From: Dan Gohman Date: Fri, 9 Apr 2010 19:14:31 +0000 (+0000) Subject: When emitting code for an add, don't force a SCEVUnknown wrapper around X-Git-Url: http://demsky.eecs.uci.edu/git/?a=commitdiff_plain;h=f8d0578e4cbd5922696c92f5068c5513d8e8d60e;p=oota-llvm.git When emitting code for an add, don't force a SCEVUnknown wrapper around a hoisted intermediate result if the intermediate result isn't an Instruction. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@100884 91177308-0d34-0410-b5e6-96231b3b80d8 --- diff --git a/lib/Analysis/ScalarEvolutionExpander.cpp b/lib/Analysis/ScalarEvolutionExpander.cpp index 24825bf5e56..1d1339078de 100644 --- a/lib/Analysis/ScalarEvolutionExpander.cpp +++ b/lib/Analysis/ScalarEvolutionExpander.cpp @@ -705,10 +705,11 @@ Value *SCEVExpander::visitAddExpr(const SCEVAddExpr *S) { Sum = expandAddToGEP(NewOps.begin(), NewOps.end(), PTy, Ty, Sum); } else if (const PointerType *PTy = dyn_cast(Op->getType())) { // The running sum is an integer, and there's a pointer at this level. - // Try to form a getelementptr. Use a SCEVUnknown so that we don't - // re-analyze the instructions that we just emitted. + // Try to form a getelementptr. If the running sum is instructions, + // use a SCEVUnknown to avoid re-analyzing them. SmallVector NewOps; - NewOps.push_back(SE.getUnknown(Sum)); + NewOps.push_back(isa(Sum) ? SE.getUnknown(Sum) : + SE.getSCEV(Sum)); for (++I; I != E && I->first == CurLoop; ++I) NewOps.push_back(I->second); Sum = expandAddToGEP(NewOps.begin(), NewOps.end(), PTy, Ty, expand(Op)); diff --git a/test/Transforms/LoopStrengthReduce/uglygep.ll b/test/Transforms/LoopStrengthReduce/uglygep.ll new file mode 100644 index 00000000000..e72364f0359 --- /dev/null +++ b/test/Transforms/LoopStrengthReduce/uglygep.ll @@ -0,0 +1,37 @@ +; RUN: opt < %s -loop-reduce -S | not grep uglygep + +; LSR shouldn't consider %t8 to be an interesting user of %t6, and it +; should be able to form pretty GEPs. + +target datalayout = "e-p:64:64:64-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:64:64-f32:32:32-f64:64:64-v64:64:64-v128:128:128-a0:0:64-s0:64:64-f80:128:128-n8:16:32:64" +target triple = "x86_64-unknown-linux-gnu" + +define void @Z4() nounwind { +bb: + br label %bb3 + +bb1: ; preds = %bb3 + br i1 undef, label %bb10, label %bb2 + +bb2: ; preds = %bb1 + %t = add i64 %t4, 1 ; [#uses=1] + br label %bb3 + +bb3: ; preds = %bb2, %bb + %t4 = phi i64 [ %t, %bb2 ], [ 0, %bb ] ; [#uses=3] + br label %bb1 + +bb10: ; preds = %bb9 + %t7 = icmp eq i64 %t4, 0 ; [#uses=1] + %t3 = add i64 %t4, 16 ; [#uses=1] + br label %bb14 + +bb14: ; preds = %bb14, %bb10 + %t2 = getelementptr inbounds i8* undef, i64 %t4 ; [#uses=1] + store i8 undef, i8* %t2 + %t6 = load float** undef + %t8 = bitcast float* %t6 to i8* ; [#uses=1] + %t9 = getelementptr inbounds i8* %t8, i64 %t3 ; [#uses=1] + store i8 undef, i8* %t9 + br label %bb14 +}