[Bug rtl-optimization/55190] [SH] ivopts causes loop setup bloat

2015-04-06 Thread amker at gcc dot gnu.org
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=55190 amker at gcc dot gnu.org changed: What|Removed |Added CC||amker at gcc dot gnu.org

[Bug rtl-optimization/55190] [SH] ivopts causes loop setup bloat

2015-04-04 Thread olegendo at gcc dot gnu.org
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=55190 --- Comment #6 from Oleg Endo --- Another example: const char* test (const char* s0, int c, int* rout) { int r = 0; for (int i = 0; i < c; ++i) //r += s0[i]; r += *s0++; *rout = r; return s0; } On SH compiles to: _test:

[Bug rtl-optimization/55190] [SH] ivopts causes loop setup bloat

2014-07-31 Thread olegendo at gcc dot gnu.org
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=55190 Oleg Endo changed: What|Removed |Added Known to fail||4.9.0 --- Comment #5 from Oleg Endo --- As

[Bug rtl-optimization/55190] [SH] ivopts causes loop setup bloat

2013-10-03 Thread olegendo at gcc dot gnu.org
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=55190 --- Comment #4 from Oleg Endo --- (In reply to bin.cheng from comment #3) > ARM can benefit from doloop structure too, but it is implemented in > different way. ARM backend defines special addsi_compare pattern and let > combine pass combine decre

[Bug rtl-optimization/55190] [SH] ivopts causes loop setup bloat

2013-09-30 Thread amker.cheng at gmail dot com
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=55190 bin.cheng changed: What|Removed |Added CC||amker.cheng at gmail dot com --- Comment #3 f

[Bug rtl-optimization/55190] [SH] ivopts causes loop setup bloat

2013-06-16 Thread olegendo at gcc dot gnu.org
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=55190 Oleg Endo changed: What|Removed |Added Component|target |rtl-optimization --- Comment #2 from Oleg End