[Bug tree-optimization/50713] SLP vs loop: code generated differs (SLP less efficient)

2012-10-25 Thread glisse at gcc dot gnu.org
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=50713 Marc Glisse changed: What|Removed |Added CC||glisse at gcc dot gnu.org --- Com

[Bug tree-optimization/50713] SLP vs loop: code generated differs (SLP less efficient)

2012-10-25 Thread vincenzo.innocente at cern dot ch
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=50713 vincenzo Innocente changed: What|Removed |Added Summary|SLP vs loop: code generated |SLP vs loop: code generated

[Bug tree-optimization/50713] SLP vs loop: code generated differs

2011-10-14 Thread rguenth at gcc dot gnu.org
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=50713 --- Comment #4 from Richard Guenther 2011-10-14 13:23:52 UTC --- (In reply to comment #3) > thanks for the splitting: it is indeed the most serious problem. > what about "complex ops in scalar code"? keep it here and shall I spawn a > specific on

[Bug tree-optimization/50713] SLP vs loop: code generated differs

2011-10-14 Thread vincenzo.innocente at cern dot ch
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=50713 --- Comment #3 from vincenzo Innocente 2011-10-14 12:54:46 UTC --- thanks for the splitting: it is indeed the most serious problem. what about "complex ops in scalar code"? keep it here and shall I spawn a specific one (maybe once PR50622 is fixe

[Bug tree-optimization/50713] SLP vs loop: code generated differs

2011-10-14 Thread rguenth at gcc dot gnu.org
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=50713 --- Comment #2 from Richard Guenther 2011-10-14 12:38:20 UTC --- I split out the inefficient by-value passing thing to PR50728

[Bug tree-optimization/50713] SLP vs loop: code generated differs

2011-10-13 Thread rguenth at gcc dot gnu.org
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=50713 Richard Guenther changed: What|Removed |Added Status|UNCONFIRMED |NEW Last reconfirmed|