llvm-mirror

mirror of https://github.com/RPCS3/llvm-mirror.git synced 2025-01-27 15:02:16 +00:00

History

Sanjay Patel 6c99435a8b [TTI CostModel] change default cost of FP ops to 1 (PR36280)

This change was mentioned at least as far back as:
https://bugs.llvm.org/show_bug.cgi?id=26837#c26
...and I found a real program that is harmed by this: 
Himeno running on AMD Jaguar gets 6% slower with SLP vectorization:
https://bugs.llvm.org/show_bug.cgi?id=36280
...but the change here appears to solve that bug only accidentally.

The div/rem costs for x86 look very wrong in some cases, but that's already true, 
so we can fix those in follow-up patches. There's also evidence that more cost model
changes are needed to solve SLP problems as shown in D42981, but that's an independent 
problem (though the solution may be adjusted after this change is made).

Differential Revision: https://reviews.llvm.org/D43079

llvm-svn: 325515

2018-02-19 16:11:44 +00:00

AArch64

Revert r314923: "Recommit : Use the basic cost if a GEP is not used as addressing mode"

2017-10-13 14:04:21 +00:00

AMDGPU

[AMDGPU] Switch to the new addr space mapping by default

2018-02-02 16:07:16 +00:00

ARM

Convert an APInt to int64_t properly in TTI::getGEPCost().

2017-10-04 20:47:33 +00:00

PowerPC

[PPC] Give unaligned memory access lower cost on processor that supports it