From 9c661853150e304b9c6604df44a0181795ae633f Mon Sep 17 00:00:00 2001 From: Nicolai Haehnle Date: Mon, 24 Apr 2017 17:08:43 +0000 Subject: InstCombine/AMDGPU: Fix constant folding of llvm.amdgcn.{icmp,fcmp} Summary: The return value of these intrinsics should always have 0 bits for inactive threads. This means that when all arguments are constant and the comparison evaluates to true, the intrinsic should return the current exec mask. Fixes some GL_ARB_shader_ballot tests. Reviewers: arsenm Subscribers: kzhuravl, wdng, yaxunl, dstuttard, tpr, llvm-commits, t-tye Differential Revision: https://reviews.llvm.org/D32344 llvm-svn: 301195 --- .../Transforms/InstCombine/InstCombineCalls.cpp | 22 ++++++++++++++++++++-- 1 file changed, 20 insertions(+), 2 deletions(-) (limited to 'llvm/lib/Transforms') diff --git a/llvm/lib/Transforms/InstCombine/InstCombineCalls.cpp b/llvm/lib/Transforms/InstCombine/InstCombineCalls.cpp index e7aa1a45737..0ea1c7be422 100644 --- a/llvm/lib/Transforms/InstCombine/InstCombineCalls.cpp +++ b/llvm/lib/Transforms/InstCombine/InstCombineCalls.cpp @@ -3432,8 +3432,26 @@ Instruction *InstCombiner::visitCallInst(CallInst &CI) { if (auto *CSrc0 = dyn_cast(Src0)) { if (auto *CSrc1 = dyn_cast(Src1)) { Constant *CCmp = ConstantExpr::getCompare(CCVal, CSrc0, CSrc1); - return replaceInstUsesWith(*II, - ConstantExpr::getSExt(CCmp, II->getType())); + if (CCmp->isNullValue()) { + return replaceInstUsesWith( + *II, ConstantExpr::getSExt(CCmp, II->getType())); + } + + // The result of V_ICMP/V_FCMP assembly instructions (which this + // intrinsic exposes) is one bit per thread, masked with the EXEC + // register (which contains the bitmask of live threads). So a + // comparison that always returns true is the same as a read of the + // EXEC register. + Value *NewF = Intrinsic::getDeclaration( + II->getModule(), Intrinsic::read_register, II->getType()); + Metadata *MDArgs[] = {MDString::get(II->getContext(), "exec")}; + MDNode *MD = MDNode::get(II->getContext(), MDArgs); + Value *Args[] = {MetadataAsValue::get(II->getContext(), MD)}; + CallInst *NewCall = Builder->CreateCall(NewF, Args); + NewCall->addAttribute(AttributeList::FunctionIndex, + Attribute::Convergent); + NewCall->takeName(II); + return replaceInstUsesWith(*II, NewCall); } // Canonicalize constants to RHS. -- cgit v1.2.3