Commit Graph

6 Commits

Author SHA1 Message Date
Stanislav Mekhanoshin 79da2a7698 [AMDGPU] Remove getBidirectionalReasonRank
This method inverts the Reason field of a scheduling candidate.
It does right comparison between RegCritical and RegExcess, but
everything else is broken. In fact it can prefer less strong reason
such as Weak over RegCritical because Weak > -RegCritical.

The CandReason enum is properly sorted, so just remove artificial
ranking.

Differential Revision: https://reviews.llvm.org/D30557

llvm-svn: 297536
2017-03-11 00:29:27 +00:00
Matt Arsenault 74a576e7d3 AMDGPU: Check nsz instead of unsafe math
llvm-svn: 293028
2017-01-25 06:27:02 +00:00
Matt Arsenault 7aad8fd8f4 Enable FeatureFlatForGlobal on Volcanic Islands
This switches to the workaround that HSA defaults to
for the mesa path.

This should be applied to the 4.0 branch.

Patch by Vedran Miletić <vedran@miletic.net>

llvm-svn: 292982
2017-01-24 22:02:15 +00:00
Matt Arsenault 3e6f9b5773 AMDGPU: Disable some fneg combines unless nsz
For -(x + y) -> (-x) + (-y), if x == -y, this would
change the result from -0.0 to 0.0. Since the fma/fmad
combine is an extension of this problem it also
applies there.

fmul should be fine, and I don't think any of the unary
operators or conversions should be a problem either.

llvm-svn: 292473
2017-01-19 06:35:27 +00:00
Matt Arsenault 45337df08f AMDGPU: Skip fneg/select combine if it can fold into other
llvm-svn: 291792
2017-01-12 18:58:15 +00:00
Matt Arsenault 2a04ff97ad AMDGPU: Pull fneg/fabs out of a select
Allows better source modifier usage.

llvm-svn: 291729
2017-01-11 23:57:38 +00:00