public inbox for gcc-bugs@sourceware.org
help / color / mirror / Atom feed
* [Bug libgomp/32789] Profiling not possible with -fopenmp
[not found] <bug-32789-4@http.gcc.gnu.org/bugzilla/>
@ 2021-12-29 7:19 ` pinskia at gcc dot gnu.org
0 siblings, 0 replies; 8+ messages in thread
From: pinskia at gcc dot gnu.org @ 2021-12-29 7:19 UTC (permalink / raw)
To: gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=32789
Andrew Pinski <pinskia at gcc dot gnu.org> changed:
What |Removed |Added
----------------------------------------------------------------------------
Resolution|--- |MOVED
Status|UNCONFIRMED |RESOLVED
--- Comment #7 from Andrew Pinski <pinskia at gcc dot gnu.org> ---
gprof and mcount is under control of glibc project.
There was an old email about this on the glibc side at
https://sourceware.org/legacy-ml/libc-hacker/1999-01/msg00105.html .
^ permalink raw reply [flat|nested] 8+ messages in thread
* [Bug libgomp/32789] Profiling not possible with -fopenmp
2007-07-17 8:49 [Bug libgomp/32789] New: " jensseidel at users dot sf dot net
` (4 preceding siblings ...)
2007-07-17 10:27 ` pinskia at gmail dot com
@ 2007-07-17 11:00 ` jensseidel at users dot sf dot net
5 siblings, 0 replies; 8+ messages in thread
From: jensseidel at users dot sf dot net @ 2007-07-17 11:00 UTC (permalink / raw)
To: gcc-bugs
------- Comment #6 from jensseidel at users dot sf dot net 2007-07-17 11:00 -------
(In reply to comment #4)
> grpof profiling is all done via a call to mcount and mcount is controlled by
> libc (in the GNU/Linux case glibc). So I doubt this is a GCC bug.>
OK, for the record: I use OpenSuse 10.2 with glibc 2.5.
According to http://www.cs.utah.edu/dept/old/texinfo/as/gprof.html#SEC1
mcount occurs in the gprof output but I haven't seen this yet (gprof
2.17.50.0.5).
(In reply to comment #5)
> And that is the reason why GCC added atomic builtins when openmp came
> in also. There are manuals for a reason :).
I don't understand this but it may be off topic. (Should I inform the Open MPI
people about atomic assembler locking code in gcc so that they can reuse it?)
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=32789
^ permalink raw reply [flat|nested] 8+ messages in thread
* [Bug libgomp/32789] Profiling not possible with -fopenmp
2007-07-17 8:49 [Bug libgomp/32789] New: " jensseidel at users dot sf dot net
` (3 preceding siblings ...)
2007-07-17 10:24 ` pinskia at gcc dot gnu dot org
@ 2007-07-17 10:27 ` pinskia at gmail dot com
2007-07-17 11:00 ` jensseidel at users dot sf dot net
5 siblings, 0 replies; 8+ messages in thread
From: pinskia at gmail dot com @ 2007-07-17 10:27 UTC (permalink / raw)
To: gcc-bugs
------- Comment #5 from pinskia at gmail dot com 2007-07-17 10:26 -------
Subject: Re: Profiling not possible with -fopenmp
On 17 Jul 2007 10:24:12 -0000, jensseidel at users dot sf dot net
<gcc-bugzilla@gcc.gnu.org> wrote:
> An Open MPI related discussion about atomic operations happened
> the last days, because architecture specific assembler code failed again
> for some exotic platforms.
And that is the reason why GCC added atomic builtins when openmp came
in also. There are manuals for a reason :).
-- Pinski
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=32789
^ permalink raw reply [flat|nested] 8+ messages in thread
* Re: [Bug libgomp/32789] Profiling not possible with -fopenmp
2007-07-17 10:24 ` jensseidel at users dot sf dot net
@ 2007-07-17 10:26 ` Andrew Pinski
0 siblings, 0 replies; 8+ messages in thread
From: Andrew Pinski @ 2007-07-17 10:26 UTC (permalink / raw)
To: gcc-bugzilla; +Cc: gcc-bugs
On 17 Jul 2007 10:24:12 -0000, jensseidel at users dot sf dot net
<gcc-bugzilla@gcc.gnu.org> wrote:
> An Open MPI related discussion about atomic operations happened
> the last days, because architecture specific assembler code failed again
> for some exotic platforms.
And that is the reason why GCC added atomic builtins when openmp came
in also. There are manuals for a reason :).
-- Pinski
^ permalink raw reply [flat|nested] 8+ messages in thread
* [Bug libgomp/32789] Profiling not possible with -fopenmp
2007-07-17 8:49 [Bug libgomp/32789] New: " jensseidel at users dot sf dot net
` (2 preceding siblings ...)
2007-07-17 10:24 ` jensseidel at users dot sf dot net
@ 2007-07-17 10:24 ` pinskia at gcc dot gnu dot org
2007-07-17 10:27 ` pinskia at gmail dot com
2007-07-17 11:00 ` jensseidel at users dot sf dot net
5 siblings, 0 replies; 8+ messages in thread
From: pinskia at gcc dot gnu dot org @ 2007-07-17 10:24 UTC (permalink / raw)
To: gcc-bugs
------- Comment #4 from pinskia at gcc dot gnu dot org 2007-07-17 10:24 -------
grpof profiling is all done via a call to mcount and mcount is controlled by
libc (in the GNU/Linux case glibc). So I doubt this is a GCC bug.
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=32789
^ permalink raw reply [flat|nested] 8+ messages in thread
* [Bug libgomp/32789] Profiling not possible with -fopenmp
2007-07-17 8:49 [Bug libgomp/32789] New: " jensseidel at users dot sf dot net
2007-07-17 9:55 ` [Bug libgomp/32789] " theodore dot papadopoulo at sophia dot inria dot fr
2007-07-17 10:11 ` theodore dot papadopoulo at sophia dot inria dot fr
@ 2007-07-17 10:24 ` jensseidel at users dot sf dot net
2007-07-17 10:26 ` Andrew Pinski
2007-07-17 10:24 ` pinskia at gcc dot gnu dot org
` (2 subsequent siblings)
5 siblings, 1 reply; 8+ messages in thread
From: jensseidel at users dot sf dot net @ 2007-07-17 10:24 UTC (permalink / raw)
To: gcc-bugs
------- Comment #3 from jensseidel at users dot sf dot net 2007-07-17 10:24 -------
(In reply to comment #2)
> And to reply to myself, it needs either to use thread local storage to hold the
> counters and then to add some piece of code to fuse the values of the various
> counters at the end of a thread (which might not be easy ?) or to use atomic
> operations (which existence depends on the architecture, but I hope that all
> multi-core processors have such instructions).
(Don't forget about SMP machines, I have a SGI Octane (2 x Mips R12000 CPUs).)
An Open MPI related discussion about atomic operations happened
the last days, because architecture specific assembler code failed again
for some exotic platforms. See e.g.
http://lists.debian.org/debian-mips/2007/07/msg00036.html
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=32789
^ permalink raw reply [flat|nested] 8+ messages in thread
* [Bug libgomp/32789] Profiling not possible with -fopenmp
2007-07-17 8:49 [Bug libgomp/32789] New: " jensseidel at users dot sf dot net
2007-07-17 9:55 ` [Bug libgomp/32789] " theodore dot papadopoulo at sophia dot inria dot fr
@ 2007-07-17 10:11 ` theodore dot papadopoulo at sophia dot inria dot fr
2007-07-17 10:24 ` jensseidel at users dot sf dot net
` (3 subsequent siblings)
5 siblings, 0 replies; 8+ messages in thread
From: theodore dot papadopoulo at sophia dot inria dot fr @ 2007-07-17 10:11 UTC (permalink / raw)
To: gcc-bugs
------- Comment #2 from theodore dot papadopoulo at sophia dot inria dot fr 2007-07-17 10:11 -------
And to reply to myself, it needs either to use thread local storage to hold the
counters and then to add some piece of code to fuse the values of the various
counters at the end of a thread (which might not be easy ?) or to use atomic
operations (which existence depends on the architecture, but I hope that all
multi-core processors have such instructions).
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=32789
^ permalink raw reply [flat|nested] 8+ messages in thread
* [Bug libgomp/32789] Profiling not possible with -fopenmp
2007-07-17 8:49 [Bug libgomp/32789] New: " jensseidel at users dot sf dot net
@ 2007-07-17 9:55 ` theodore dot papadopoulo at sophia dot inria dot fr
2007-07-17 10:11 ` theodore dot papadopoulo at sophia dot inria dot fr
` (4 subsequent siblings)
5 siblings, 0 replies; 8+ messages in thread
From: theodore dot papadopoulo at sophia dot inria dot fr @ 2007-07-17 9:55 UTC (permalink / raw)
To: gcc-bugs
------- Comment #1 from theodore dot papadopoulo at sophia dot inria dot fr 2007-07-17 09:55 -------
This is similar to the comment (maybe misplaced) of
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=31862
The problem, as far as I understand it is that any kind of profiling (gprof,
profile-arcs, probably mudflap, ...) rely on some global variables that would
need
to be made thread local.
I do not know how difficult this would be however....
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=32789
^ permalink raw reply [flat|nested] 8+ messages in thread
end of thread, other threads:[~2021-12-29 7:19 UTC | newest]
Thread overview: 8+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
[not found] <bug-32789-4@http.gcc.gnu.org/bugzilla/>
2021-12-29 7:19 ` [Bug libgomp/32789] Profiling not possible with -fopenmp pinskia at gcc dot gnu.org
2007-07-17 8:49 [Bug libgomp/32789] New: " jensseidel at users dot sf dot net
2007-07-17 9:55 ` [Bug libgomp/32789] " theodore dot papadopoulo at sophia dot inria dot fr
2007-07-17 10:11 ` theodore dot papadopoulo at sophia dot inria dot fr
2007-07-17 10:24 ` jensseidel at users dot sf dot net
2007-07-17 10:26 ` Andrew Pinski
2007-07-17 10:24 ` pinskia at gcc dot gnu dot org
2007-07-17 10:27 ` pinskia at gmail dot com
2007-07-17 11:00 ` jensseidel at users dot sf dot net
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).