[PATCH] D68873: [AMDGPU] Amend target loop unroll defaults

[Tim Corringham via Phabricator via llvm-commits]

timcorringham updated this revision to Diff 225431.
timcorringham marked 6 inline comments as done.
timcorringham added a comment.

Changes to address review comments


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D68873/new/

https://reviews.llvm.org/D68873

Files:
  llvm/lib/Target/AMDGPU/AMDGPUTargetTransformInfo.cpp
  llvm/lib/Target/AMDGPU/AMDGPUTargetTransformInfo.h


Index: llvm/lib/Target/AMDGPU/AMDGPUTargetTransformInfo.h
===================================================================
--- llvm/lib/Target/AMDGPU/AMDGPUTargetTransformInfo.h
+++ llvm/lib/Target/AMDGPU/AMDGPUTargetTransformInfo.h
@@ -46,7 +46,7 @@
 
   Triple TargetTriple;
 
-  const TargetSubtargetInfo *ST;
+  const GCNSubtarget *ST;
   const TargetLoweringBase *TLI;
 
   const TargetSubtargetInfo *getST() const { return ST; }
Index: llvm/lib/Target/AMDGPU/AMDGPUTargetTransformInfo.cpp
===================================================================
--- llvm/lib/Target/AMDGPU/AMDGPUTargetTransformInfo.cpp
+++ llvm/lib/Target/AMDGPU/AMDGPUTargetTransformInfo.cpp
@@ -96,10 +96,23 @@
 
   // TODO: Do we want runtime unrolling?
 
+  // Set more aggressive defaults for PAL shaders
+  if (TargetTriple.getOS() == Triple::AMDPAL) {
+    UP.MaxPercentThresholdBoost = 1000;
+    // and even more aggressive for GFX10
+    if (ST->getGeneration() >= AMDGPUSubtarget::GFX10) {
+      UP.Threshold = 1100;
+      UP.PartialThreshold = 1100;
+    } else {
+      UP.Threshold = 700;
+      UP.PartialThreshold = 700;
+    }
+  }
+
   // Maximum alloca size than can fit registers. Reserve 16 registers.
   const unsigned MaxAlloca = (256 - 16) * 4;
   unsigned ThresholdPrivate = UnrollThresholdPrivate;
-  unsigned ThresholdLocal = UnrollThresholdLocal;
+  unsigned ThresholdLocal = std::max((unsigned)UnrollThresholdLocal, UP.Threshold);
   unsigned MaxBoost = std::max(ThresholdPrivate, ThresholdLocal);
   for (const BasicBlock *BB : L->getBlocks()) {
     const DataLayout &DL = BB->getModule()->getDataLayout();


-------------- next part --------------
A non-text attachment was scrubbed...
Name: D68873.225431.patch
Type: text/x-patch
Size: 1627 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20191017/591397d4/attachment.bin>