[SLPVectorizer] Change arguments of 'isStridedLoad' (NFC) #160401
Merged
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Member
|
@llvm/pr-subscribers-vectorizers Author: Mikhail Gudim (mgudim) ChangesThis is needed to reduce the diff for the future work on widening strided loads. Also, with this change we'll be able to re-use this for the case when each pointer represents a start of a group of contiguous loads. Full diff: https://github.com/llvm/llvm-project/pull/160401.diff 1 Files Affected:
diff --git a/llvm/lib/Transforms/Vectorize/SLPVectorizer.cpp b/llvm/lib/Transforms/Vectorize/SLPVectorizer.cpp
index 1814d9a6811c0..4f9ec424da6d4 100644
--- a/llvm/lib/Transforms/Vectorize/SLPVectorizer.cpp
+++ b/llvm/lib/Transforms/Vectorize/SLPVectorizer.cpp
@@ -2234,10 +2234,9 @@ class BoUpSLP {
/// TODO: If load combining is allowed in the IR optimizer, this analysis
/// may not be necessary.
bool isLoadCombineCandidate(ArrayRef<Value *> Stores) const;
- bool isStridedLoad(ArrayRef<Value *> VL, ArrayRef<Value *> PointerOps,
- ArrayRef<unsigned> Order, const TargetTransformInfo &TTI,
- const DataLayout &DL, ScalarEvolution &SE,
- const int64_t Diff, StridedPtrInfo &SPtrInfo) const;
+ bool isStridedLoad(ArrayRef<Value *> PointerOps, Type *ScalarTy,
+ Align Alignment, int64_t Diff, Value *Ptr0, Value *PtrN,
+ StridedPtrInfo &SPtrInfo) const;
/// Checks if the given array of loads can be represented as a vectorized,
/// scatter or just simple gather.
@@ -6817,13 +6816,10 @@ isMaskedLoadCompress(ArrayRef<Value *> VL, ArrayRef<Value *> PointerOps,
/// 4. Any pointer operand is an instruction with the users outside of the
/// current graph (for masked gathers extra extractelement instructions
/// might be required).
-bool BoUpSLP::isStridedLoad(ArrayRef<Value *> VL, ArrayRef<Value *> PointerOps,
- ArrayRef<unsigned> Order,
- const TargetTransformInfo &TTI,
- const DataLayout &DL, ScalarEvolution &SE,
- const int64_t Diff,
- StridedPtrInfo &SPtrInfo) const {
- const size_t Sz = VL.size();
+bool BoUpSLP::isStridedLoad(ArrayRef<Value *> PointerOps, Type *ScalarTy,
+ Align Alignment, int64_t Diff, Value *Ptr0,
+ Value *PtrN, StridedPtrInfo &SPtrInfo) const {
+ const size_t Sz = PointerOps.size();
if (Diff % (Sz - 1) != 0)
return false;
@@ -6835,7 +6831,6 @@ bool BoUpSLP::isStridedLoad(ArrayRef<Value *> VL, ArrayRef<Value *> PointerOps,
});
const uint64_t AbsoluteDiff = std::abs(Diff);
- Type *ScalarTy = VL.front()->getType();
auto *VecTy = getWidenedType(ScalarTy, Sz);
if (IsAnyPointerUsedOutGraph ||
(AbsoluteDiff > Sz &&
@@ -6846,20 +6841,9 @@ bool BoUpSLP::isStridedLoad(ArrayRef<Value *> VL, ArrayRef<Value *> PointerOps,
int64_t Stride = Diff / static_cast<int64_t>(Sz - 1);
if (Diff != Stride * static_cast<int64_t>(Sz - 1))
return false;
- Align Alignment =
- cast<LoadInst>(Order.empty() ? VL.front() : VL[Order.front()])
- ->getAlign();
- if (!TTI.isLegalStridedLoadStore(VecTy, Alignment))
+ if (!TTI->isLegalStridedLoadStore(VecTy, Alignment))
return false;
- Value *Ptr0;
- Value *PtrN;
- if (Order.empty()) {
- Ptr0 = PointerOps.front();
- PtrN = PointerOps.back();
- } else {
- Ptr0 = PointerOps[Order.front()];
- PtrN = PointerOps[Order.back()];
- }
+
// Iterate through all pointers and check if all distances are
// unique multiple of Dist.
SmallSet<int64_t, 4> Dists;
@@ -6868,14 +6852,14 @@ bool BoUpSLP::isStridedLoad(ArrayRef<Value *> VL, ArrayRef<Value *> PointerOps,
if (Ptr == PtrN)
Dist = Diff;
else if (Ptr != Ptr0)
- Dist = *getPointersDiff(ScalarTy, Ptr0, ScalarTy, Ptr, DL, SE);
+ Dist = *getPointersDiff(ScalarTy, Ptr0, ScalarTy, Ptr, *DL, *SE);
// If the strides are not the same or repeated, we can't
// vectorize.
if (((Dist / Stride) * Stride) != Dist || !Dists.insert(Dist).second)
break;
}
if (Dists.size() == Sz) {
- Type *StrideTy = DL.getIndexType(Ptr0->getType());
+ Type *StrideTy = DL->getIndexType(Ptr0->getType());
SPtrInfo.StrideVal = ConstantInt::get(StrideTy, Stride);
SPtrInfo.Ty = getWidenedType(ScalarTy, Sz);
return true;
@@ -6964,7 +6948,8 @@ BoUpSLP::LoadsState BoUpSLP::canVectorizeLoads(
cast<Instruction>(V), UserIgnoreList);
}))
return LoadsState::CompressVectorize;
- if (isStridedLoad(VL, PointerOps, Order, *TTI, *DL, *SE, *Diff, SPtrInfo))
+ if (isStridedLoad(PointerOps, ScalarTy, CommonAlignment, *Diff, Ptr0, PtrN,
+ SPtrInfo))
return LoadsState::StridedVectorize;
}
if (!TTI->isLegalMaskedGather(VecTy, CommonAlignment) ||
|
Member
|
@llvm/pr-subscribers-llvm-transforms Author: Mikhail Gudim (mgudim) ChangesThis is needed to reduce the diff for the future work on widening strided loads. Also, with this change we'll be able to re-use this for the case when each pointer represents a start of a group of contiguous loads. Full diff: https://github.com/llvm/llvm-project/pull/160401.diff 1 Files Affected:
diff --git a/llvm/lib/Transforms/Vectorize/SLPVectorizer.cpp b/llvm/lib/Transforms/Vectorize/SLPVectorizer.cpp
index 1814d9a6811c0..4f9ec424da6d4 100644
--- a/llvm/lib/Transforms/Vectorize/SLPVectorizer.cpp
+++ b/llvm/lib/Transforms/Vectorize/SLPVectorizer.cpp
@@ -2234,10 +2234,9 @@ class BoUpSLP {
/// TODO: If load combining is allowed in the IR optimizer, this analysis
/// may not be necessary.
bool isLoadCombineCandidate(ArrayRef<Value *> Stores) const;
- bool isStridedLoad(ArrayRef<Value *> VL, ArrayRef<Value *> PointerOps,
- ArrayRef<unsigned> Order, const TargetTransformInfo &TTI,
- const DataLayout &DL, ScalarEvolution &SE,
- const int64_t Diff, StridedPtrInfo &SPtrInfo) const;
+ bool isStridedLoad(ArrayRef<Value *> PointerOps, Type *ScalarTy,
+ Align Alignment, int64_t Diff, Value *Ptr0, Value *PtrN,
+ StridedPtrInfo &SPtrInfo) const;
/// Checks if the given array of loads can be represented as a vectorized,
/// scatter or just simple gather.
@@ -6817,13 +6816,10 @@ isMaskedLoadCompress(ArrayRef<Value *> VL, ArrayRef<Value *> PointerOps,
/// 4. Any pointer operand is an instruction with the users outside of the
/// current graph (for masked gathers extra extractelement instructions
/// might be required).
-bool BoUpSLP::isStridedLoad(ArrayRef<Value *> VL, ArrayRef<Value *> PointerOps,
- ArrayRef<unsigned> Order,
- const TargetTransformInfo &TTI,
- const DataLayout &DL, ScalarEvolution &SE,
- const int64_t Diff,
- StridedPtrInfo &SPtrInfo) const {
- const size_t Sz = VL.size();
+bool BoUpSLP::isStridedLoad(ArrayRef<Value *> PointerOps, Type *ScalarTy,
+ Align Alignment, int64_t Diff, Value *Ptr0,
+ Value *PtrN, StridedPtrInfo &SPtrInfo) const {
+ const size_t Sz = PointerOps.size();
if (Diff % (Sz - 1) != 0)
return false;
@@ -6835,7 +6831,6 @@ bool BoUpSLP::isStridedLoad(ArrayRef<Value *> VL, ArrayRef<Value *> PointerOps,
});
const uint64_t AbsoluteDiff = std::abs(Diff);
- Type *ScalarTy = VL.front()->getType();
auto *VecTy = getWidenedType(ScalarTy, Sz);
if (IsAnyPointerUsedOutGraph ||
(AbsoluteDiff > Sz &&
@@ -6846,20 +6841,9 @@ bool BoUpSLP::isStridedLoad(ArrayRef<Value *> VL, ArrayRef<Value *> PointerOps,
int64_t Stride = Diff / static_cast<int64_t>(Sz - 1);
if (Diff != Stride * static_cast<int64_t>(Sz - 1))
return false;
- Align Alignment =
- cast<LoadInst>(Order.empty() ? VL.front() : VL[Order.front()])
- ->getAlign();
- if (!TTI.isLegalStridedLoadStore(VecTy, Alignment))
+ if (!TTI->isLegalStridedLoadStore(VecTy, Alignment))
return false;
- Value *Ptr0;
- Value *PtrN;
- if (Order.empty()) {
- Ptr0 = PointerOps.front();
- PtrN = PointerOps.back();
- } else {
- Ptr0 = PointerOps[Order.front()];
- PtrN = PointerOps[Order.back()];
- }
+
// Iterate through all pointers and check if all distances are
// unique multiple of Dist.
SmallSet<int64_t, 4> Dists;
@@ -6868,14 +6852,14 @@ bool BoUpSLP::isStridedLoad(ArrayRef<Value *> VL, ArrayRef<Value *> PointerOps,
if (Ptr == PtrN)
Dist = Diff;
else if (Ptr != Ptr0)
- Dist = *getPointersDiff(ScalarTy, Ptr0, ScalarTy, Ptr, DL, SE);
+ Dist = *getPointersDiff(ScalarTy, Ptr0, ScalarTy, Ptr, *DL, *SE);
// If the strides are not the same or repeated, we can't
// vectorize.
if (((Dist / Stride) * Stride) != Dist || !Dists.insert(Dist).second)
break;
}
if (Dists.size() == Sz) {
- Type *StrideTy = DL.getIndexType(Ptr0->getType());
+ Type *StrideTy = DL->getIndexType(Ptr0->getType());
SPtrInfo.StrideVal = ConstantInt::get(StrideTy, Stride);
SPtrInfo.Ty = getWidenedType(ScalarTy, Sz);
return true;
@@ -6964,7 +6948,8 @@ BoUpSLP::LoadsState BoUpSLP::canVectorizeLoads(
cast<Instruction>(V), UserIgnoreList);
}))
return LoadsState::CompressVectorize;
- if (isStridedLoad(VL, PointerOps, Order, *TTI, *DL, *SE, *Diff, SPtrInfo))
+ if (isStridedLoad(PointerOps, ScalarTy, CommonAlignment, *Diff, Ptr0, PtrN,
+ SPtrInfo))
return LoadsState::StridedVectorize;
}
if (!TTI->isLegalMaskedGather(VecTy, CommonAlignment) ||
|
mgudim
commented
Sep 23, 2025
Contributor
Author
|
Next, I was planning to do this: https://github.com/llvm/llvm-project/pull/160411/files I think this diff looks much more clear than before. |
|
✅ With the latest revision this PR passed the C/C++ code formatter. |
mgudim
force-pushed
the
is_strided_load
branch
from
99e32d7 to
717eea9
Compare
added 3 commits
September 25, 2025 12:41
This is needed to reduce the diff for the future work on widening strided loads. Also, with this change we'll be able to re-use this for the case when each pointer represents a start of a group of contiguous loads.
mgudim
force-pushed
the
is_strided_load
branch
from
388f51b to
11bd9a2
Compare
Contributor
Author
|
Is this OK to merge? |
alexey-bataev
approved these changes
Oct 1, 2025
Collaborator
|
LLVM Buildbot has detected a new failure on builder Full details are available at: https://lab.llvm.org/buildbot/#/builders/66/builds/20048 Here is the relevant piece of the build log for the reference |
mahesh-attarde
pushed a commit
to mahesh-attarde/llvm-project
that referenced
this pull request
Oct 3, 2025
This is needed to reduce the diff for the future work on widening strided loads. Also, with this change we'll be able to re-use this for the case when each pointer represents a start of a group of contiguous loads.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
4 participants
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
This is needed to reduce the diff for the future work on widening strided loads. Also, with this change we'll be able to re-use this for the case when each pointer represents a start of a group of contiguous loads.