and string.h.
* gdb.base/siginfo.c: Include string.h.
* gdb.base/sigstep.c: Include string.h.
* gdb.mi/basics.c: Include stdio.h.
* gdb.mi/gdb701.c: Declare exit.
* gdb.mi/var-cmd.c: Declare exit/
===================================================================
Index: testsuite/gdb.base/auxv.c
--- testsuite/gdb.base/auxv.c 16 Mar 2004 21:47:03 -0000 1.1
+++ testsuite/gdb.base/auxv.c 14 Oct 2004 07:40:00 -0000
@@ -11,6 +11,7 @@
#endif
#if HAVE_ABORT
+extern void abort (void);
#define ABORT abort()
#else
#define ABORT {char *invalid = 0; *invalid = 0xFF;}
===================================================================
Index: testsuite/gdb.base/charset.c
--- testsuite/gdb.base/charset.c 26 Feb 2004 17:23:23 -0000 1.3
+++ testsuite/gdb.base/charset.c 14 Oct 2004 07:40:00 -0000
@@ -22,6 +22,7 @@
bug-gdb@gnu.org */
#include <stdio.h>
+#include <stdlib.h>
#include <string.h>
===================================================================
Index: testsuite/gdb.base/complex.c
--- testsuite/gdb.base/complex.c 29 Jun 2003 20:32:23 -0000 1.1
+++ testsuite/gdb.base/complex.c 14 Oct 2004 07:40:00 -0000
@@ -1,6 +1,9 @@
/* Test taken from GCC. Verify that we can print a structure containing
a complex number. */
+extern void abort (void);
+extern void exit (int);
+
typedef __complex__ float cf;
struct x { char c; cf f; } __attribute__ ((__packed__));
struct unpacked_x { char c; cf f; };
===================================================================
Index: testsuite/gdb.base/freebpcmd.c
--- testsuite/gdb.base/freebpcmd.c 17 Dec 2003 21:34:51 -0000 1.2
+++ testsuite/gdb.base/freebpcmd.c 14 Oct 2004 07:40:00 -0000
@@ -18,6 +18,8 @@
Foundation, Inc., 59 Temple Place - Suite 330,
Boston, MA 02111-1307, USA. */
+#include <stdio.h>
+
int
main (int argc, char **argv)
{
===================================================================
Index: testsuite/gdb.base/gcore.c
--- testsuite/gdb.base/gcore.c 9 Jan 2002 00:37:41 -0000 1.1
+++ testsuite/gdb.base/gcore.c 14 Oct 2004 07:40:00 -0000
@@ -3,6 +3,7 @@
*/
#include <stdlib.h>
+#include <string.h>
int extern_array[4] = {1, 2, 3, 4};
static int static_array[4] = {5, 6, 7, 8};
===================================================================
Index: testsuite/gdb.base/long_long.c
--- testsuite/gdb.base/long_long.c 16 Jun 2004 15:19:41 -0000 1.3
+++ testsuite/gdb.base/long_long.c 14 Oct 2004 07:40:00 -0000
@@ -26,6 +26,7 @@
*
* cc +e +DA2.0 -g -o long_long long_long.c
*/
+#include <string.h>
enum { MAX_BYTES = 16 };
===================================================================
Index: testsuite/gdb.base/sigaltstack.c
--- testsuite/gdb.base/sigaltstack.c 5 Apr 2004 21:00:21 -0000 1.1
+++ testsuite/gdb.base/sigaltstack.c 14 Oct 2004 07:40:00 -0000
@@ -21,6 +21,8 @@
#include <signal.h>
#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
#include <sys/time.h>
enum level { MAIN, OUTER, INNER, LEAF, NR_LEVELS };
===================================================================
Index: testsuite/gdb.base/siginfo.c
--- testsuite/gdb.base/siginfo.c 15 Apr 2004 21:14:00 -0000 1.1
+++ testsuite/gdb.base/siginfo.c 14 Oct 2004 07:40:00 -0000
@@ -20,6 +20,7 @@
#include <stdio.h>
#include <signal.h>
+#include <string.h>
#include <sys/time.h>
static volatile int done;
===================================================================
Index: testsuite/gdb.base/sigstep.c
--- testsuite/gdb.base/sigstep.c 21 Apr 2004 17:19:28 -0000 1.1
+++ testsuite/gdb.base/sigstep.c 14 Oct 2004 07:40:00 -0000
@@ -19,6 +19,7 @@
*/
#include <stdio.h>
+#include <string.h>
#include <signal.h>
#include <sys/time.h>
===================================================================
Index: testsuite/gdb.mi/basics.c
--- testsuite/gdb.mi/basics.c 14 Mar 2000 05:02:03 -0000 1.3
+++ testsuite/gdb.mi/basics.c 14 Oct 2004 07:40:00 -0000
@@ -2,6 +2,7 @@
* This simple program that passes different types of arguments
* on function calls. Useful to test printing frames, stepping, etc.
*/
+#include <stdio.h>
int callee4 (void)
{
===================================================================
Index: testsuite/gdb.mi/gdb701.c
--- testsuite/gdb.mi/gdb701.c 16 Sep 2002 19:01:43 -0000 1.1
+++ testsuite/gdb.mi/gdb701.c 14 Oct 2004 07:40:00 -0000
@@ -1,3 +1,5 @@
+extern void exit(int);
+
struct _foo
{
int x;
===================================================================
Index: testsuite/gdb.mi/var-cmd.c
--- testsuite/gdb.mi/var-cmd.c 23 Feb 2000 00:25:43 -0000 1.1
+++ testsuite/gdb.mi/var-cmd.c 14 Oct 2004 07:40:00 -0000
@@ -1,3 +1,5 @@
+extern void exit (int);
+
struct _simple_struct {
int integer;
unsigned int unsigned_integer;
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: testsuite fixes for bleeding edge gcc
2004-10-14 7:53 testsuite fixes for bleeding edge gcc Zack Weinberg
@ 2004-10-14 12:04 ` Michael Chastain
2004-10-14 16:59 ` Zack Weinberg
2004-10-14 19:14 ` Andrew Cagney
0 siblings, 2 replies; 6+ messages in thread
From: Michael Chastain @ 2004-10-14 12:04 UTC (permalink / raw)
To: zack, gdb-patches
Hi Zack!
> Current development versions of GCC are substantially pickier about
> "built-in" library functions being properly declared, which causes a
> large number of GDB testsuite failures.
Yes.
> The appended patch adds all the missing declarations to the 6.2-branch
> version of the testsuite. I haven't tried mainline. Thoughts?
Except for gdb.mi/gdb701, these are all fixed in gdb HEAD.
Also fixed in gdb HEAD are gdb.thread/*.exp, plus two tests
that check the gcc version number that needed 'gcc-4-*'.
gdb_6_2-branch is dead. Andrew is going to create gdb_6_3-branch
in a few more days or so.
If you really do have a need for it, I could merge the whole test suite
from HEAD to gdb_6_2-branch. I stopped testing gdb_6_2-branch just
recently, but before I stopped, I was regularly testing gdb from the
branch with the test suite from gdb HEAD.
Michael
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: testsuite fixes for bleeding edge gcc
2004-10-14 12:04 ` Michael Chastain
@ 2004-10-14 16:59 ` Zack Weinberg
2004-10-14 18:55 ` Michael Chastain
2004-10-14 19:14 ` Andrew Cagney
1 sibling, 1 reply; 6+ messages in thread
From: Zack Weinberg @ 2004-10-14 16:59 UTC (permalink / raw)
To: Michael Chastain; +Cc: gdb-patches
Michael Chastain <mec.gnu@mindspring.com> writes:
>> The appended patch adds all the missing declarations to the 6.2-branch
>> version of the testsuite. I haven't tried mainline. Thoughts?
>
> Except for gdb.mi/gdb701, these are all fixed in gdb HEAD.
> Also fixed in gdb HEAD are gdb.thread/*.exp, plus two tests
> that check the gcc version number that needed 'gcc-4-*'.
>
> gdb_6_2-branch is dead. Andrew is going to create gdb_6_3-branch
> in a few more days or so.
>
> If you really do have a need for it, I could merge the whole test suite
> from HEAD to gdb_6_2-branch. I stopped testing gdb_6_2-branch just
> recently, but before I stopped, I was regularly testing gdb from the
> branch with the test suite from gdb HEAD.
I was only testing gdb_6_2-branch because I wanted to avoid
hypothetical breakage in gdb HEAD which would confuse the issue of
whether my GCC changes worked. If y'all are about to branch for a 6.3
release, perhaps I do not need to be doing that right now. Would be
nice to get gdb.mi/gdb701 fixed in HEAD though.
zw
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: testsuite fixes for bleeding edge gcc
2004-10-14 16:59 ` Zack Weinberg
@ 2004-10-14 18:55 ` Michael Chastain
2004-10-14 21:32 ` Zack Weinberg
0 siblings, 1 reply; 6+ messages in thread
From: Michael Chastain @ 2004-10-14 18:55 UTC (permalink / raw)
To: zack; +Cc: gdb-patches
> whether my GCC changes worked.
Ah. That is a good reason for me to keep up maintenance on the
6.2 branch, if it happens often enough. Just because we don't
release from the branch doesn't mean it's useless.
Also if you are doing before-and-after comparisons with gdb,
watch out for some tests that are unstable from run to run.
http://sources.redhat.com/ml/gdb-testers
... and search for particular test names. Or just google the
test names.
> Would be nice to get gdb.mi/gdb701 fixed in HEAD though.
Yeah. Andrew or I have to pick up that little nit.
Michael
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: testsuite fixes for bleeding edge gcc
2004-10-14 12:04 ` Michael Chastain
2004-10-14 16:59 ` Zack Weinberg
@ 2004-10-14 19:14 ` Andrew Cagney
1 sibling, 0 replies; 6+ messages in thread
From: Andrew Cagney @ 2004-10-14 19:14 UTC (permalink / raw)
To: Michael Chastain; +Cc: zack, gdb-patches
[-- Attachment #1: Type: text/plain, Size: 313 bytes --]
> Except for gdb.mi/gdb701, these are all fixed in gdb HEAD.
> Also fixed in gdb HEAD are gdb.thread/*.exp, plus two tests
> that check the gcc version number that needed 'gcc-4-*'.
>
> gdb_6_2-branch is dead. Andrew is going to create gdb_6_3-branch
> in a few more days or so.
I've checked this in.
Andrew
[-- Attachment #2: diffs --]
[-- Type: text/plain, Size: 479 bytes --]
2004-10-14 Andrew Cagney <cagney@gnu.org>
* gdb.mi/gdb701.c (main): Return 0.
Index: gdb.mi/gdb701.c
===================================================================
RCS file: /cvs/src/src/gdb/testsuite/gdb.mi/gdb701.c,v
retrieving revision 1.1
diff -p -u -r1.1 gdb701.c
--- gdb.mi/gdb701.c 16 Sep 2002 19:01:43 -0000 1.1
+++ gdb.mi/gdb701.c 14 Oct 2004 19:12:43 -0000
@@ -11,5 +11,5 @@ int
main (int argc, char *argv[])
{
Foo *foo = 0;
- exit (0);
+ return 0;
}
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: testsuite fixes for bleeding edge gcc
2004-10-14 18:55 ` Michael Chastain
@ 2004-10-14 21:32 ` Zack Weinberg
0 siblings, 0 replies; 6+ messages in thread
From: Zack Weinberg @ 2004-10-14 21:32 UTC (permalink / raw)
To: Michael Chastain; +Cc: gdb-patches
Michael Chastain <mec.gnu@mindspring.com> writes:
>> whether my GCC changes worked.
>
> Ah. That is a good reason for me to keep up maintenance on the
> 6.2 branch, if it happens often enough. Just because we don't
> release from the branch doesn't mean it's useless.
I don't myself have plans to do this on a regular basis, but it would
be nice if it were something that GCC developers could do easily,
because (as you know) GCC's own testsuite doesn't catch most bugs in
debugging-info output.
This and the use of KFAIL (forcing me to upgrade dejagnu) were the
only major problems I had.
zw
^ permalink raw reply [flat|nested] 6+ messages in thread
end of thread, other threads:[~2004-10-14 21:32 UTC | newest]
Thread overview: 6+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2004-10-14 7:53 testsuite fixes for bleeding edge gcc Zack Weinberg
2004-10-14 12:04 ` Michael Chastain
2004-10-14 16:59 ` Zack Weinberg
2004-10-14 18:55 ` Michael Chastain
2004-10-14 21:32 ` Zack Weinberg
2004-10-14 19:14 ` Andrew Cagney
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox