http://gcc.gnu.org/bugzilla/show_bug.cgi?id=53346
Uros Bizjak <ubizjak at gmail dot com> changed:
What |Removed |Added
----------------------------------------------------------------------------
Status|RESOLVED |NEW
Resolution|FIXED |
--- Comment #8 from Uros Bizjak <ubizjak at gmail dot com> 2012-05-18 14:46:01
UTC ---
(In reply to comment #7)
> Fixed.
Unfortunately, the loop in original rnflow test still gets vectorized, with no
change in the runtime:
Each sample counts as 0.01 seconds.
% cumulative self self total
time seconds seconds calls s/call s/call name
43.46 9.69 9.69 64 0.15 0.15 cptrf2_
40.63 18.75 9.06 6685 0.00 0.00 trs2a2.2054
7.89 20.51 1.76 64 0.03 0.03 gentrs_
2.02 20.96 0.45 6685 0.00 0.00 invima.2045
1.93 21.39 0.43 64 0.01 0.01 cptrf1_
1.17 21.65 0.26 1 0.26 12.36 matsim_
0.99 21.87 0.22 64 0.00 0.00 cmpcpt_
GNU Fortran (GCC) version 4.8.0 20120518 (experimental) [trunk revision 187655]
