RPCS3/llvm-mirror
1
0
Fork 0
mirror of https://github.com/RPCS3/llvm-mirror.git synced 2024-11-23 19:23:23 +01:00
Code Issues Projects Releases Wiki Activity

[LV] Minor savings to Sink casts to unravel first order recurrence

Browse Source 
Two minor savings: avoid copying the SinkAfter map and avoid moving a cast if it
is not needed.

Differential Revision: https://reviews.llvm.org/D36408

llvm-svn: 310910
This commit is contained in:
Ayal Zaks 2017-08-15 08:32:59 +00:00
parent b9919e8ad8
commit 43ce19524f
3 changed files with 7 additions and 3 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

3
lib/Transforms/Utils/LoopUtils.cpp
View File

@ -565,7 +565,8 @@ bool RecurrenceDescriptor::isFirstOrderRecurrence(
auto *I = Phi->user_back();
if (I->isCast() && (I->getParent() == Phi->getParent()) && I->hasOneUse() &&
DT->dominates(Previous, I->user_back())) {
SinkAfter[I] = Previous;
if (!DT->dominates(Previous, I)) // Otherwise we're good w/o sinking.
SinkAfter[I] = Previous;
return true;
}
}

2
lib/Transforms/Vectorize/LoopVectorize.cpp
View File

@ -7655,7 +7655,7 @@ void LoopVectorizationPlanner::executePlan(InnerLoopVectorizer &ILV) {
// 2. Copy and widen instructions from the old loop into the new loop.
// Move instructions to handle first-order recurrences.
DenseMap<Instruction *, Instruction *> SinkAfter = Legal->getSinkAfter();
DenseMap<Instruction *, Instruction *> &SinkAfter = Legal->getSinkAfter();
for (auto &Entry : SinkAfter) {
Entry.first->removeFromParent();
Entry.first->insertAfter(Entry.second);

5
test/Transforms/LoopVectorize/first-order-recurrence.ll
View File

@ -140,7 +140,10 @@ scalar.body:
; CHECK: vector.body:
; CHECK: %vector.recur = phi <4 x i16> [ %vector.recur.init, %vector.ph ], [ [[L1:%[a-zA-Z0-9.]+]], %vector.body ]
; CHECK: [[L1]] = load <4 x i16>
; CHECK: {{.*}} = shufflevector <4 x i16> %vector.recur, <4 x i16> [[L1]], <4 x i32> <i32 3, i32 4, i32 5, i32 6>
; CHECK: [[SHUF:%[a-zA-Z0-9.]+]] = shufflevector <4 x i16> %vector.recur, <4 x i16> [[L1]], <4 x i32> <i32 3, i32 4, i32 5, i32 6>
; Check also that the casts were not moved needlessly.
; CHECK: sitofp <4 x i16> [[L1]] to <4 x double>
; CHECK: sitofp <4 x i16> [[SHUF]] to <4 x double>
; CHECK: middle.block:
; CHECK: %vector.recur.extract = extractelement <4 x i16> [[L1]], i32 3
; CHECK: scalar.ph: