summaryrefslogtreecommitdiffstats
path: root/llvm/lib
diff options
context:
space:
mode:
authorSanjay Patel <spatel@rotateright.com>2017-06-07 00:17:08 +0000
committerSanjay Patel <spatel@rotateright.com>2017-06-07 00:17:08 +0000
commitf57015d4cc92ad7721ab52be4ba26e94b3a379e9 (patch)
treecd65333c6865935b8397fd4c4ac5ce5e1a21cd64 /llvm/lib
parentd9c251f46c83dcdbb1a949029a118771c594bb45 (diff)
downloadbcm5719-llvm-f57015d4cc92ad7721ab52be4ba26e94b3a379e9.tar.gz
bcm5719-llvm-f57015d4cc92ad7721ab52be4ba26e94b3a379e9.zip
[CGP / PowerPC] use direct compares if there's only one load per block in memcmp() expansion
I'd like to enable CGP memcmp expansion for x86, but the output from CGP would regress the special cases (memcmp(x,y,N) != 0 for N=1,2,4,8,16,32 bytes) that we already handle. I'm not sure if we'll actually be able to produce the optimal code given the block-at-a-time limitation in the DAG. We might have to just avoid those special-cases here in CGP. But regardless of that, I think this is a win for the more general cases. http://rise4fun.com/Alive/cbQ Differential Revision: https://reviews.llvm.org/D33963 llvm-svn: 304849
Diffstat (limited to 'llvm/lib')
-rw-r--r--llvm/lib/CodeGen/CodeGenPrepare.cpp30
1 files changed, 19 insertions, 11 deletions
diff --git a/llvm/lib/CodeGen/CodeGenPrepare.cpp b/llvm/lib/CodeGen/CodeGenPrepare.cpp
index 2d6d3ada212..12b00eaf6d7 100644
--- a/llvm/lib/CodeGen/CodeGenPrepare.cpp
+++ b/llvm/lib/CodeGen/CodeGenPrepare.cpp
@@ -1812,7 +1812,7 @@ void MemCmpExpansion::emitLoadCompareBlockMultipleLoads(
unsigned NumLoads = std::min(NumLoadsRemaining, NumLoadsPerBlock);
Builder.SetInsertPoint(LoadCmpBlocks[Index]);
-
+ Value *Cmp = nullptr;
for (unsigned i = 0; i < NumLoads; ++i) {
unsigned LoadSize = getLoadSize(RemainingBytes);
unsigned GEPIndex = NumBytesProcessed / LoadSize;
@@ -1846,9 +1846,16 @@ void MemCmpExpansion::emitLoadCompareBlockMultipleLoads(
LoadSrc1 = Builder.CreateZExtOrTrunc(LoadSrc1, MaxLoadType);
LoadSrc2 = Builder.CreateZExtOrTrunc(LoadSrc2, MaxLoadType);
}
- Diff = Builder.CreateXor(LoadSrc1, LoadSrc2);
- Diff = Builder.CreateZExtOrTrunc(Diff, MaxLoadType);
- XorList.push_back(Diff);
+ if (NumLoads != 1) {
+ // If we have multiple loads per block, we need to generate a composite
+ // comparison using xor+or.
+ Diff = Builder.CreateXor(LoadSrc1, LoadSrc2);
+ Diff = Builder.CreateZExtOrTrunc(Diff, MaxLoadType);
+ XorList.push_back(Diff);
+ } else {
+ // If there's only one load per block, we just compare the loaded values.
+ Cmp = Builder.CreateICmpNE(LoadSrc1, LoadSrc2);
+ }
}
auto pairWiseOr = [&](std::vector<Value *> &InList) -> std::vector<Value *> {
@@ -1862,16 +1869,17 @@ void MemCmpExpansion::emitLoadCompareBlockMultipleLoads(
return OutList;
};
- // Pairwise OR the XOR results.
- OrList = pairWiseOr(XorList);
+ if (!Cmp) {
+ // Pairwise OR the XOR results.
+ OrList = pairWiseOr(XorList);
- // Pairwise OR the OR results until one result left.
- while (OrList.size() != 1) {
- OrList = pairWiseOr(OrList);
+ // Pairwise OR the OR results until one result left.
+ while (OrList.size() != 1) {
+ OrList = pairWiseOr(OrList);
+ }
+ Cmp = Builder.CreateICmpNE(OrList[0], ConstantInt::get(Diff->getType(), 0));
}
- Value *Cmp = Builder.CreateICmp(ICmpInst::ICMP_NE, OrList[0],
- ConstantInt::get(Diff->getType(), 0));
BasicBlock *NextBB = (Index == (LoadCmpBlocks.size() - 1))
? EndBlock
: LoadCmpBlocks[Index + 1];
OpenPOWER on IntegriCloud