public inbox for gcc-bugs@sourceware.org
help / color / mirror / Atom feed
* [Bug c/13985] New: [3.4/3.5 regression] ICE in gcc.c-torture/compile/930621-1.c
@ 2004-02-02 21:35 cato at df dot lth dot se
2004-02-02 21:52 ` [Bug optimization/13985] " pinskia at gcc dot gnu dot org
` (26 more replies)
0 siblings, 27 replies; 28+ messages in thread
From: cato at df dot lth dot se @ 2004-02-02 21:35 UTC (permalink / raw)
To: gcc-bugs
The gcc.c-torture/compile/930621-1.c ICE when using -O3 for
i386-unknown-netbsdelf1.6 on mainline and the 3.4 branch.
The failure (compiled from today's mainline):
/tmp/build/gcc/xgcc -B/tmp/build/gcc -O3 -c 930621-1.c
930621-1.c: In function `modlookup':
930621-1.c:38: internal compiler error: in get_loop_body, at cfgloop.c:958
Please submit a full bug report,
with preprocessed source if appropriate.
See <URL:http://gcc.gnu.org/bugs.html> for instructions.
Compiler version:
/tmp/build/gcc/xgcc -v
Using built-in specs.
Configured with: ../gcc/configure --disable-werror
Thread model: single
gcc version 3.5.0 20040202 (experimental)
The commit that broke this test is:
CVSROOT: /cvs/gcc
Module name: gcc
Changes by: steven@gcc.gnu.org 2004-01-05 09:35:06
Modified files:
gcc : toplev.c
Log message:
I am a moron.
Patches:
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/toplev.c.diff?cvsroot=gcc&r1=1.857&r2=1.858
--
Summary: [3.4/3.5 regression] ICE in gcc.c-
torture/compile/930621-1.c
Product: gcc
Version: 3.5.0
Status: UNCONFIRMED
Severity: normal
Priority: P2
Component: c
AssignedTo: unassigned at gcc dot gnu dot org
ReportedBy: cato at df dot lth dot se
CC: gcc-bugs at gcc dot gnu dot org
GCC build triplet: i386-unknown-netbsdelf1.6
GCC host triplet: i386-unknown-netbsdelf1.6
GCC target triplet: i386-unknown-netbsdelf1.6
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=13985
^ permalink raw reply [flat|nested] 28+ messages in thread
* [Bug optimization/13985] [3.4/3.5 regression] ICE in gcc.c-torture/compile/930621-1.c
2004-02-02 21:35 [Bug c/13985] New: [3.4/3.5 regression] ICE in gcc.c-torture/compile/930621-1.c cato at df dot lth dot se
@ 2004-02-02 21:52 ` pinskia at gcc dot gnu dot org
2004-02-02 23:35 ` steven at gcc dot gnu dot org
` (25 subsequent siblings)
26 siblings, 0 replies; 28+ messages in thread
From: pinskia at gcc dot gnu dot org @ 2004-02-02 21:52 UTC (permalink / raw)
To: gcc-bugs
--
What |Removed |Added
----------------------------------------------------------------------------
CC| |steven at gcc dot gnu dot
| |org
Severity|normal |critical
Component|c |optimization
Keywords| |ice-on-valid-code
Target Milestone|--- |3.4.0
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=13985
^ permalink raw reply [flat|nested] 28+ messages in thread
* [Bug optimization/13985] [3.4/3.5 regression] ICE in gcc.c-torture/compile/930621-1.c
2004-02-02 21:35 [Bug c/13985] New: [3.4/3.5 regression] ICE in gcc.c-torture/compile/930621-1.c cato at df dot lth dot se
2004-02-02 21:52 ` [Bug optimization/13985] " pinskia at gcc dot gnu dot org
@ 2004-02-02 23:35 ` steven at gcc dot gnu dot org
2004-02-03 0:53 ` steven at gcc dot gnu dot org
` (24 subsequent siblings)
26 siblings, 0 replies; 28+ messages in thread
From: steven at gcc dot gnu dot org @ 2004-02-02 23:35 UTC (permalink / raw)
To: gcc-bugs
------- Additional Comments From steven at gcc dot gnu dot org 2004-02-02 23:35 -------
This checkin of mine is a good candidate for the worst checkin ever. The fix
has been on mainline for a couple of days now, and I'll put it on the 3.4
branch later today.
--
What |Removed |Added
----------------------------------------------------------------------------
Status|UNCONFIRMED |NEW
Ever Confirmed| |1
Last reconfirmed|0000-00-00 00:00:00 |2004-02-02 23:35:35
date| |
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=13985
^ permalink raw reply [flat|nested] 28+ messages in thread
* [Bug optimization/13985] [3.4/3.5 regression] ICE in gcc.c-torture/compile/930621-1.c
2004-02-02 21:35 [Bug c/13985] New: [3.4/3.5 regression] ICE in gcc.c-torture/compile/930621-1.c cato at df dot lth dot se
2004-02-02 21:52 ` [Bug optimization/13985] " pinskia at gcc dot gnu dot org
2004-02-02 23:35 ` steven at gcc dot gnu dot org
@ 2004-02-03 0:53 ` steven at gcc dot gnu dot org
2004-02-03 5:51 ` cato at df dot lth dot se
` (23 subsequent siblings)
26 siblings, 0 replies; 28+ messages in thread
From: steven at gcc dot gnu dot org @ 2004-02-03 0:53 UTC (permalink / raw)
To: gcc-bugs
------- Additional Comments From steven at gcc dot gnu dot org 2004-02-03 00:53 -------
This checking should fix it:
http://gcc.gnu.org/ml/gcc-cvs/2004-02/msg00086.html
Can you confirm that please?
--
What |Removed |Added
----------------------------------------------------------------------------
AssignedTo|unassigned at gcc dot gnu |steven at gcc dot gnu dot
|dot org |org
Status|NEW |ASSIGNED
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=13985
^ permalink raw reply [flat|nested] 28+ messages in thread
* [Bug optimization/13985] [3.4/3.5 regression] ICE in gcc.c-torture/compile/930621-1.c
2004-02-02 21:35 [Bug c/13985] New: [3.4/3.5 regression] ICE in gcc.c-torture/compile/930621-1.c cato at df dot lth dot se
` (2 preceding siblings ...)
2004-02-03 0:53 ` steven at gcc dot gnu dot org
@ 2004-02-03 5:51 ` cato at df dot lth dot se
2004-02-03 8:39 ` stevenb at suse dot de
` (22 subsequent siblings)
26 siblings, 0 replies; 28+ messages in thread
From: cato at df dot lth dot se @ 2004-02-03 5:51 UTC (permalink / raw)
To: gcc-bugs
------- Additional Comments From cato at df dot lth dot se 2004-02-03 05:51 -------
Subject: Re: [3.4/3.5 regression] ICE in
gcc.c-torture/compile/930621-1.c
On Tue, 2 Feb 2004, steven at gcc dot gnu dot org wrote:
> ------- Additional Comments From steven at gcc dot gnu dot org 2004-02-02 23:35 -------
> This checkin of mine is a good candidate for the worst checkin ever. The fix
> has been on mainline for a couple of days now, and I'll put it on the 3.4
> branch later today.
The problem is still present on mainline -- the test did fail when I last
bootstrapped and tested, 10 hours ago...
/Krister
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=13985
^ permalink raw reply [flat|nested] 28+ messages in thread
* [Bug optimization/13985] [3.4/3.5 regression] ICE in gcc.c-torture/compile/930621-1.c
2004-02-02 21:35 [Bug c/13985] New: [3.4/3.5 regression] ICE in gcc.c-torture/compile/930621-1.c cato at df dot lth dot se
` (3 preceding siblings ...)
2004-02-03 5:51 ` cato at df dot lth dot se
@ 2004-02-03 8:39 ` stevenb at suse dot de
2004-02-03 8:59 ` steven at gcc dot gnu dot org
` (21 subsequent siblings)
26 siblings, 0 replies; 28+ messages in thread
From: stevenb at suse dot de @ 2004-02-03 8:39 UTC (permalink / raw)
To: gcc-bugs
------- Additional Comments From stevenb at suse dot de 2004-02-03 08:39 -------
Subject: Re: [3.4/3.5 regression] ICE in gcc.c-torture/compile/930621-1.c
On Tuesday 03 February 2004 06:51, cato at df dot lth dot se wrote:
> The problem is still present on mainline -- the test did fail when I last
> bootstrapped and tested, 10 hours ago...
Then my broken checkin cannot be the cause of this because it has been
corrected on mainline several days ago. Something must have broken it
before December 30, 2003.
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=13985
^ permalink raw reply [flat|nested] 28+ messages in thread
* [Bug optimization/13985] [3.4/3.5 regression] ICE in gcc.c-torture/compile/930621-1.c
2004-02-02 21:35 [Bug c/13985] New: [3.4/3.5 regression] ICE in gcc.c-torture/compile/930621-1.c cato at df dot lth dot se
` (4 preceding siblings ...)
2004-02-03 8:39 ` stevenb at suse dot de
@ 2004-02-03 8:59 ` steven at gcc dot gnu dot org
2004-02-03 23:01 ` cato at df dot lth dot se
` (20 subsequent siblings)
26 siblings, 0 replies; 28+ messages in thread
From: steven at gcc dot gnu dot org @ 2004-02-03 8:59 UTC (permalink / raw)
To: gcc-bugs
------- Additional Comments From steven at gcc dot gnu dot org 2004-02-03 08:59 -------
Indeed, according to the posted test results, this was already
failing before December 30, 2003. In fact it also FAILed in
October. See the following two messages:
http://gcc.gnu.org/ml/gcc-testresults/2003-12/msg00853.html
http://gcc.gnu.org/ml/gcc-testresults/2003-10/msg01343.html
So this is not a new FAIL, just one that was hidden because
of a broken checkin that moved loop2.
--
What |Removed |Added
----------------------------------------------------------------------------
Status|ASSIGNED |RESOLVED
Resolution| |INVALID
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=13985
^ permalink raw reply [flat|nested] 28+ messages in thread
* [Bug optimization/13985] [3.4/3.5 regression] ICE in gcc.c-torture/compile/930621-1.c
2004-02-02 21:35 [Bug c/13985] New: [3.4/3.5 regression] ICE in gcc.c-torture/compile/930621-1.c cato at df dot lth dot se
` (5 preceding siblings ...)
2004-02-03 8:59 ` steven at gcc dot gnu dot org
@ 2004-02-03 23:01 ` cato at df dot lth dot se
2004-02-07 1:42 ` cato at df dot lth dot se
` (19 subsequent siblings)
26 siblings, 0 replies; 28+ messages in thread
From: cato at df dot lth dot se @ 2004-02-03 23:01 UTC (permalink / raw)
To: gcc-bugs
------- Additional Comments From cato at df dot lth dot se 2004-02-03 23:01 -------
Subject: Re: [3.4/3.5 regression] ICE in
gcc.c-torture/compile/930621-1.c
On Tue, 3 Feb 2004, steven at gcc dot gnu dot org wrote:
> So this is not a new FAIL, just one that was hidden because
> of a broken checkin that moved loop2.
OK. I'll find out what broke it the first time...
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=13985
^ permalink raw reply [flat|nested] 28+ messages in thread
* [Bug optimization/13985] [3.4/3.5 regression] ICE in gcc.c-torture/compile/930621-1.c
2004-02-02 21:35 [Bug c/13985] New: [3.4/3.5 regression] ICE in gcc.c-torture/compile/930621-1.c cato at df dot lth dot se
` (6 preceding siblings ...)
2004-02-03 23:01 ` cato at df dot lth dot se
@ 2004-02-07 1:42 ` cato at df dot lth dot se
2004-02-07 2:19 ` pinskia at gcc dot gnu dot org
` (18 subsequent siblings)
26 siblings, 0 replies; 28+ messages in thread
From: cato at df dot lth dot se @ 2004-02-07 1:42 UTC (permalink / raw)
To: gcc-bugs
------- Additional Comments From cato at df dot lth dot se 2004-02-07 01:42 -------
Subject: Re: [3.4/3.5 regression] ICE in
gcc.c-torture/compile/930621-1.c
The change that introduced this failure is:
2003-10-25 Roger Sayle <roger@eyesopen.com>
* simplify-rtx.c (simplify_replace_rtx): Avoid allocating duplicate
RTL nodes. If an operator's operands are unchanged, return the
original argument unchanged.
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=13985
^ permalink raw reply [flat|nested] 28+ messages in thread
* [Bug optimization/13985] [3.4/3.5 regression] ICE in gcc.c-torture/compile/930621-1.c
2004-02-02 21:35 [Bug c/13985] New: [3.4/3.5 regression] ICE in gcc.c-torture/compile/930621-1.c cato at df dot lth dot se
` (7 preceding siblings ...)
2004-02-07 1:42 ` cato at df dot lth dot se
@ 2004-02-07 2:19 ` pinskia at gcc dot gnu dot org
2004-02-07 2:22 ` pinskia at gcc dot gnu dot org
` (17 subsequent siblings)
26 siblings, 0 replies; 28+ messages in thread
From: pinskia at gcc dot gnu dot org @ 2004-02-07 2:19 UTC (permalink / raw)
To: gcc-bugs
------- Additional Comments From pinskia at gcc dot gnu dot org 2004-02-07 02:19 -------
Reopening based on it is not fixed.
--
What |Removed |Added
----------------------------------------------------------------------------
Status|RESOLVED |REOPENED
Resolution|INVALID |
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=13985
^ permalink raw reply [flat|nested] 28+ messages in thread
* [Bug optimization/13985] [3.4/3.5 regression] ICE in gcc.c-torture/compile/930621-1.c
2004-02-02 21:35 [Bug c/13985] New: [3.4/3.5 regression] ICE in gcc.c-torture/compile/930621-1.c cato at df dot lth dot se
` (8 preceding siblings ...)
2004-02-07 2:19 ` pinskia at gcc dot gnu dot org
@ 2004-02-07 2:22 ` pinskia at gcc dot gnu dot org
2004-03-04 1:57 ` pinskia at gcc dot gnu dot org
` (16 subsequent siblings)
26 siblings, 0 replies; 28+ messages in thread
From: pinskia at gcc dot gnu dot org @ 2004-02-07 2:22 UTC (permalink / raw)
To: gcc-bugs
------- Additional Comments From pinskia at gcc dot gnu dot org 2004-02-07 02:22 -------
Roger this is caused by the same patch as before, but this is for i386 instead of i686.
--
What |Removed |Added
----------------------------------------------------------------------------
AssignedTo|steven at gcc dot gnu dot |roger at eyesopen dot com
|org |
Status|REOPENED |ASSIGNED
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=13985
^ permalink raw reply [flat|nested] 28+ messages in thread
* [Bug optimization/13985] [3.4/3.5 regression] ICE in gcc.c-torture/compile/930621-1.c
2004-02-02 21:35 [Bug c/13985] New: [3.4/3.5 regression] ICE in gcc.c-torture/compile/930621-1.c cato at df dot lth dot se
` (9 preceding siblings ...)
2004-02-07 2:22 ` pinskia at gcc dot gnu dot org
@ 2004-03-04 1:57 ` pinskia at gcc dot gnu dot org
2004-03-06 11:36 ` ebotcazou at gcc dot gnu dot org
` (15 subsequent siblings)
26 siblings, 0 replies; 28+ messages in thread
From: pinskia at gcc dot gnu dot org @ 2004-03-04 1:57 UTC (permalink / raw)
To: gcc-bugs
------- Additional Comments From pinskia at gcc dot gnu dot org 2004-03-04 01:57 -------
Roger are you still working on this, I think this is not yours though.
--
What |Removed |Added
----------------------------------------------------------------------------
AssignedTo|roger at eyesopen dot com |unassigned at gcc dot gnu
| |dot org
Severity|critical |minor
Status|ASSIGNED |NEW
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=13985
^ permalink raw reply [flat|nested] 28+ messages in thread
* [Bug optimization/13985] [3.4/3.5 regression] ICE in gcc.c-torture/compile/930621-1.c
2004-02-02 21:35 [Bug c/13985] New: [3.4/3.5 regression] ICE in gcc.c-torture/compile/930621-1.c cato at df dot lth dot se
` (10 preceding siblings ...)
2004-03-04 1:57 ` pinskia at gcc dot gnu dot org
@ 2004-03-06 11:36 ` ebotcazou at gcc dot gnu dot org
2004-03-11 6:31 ` ebotcazou at gcc dot gnu dot org
` (14 subsequent siblings)
26 siblings, 0 replies; 28+ messages in thread
From: ebotcazou at gcc dot gnu dot org @ 2004-03-06 11:36 UTC (permalink / raw)
To: gcc-bugs
------- Additional Comments From ebotcazou at gcc dot gnu dot org 2004-03-06 11:36 -------
Investigating.
--
What |Removed |Added
----------------------------------------------------------------------------
AssignedTo|unassigned at gcc dot gnu |ebotcazou at gcc dot gnu dot
|dot org |org
Status|NEW |ASSIGNED
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=13985
^ permalink raw reply [flat|nested] 28+ messages in thread
* [Bug optimization/13985] [3.4/3.5 regression] ICE in gcc.c-torture/compile/930621-1.c
2004-02-02 21:35 [Bug c/13985] New: [3.4/3.5 regression] ICE in gcc.c-torture/compile/930621-1.c cato at df dot lth dot se
` (11 preceding siblings ...)
2004-03-06 11:36 ` ebotcazou at gcc dot gnu dot org
@ 2004-03-11 6:31 ` ebotcazou at gcc dot gnu dot org
2004-03-13 19:46 ` ebotcazou at gcc dot gnu dot org
` (13 subsequent siblings)
26 siblings, 0 replies; 28+ messages in thread
From: ebotcazou at gcc dot gnu dot org @ 2004-03-11 6:31 UTC (permalink / raw)
To: gcc-bugs
------- Additional Comments From ebotcazou at gcc dot gnu dot org 2004-03-11 06:31 -------
Patch: http://gcc.gnu.org/ml/gcc-patches/2004-03/msg00672.html
--
What |Removed |Added
----------------------------------------------------------------------------
CC| |mmitchel at gcc dot gnu dot
| |org
Keywords| |patch
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=13985
^ permalink raw reply [flat|nested] 28+ messages in thread
* [Bug optimization/13985] [3.4/3.5 regression] ICE in gcc.c-torture/compile/930621-1.c
2004-02-02 21:35 [Bug c/13985] New: [3.4/3.5 regression] ICE in gcc.c-torture/compile/930621-1.c cato at df dot lth dot se
` (12 preceding siblings ...)
2004-03-11 6:31 ` ebotcazou at gcc dot gnu dot org
@ 2004-03-13 19:46 ` ebotcazou at gcc dot gnu dot org
2004-03-21 18:56 ` mmitchel at gcc dot gnu dot org
` (12 subsequent siblings)
26 siblings, 0 replies; 28+ messages in thread
From: ebotcazou at gcc dot gnu dot org @ 2004-03-13 19:46 UTC (permalink / raw)
To: gcc-bugs
--
What |Removed |Added
----------------------------------------------------------------------------
Severity|minor |normal
GCC build triplet|i386-unknown-netbsdelf1.6 |i386-*-*
GCC host triplet|i386-unknown-netbsdelf1.6 |i386-*-*
GCC target triplet|i386-unknown-netbsdelf1.6 |i386-*-*
Version|3.5.0 |3.4.0
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=13985
^ permalink raw reply [flat|nested] 28+ messages in thread
* [Bug optimization/13985] [3.4/3.5 regression] ICE in gcc.c-torture/compile/930621-1.c
2004-02-02 21:35 [Bug c/13985] New: [3.4/3.5 regression] ICE in gcc.c-torture/compile/930621-1.c cato at df dot lth dot se
` (13 preceding siblings ...)
2004-03-13 19:46 ` ebotcazou at gcc dot gnu dot org
@ 2004-03-21 18:56 ` mmitchel at gcc dot gnu dot org
2004-04-18 22:45 ` mmitchel at gcc dot gnu dot org
` (11 subsequent siblings)
26 siblings, 0 replies; 28+ messages in thread
From: mmitchel at gcc dot gnu dot org @ 2004-03-21 18:56 UTC (permalink / raw)
To: gcc-bugs
------- Additional Comments From mmitchel at gcc dot gnu dot org 2004-03-21 18:56 -------
I keep looking at this patch, and I keep not knowing whether I know enough to
approve it.
Eric, would you run this patch past RTH and/or Jeff Law and see what their
reaction is?
If they think it's OK for 3.4.0, put it in. Otherwise, let's move this back to
3.4.1.
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=13985
^ permalink raw reply [flat|nested] 28+ messages in thread
* [Bug optimization/13985] [3.4/3.5 regression] ICE in gcc.c-torture/compile/930621-1.c
2004-02-02 21:35 [Bug c/13985] New: [3.4/3.5 regression] ICE in gcc.c-torture/compile/930621-1.c cato at df dot lth dot se
` (14 preceding siblings ...)
2004-03-21 18:56 ` mmitchel at gcc dot gnu dot org
@ 2004-04-18 22:45 ` mmitchel at gcc dot gnu dot org
2004-04-19 11:14 ` ebotcazou at gcc dot gnu dot org
` (10 subsequent siblings)
26 siblings, 0 replies; 28+ messages in thread
From: mmitchel at gcc dot gnu dot org @ 2004-04-18 22:45 UTC (permalink / raw)
To: gcc-bugs
------- Additional Comments From mmitchel at gcc dot gnu dot org 2004-04-18 21:53 -------
It appears that Richard/Jeff were either never asked or never approved this
patch. Therefore, I've postponed it until GCC 3.4.1.
--
What |Removed |Added
----------------------------------------------------------------------------
Target Milestone|3.4.0 |3.4.1
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=13985
^ permalink raw reply [flat|nested] 28+ messages in thread
* [Bug optimization/13985] [3.4/3.5 regression] ICE in gcc.c-torture/compile/930621-1.c
2004-02-02 21:35 [Bug c/13985] New: [3.4/3.5 regression] ICE in gcc.c-torture/compile/930621-1.c cato at df dot lth dot se
` (15 preceding siblings ...)
2004-04-18 22:45 ` mmitchel at gcc dot gnu dot org
@ 2004-04-19 11:14 ` ebotcazou at gcc dot gnu dot org
2004-04-19 17:31 ` law at redhat dot com
` (9 subsequent siblings)
26 siblings, 0 replies; 28+ messages in thread
From: ebotcazou at gcc dot gnu dot org @ 2004-04-19 11:14 UTC (permalink / raw)
To: gcc-bugs
------- Additional Comments From ebotcazou at gcc dot gnu dot org 2004-04-19 05:29 -------
For the sake of completeness: they were both asked.
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=13985
^ permalink raw reply [flat|nested] 28+ messages in thread
* [Bug optimization/13985] [3.4/3.5 regression] ICE in gcc.c-torture/compile/930621-1.c
2004-02-02 21:35 [Bug c/13985] New: [3.4/3.5 regression] ICE in gcc.c-torture/compile/930621-1.c cato at df dot lth dot se
` (16 preceding siblings ...)
2004-04-19 11:14 ` ebotcazou at gcc dot gnu dot org
@ 2004-04-19 17:31 ` law at redhat dot com
2004-04-19 23:45 ` ebotcazou at gcc dot gnu dot org
` (8 subsequent siblings)
26 siblings, 0 replies; 28+ messages in thread
From: law at redhat dot com @ 2004-04-19 17:31 UTC (permalink / raw)
To: gcc-bugs
------- Additional Comments From law at redhat dot com 2004-04-19 16:25 -------
Subject: Re: [3.4/3.5 regression] ICE in
gcc.c-torture/compile/930621-1.c
In message <20040419052909.23755.qmail@sources.redhat.com>, "ebotcazou at gcc d
ot gnu dot org" writes:
>
>------- Additional Comments From ebotcazou at gcc dot gnu dot org 2004-04-19
> 05:29 -------
>For the sake of completeness: they were both asked.
Sorry, I must have missed that email message (Eric certainly sent it back in
March).
The patch looks reasonable to me -- though I must admit, I'm not very
familiar with the code in cfgloopmanip.c.
You might ping Zdenek directly for a sanity check -- he wrote much of that
code originally and is pretty sharp regarding updates of datastructures
derived from the CFG such as the loop tree.
jeff
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=13985
^ permalink raw reply [flat|nested] 28+ messages in thread
* [Bug optimization/13985] [3.4/3.5 regression] ICE in gcc.c-torture/compile/930621-1.c
2004-02-02 21:35 [Bug c/13985] New: [3.4/3.5 regression] ICE in gcc.c-torture/compile/930621-1.c cato at df dot lth dot se
` (17 preceding siblings ...)
2004-04-19 17:31 ` law at redhat dot com
@ 2004-04-19 23:45 ` ebotcazou at gcc dot gnu dot org
2004-04-20 0:07 ` rakdver at atrey dot karlin dot mff dot cuni dot cz
` (7 subsequent siblings)
26 siblings, 0 replies; 28+ messages in thread
From: ebotcazou at gcc dot gnu dot org @ 2004-04-19 23:45 UTC (permalink / raw)
To: gcc-bugs
------- Additional Comments From ebotcazou at gcc dot gnu dot org 2004-04-19 21:27 -------
Thanks Jeff.
Zdenek, would you mind taking a look at the patch linked to in comment #12? Do
you think the problem is correctly assessed? If so, do you think it's the right
solution to the problem? Thanks in advance.
--
What |Removed |Added
----------------------------------------------------------------------------
CC| |rakdver at gcc dot gnu dot
| |org
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=13985
^ permalink raw reply [flat|nested] 28+ messages in thread
* [Bug optimization/13985] [3.4/3.5 regression] ICE in gcc.c-torture/compile/930621-1.c
2004-02-02 21:35 [Bug c/13985] New: [3.4/3.5 regression] ICE in gcc.c-torture/compile/930621-1.c cato at df dot lth dot se
` (18 preceding siblings ...)
2004-04-19 23:45 ` ebotcazou at gcc dot gnu dot org
@ 2004-04-20 0:07 ` rakdver at atrey dot karlin dot mff dot cuni dot cz
2004-04-20 9:06 ` ebotcazou at gcc dot gnu dot org
` (6 subsequent siblings)
26 siblings, 0 replies; 28+ messages in thread
From: rakdver at atrey dot karlin dot mff dot cuni dot cz @ 2004-04-20 0:07 UTC (permalink / raw)
To: gcc-bugs
------- Additional Comments From rakdver at atrey dot karlin dot mff dot cuni dot cz 2004-04-19 22:53 -------
Subject: Re: [3.4/3.5 regression] ICE in gcc.c-torture/compile/930621-1.c
Hello,
> Zdenek, would you mind taking a look at the patch linked to in comment #12? Do
> you think the problem is correctly assessed? If so, do you think it's the right
> solution to the problem? Thanks in advance.
the patch seems OK to me (good work -- I was not really able to write this part of
loop manipulation in a clear way :-( ). Perhaps I would suggest to formulate the
comment
+ /* Changing the placement of a loop in the loop tree may have an
+ effect on its preheader with regard to the condition stated in
+ the description of fix_bb_placement. */
+ fix_bb_placements (loops, loop_preheader_edge (loop)->src);
a bit more precisely:
We have determined that LOOP is no longer a subloop of its original
LOOP->FATHER. This means that we removed some edge(s) and thus
cancelled all paths from the LOOP to LOOP->FATHER's latch. This might
however cause also cancelation of all paths from the preheader of the
loop (and possibly some of its predecessors) to this latch, so we need
to update their placement inside loops as well.
Zdenek
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=13985
^ permalink raw reply [flat|nested] 28+ messages in thread
* [Bug optimization/13985] [3.4/3.5 regression] ICE in gcc.c-torture/compile/930621-1.c
2004-02-02 21:35 [Bug c/13985] New: [3.4/3.5 regression] ICE in gcc.c-torture/compile/930621-1.c cato at df dot lth dot se
` (19 preceding siblings ...)
2004-04-20 0:07 ` rakdver at atrey dot karlin dot mff dot cuni dot cz
@ 2004-04-20 9:06 ` ebotcazou at gcc dot gnu dot org
2004-04-20 9:07 ` rakdver at atrey dot karlin dot mff dot cuni dot cz
` (5 subsequent siblings)
26 siblings, 0 replies; 28+ messages in thread
From: ebotcazou at gcc dot gnu dot org @ 2004-04-20 9:06 UTC (permalink / raw)
To: gcc-bugs
------- Additional Comments From ebotcazou at gcc dot gnu dot org 2004-04-20 06:48 -------
Thanks Zdenek.
However, as a GCC hacker who didn't know anything about the code in
cfgloopmanip.c before trying to fix the bug, I think your comment is too
detailed(!) and assume too much of a familiarity with the cfg*.c modules.
I think we must comment on the same level of abstraction as the function itself.
The description of fix_bb_placement is nice:
/* Fix placement of basic block BB inside loop hierarchy stored in LOOPS --
Let L be a loop to that BB belongs. Then every successor of BB must either
1) belong to some superloop of loop L, or
2) be a header of loop K such that K->outer is superloop of L
Returns true if we had to move BB into other loop to enforce this condition,
false if the placement of BB was already correct (provided that placements
of its successors are correct). */
I think it is obvious to see (well, modulo a few seconds :-) that if a loop is
reparented, condition 2) is altered for its preheader. You don't need to know
anything about the latch and the search order.
What about the following wording instead?
/* Changing the placement of a loop in the loop tree may alter the validity
of condition 2) stated in the description of fix_bb_placement for its
preheader, because the successor is the header and belongs to the loop.
So call fix_bb_placements to fix up the placement of the preheader and
(possibly) of its predecessors. */
[As you have probably guessed, this whole comment is primarily aimed at being
considered as a nominee for the next hair-splitting contest :-)]
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=13985
^ permalink raw reply [flat|nested] 28+ messages in thread
* [Bug optimization/13985] [3.4/3.5 regression] ICE in gcc.c-torture/compile/930621-1.c
2004-02-02 21:35 [Bug c/13985] New: [3.4/3.5 regression] ICE in gcc.c-torture/compile/930621-1.c cato at df dot lth dot se
` (20 preceding siblings ...)
2004-04-20 9:06 ` ebotcazou at gcc dot gnu dot org
@ 2004-04-20 9:07 ` rakdver at atrey dot karlin dot mff dot cuni dot cz
2004-04-20 11:03 ` ebotcazou at gcc dot gnu dot org
` (4 subsequent siblings)
26 siblings, 0 replies; 28+ messages in thread
From: rakdver at atrey dot karlin dot mff dot cuni dot cz @ 2004-04-20 9:07 UTC (permalink / raw)
To: gcc-bugs
------- Additional Comments From rakdver at atrey dot karlin dot mff dot cuni dot cz 2004-04-20 07:18 -------
Subject: Re: [3.4/3.5 regression] ICE in gcc.c-torture/compile/930621-1.c
Hello,
> However, as a GCC hacker who didn't know anything about the code in
> cfgloopmanip.c before trying to fix the bug, I think your comment is too
> detailed(!) and assume too much of a familiarity with the cfg*.c modules.
I don't understand? The comment as I formulated it assumes very little
knowledge about our implementation (basically just that loop->father is
superloop of the loop). It only refers to the well-known definition of
the natural loop.
> I think we must comment on the same level of abstraction as the function itself.
I was sort of thinking about extending the comments at the
fix_{bb,loop}_placement in a similar way (i.e. to explain not only what
they are doing, but also why it is so).
> The description of fix_bb_placement is nice:
> /* Fix placement of basic block BB inside loop hierarchy stored in LOOPS --
> Let L be a loop to that BB belongs. Then every successor of BB must either
> 1) belong to some superloop of loop L, or
> 2) be a header of loop K such that K->outer is superloop of L
> Returns true if we had to move BB into other loop to enforce this condition,
> false if the placement of BB was already correct (provided that placements
> of its successors are correct). */
>
> I think it is obvious to see (well, modulo a few seconds :-) that if a loop is
> reparented, condition 2) is altered for its preheader. You don't need to know
> anything about the latch and the search order.
>
> What about the following wording instead?
>
> /* Changing the placement of a loop in the loop tree may alter the validity
> of condition 2) stated in the description of fix_bb_placement for its
> preheader, because the successor is the header and belongs to the loop.
> So call fix_bb_placements to fix up the placement of the preheader and
> (possibly) of its predecessors. */
>
>
> [As you have probably guessed, this whole comment is primarily aimed at being
> considered as a nominee for the next hair-splitting contest :-)]
Zdenek
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=13985
^ permalink raw reply [flat|nested] 28+ messages in thread
* [Bug optimization/13985] [3.4/3.5 regression] ICE in gcc.c-torture/compile/930621-1.c
2004-02-02 21:35 [Bug c/13985] New: [3.4/3.5 regression] ICE in gcc.c-torture/compile/930621-1.c cato at df dot lth dot se
` (21 preceding siblings ...)
2004-04-20 9:07 ` rakdver at atrey dot karlin dot mff dot cuni dot cz
@ 2004-04-20 11:03 ` ebotcazou at gcc dot gnu dot org
2004-04-23 22:17 ` cvs-commit at gcc dot gnu dot org
` (3 subsequent siblings)
26 siblings, 0 replies; 28+ messages in thread
From: ebotcazou at gcc dot gnu dot org @ 2004-04-20 11:03 UTC (permalink / raw)
To: gcc-bugs
------- Additional Comments From ebotcazou at gcc dot gnu dot org 2004-04-20 09:24 -------
> I don't understand? The comment as I formulated it assumes very little
> knowledge about our implementation (basically just that loop->father is
> superloop of the loop). It only refers to the well-known definition of
> the natural loop.
That loop->father is a superloop of the loop is hardly problematic indeed :-)
But "all paths from the LOOP to LOOP->FATHER's latch" implicitly assumes that
you know the criterion based on which the loop are sorted and, given the lack of
hurry to approve the patch :-), I don't think everyone is as familiar as you
with this stuff.
> I was sort of thinking about extending the comments at the
> fix_{bb,loop}_placement in a similar way (i.e. to explain not only what
> they are doing, but also why it is so).
IMHO such a general comment would rather belong in the class of file-scope
comments (that cfgloopmanip.c lacks btw). As I said, the comment of
fix_bb_placement is nice because it matches the level of abstraction of the
function itself (which doesn't deal with latches).
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=13985
^ permalink raw reply [flat|nested] 28+ messages in thread
* [Bug optimization/13985] [3.4/3.5 regression] ICE in gcc.c-torture/compile/930621-1.c
2004-02-02 21:35 [Bug c/13985] New: [3.4/3.5 regression] ICE in gcc.c-torture/compile/930621-1.c cato at df dot lth dot se
` (22 preceding siblings ...)
2004-04-20 11:03 ` ebotcazou at gcc dot gnu dot org
@ 2004-04-23 22:17 ` cvs-commit at gcc dot gnu dot org
2004-04-23 22:27 ` cvs-commit at gcc dot gnu dot org
` (2 subsequent siblings)
26 siblings, 0 replies; 28+ messages in thread
From: cvs-commit at gcc dot gnu dot org @ 2004-04-23 22:17 UTC (permalink / raw)
To: gcc-bugs
------- Additional Comments From cvs-commit at gcc dot gnu dot org 2004-04-23 22:01 -------
Subject: Bug 13985
CVSROOT: /cvs/gcc
Module name: gcc
Changes by: ebotcazou@gcc.gnu.org 2004-04-23 22:01:36
Modified files:
gcc : ChangeLog cfgloopmanip.c
gcc/testsuite : ChangeLog
Added files:
gcc/testsuite/gcc.dg: loop-3.c
Log message:
PR optimization/13985
* cfgloopmanip.c (fix_loop_placements): New prototype.
Call fix_bb_placements on the preheader of loops that have
been reparented.
(remove_path): Adjust call to fix_loop_placements.
Patches:
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/ChangeLog.diff?cvsroot=gcc&r1=2.3473&r2=2.3474
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/cfgloopmanip.c.diff?cvsroot=gcc&r1=1.24&r2=1.25
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/testsuite/ChangeLog.diff?cvsroot=gcc&r1=1.3699&r2=1.3700
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/testsuite/gcc.dg/loop-3.c.diff?cvsroot=gcc&r1=NONE&r2=1.1
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=13985
^ permalink raw reply [flat|nested] 28+ messages in thread
* [Bug optimization/13985] [3.4/3.5 regression] ICE in gcc.c-torture/compile/930621-1.c
2004-02-02 21:35 [Bug c/13985] New: [3.4/3.5 regression] ICE in gcc.c-torture/compile/930621-1.c cato at df dot lth dot se
` (23 preceding siblings ...)
2004-04-23 22:17 ` cvs-commit at gcc dot gnu dot org
@ 2004-04-23 22:27 ` cvs-commit at gcc dot gnu dot org
2004-04-23 22:45 ` ebotcazou at gcc dot gnu dot org
2004-05-18 22:16 ` [Bug rtl-optimization/13985] " ebotcazou at gcc dot gnu dot org
26 siblings, 0 replies; 28+ messages in thread
From: cvs-commit at gcc dot gnu dot org @ 2004-04-23 22:27 UTC (permalink / raw)
To: gcc-bugs
------- Additional Comments From cvs-commit at gcc dot gnu dot org 2004-04-23 22:10 -------
Subject: Bug 13985
CVSROOT: /cvs/gcc
Module name: gcc
Branch: gcc-3_4-branch
Changes by: ebotcazou@gcc.gnu.org 2004-04-23 22:10:47
Modified files:
gcc : ChangeLog cfgloopmanip.c
gcc/testsuite : ChangeLog
Added files:
gcc/testsuite/gcc.dg: loop-3.c
Log message:
PR optimization/13985
* cfgloopmanip.c (fix_loop_placements): New prototype.
Call fix_bb_placements on the preheader of loops that have
been reparented.
(remove_path): Adjust call to fix_loop_placements.
Patches:
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/ChangeLog.diff?cvsroot=gcc&only_with_tag=gcc-3_4-branch&r1=2.2326.2.415&r2=2.2326.2.416
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/cfgloopmanip.c.diff?cvsroot=gcc&only_with_tag=gcc-3_4-branch&r1=1.19&r2=1.19.4.1
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/testsuite/ChangeLog.diff?cvsroot=gcc&only_with_tag=gcc-3_4-branch&r1=1.3389.2.172&r2=1.3389.2.173
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/testsuite/gcc.dg/loop-3.c.diff?cvsroot=gcc&only_with_tag=gcc-3_4-branch&r1=NONE&r2=1.1.2.1
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=13985
^ permalink raw reply [flat|nested] 28+ messages in thread
* [Bug optimization/13985] [3.4/3.5 regression] ICE in gcc.c-torture/compile/930621-1.c
2004-02-02 21:35 [Bug c/13985] New: [3.4/3.5 regression] ICE in gcc.c-torture/compile/930621-1.c cato at df dot lth dot se
` (24 preceding siblings ...)
2004-04-23 22:27 ` cvs-commit at gcc dot gnu dot org
@ 2004-04-23 22:45 ` ebotcazou at gcc dot gnu dot org
2004-05-18 22:16 ` [Bug rtl-optimization/13985] " ebotcazou at gcc dot gnu dot org
26 siblings, 0 replies; 28+ messages in thread
From: ebotcazou at gcc dot gnu dot org @ 2004-04-23 22:45 UTC (permalink / raw)
To: gcc-bugs
------- Additional Comments From ebotcazou at gcc dot gnu dot org 2004-04-23 22:12 -------
Fixed, finally.
--
What |Removed |Added
----------------------------------------------------------------------------
Status|ASSIGNED |RESOLVED
Resolution| |FIXED
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=13985
^ permalink raw reply [flat|nested] 28+ messages in thread
* [Bug rtl-optimization/13985] [3.4/3.5 regression] ICE in gcc.c-torture/compile/930621-1.c
2004-02-02 21:35 [Bug c/13985] New: [3.4/3.5 regression] ICE in gcc.c-torture/compile/930621-1.c cato at df dot lth dot se
` (25 preceding siblings ...)
2004-04-23 22:45 ` ebotcazou at gcc dot gnu dot org
@ 2004-05-18 22:16 ` ebotcazou at gcc dot gnu dot org
26 siblings, 0 replies; 28+ messages in thread
From: ebotcazou at gcc dot gnu dot org @ 2004-05-18 22:16 UTC (permalink / raw)
To: gcc-bugs
------- Additional Comments From ebotcazou at gcc dot gnu dot org 2004-05-18 11:19 -------
*** Bug 15512 has been marked as a duplicate of this bug. ***
--
What |Removed |Added
----------------------------------------------------------------------------
CC| |niva at niisi dot msk dot ru
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=13985
^ permalink raw reply [flat|nested] 28+ messages in thread
end of thread, other threads:[~2004-05-18 11:19 UTC | newest]
Thread overview: 28+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2004-02-02 21:35 [Bug c/13985] New: [3.4/3.5 regression] ICE in gcc.c-torture/compile/930621-1.c cato at df dot lth dot se
2004-02-02 21:52 ` [Bug optimization/13985] " pinskia at gcc dot gnu dot org
2004-02-02 23:35 ` steven at gcc dot gnu dot org
2004-02-03 0:53 ` steven at gcc dot gnu dot org
2004-02-03 5:51 ` cato at df dot lth dot se
2004-02-03 8:39 ` stevenb at suse dot de
2004-02-03 8:59 ` steven at gcc dot gnu dot org
2004-02-03 23:01 ` cato at df dot lth dot se
2004-02-07 1:42 ` cato at df dot lth dot se
2004-02-07 2:19 ` pinskia at gcc dot gnu dot org
2004-02-07 2:22 ` pinskia at gcc dot gnu dot org
2004-03-04 1:57 ` pinskia at gcc dot gnu dot org
2004-03-06 11:36 ` ebotcazou at gcc dot gnu dot org
2004-03-11 6:31 ` ebotcazou at gcc dot gnu dot org
2004-03-13 19:46 ` ebotcazou at gcc dot gnu dot org
2004-03-21 18:56 ` mmitchel at gcc dot gnu dot org
2004-04-18 22:45 ` mmitchel at gcc dot gnu dot org
2004-04-19 11:14 ` ebotcazou at gcc dot gnu dot org
2004-04-19 17:31 ` law at redhat dot com
2004-04-19 23:45 ` ebotcazou at gcc dot gnu dot org
2004-04-20 0:07 ` rakdver at atrey dot karlin dot mff dot cuni dot cz
2004-04-20 9:06 ` ebotcazou at gcc dot gnu dot org
2004-04-20 9:07 ` rakdver at atrey dot karlin dot mff dot cuni dot cz
2004-04-20 11:03 ` ebotcazou at gcc dot gnu dot org
2004-04-23 22:17 ` cvs-commit at gcc dot gnu dot org
2004-04-23 22:27 ` cvs-commit at gcc dot gnu dot org
2004-04-23 22:45 ` ebotcazou at gcc dot gnu dot org
2004-05-18 22:16 ` [Bug rtl-optimization/13985] " ebotcazou at gcc dot gnu dot 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).