[PATCH] D29473: [AMDGPU] Unroll preferences improvements
Valery Pykhtin via Phabricator via llvm-commits
llvm-commits at lists.llvm.org
Fri Feb 3 07:08:58 PST 2017
vpykhtin added inline comments.
================
Comment at: llvm/trunk/lib/Target/AMDGPU/AMDGPUTargetTransformInfo.cpp:88
// programs way too big.
- UP.Threshold = 800;
+ UP.Threshold = UnrollThresholdPrivate;
+ return;
----------------
tstellarAMD wrote:
> Do you also want to set PartialThreshold here?
I thought partialy unrolled loops won't make it possible to SROA private arrays. What are the benefits of partial unrolling on AMDGPU btw? What comes in mind: mem ops clustering/widening, less branches? What else?
Repository:
rL LLVM
https://reviews.llvm.org/D29473
More information about the llvm-commits
mailing list