[Bug rtl-optimization/53190] CSE (?) CSEs asms

2012-05-03 Thread ubizjak at gmail dot com
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=53190 --- Comment #11 from Uros Bizjak 2012-05-03 11:43:16 UTC --- (In reply to comment #10) > So, should somebody file a bug against GCC that to make the SSE cr available? Er, gcc doesn't need it, all SSE conversion builtins are implemented without t

[Bug rtl-optimization/53190] CSE (?) CSEs asms

2012-05-03 Thread aj at gcc dot gnu.org
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=53190 --- Comment #10 from Andreas Jaeger 2012-05-03 11:02:43 UTC --- So, should somebody file a bug against GCC that to make the SSE cr available?

[Bug rtl-optimization/53190] CSE (?) CSEs asms

2012-05-03 Thread ubizjak at gmail dot com
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=53190 --- Comment #9 from Uros Bizjak 2012-05-03 09:38:59 UTC --- (In reply to comment #8) > So, how should the inline rewritten? > > Adding volatile is one option: > extern __inline __attribute__ ((__always_inline__)) long int > __attribute__ ((__not

[Bug rtl-optimization/53190] CSE (?) CSEs asms

2012-05-03 Thread aj at gcc dot gnu.org
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=53190 --- Comment #8 from Andreas Jaeger 2012-05-03 09:19:16 UTC --- So, how should the inline rewritten? Adding volatile is one option: extern __inline __attribute__ ((__always_inline__)) long int __attribute__ ((__nothrow__ )) lrintf (float __x) {

[Bug rtl-optimization/53190] CSE (?) CSEs asms

2012-05-03 Thread ubizjak at gmail dot com
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=53190 --- Comment #7 from Uros Bizjak 2012-05-03 08:39:19 UTC --- (In reply to comment #6) > Well, the asm in glibc is simply > > extern __inline __attribute__ ((__always_inline__)) long int > __attribute__ ((__nothrow__ )) lrintf (float __x) > { >

[Bug rtl-optimization/53190] CSE (?) CSEs asms

2012-05-03 Thread rguenther at suse dot de
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=53190 --- Comment #6 from rguenther at suse dot de 2012-05-03 08:19:17 UTC --- On Thu, 3 May 2012, ubizjak at gmail dot com wrote: > http://gcc.gnu.org/bugzilla/show_bug.cgi?id=53190 > > --- Comment #5 from Uros Bizjak 2012-05-03 > 08:15:00 UTC ---

[Bug rtl-optimization/53190] CSE (?) CSEs asms

2012-05-03 Thread ubizjak at gmail dot com
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=53190 --- Comment #5 from Uros Bizjak 2012-05-03 08:15:00 UTC --- (In reply to comment #4) > So you say that it's correct for CSE to CSE asm()s if they have the same > asm string and the same asm operands in case the asm is not volatile? I was > not

[Bug rtl-optimization/53190] CSE (?) CSEs asms

2012-05-03 Thread rguenth at gcc dot gnu.org
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=53190 --- Comment #4 from Richard Guenther 2012-05-03 07:59:40 UTC --- (In reply to comment #3) > (In reply to comment #2) > > (clobber (reg:QI 18 fpsr)) > > (clobber (reg:QI 17 flags)) > > > > > > It might clobber those but t

[Bug rtl-optimization/53190] CSE (?) CSEs asms

2012-05-02 Thread ubizjak at gmail dot com
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=53190 Uros Bizjak changed: What|Removed |Added Status|UNCONFIRMED |RESOLVED Resolution|

[Bug rtl-optimization/53190] CSE (?) CSEs asms

2012-05-02 Thread pinskia at gcc dot gnu.org
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=53190 --- Comment #2 from Andrew Pinski 2012-05-02 17:09:02 UTC --- (clobber (reg:QI 18 fpsr)) (clobber (reg:QI 17 flags)) It might clobber those but that is not considered an use. So CSE is doing the correct thing as there i

[Bug rtl-optimization/53190] CSE (?) CSEs asms

2012-05-02 Thread rguenth at gcc dot gnu.org
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=53190 --- Comment #1 from Richard Guenther 2012-05-02 12:47:58 UTC --- Created attachment 27287 --> http://gcc.gnu.org/bugzilla/attachment.cgi?id=27287 preprocessed source Preprocessed source from glibc test-ifloat testcase.