Recommit "[LV][VPlan] Remove any-of reduction from precomputeCost. NFC (#117109)" #117289
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
The any-of reduction contains phi and select instructions. The select instruction might be optimized and removed in the vplan which may cause VF difference between legacy and VPlan-based model. But if the select instruction be removed, `planContainsAdditionalSimplifications()` will catch it and disable the assertion. Therefore, we can just remove the ayn-of reduction calculation in the precomputeCost().
|
@llvm/pr-subscribers-vectorizers @llvm/pr-subscribers-llvm-transforms Author: Elvis Wang (ElvisWang123) ChangesUpdate the test cases contains Origin message: The any-of reduction contains phi and select instructions. The select instruction might be optimized and removed in the vplan which may cause VF difference between legacy and VPlan-based model. But if the select instruction be removed, planContainsAdditionalSimplifications() will catch it and disable the assertion. Therefore, we can just remove the ayn-of reduction calculation in the precomputeCost(). Recommit "[LV][VPlan] Remove any-of reduction from precomputeCost. NFC (#117109)" Full diff: https://github.com/llvm/llvm-project/pull/117289.diff 2 Files Affected:
diff --git a/llvm/lib/Transforms/Vectorize/LoopVectorize.cpp b/llvm/lib/Transforms/Vectorize/LoopVectorize.cpp
index d13770a35c108f..5b556058cc762c 100644
--- a/llvm/lib/Transforms/Vectorize/LoopVectorize.cpp
+++ b/llvm/lib/Transforms/Vectorize/LoopVectorize.cpp
@@ -7303,34 +7303,14 @@ LoopVectorizationPlanner::precomputeCosts(VPlan &Plan, ElementCount VF,
// The legacy cost model has special logic to compute the cost of in-loop
// reductions, which may be smaller than the sum of all instructions involved
- // in the reduction. For AnyOf reductions, VPlan codegen may remove the select
- // which the legacy cost model uses to assign cost. Pre-compute their costs
- // for now.
+ // in the reduction.
// TODO: Switch to costing based on VPlan once the logic has been ported.
for (const auto &[RedPhi, RdxDesc] : Legal->getReductionVars()) {
if (ForceTargetInstructionCost.getNumOccurrences())
continue;
- if (!CM.isInLoopReduction(RedPhi) &&
- !RecurrenceDescriptor::isAnyOfRecurrenceKind(
- RdxDesc.getRecurrenceKind()))
- continue;
-
- // AnyOf reduction codegen may remove the select. To match the legacy cost
- // model, pre-compute the cost for AnyOf reductions here.
- if (RecurrenceDescriptor::isAnyOfRecurrenceKind(
- RdxDesc.getRecurrenceKind())) {
- auto *Select = cast<SelectInst>(*find_if(
- RedPhi->users(), [](User *U) { return isa<SelectInst>(U); }));
- assert(!CostCtx.SkipCostComputation.contains(Select) &&
- "reduction op visited multiple times");
- CostCtx.SkipCostComputation.insert(Select);
- auto ReductionCost = CostCtx.getLegacyCost(Select, VF);
- LLVM_DEBUG(dbgs() << "Cost of " << ReductionCost << " for VF " << VF
- << ":\n any-of reduction " << *Select << "\n");
- Cost += ReductionCost;
+ if (!CM.isInLoopReduction(RedPhi))
continue;
- }
const auto &ChainOps = RdxDesc.getReductionOpChain(RedPhi, OrigLoop);
SetVector<Instruction *> ChainOpsAndOperands(ChainOps.begin(),
diff --git a/llvm/test/Transforms/LoopVectorize/AArch64/select-costs.ll b/llvm/test/Transforms/LoopVectorize/AArch64/select-costs.ll
index 6165a73e77f238..403fc9f316d35c 100644
--- a/llvm/test/Transforms/LoopVectorize/AArch64/select-costs.ll
+++ b/llvm/test/Transforms/LoopVectorize/AArch64/select-costs.ll
@@ -50,10 +50,6 @@ for.cond.cleanup: ; preds = %for.cond.cleanup.lo
define i32 @multi_user_cmp(ptr readonly %a, i64 noundef %n) {
; CHECK: LV: Checking a loop in 'multi_user_cmp'
-; CHECK: Cost of 1 for VF 16:
-; CHECK: any-of reduction %all.off = select i1 %cmp1, i1 %all.off.next, i1 false
-; CHECK: Cost of 1 for VF 16:
-; CHECK: any-of reduction %.any.0.off0 = select i1 %cmp1, i1 true, i1 %any.0.off09
; CHECK: Cost of 4 for VF 16: WIDEN ir<%cmp1> = fcmp olt ir<%load1>, ir<0.000000e+00>
; CHECK: LV: Selecting VF: 16.
entry:
|
|
LLVM Buildbot has detected a new failure on builder Full details are available at: https://lab.llvm.org/buildbot/#/builders/171/builds/11246 Here is the relevant piece of the build log for the reference |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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.
Update the test cases contains
any-ofprintings from the precomputeCost().Origin message:
The any-of reduction contains phi and select instructions.
The select instruction might be optimized and removed in the vplan which may cause VF difference between legacy and VPlan-based model. But if the select instruction be removed, planContainsAdditionalSimplifications() will catch it and disable the assertion.
Therefore, we can just remove the ayn-of reduction calculation in the precomputeCost().
Recommit "[LV][VPlan] Remove any-of reduction from precomputeCost. NFC (#117109)"