[patch] Fix watchpoint-at-epilogue 7.0 regression (s390x)

[patch] Fix watchpoint-at-epilogue 7.0 regression (s390x)

[Jan Kratochvil]

Hi,

attached testcase regressed 6.8->7.0 on some s390x systems.

At the moment PC is at epilogue (after frame pointer got destroyed and PC is
still in the calleed function) we must not try any unwinding as it can try to
unwind a garbage.

Correct way would be to implement epilogue unwinder (such as
amd64_epilogue_frame_unwind or i386_epilogue_frame_unwind ones) but it would
not be so easy on s390x as it would need full prologue-analyzer: After SP gets
restored there can be various instructions before the last jump - GDB needs to
interpret the jump to find the register value where it is going to jump to.
00000048279aa09c <round_and_return>:
[...]
  48279aa1b8: eb 6f f0 d0 00 04 lmg  %r6,%r15,208(%r15)
  48279aa1be: 50 32 10 00       st   %r3,0(%r2,%r1)	# SP (%r15) is gone hone
  48279aa1c2: b3 74 00 00       lzer %r0
  48279aa1c6: 07 f4             br   %r4		# %r4 is the target
[...]

Just an easy fix is to be more careful and do not try any unwinding if we find
PC is in the epilogue (previous instruction modifies SP=%r15).

It is even a performance optimization and I see no regression risk there.

The specific s390x error is:
#0  throw_error (error=MEMORY_ERROR, fmt=0x8042676a "Cannot access memory at address %s") at exceptions.c:415
#1  in memory_error (status=5, memaddr=0) at corefile.c:220
#2  in read_memory (memaddr=0, myaddr=0x3ffffc7f7f0 "", len=8) at corefile.c:238
#3  in read_memory_unsigned_integer (memaddr=0, len=8, byte_order=BFD_ENDIAN_BIG) at corefile.c:321
#4  in s390_backchain_frame_unwind_cache (this_frame=0x80608958, info=0x80608a08) at s390-tdep.c:1525
#5  in s390_frame_unwind_cache (this_frame=0x80608958, this_prologue_cache=0x80608970) at s390-tdep.c:1572
#6  in s390_frame_this_id (this_frame=0x80608958, this_prologue_cache=0x80608970, this_id=0x806089b8) at s390-tdep.c:1583
#7  in get_frame_id (fi=0x80608958) at frame.c:335
#8  in frame_find_by_id (id={stack_addr = 4398044824952, code_addr = 2147484952, special_addr = 0, stack_addr_p = 1, code_addr_p = 1, special_addr_p = 0, inline_depth = 0}) at frame.c:587 
#9  in watchpoint_check (p=0x8095bbc0) at breakpoint.c:3203
#10 in catch_errors (func=0x801262c0 <watchpoint_check>, func_args=0x8095bbc0, errstring=0x80a26410 "Error evaluating expression for watchpoint 3\n", mask=6) at exceptions.c:510
#11 in bpstat_check_watchpoint (bs=0x8095bbc0) at breakpoint.c:3404
#12 in bpstat_stop_status (aspace=0x8061fcb0, bp_addr=2147485040, ptid={pid = 17372, lwp = 17372, tid = 0}) at breakpoint.c:3594
#13 in handle_inferior_event (ecs=0x3ffffc80520) at infrun.c:3588

I did not analyze why 6.8 did not error out, the GDB code is similar there.

No regressions on:
{x86_64,x86_64-m32,i686}-fedora12-linux-gnu (CVS HEAD GDB)
s390x-rhel48-linux-gnu (CVS HEAD GDB)
s390-rhel48-linux-gnu (CVS HEAD GDB)
s390x-rhel54-linux-gnu (Fedora 12 GDB)
s390-rhel54-linux-gnu (Fedora 12 GDB)


Thanks,
Jan


gdb/
2009-12-25  Jan Kratochvil  <jan.kratochvil@redhat.com>

	* breakpoint.c (watchpoint_check): Check the call
	gdbarch_in_function_epilogue_p before calling frame_find_by_id.
	Extend the comment.

gdb/testsuite/
2009-12-25  Jan Kratochvil  <jan.kratochvil@redhat.com>

	* gdb.base/watchpoint-cond-gone.exp, gdb.base/watchpoint-cond-gone.c,
	gdb.base/watchpoint-cond-gone-stripped.c: New.

--- a/gdb/breakpoint.c
+++ b/gdb/breakpoint.c
@@ -3200,6 +3200,17 @@ watchpoint_check (void *p)
       struct gdbarch *frame_arch = get_frame_arch (frame);
       CORE_ADDR frame_pc = get_frame_pc (frame);
 
+      /* in_function_epilogue_p() returns a non-zero value if we're still
+	 in the function but the stack frame has already been invalidated.
+	 Since we can't rely on the values of local variables after the
+	 stack has been destroyed, we are treating the watchpoint in that
+	 state as `not changed' without further checking.  Don't mark
+	 watchpoints as changed if the current frame is in an epilogue -
+	 even if they are in some other frame, our view of the stack
+	 is likely to be wrong and frame_find_by_id could error out.  */
+      if (gdbarch_in_function_epilogue_p (frame_arch, frame_pc))
+	return WP_VALUE_NOT_CHANGED;
+
       fr = frame_find_by_id (b->watchpoint_frame);
       within_current_scope = (fr != NULL);
 
@@ -3216,17 +3227,6 @@ watchpoint_check (void *p)
 	    within_current_scope = 0;
 	}
 
-      /* in_function_epilogue_p() returns a non-zero value if we're still
-	 in the function but the stack frame has already been invalidated.
-	 Since we can't rely on the values of local variables after the
-	 stack has been destroyed, we are treating the watchpoint in that
-	 state as `not changed' without further checking.  Don't mark
-	 watchpoints as changed if the current frame is in an epilogue -
-	 even if they are in some other frame, our view of the stack
-	 is likely to be wrong.  */
-      if (gdbarch_in_function_epilogue_p (frame_arch, frame_pc))
-	return WP_VALUE_NOT_CHANGED;
-
       if (within_current_scope)
 	/* If we end up stopping, the current frame will get selected
 	   in normal_stop.  So this call to select_frame won't affect
--- /dev/null
+++ b/gdb/testsuite/gdb.base/watchpoint-cond-gone-stripped.c
@@ -0,0 +1,23 @@
+/* This testcase is part of GDB, the GNU debugger.
+
+   Copyright 2009 Free Software Foundation, Inc.
+
+   This program is free software; you can redistribute it and/or modify
+   it under the terms of the GNU General Public License as published by
+   the Free Software Foundation; either version 2 of the License, or
+   (at your option) any later version.
+
+   This program is distributed in the hope that it will be useful,
+   but WITHOUT ANY WARRANTY; without even the implied warranty of
+   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+   GNU General Public License for more details.
+ 
+   You should have received a copy of the GNU General Public License
+   along with this program; if not, write to the Free Software
+   Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA. */
+
+void
+jumper (void (*jumpto) (void))
+{
+  (*jumpto) ();
+}
--- /dev/null
+++ b/gdb/testsuite/gdb.base/watchpoint-cond-gone.c
@@ -0,0 +1,37 @@
+/* This testcase is part of GDB, the GNU debugger.
+
+   Copyright 2009 Free Software Foundation, Inc.
+
+   This program is free software; you can redistribute it and/or modify
+   it under the terms of the GNU General Public License as published by
+   the Free Software Foundation; either version 2 of the License, or
+   (at your option) any later version.
+
+   This program is distributed in the hope that it will be useful,
+   but WITHOUT ANY WARRANTY; without even the implied warranty of
+   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+   GNU General Public License for more details.
+ 
+   You should have received a copy of the GNU General Public License
+   along with this program; if not, write to the Free Software
+   Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA. */
+
+extern void jumper (void (*jumpto) (void));
+
+static void
+func (void)
+{
+  volatile int c;
+
+  c = 5;
+  c = 10;	/* watchpoint-here */
+  c = 20;
+}
+
+int
+main (void)
+{
+  jumper (func);
+
+  return 0;
+}
--- /dev/null
+++ b/gdb/testsuite/gdb.base/watchpoint-cond-gone.exp
@@ -0,0 +1,51 @@
+# Copyright 2009 Free Software Foundation, Inc.
+
+# This program is free software; you can redistribute it and/or modify
+# it under the terms of the GNU General Public License as published by
+# the Free Software Foundation; either version 2 of the License, or
+# (at your option) any later version.
+#
+# This program is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+# GNU General Public License for more details.
+#
+# You should have received a copy of the GNU General Public License
+# along with this program; if not, write to the Free Software
+# Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
+
+set testfile "watchpoint-cond-gone"
+set srcfile ${testfile}.c
+set srcfilestripped ${testfile}-stripped.c
+set objfilestripped ${objdir}/${subdir}/${testfile}-stripped.o
+set binfile ${objdir}/${subdir}/${testfile}
+
+# We need to generate a function without DWARF to crash older GDB.
+# Stepping into a dynamic function trampoline or stepping out of MAIN may work
+# but it is not a reliable FAIL case.
+
+if  { [gdb_compile "${srcdir}/${subdir}/${srcfilestripped}" "${objfilestripped}" object {}] != ""
+      || [gdb_compile "${srcdir}/${subdir}/${srcfile} ${objfilestripped}" "${binfile}" executable {debug}] != "" } {
+    gdb_suppress_entire_file "Testcase compile failed, so all tests in this file will automatically fail."
+}
+
+clean_restart ${testfile}
+
+# Problem does not occur otherwise.
+gdb_test "set can-use-hw-watchpoints 0"
+
+if ![runto_main] {
+    return -1
+}
+
+gdb_breakpoint [gdb_get_line_number "watchpoint-here"]
+gdb_continue_to_breakpoint "Place to set the watchpoint"
+
+# The condition `c == 30' is the subject being tested.
+gdb_test "watch c if c == 30" "" "Place the watchpoint"
+
+# We may stay either in the function itself or only at the first instruction of
+# its caller depending on the epilogue unwinder (or valid epilogue CFI) presence.
+gdb_test "finish" \
+	 "Watchpoint .* deleted because the program has left the block in.*which its expression is valid..*in (jumper|func).*" \
+	 "Catch the no longer valid watchpoint"

Re: [patch] Fix watchpoint-at-epilogue 7.0 regression (s390x)

[Joel Brobecker]

> gdb/
> 2009-12-25  Jan Kratochvil  <jan.kratochvil@redhat.com>
> 
> 	* breakpoint.c (watchpoint_check): Check the call
> 	gdbarch_in_function_epilogue_p before calling frame_find_by_id.
> 	Extend the comment.

This part looks OK to me.

On a more general side, I wondered for a while if frame_find_by_id
should be modified to never raise an exception (and return NULL instead).
But decided that this was probably a bad idea. Using the same kind
of example as provided by Jan, if our program was to stop inside
the epilogue a function called by the block where the expression
is valid, we might end up deleting the watchpoint prematurely.

> gdb/testsuite/
> 2009-12-25  Jan Kratochvil  <jan.kratochvil@redhat.com>
> 
> 	* gdb.base/watchpoint-cond-gone.exp, gdb.base/watchpoint-cond-gone.c,
> 	gdb.base/watchpoint-cond-gone-stripped.c: New.

OK, with a few changes:
  - The copyright notices needs to be GPL v3;
  - One nit, see below.

> +    gdb_suppress_entire_file "Testcase compile failed, so all tests in this file will automatically fail."

We are now trying to get rid of gdb_suppress_entire_file. Can you just
"return" instead?

BTW: Thank you for using some of the convenience routines. This reminded
me that I wanted to start writing a GDB testcase writing cookbook.  Your
testcase had a lot of stuff that I wanted to put in there, so I just
finally started it!  Will send an email to gdb@ when I'm done...

-- 
Joel

Re: [patch] Fix watchpoint-at-epilogue 7.0 regression (s390x)

[Joel Brobecker]

Something I forgot to mention:

> > 	* gdb.base/watchpoint-cond-gone.exp, gdb.base/watchpoint-cond-gone.c,
> > 	gdb.base/watchpoint-cond-gone-stripped.c: New.

This might be a problem for DJGPP (8+3)...

-- 
Joel

Re: [patch] Fix watchpoint-at-epilogue 7.0 regression (s390x)

[Jan Kratochvil]

On Sun, 27 Dec 2009 06:33:13 +0100, Joel Brobecker wrote:
> Something I forgot to mention:
> 
> > > 	* gdb.base/watchpoint-cond-gone.exp, gdb.base/watchpoint-cond-gone.c,
> > > 	gdb.base/watchpoint-cond-gone-stripped.c: New.
> 
> This might be a problem for DJGPP (8+3)...

gdb/
2009-12-27  Jan Kratochvil  <jan.kratochvil@redhat.com>

	* config/djgpp/fnchange.lst: Add translations for 
	watchpoint-cond-gone.exp, watchpoint-cond-gone.c and
	watchpoint-cond-gone-stripped.c.

(not tested)

I will include it on its check-in.

Still I do not understand it fully - it cannot work after the rename due to the
- rename - and it gets extracted fine by DJGPP tar.exe:
watch-~1.c   watch_~1.exp watchp~2.c   watchp~3.exp watchp~5.c   watchp~6.exp
watch-~1.exp watchp~1.c   watchp~2.exp watchp~4.c   watchp~5.exp watchp~7.c
watch_~1.c   watchp~1.exp watchp~3.c   watchp~4.exp watchp~6.c   watchp~8.c


Thanks,
Jan

--- a/gdb/config/djgpp/fnchange.lst
+++ b/gdb/config/djgpp/fnchange.lst
@@ -400,6 +400,9 @@
 @V@/gdb/testsuite/gdb.base/watchpoint-solib.c @V@/gdb/testsuite/gdb.base/wp-solib.c
 @V@/gdb/testsuite/gdb.base/watchpoint-hw.exp @V@/gdb/testsuite/gdb.base/wp-hw.exp
 @V@/gdb/testsuite/gdb.base/watchpoint-solib.exp @V@/gdb/testsuite/gdb.base/wp-solib.exp
+@V@/gdb/testsuite/gdb.base/watchpoint-cond-gone.exp @V@/gdb/testsuite/gdb.base/wpcondg.exp
+@V@/gdb/testsuite/gdb.base/watchpoint-cond-gone.c @V@/gdb/testsuite/gdb.base/wpcondg.c
+@V@/gdb/testsuite/gdb.base/watchpoint-cond-gone-stripped.c @V@/gdb/testsuite/gdb.base/wpcondgs.c
 @V@/gdb/testsuite/gdb.cell/coremaker-spu.c @V@/gdb/testsuite/gdb.cell/core-spu.c
 @V@/gdb/testsuite/gdb.cell/ea-cache-spu.c @V@/gdb/testsuite/gdb.cell/ea-spu.c
 @V@/gdb/testsuite/gdb.cell/mem-access-spu.c @V@/gdb/testsuite/gdb.cell/mem-spu.c

Re: [patch] Fix watchpoint-at-epilogue 7.0 regression (s390x)

[Eli Zaretskii]

> Date: Sun, 27 Dec 2009 20:29:16 +0100
> From: Jan Kratochvil <jan.kratochvil@redhat.com>
> Cc: gdb-patches@sourceware.org
> 
> gdb/
> 2009-12-27  Jan Kratochvil  <jan.kratochvil@redhat.com>
> 
> 	* config/djgpp/fnchange.lst: Add translations for 
> 	watchpoint-cond-gone.exp, watchpoint-cond-gone.c and
> 	watchpoint-cond-gone-stripped.c.
> 
> (not tested)
> 
> I will include it on its check-in.

Thanks.

> Still I do not understand it fully - it cannot work after the rename

The test suite does not work with DJGPP, anyway, since there's no
`expect' (no async subprocesses).  So renaming does not break anything
that isn't already broken.

> and it gets extracted fine by DJGPP tar.exe:
> watch-~1.c   watch_~1.exp watchp~2.c   watchp~3.exp watchp~5.c   watchp~6.exp
> watch-~1.exp watchp~1.c   watchp~2.exp watchp~4.c   watchp~5.exp watchp~7.c
> watch_~1.c   watchp~1.exp watchp~3.c   watchp~4.exp watchp~6.c   watchp~8.c

When GDB is built for DJGPP, it needs to be extracted by djunpack.bat,
not by Tar.  That is because Tar (and gzip/bzip2) can not be
guaranteed to be available on DJGPP platforms.  djunpack.bat invokes
djtar.exe, which is part of the core DJGPP distro and includes gzip
and bzip2 decompressors built-in.  djtar.exe will complain when it
encounters file-name clashes, and will ask for a different file name.
Typing a new name for a lot of files is annoying.  The file produced
from fnchange.lst is submitted to djtar.exe to tell it how to rename
each file whose name will otherwise clash, so the user is not
bothered.  Using fnchange.lst also ensures all DJGPP users have the
same files after unpacking, instead of relying on various tools each
of which does its own trickery to cope with 8+3 namespace.

> --- a/gdb/config/djgpp/fnchange.lst
> +++ b/gdb/config/djgpp/fnchange.lst
> @@ -400,6 +400,9 @@
>  @V@/gdb/testsuite/gdb.base/watchpoint-solib.c @V@/gdb/testsuite/gdb.base/wp-solib.c
>  @V@/gdb/testsuite/gdb.base/watchpoint-hw.exp @V@/gdb/testsuite/gdb.base/wp-hw.exp
>  @V@/gdb/testsuite/gdb.base/watchpoint-solib.exp @V@/gdb/testsuite/gdb.base/wp-solib.exp
> +@V@/gdb/testsuite/gdb.base/watchpoint-cond-gone.exp @V@/gdb/testsuite/gdb.base/wpcondg.exp
> +@V@/gdb/testsuite/gdb.base/watchpoint-cond-gone.c @V@/gdb/testsuite/gdb.base/wpcondg.c
> +@V@/gdb/testsuite/gdb.base/watchpoint-cond-gone-stripped.c @V@/gdb/testsuite/gdb.base/wpcondgs.c
>  @V@/gdb/testsuite/gdb.cell/coremaker-spu.c @V@/gdb/testsuite/gdb.cell/core-spu.c
>  @V@/gdb/testsuite/gdb.cell/ea-cache-spu.c @V@/gdb/testsuite/gdb.cell/ea-spu.c
>  @V@/gdb/testsuite/gdb.cell/mem-access-spu.c @V@/gdb/testsuite/gdb.cell/mem-spu.c

This is fine.

Re: [patch] Fix watchpoint-at-epilogue 7.0 regression (s390x)

[Jan Kratochvil]

On Sun, 27 Dec 2009 20:59:05 +0100, Eli Zaretskii wrote:
> djtar.exe will complain when it encounters file-name clashes, and will ask
> for a different file name.

ah, thanks, I tried just that bare DJGPP tar.exe.


Regards,
Jan

Re: [patch] Fix watchpoint-at-epilogue 7.0 regression (s390x)

[Jan Kratochvil]

On Sat, 26 Dec 2009 06:35:55 +0100, Joel Brobecker wrote:
> > gdb/
> > 2009-12-25  Jan Kratochvil  <jan.kratochvil@redhat.com>
> > 
> > 	* breakpoint.c (watchpoint_check): Check the call
> > 	gdbarch_in_function_epilogue_p before calling frame_find_by_id.
> > 	Extend the comment.
> 
> This part looks OK to me.

> > gdb/testsuite/
> > 2009-12-25  Jan Kratochvil  <jan.kratochvil@redhat.com>
> > 
> > 	* gdb.base/watchpoint-cond-gone.exp, gdb.base/watchpoint-cond-gone.c,
> > 	gdb.base/watchpoint-cond-gone-stripped.c: New.
> 
> OK, with a few changes:
>   - The copyright notices needs to be GPL v3;

Done.

> > +    gdb_suppress_entire_file "Testcase compile failed, so all tests in this file will automatically fail."
> 
> We are now trying to get rid of gdb_suppress_entire_file. Can you just
> "return" instead?

Done.


> This reminded me that I wanted to start writing a GDB testcase writing
> cookbook.  Your testcase had a lot of stuff that I wanted to put in there,
> so I just finally started it!  Will send an email to gdb@ when I'm done...

http://sourceware.org/gdb/wiki/GDBTestcaseCookbook

Updated it now with a testcase template.  Unfortunately the GDB wiki always
hangs during check-in and also I never got any change notification.  Therefore
guessing the notifications do not work.


On Sun, 27 Dec 2009 20:59:05 +0100, Eli Zaretskii wrote:
> > --- a/gdb/config/djgpp/fnchange.lst
> > +++ b/gdb/config/djgpp/fnchange.lst
> 
> This is fine.


Checked-in.


Thanks,
Jan


http://sourceware.org/ml/gdb-cvs/2010-01/msg00169.html

--- src/gdb/ChangeLog	2010/01/19 18:11:17	1.11276
+++ src/gdb/ChangeLog	2010/01/19 20:31:32	1.11277
@@ -1,3 +1,12 @@
+2010-01-19  Jan Kratochvil  <jan.kratochvil@redhat.com>
+
+	* breakpoint.c (watchpoint_check): Check the call
+	gdbarch_in_function_epilogue_p before calling frame_find_by_id.
+	Extend the comment.
+	* config/djgpp/fnchange.lst: Add translations for 
+	watchpoint-cond-gone.exp, watchpoint-cond-gone.c and
+	watchpoint-cond-gone-stripped.c.
+
 2010-01-19  Tom Tromey  <tromey@redhat.com>
 
 	PR c++/8000:
--- src/gdb/breakpoint.c	2010/01/19 09:39:11	1.450
+++ src/gdb/breakpoint.c	2010/01/19 20:31:34	1.451
@@ -3217,6 +3217,17 @@
       struct gdbarch *frame_arch = get_frame_arch (frame);
       CORE_ADDR frame_pc = get_frame_pc (frame);
 
+      /* in_function_epilogue_p() returns a non-zero value if we're still
+	 in the function but the stack frame has already been invalidated.
+	 Since we can't rely on the values of local variables after the
+	 stack has been destroyed, we are treating the watchpoint in that
+	 state as `not changed' without further checking.  Don't mark
+	 watchpoints as changed if the current frame is in an epilogue -
+	 even if they are in some other frame, our view of the stack
+	 is likely to be wrong and frame_find_by_id could error out.  */
+      if (gdbarch_in_function_epilogue_p (frame_arch, frame_pc))
+	return WP_VALUE_NOT_CHANGED;
+
       fr = frame_find_by_id (b->watchpoint_frame);
       within_current_scope = (fr != NULL);
 
@@ -3233,17 +3244,6 @@
 	    within_current_scope = 0;
 	}
 
-      /* in_function_epilogue_p() returns a non-zero value if we're still
-	 in the function but the stack frame has already been invalidated.
-	 Since we can't rely on the values of local variables after the
-	 stack has been destroyed, we are treating the watchpoint in that
-	 state as `not changed' without further checking.  Don't mark
-	 watchpoints as changed if the current frame is in an epilogue -
-	 even if they are in some other frame, our view of the stack
-	 is likely to be wrong.  */
-      if (gdbarch_in_function_epilogue_p (frame_arch, frame_pc))
-	return WP_VALUE_NOT_CHANGED;
-
       if (within_current_scope)
 	/* If we end up stopping, the current frame will get selected
 	   in normal_stop.  So this call to select_frame won't affect
--- src/gdb/config/djgpp/fnchange.lst	2010/01/01 06:53:34	1.106
+++ src/gdb/config/djgpp/fnchange.lst	2010/01/19 20:31:35	1.107
@@ -401,6 +401,9 @@
 @V@/gdb/testsuite/gdb.base/watchpoint-solib.c @V@/gdb/testsuite/gdb.base/wp-solib.c
 @V@/gdb/testsuite/gdb.base/watchpoint-hw.exp @V@/gdb/testsuite/gdb.base/wp-hw.exp
 @V@/gdb/testsuite/gdb.base/watchpoint-solib.exp @V@/gdb/testsuite/gdb.base/wp-solib.exp
+@V@/gdb/testsuite/gdb.base/watchpoint-cond-gone.exp @V@/gdb/testsuite/gdb.base/wpcondg.exp
+@V@/gdb/testsuite/gdb.base/watchpoint-cond-gone.c @V@/gdb/testsuite/gdb.base/wpcondg.c
+@V@/gdb/testsuite/gdb.base/watchpoint-cond-gone-stripped.c @V@/gdb/testsuite/gdb.base/wpcondgs.c
 @V@/gdb/testsuite/gdb.cell/coremaker-spu.c @V@/gdb/testsuite/gdb.cell/core-spu.c
 @V@/gdb/testsuite/gdb.cell/ea-cache-spu.c @V@/gdb/testsuite/gdb.cell/ea-spu.c
 @V@/gdb/testsuite/gdb.cell/mem-access-spu.c @V@/gdb/testsuite/gdb.cell/mem-spu.c
--- src/gdb/testsuite/ChangeLog	2010/01/19 18:11:19	1.2094
+++ src/gdb/testsuite/ChangeLog	2010/01/19 20:31:37	1.2095
@@ -1,3 +1,8 @@
+2010-01-19  Jan Kratochvil  <jan.kratochvil@redhat.com>
+
+	* gdb.base/watchpoint-cond-gone.exp, gdb.base/watchpoint-cond-gone.c,
+	gdb.base/watchpoint-cond-gone-stripped.c: New.
+
 2010-01-19  Tom Tromey  <tromey@redhat.com>
 
 	PR c++/8000:
--- src/gdb/testsuite/gdb.base/watchpoint-cond-gone-stripped.c
+++ src/gdb/testsuite/gdb.base/watchpoint-cond-gone-stripped.c	2010-01-19 20:47:03.774343000 +0000
@@ -0,0 +1,22 @@
+/* This testcase is part of GDB, the GNU debugger.
+
+   Copyright 2010 Free Software Foundation, Inc.
+
+   This program is free software; you can redistribute it and/or modify
+   it under the terms of the GNU General Public License as published by
+   the Free Software Foundation; either version 3 of the License, or
+   (at your option) any later version.
+
+   This program is distributed in the hope that it will be useful,
+   but WITHOUT ANY WARRANTY; without even the implied warranty of
+   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+   GNU General Public License for more details.
+
+   You should have received a copy of the GNU General Public License
+   along with this program.  If not, see <http://www.gnu.org/licenses/>.  */
+
+void
+jumper (void (*jumpto) (void))
+{
+  (*jumpto) ();
+}
--- src/gdb/testsuite/gdb.base/watchpoint-cond-gone.c
+++ src/gdb/testsuite/gdb.base/watchpoint-cond-gone.c	2010-01-19 20:47:04.302028000 +0000
@@ -0,0 +1,36 @@
+/* This testcase is part of GDB, the GNU debugger.
+
+   Copyright 2010 Free Software Foundation, Inc.
+
+   This program is free software; you can redistribute it and/or modify
+   it under the terms of the GNU General Public License as published by
+   the Free Software Foundation; either version 3 of the License, or
+   (at your option) any later version.
+
+   This program is distributed in the hope that it will be useful,
+   but WITHOUT ANY WARRANTY; without even the implied warranty of
+   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+   GNU General Public License for more details.
+
+   You should have received a copy of the GNU General Public License
+   along with this program.  If not, see <http://www.gnu.org/licenses/>.  */
+
+extern void jumper (void (*jumpto) (void));
+
+static void
+func (void)
+{
+  volatile int c;
+
+  c = 5;
+  c = 10;	/* watchpoint-here */
+  c = 20;
+}
+
+int
+main (void)
+{
+  jumper (func);
+
+  return 0;
+}
--- src/gdb/testsuite/gdb.base/watchpoint-cond-gone.exp
+++ src/gdb/testsuite/gdb.base/watchpoint-cond-gone.exp	2010-01-19 20:47:04.789355000 +0000
@@ -0,0 +1,51 @@
+# Copyright 2010 Free Software Foundation, Inc.
+
+# This program is free software; you can redistribute it and/or modify
+# it under the terms of the GNU General Public License as published by
+# the Free Software Foundation; either version 3 of the License, or
+# (at your option) any later version.
+#
+# This program is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+# GNU General Public License for more details.
+#
+# You should have received a copy of the GNU General Public License
+# along with this program.  If not, see <http://www.gnu.org/licenses/>.
+
+set testfile "watchpoint-cond-gone"
+set srcfile ${testfile}.c
+set srcfilestripped ${testfile}-stripped.c
+set objfilestripped ${objdir}/${subdir}/${testfile}-stripped.o
+set binfile ${objdir}/${subdir}/${testfile}
+
+# We need to generate a function without DWARF to crash older GDB.
+# Stepping into a dynamic function trampoline or stepping out of MAIN may work
+# but it is not a reliable FAIL case.
+
+if  { [gdb_compile "${srcdir}/${subdir}/${srcfilestripped}" "${objfilestripped}" object {}] != ""
+      || [gdb_compile "${srcdir}/${subdir}/${srcfile} ${objfilestripped}" "${binfile}" executable {debug}] != "" } {
+    untested watchpoint-cond-gone.exp
+    return -1
+}
+
+clean_restart ${testfile}
+
+# Problem does not occur otherwise.
+gdb_test "set can-use-hw-watchpoints 0"
+
+if ![runto_main] {
+    return -1
+}
+
+gdb_breakpoint [gdb_get_line_number "watchpoint-here"]
+gdb_continue_to_breakpoint "Place to set the watchpoint"
+
+# The condition `c == 30' is the subject being tested.
+gdb_test "watch c if c == 30" "" "Place the watchpoint"
+
+# We may stay either in the function itself or only at the first instruction of
+# its caller depending on the epilogue unwinder (or valid epilogue CFI) presence.
+gdb_test "finish" \
+	 "Watchpoint .* deleted because the program has left the block in.*which its expression is valid..*in (jumper|func).*" \
+	 "Catch the no longer valid watchpoint"

Re: [patch] Fix watchpoint-at-epilogue 7.0 regression (s390x)

[Joel Brobecker]

> http://sourceware.org/gdb/wiki/GDBTestcaseCookbook
> 
> Updated it now with a testcase template.  Unfortunately the GDB wiki always
> hangs during check-in and also I never got any change notification.  Therefore
> guessing the notifications do not work.

Thanks. I got the notification, they work. One possible reason for
not sending it is that you are the author. Authors never get emails
for their own changes.

-- 
Joel