| commit | 987322bc0b170570a7bd539480252453fcc7a6f5 | |
| author | Adhemerval Zanella <azanella@linux.vnet.ibm.com> | |
| Fri, 29 Mar 2013 23:15:28 +0000 (29 18:15 -0500) | ||
| committer | Ryan S. Arnold <rsa@linux.vnet.ibm.com> | |
| Thu, 25 Jul 2013 20:35:22 +0000 (25 15:35 -0500) | ||
| tree | 7270e0359a3f15e2d3938d6ece41f5cfe1245ec0 | treesnapshot (tar.gz zip) |
| parent | ace889170837e550871cf0d33f88ba2470c7d834 | commitdiff |
PowerPC: remove branch prediction from rint implementation
The branch prediction hints is actually hurts performance in this case.
The assembly implementation make two assumptions: 1. 'fabs (x) < 2^52'
is unlikely and 2. 'x > 0.0' is unlike (if 1. is true). Since it a
general floating point function, expected input is not bounded and then
it is better to let the hardware handle the branches.
(backported from commit 60c414c346a1d5ef0510ffbdc0ab75f288ee4d3f)
This backport does not include the benchmark tests from the original
commit.
The branch prediction hints is actually hurts performance in this case.
The assembly implementation make two assumptions: 1. 'fabs (x) < 2^52'
is unlikely and 2. 'x > 0.0' is unlike (if 1. is true). Since it a
general floating point function, expected input is not bounded and then
it is better to let the hardware handle the branches.
(backported from commit 60c414c346a1d5ef0510ffbdc0ab75f288ee4d3f)
This backport does not include the benchmark tests from the original
commit.