[llvm] aad9365 - [SVE] Eliminate calls to default-false VectorType::get() from AMDGPU

[Christopher Tetreault via llvm-commits]

Author: Christopher Tetreault
Date: 2020-05-29T17:54:17-07:00
New Revision: aad936548247107b1afd92ea9229b396564063b9

URL: https://github.com/llvm/llvm-project/commit/aad936548247107b1afd92ea9229b396564063b9
DIFF: https://github.com/llvm/llvm-project/commit/aad936548247107b1afd92ea9229b396564063b9.diff

LOG: [SVE] Eliminate calls to default-false VectorType::get() from AMDGPU

Reviewers: efriedma, david-arm, fpetrogalli, arsenm

Reviewed By: david-arm

Subscribers: arsenm, kzhuravl, jvesely, wdng, nhaehnle, yaxunl, dstuttard, tpr, t-tye, tschuett, hiraditya, rkruppe, psnobl, kerbowa, llvm-commits

Tags: #llvm

Differential Revision: https://reviews.llvm.org/D80328

Added: 
    

Modified: 
    llvm/lib/Target/AMDGPU/AMDGPUAtomicOptimizer.cpp
    llvm/lib/Target/AMDGPU/AMDGPULibCalls.cpp
    llvm/lib/Target/AMDGPU/AMDGPULibFunc.cpp
    llvm/lib/Target/AMDGPU/AMDGPULowerKernelArguments.cpp
    llvm/lib/Target/AMDGPU/AMDGPUPrintfRuntimeBinding.cpp
    llvm/lib/Target/AMDGPU/AMDGPUTargetTransformInfo.cpp

Removed: 
    


################################################################################
diff  --git a/llvm/lib/Target/AMDGPU/AMDGPUAtomicOptimizer.cpp b/llvm/lib/Target/AMDGPU/AMDGPUAtomicOptimizer.cpp
index 46ffc77d8edd..c9d25d4250d5 100644
--- a/llvm/lib/Target/AMDGPU/AMDGPUAtomicOptimizer.cpp
+++ b/llvm/lib/Target/AMDGPU/AMDGPUAtomicOptimizer.cpp
@@ -438,7 +438,7 @@ void AMDGPUAtomicOptimizer::optimizeAtomic(Instruction &I,
 
   Type *const Ty = I.getType();
   const unsigned TyBitWidth = DL->getTypeSizeInBits(Ty);
-  Type *const VecTy = VectorType::get(B.getInt32Ty(), 2);
+  auto *const VecTy = FixedVectorType::get(B.getInt32Ty(), 2);
 
   // This is the value in the atomic operation we need to combine in order to
   // reduce the number of atomic operations.

diff  --git a/llvm/lib/Target/AMDGPU/AMDGPULibCalls.cpp b/llvm/lib/Target/AMDGPU/AMDGPULibCalls.cpp
index 956508e12227..0ef858693027 100644
--- a/llvm/lib/Target/AMDGPU/AMDGPULibCalls.cpp
+++ b/llvm/lib/Target/AMDGPU/AMDGPULibCalls.cpp
@@ -598,7 +598,7 @@ bool AMDGPULibCalls::fold_read_write_pipe(CallInst *CI, IRBuilder<> &B,
   if (Size <= 8)
     PtrElemTy = Type::getIntNTy(Ctx, Size * 8);
   else
-    PtrElemTy = VectorType::get(Type::getInt64Ty(Ctx), Size / 8);
+    PtrElemTy = FixedVectorType::get(Type::getInt64Ty(Ctx), Size / 8);
   unsigned PtrArgLoc = CI->getNumArgOperands() - 3;
   auto PtrArg = CI->getArgOperand(PtrArgLoc);
   unsigned PtrArgAS = PtrArg->getType()->getPointerAddressSpace();

diff  --git a/llvm/lib/Target/AMDGPU/AMDGPULibFunc.cpp b/llvm/lib/Target/AMDGPU/AMDGPULibFunc.cpp
index b39039861f51..2b5143ba7506 100644
--- a/llvm/lib/Target/AMDGPU/AMDGPULibFunc.cpp
+++ b/llvm/lib/Target/AMDGPU/AMDGPULibFunc.cpp
@@ -902,7 +902,7 @@ static Type* getIntrinsicParamType(
     return nullptr;
   }
   if (P.VectorSize > 1)
-    T = VectorType::get(T, P.VectorSize);
+    T = FixedVectorType::get(T, P.VectorSize);
   if (P.PtrKind != AMDGPULibFunc::BYVALUE)
     T = useAddrSpace ? T->getPointerTo((P.PtrKind & AMDGPULibFunc::ADDR_SPACE)
                                        - 1)

diff  --git a/llvm/lib/Target/AMDGPU/AMDGPULowerKernelArguments.cpp b/llvm/lib/Target/AMDGPU/AMDGPULowerKernelArguments.cpp
index 52e192e576dd..58bd6e5f3b2b 100644
--- a/llvm/lib/Target/AMDGPU/AMDGPULowerKernelArguments.cpp
+++ b/llvm/lib/Target/AMDGPU/AMDGPULowerKernelArguments.cpp
@@ -167,7 +167,7 @@ bool AMDGPULowerKernelArguments::runOnFunction(Function &F) {
     }
 
     if (IsV3 && Size >= 32) {
-      V4Ty = VectorType::get(VT->getElementType(), 4);
+      V4Ty = FixedVectorType::get(VT->getElementType(), 4);
       // Use the hack that clang uses to avoid SelectionDAG ruining v3 loads
       AdjustedArgTy = V4Ty;
     }

diff  --git a/llvm/lib/Target/AMDGPU/AMDGPUPrintfRuntimeBinding.cpp b/llvm/lib/Target/AMDGPU/AMDGPUPrintfRuntimeBinding.cpp
index ab5b62ccf82e..524a34be876f 100644
--- a/llvm/lib/Target/AMDGPU/AMDGPUPrintfRuntimeBinding.cpp
+++ b/llvm/lib/Target/AMDGPU/AMDGPUPrintfRuntimeBinding.cpp
@@ -516,7 +516,7 @@ bool AMDGPUPrintfRuntimeBinding::lowerPrintfForGpu(
             break;
           }
           if (EleCount > 1) {
-            IType = dyn_cast<Type>(VectorType::get(IType, EleCount));
+            IType = FixedVectorType::get(IType, EleCount);
           }
           Arg = new BitCastInst(Arg, IType, "PrintArgVect", Brnch);
           WhatToStore.push_back(Arg);

diff  --git a/llvm/lib/Target/AMDGPU/AMDGPUTargetTransformInfo.cpp b/llvm/lib/Target/AMDGPU/AMDGPUTargetTransformInfo.cpp
index eeedfe7a8c02..9e738dd6fdb3 100644
--- a/llvm/lib/Target/AMDGPU/AMDGPUTargetTransformInfo.cpp
+++ b/llvm/lib/Target/AMDGPU/AMDGPUTargetTransformInfo.cpp
@@ -334,12 +334,12 @@ Type *GCNTTIImpl::getMemcpyLoopLoweringType(LLVMContext &Context, Value *Length,
       SrcAddrSpace == AMDGPUAS::REGION_ADDRESS ||
       DestAddrSpace == AMDGPUAS::LOCAL_ADDRESS ||
       DestAddrSpace == AMDGPUAS::REGION_ADDRESS) {
-    return VectorType::get(Type::getInt32Ty(Context), 2);
+    return FixedVectorType::get(Type::getInt32Ty(Context), 2);
   }
 
   // Global memory works best with 16-byte accesses. Private memory will also
   // hit this, although they'll be decomposed.
-  return VectorType::get(Type::getInt32Ty(Context), 4);
+  return FixedVectorType::get(Type::getInt32Ty(Context), 4);
 }
 
 void GCNTTIImpl::getMemcpyLoopResidualLoweringType(