[Bug target/53334] [4.8 Regression] ICE in extract_insn, at recog.c:2131

2012-05-26 Thread pinskia at gcc dot gnu.org 
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=53334

Andrew Pinski  changed:

   What|Removed |Added

 Status|NEW |RESOLVED
 Resolution||FIXED

--- Comment #7 from Andrew Pinski  2012-05-26 
18:14:25 UTC ---
Fixed.

[Bug target/53334] [4.8 Regression] ICE in extract_insn, at recog.c:2131

2012-05-22 Thread ramana at gcc dot gnu.org 
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=53334

--- Comment #6 from Ramana Radhakrishnan  2012-05-22 
09:07:03 UTC ---
Author: ramana
Date: Tue May 22 09:06:55 2012
New Revision: 187761

URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=187761
Log:
Fix PR target/53334


2012-05-22  Ramana Radhakrishnan  

PR target/53334
* config/arm/arm-protos.h (arm_validize_comparison): Declare.
* config/arm/arm.c (arm_validize_comparison): Define.
* config/arm/arm.md ("cbranchsi4"): Cleanup expansion and use
arm_validize_comparison.
("cbranchdi4"): Likewise.
("cstoredi4"): Likewise.
("movsicc"): Likewise.
("movsfcc"): Likewise.
("movdfcc"): Likewise.


Modified:
trunk/gcc/ChangeLog
trunk/gcc/config/arm/arm-protos.h
trunk/gcc/config/arm/arm.c
trunk/gcc/config/arm/arm.md

[Bug target/53334] [4.8 Regression] ICE in extract_insn, at recog.c:2131

2012-05-16 Thread pinskia at gcc dot gnu.org 
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=53334

Andrew Pinski  changed:

   What|Removed |Added

 CC||ramana at gcc dot gnu.org

--- Comment #5 from Andrew Pinski  2012-05-16 
16:15:21 UTC ---
*** Bug 53376 has been marked as a duplicate of this bug. ***

[Bug target/53334] [4.8 Regression] ICE in extract_insn, at recog.c:2131

2012-05-14 Thread pinskia at gcc dot gnu.org 
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=53334

Andrew Pinski  changed:

   What|Removed |Added

 CC||pinskia at gcc dot gnu.org
   Target Milestone|--- |4.8.0
Summary|ICE in extract_insn, at |[4.8 Regression] ICE in
   |recog.c:2131|extract_insn, at
   ||recog.c:2131

--- Comment #4 from Andrew Pinski  2012-05-14 
16:55:55 UTC ---
(In reply to comment #3)
> The movsicc (and related) patterns are not validating the arguments to the
> comparison properly.

This latent bug was exposed by my patch which uses the mov*cc patterns during
expanding rather than just during ifcvt (where the arguments are most likely
already validated).