--- Comment #28 from rguenth at gcc dot gnu dot org 2010-03-18 17:59
---
All fine again. Fixed.
--
rguenth at gcc dot gnu dot org changed:
What|Removed |Added
--- Comment #27 from bernds at gcc dot gnu dot org 2010-03-17 09:25 ---
Subject: Bug 42216
Author: bernds
Date: Wed Mar 17 09:25:35 2010
New Revision: 157511
URL: http://gcc.gnu.org/viewcvs?root=gccview=revrev=157511
Log:
PR rtl-optimization/42216
* regrename.c
--- Comment #26 from dominiq at lps dot ens dot fr 2010-03-14 09:33 ---
I have posted a full regression testing with the patch in comment #23 at
http://gcc.gnu.org/ml/gcc-testresults/2010-03/msg01182.html
The failures are the usual ones and pr42220 is still fixed.
--
--- Comment #24 from dominiq at lps dot ens dot fr 2010-03-09 08:35 ---
If you wouldn't mind, please test the attached patch which should be an
alternative fix for 42220 and avoids the need to set fail_block for some
cases.
I cannot test if the patch in comment #23 fixes the
--- Comment #25 from rguenther at suse dot de 2010-03-09 10:31 ---
Subject: Re: [4.5 Regression] changes in scheduling
regress 464.h264ref 20%
On Mon, 8 Mar 2010, bernds_cb1 at t-online dot de wrote:
--- Comment #23 from bernds_cb1 at t-online dot de 2010-03-08 23:04
---
--- Comment #22 from rguenth at gcc dot gnu dot org 2010-03-08 11:48
---
All is bad again after the fix for PR42220:
2010-03-07 Bernd Schmidt bernd.schm...@analog.com
PR rtl-optimization/42220
* regrename.c (scan_rtx) case STRICT_LOW_PART, ZERO_EXTRACT:
Use
--- Comment #23 from bernds_cb1 at t-online dot de 2010-03-08 23:04 ---
Created an attachment (id=20048)
-- (http://gcc.gnu.org/bugzilla/attachment.cgi?id=20048action=view)
Alternative fix for 42220
If you wouldn't mind, please test the attached patch which should be an
alternative
--- Comment #21 from rguenth at gcc dot gnu dot org 2009-12-11 21:39
---
All is well again.
--
rguenth at gcc dot gnu dot org changed:
What|Removed |Added
--- Comment #20 from bernds at gcc dot gnu dot org 2009-12-10 18:03 ---
Subject: Bug 42216
Author: bernds
Date: Thu Dec 10 18:03:05 2009
New Revision: 155134
URL: http://gcc.gnu.org/viewcvs?root=gccview=revrev=155134
Log:
PR rtl-opt/42216
* regrename.c: Error out if
--- Comment #18 from bernds_cb1 at t-online dot de 2009-12-04 14:26 ---
Unfortunately it causes failures. Tracking these mismatches really is quite
tricky.
I'll try to fix it.
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=42216
--- Comment #19 from rguenther at suse dot de 2009-12-04 14:34 ---
Subject: Re: [4.5 Regression] changes in scheduling
regress 464.h264ref 20%
On Fri, 4 Dec 2009, bernds_cb1 at t-online dot de wrote:
--- Comment #18 from bernds_cb1 at t-online dot de 2009-12-04 14:26
---
--
rguenth at gcc dot gnu dot org changed:
What|Removed |Added
Priority|P3 |P2
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=42216
--- Comment #11 from rguenth at gcc dot gnu dot org 2009-12-02 13:16
---
The patch didn't change performance. For the relevant functions I see
mv-search.c:452:1: warning: failing block due to bad pattern
(the location seems to be always the end of the function though).
The insns
--- Comment #12 from rguenth at gcc dot gnu dot org 2009-12-02 13:16
---
Created an attachment (id=19208)
-- (http://gcc.gnu.org/bugzilla/attachment.cgi?id=19208action=view)
basic-block that is rejected
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=42216
--- Comment #13 from bernds_cb1 at t-online dot de 2009-12-02 16:28 ---
Created an attachment (id=19211)
-- (http://gcc.gnu.org/bugzilla/attachment.cgi?id=19211action=view)
Patch to make it less conservative when accepting matching constraints with
different modes
--
--- Comment #14 from bernds_cb1 at t-online dot de 2009-12-02 16:29 ---
Would you mind trying another patch, both for testing performance, and if that
fails, for interesting warnings?
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=42216
--- Comment #15 from rguenth at gcc dot gnu dot org 2009-12-02 16:57
---
With the second patch there are no warnings when building 464.h264ref (I have
reverted the first patch before installing the 2nd, it didn't apply cleanly
otherwise). Performance is back to normal with the patch
--- Comment #16 from bernds_cb1 at t-online dot de 2009-12-02 17:06 ---
Thanks for testing. This means the regression is fixed by the patch? I'll do
a full test run then.
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=42216
--- Comment #17 from rguenther at suse dot de 2009-12-02 17:07 ---
Subject: Re: [4.5 Regression] changes in scheduling
regress 464.h264ref 20%
On Wed, 2 Dec 2009, bernds_cb1 at t-online dot de wrote:
--- Comment #16 from bernds_cb1 at t-online dot de 2009-12-02 17:06
---
19 matches
Mail list logo
