[Bug target/55461] _mm_loadu_si128 generates wrong instruction on 4.8

2012-11-25 Thread andi-gcc at firstfloor dot org
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=55461 Andi Kleen changed: What|Removed |Added Status|UNCONFIRMED |RESOLVED Resolution|

[Bug target/55461] _mm_loadu_si128 generates wrong instruction on 4.8

2012-11-25 Thread ubizjak at gmail dot com
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=55461 --- Comment #3 from Uros Bizjak 2012-11-25 18:48:25 UTC --- (In reply to comment #2) > That's a design not matching the hardware then -- read up on reformatting > penalties. > > Also the way the rule in the md file is written it more lo

[Bug target/55461] _mm_loadu_si128 generates wrong instruction on 4.8

2012-11-25 Thread andi-gcc at firstfloor dot org
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=55461 --- Comment #2 from Andi Kleen 2012-11-25 18:07:49 UTC --- That's a design not matching the hardware then -- read up on reformatting penalties. Also the way the rule in the md file is written it more looks like a bug to me. It seems to

[Bug target/55461] _mm_loadu_si128 generates wrong instruction on 4.8

2012-11-25 Thread ubizjak at gmail dot com
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=55461 --- Comment #1 from Uros Bizjak 2012-11-25 08:59:13 UTC --- This is by design, movups is one byte shorter than movupd.