public inbox for gdb-prs@sourceware.org
help / color / mirror / Atom feed
* [Bug testsuite/13859] New: `set non-stop 1' triggers some fails in gdb.threads/watchthreads-reorder.exp
@ 2012-03-16 8:05 qiyao at gcc dot gnu.org
2012-03-16 10:23 ` [Bug testsuite/13859] " palves at redhat dot com
` (4 more replies)
0 siblings, 5 replies; 6+ messages in thread
From: qiyao at gcc dot gnu.org @ 2012-03-16 8:05 UTC (permalink / raw)
To: gdb-prs
http://sourceware.org/bugzilla/show_bug.cgi?id=13859
Bug #: 13859
Summary: `set non-stop 1' triggers some fails in
gdb.threads/watchthreads-reorder.exp
Product: gdb
Version: HEAD
Status: NEW
Severity: normal
Priority: P2
Component: testsuite
AssignedTo: unassigned@sourceware.org
ReportedBy: qiyao@gcc.gnu.org
Classification: Unclassified
When running gdb.threads/watchthreads-reorder.exp in native gdb when displaced
stepping turned on, we got following fails,
FAIL: gdb.threads/watchthreads-reorder.exp: reorder0: continue a (timeout)
FAIL: gdb.threads/watchthreads-reorder.exp: reorder0: continue to breakpoint:
break-at-exit
FAIL: gdb.threads/watchthreads-reorder.exp: reorder1: continue a (timeout)
FAIL: gdb.threads/watchthreads-reorder.exp: reorder1: rwatch unused1_rwatch
FAIL: gdb.threads/watchthreads-reorder.exp: reorder1: rwatch unused2_rwatch
FAIL: gdb.threads/watchthreads-reorder.exp: reorder1: continue b
FAIL: gdb.threads/watchthreads-reorder.exp: reorder1: continue to breakpoint:
break-at-exit
(gdb) PASS: gdb.threads/watchthreads-reorder.exp: reorder0: rwatch
thread2_rwatch
break 364^M
Breakpoint 4 at 0x8049942: file
../../../gdb/gdb/testsuite/gdb.threads/watchthreads-reorder.c, line 364.^M
(gdb) continue^M
Continuing.^M
[New Thread 0x403d9b70 (LWP 1823)]^M
[New Thread 0x405dab70 (LWP 1824)]^M
Stopping GDB PID 1816.^M
Waiting till the threads initialize their TIDs.^M
Thread 1 TID = 1823, thread 2 TID = 1824, PID = 1820.^M
Waiting till the threads get trapped by the watchpoints.^M
Resuming GDB PID 1816.^M
Joining the threads.^M
Hardware read watchpoint 3: thread2_rwatch^M
^M
Value = 0^M
0x08048f9f in thread2_func (unused=0x0) at
../../../gdb/gdb/testsuite/gdb.threads/watchthreads-reorder.c:126^M
126 rwatch_store = thread2_rwatch;^M
(gdb) Hardware read watchpoint 2: thread1_rwatch^M
^M
Value = 0^M
0x08048eb6 in thread1_func (unused=0x0) at
../../../gdb/gdb/testsuite/gdb.threads/watchthreads-reorder.c:100^M
100 rwatch_store = thread1_rwatch;^M
FAIL: gdb.threads/watchthreads-reorder.exp: reorder0: continue a (timeout)
Haven't analyzed these fails carefully yet, but looks like they are caused by
the test case that is unable to parse/match output after displaced stepping is
on. Open this PR against component `testsuite'.
--
Configure bugmail: http://sourceware.org/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.
^ permalink raw reply [flat|nested] 6+ messages in thread
* [Bug testsuite/13859] `set non-stop 1' triggers some fails in gdb.threads/watchthreads-reorder.exp
2012-03-16 8:05 [Bug testsuite/13859] New: `set non-stop 1' triggers some fails in gdb.threads/watchthreads-reorder.exp qiyao at gcc dot gnu.org
@ 2012-03-16 10:23 ` palves at redhat dot com
2012-03-16 11:07 ` [Bug testsuite/13859] Fails in gdb.threads/watchthreads-reorder.exp when run it in non-stop + async qiyao at gcc dot gnu.org
` (3 subsequent siblings)
4 siblings, 0 replies; 6+ messages in thread
From: palves at redhat dot com @ 2012-03-16 10:23 UTC (permalink / raw)
To: gdb-prs
http://sourceware.org/bugzilla/show_bug.cgi?id=13859
Pedro Alves <palves at redhat dot com> changed:
What |Removed |Added
----------------------------------------------------------------------------
CC| |palves at redhat dot com
--- Comment #1 from Pedro Alves <palves at redhat dot com> 2012-03-16 10:22:36 UTC ---
> Haven't analyzed these fails carefully yet, but looks like they are caused by
> the test case that is unable to parse/match output after displaced stepping is
> on. Open this PR against component `testsuite'.
Weird, just displaced stepping ("set displaced-stepping on") should not affect
the output. non-stop+async has different behavior and output by design. Is
that what you're really trying? These tests have been written for all-stop.
Why would you be running them with "set non-stop 1", and expecting them to
work?
--
Configure bugmail: http://sourceware.org/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.
^ permalink raw reply [flat|nested] 6+ messages in thread
* [Bug testsuite/13859] Fails in gdb.threads/watchthreads-reorder.exp when run it in non-stop + async
2012-03-16 8:05 [Bug testsuite/13859] New: `set non-stop 1' triggers some fails in gdb.threads/watchthreads-reorder.exp qiyao at gcc dot gnu.org
2012-03-16 10:23 ` [Bug testsuite/13859] " palves at redhat dot com
@ 2012-03-16 11:07 ` qiyao at gcc dot gnu.org
2012-03-16 11:25 ` palves at redhat dot com
` (2 subsequent siblings)
4 siblings, 0 replies; 6+ messages in thread
From: qiyao at gcc dot gnu.org @ 2012-03-16 11:07 UTC (permalink / raw)
To: gdb-prs
http://sourceware.org/bugzilla/show_bug.cgi?id=13859
Yao Qi <qiyao at gcc dot gnu.org> changed:
What |Removed |Added
----------------------------------------------------------------------------
Summary|`set non-stop 1' triggers |Fails in
|some fails in |gdb.threads/watchthreads-re
|gdb.threads/watchthreads-re |order.exp when run it in
|order.exp |non-stop + async
--- Comment #2 from Yao Qi <qiyao at gcc dot gnu.org> 2012-03-16 11:06:38 UTC ---
(In reply to comment #1)
> > Haven't analyzed these fails carefully yet, but looks like they are caused by
> > the test case that is unable to parse/match output after displaced stepping is
> > on. Open this PR against component `testsuite'.
>
> Weird, just displaced stepping ("set displaced-stepping on") should not affect
I was running it in non-stop+async.
> the output. non-stop+async has different behavior and output by design. Is
> that what you're really trying? These tests have been written for all-stop.
Yes, that is what I really want. I ported your patch to CVS trunk, and run
test suite,
[RFC/WIP PATCH v2 02/14] Implement all-stop on top of a target running
non-stop mode
http://sourceware.org/ml/gdb-patches/2011-12/msg00564.html
In order to differentiate these regressions fails are caused this patch above
or just non-stop+async, I run them in non-stop+async mode, see these fails, and
open these PRs.
--
Configure bugmail: http://sourceware.org/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.
^ permalink raw reply [flat|nested] 6+ messages in thread
* [Bug testsuite/13859] Fails in gdb.threads/watchthreads-reorder.exp when run it in non-stop + async
2012-03-16 8:05 [Bug testsuite/13859] New: `set non-stop 1' triggers some fails in gdb.threads/watchthreads-reorder.exp qiyao at gcc dot gnu.org
2012-03-16 10:23 ` [Bug testsuite/13859] " palves at redhat dot com
2012-03-16 11:07 ` [Bug testsuite/13859] Fails in gdb.threads/watchthreads-reorder.exp when run it in non-stop + async qiyao at gcc dot gnu.org
@ 2012-03-16 11:25 ` palves at redhat dot com
2012-03-16 12:09 ` palves at redhat dot com
2012-03-16 13:34 ` qiyao at gcc dot gnu.org
4 siblings, 0 replies; 6+ messages in thread
From: palves at redhat dot com @ 2012-03-16 11:25 UTC (permalink / raw)
To: gdb-prs
http://sourceware.org/bugzilla/show_bug.cgi?id=13859
--- Comment #3 from Pedro Alves <palves at redhat dot com> 2012-03-16 11:24:22 UTC ---
> Yes, that is what I really want. I ported your patch to CVS trunk, and run
> test suite,
No it's not what you want. :-)
That patch implements *all-stop* on top of non-stop. There should be no user
visible change compared to an unpatched gdb, *when you run in all-stop*.
Running an all-stop test with "set non-stop 1" tends to only show red herrings,
as it changes _user-visible behavior and output_.
non-stop enables a collection of secondary modes: async, and
displaced-stepping, so test those individually with an unpatched gdb:
set GDBFLAGS "-ex \"set target-async on\""
and:
set GDBFLAGS "-ex \"set displaced-stepping on\""
--
Configure bugmail: http://sourceware.org/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.
^ permalink raw reply [flat|nested] 6+ messages in thread
* [Bug testsuite/13859] Fails in gdb.threads/watchthreads-reorder.exp when run it in non-stop + async
2012-03-16 8:05 [Bug testsuite/13859] New: `set non-stop 1' triggers some fails in gdb.threads/watchthreads-reorder.exp qiyao at gcc dot gnu.org
` (2 preceding siblings ...)
2012-03-16 11:25 ` palves at redhat dot com
@ 2012-03-16 12:09 ` palves at redhat dot com
2012-03-16 13:34 ` qiyao at gcc dot gnu.org
4 siblings, 0 replies; 6+ messages in thread
From: palves at redhat dot com @ 2012-03-16 12:09 UTC (permalink / raw)
To: gdb-prs
http://sourceware.org/bugzilla/show_bug.cgi?id=13859
--- Comment #4 from Pedro Alves <palves at redhat dot com> 2012-03-16 12:08:49 UTC ---
Oh, and non-stop also enables always-inserted mode, so also run the testsuite
on the pristine gdb with just:
set GDBFLAGS "-ex \"set breakpoint always-inserted on\""
--
Configure bugmail: http://sourceware.org/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.
^ permalink raw reply [flat|nested] 6+ messages in thread
* [Bug testsuite/13859] Fails in gdb.threads/watchthreads-reorder.exp when run it in non-stop + async
2012-03-16 8:05 [Bug testsuite/13859] New: `set non-stop 1' triggers some fails in gdb.threads/watchthreads-reorder.exp qiyao at gcc dot gnu.org
` (3 preceding siblings ...)
2012-03-16 12:09 ` palves at redhat dot com
@ 2012-03-16 13:34 ` qiyao at gcc dot gnu.org
4 siblings, 0 replies; 6+ messages in thread
From: qiyao at gcc dot gnu.org @ 2012-03-16 13:34 UTC (permalink / raw)
To: gdb-prs
http://sourceware.org/bugzilla/show_bug.cgi?id=13859
Yao Qi <qiyao at gcc dot gnu.org> changed:
What |Removed |Added
----------------------------------------------------------------------------
Status|NEW |RESOLVED
Resolution| |INVALID
--- Comment #5 from Yao Qi <qiyao at gcc dot gnu.org> 2012-03-16 13:34:01 UTC ---
(In reply to comment #3)
> > Yes, that is what I really want. I ported your patch to CVS trunk, and run
> > test suite,
>
> No it's not what you want. :-)
>
> That patch implements *all-stop* on top of non-stop. There should be no user
> visible change compared to an unpatched gdb, *when you run in all-stop*.
>
I see. That makes sense to me.
> Running an all-stop test with "set non-stop 1" tends to only show red herrings,
> as it changes _user-visible behavior and output_.
>
> non-stop enables a collection of secondary modes: async, and
> displaced-stepping, so test those individually with an unpatched gdb:
>
> set GDBFLAGS "-ex \"set target-async on\""
>
> and:
>
> set GDBFLAGS "-ex \"set displaced-stepping on\""
Re-run this test case with asyn on and displaced-stepping on respectively. No
fails. Close this one as it is invalid.
--
Configure bugmail: http://sourceware.org/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.
^ permalink raw reply [flat|nested] 6+ messages in thread
end of thread, other threads:[~2012-03-16 13:34 UTC | newest]
Thread overview: 6+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2012-03-16 8:05 [Bug testsuite/13859] New: `set non-stop 1' triggers some fails in gdb.threads/watchthreads-reorder.exp qiyao at gcc dot gnu.org
2012-03-16 10:23 ` [Bug testsuite/13859] " palves at redhat dot com
2012-03-16 11:07 ` [Bug testsuite/13859] Fails in gdb.threads/watchthreads-reorder.exp when run it in non-stop + async qiyao at gcc dot gnu.org
2012-03-16 11:25 ` palves at redhat dot com
2012-03-16 12:09 ` palves at redhat dot com
2012-03-16 13:34 ` qiyao at gcc dot gnu.org
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).