This is an archive of the discontinued LLVM Phabricator instance.

Differential D129843

[AArch64][NFC] Simplify loop vectoriser tail-folding tests
ClosedPublic

Authored by david-arm on Jul 15 2022, 3:07 AM.

Details

Reviewers
sdesmalen
kmclaughlin
RosieSumpter
Commits
rG1e77b0c871a7: [AArch64][NFC] Simplify loop vectoriser tail-folding tests
Summary

I've simplified all of the SVE vectoriser tail-folding tests to
only care about testing the flag:

-prefer-predicate-over-epiloge=predicate-else-scalar-epilogue

In practice we always want to fall back on unpredicated vector
loops if tail-folding is not possible.

Diff Detail

Event Timeline

david-arm created this revision.Jul 15 2022, 3:07 AM
Herald added a project: Restricted Project. · View Herald TranscriptJul 15 2022, 3:07 AM
Herald added subscribers: ctetreau, zzheng, kristof.beyls. · View Herald Transcript
david-arm requested review of this revision.Jul 15 2022, 3:07 AM
Herald added a project: Restricted Project. · View Herald TranscriptJul 15 2022, 3:07 AM
Herald added a subscriber: llvm-commits. · View Herald Transcript
Harbormaster completed remote builds in B175598: Diff 444924.Jul 15 2022, 4:49 AM
Matt added a subscriber: Matt.Jul 15 2022, 10:54 AM
kmclaughlin accepted this revision.Jul 18 2022, 3:32 AM
This revision is now accepted and ready to land.Jul 18 2022, 3:32 AM
This revision was landed with ongoing or failed builds.Jul 18 2022, 5:37 AM
Closed by commit rG1e77b0c871a7: [AArch64][NFC] Simplify loop vectoriser tail-folding tests (authored by david-arm). · Explain Why
This revision was automatically updated to reflect the committed changes.
david-arm added a commit: rG1e77b0c871a7: [AArch64][NFC] Simplify loop vectoriser tail-folding tests.

Revision Contents

PathSize
llvm/
test/
Transforms/
LoopVectorize/
AArch64/
2 lines
2 lines
1 line
1 line

Diff 445465

llvm/test/Transforms/LoopVectorize/AArch64/scalable-reductions-tf.ll

; RUN: opt < %s -loop-vectorize -prefer-predicate-over-epilogue=predicate-dont-vectorize \ ; RUN: opt < %s -loop-vectorize -prefer-predicate-over-epilogue=predicate-else-scalar-epilogue \
; RUN: -mtriple aarch64-unknown-linux-gnu -mattr=+sve -S | FileCheck %s ; RUN: -mtriple aarch64-unknown-linux-gnu -mattr=+sve -S | FileCheck %s
define void @invariant_store_red_exit_is_phi(i32* %dst, i32* readonly %src, i64 %n) { define void @invariant_store_red_exit_is_phi(i32* %dst, i32* readonly %src, i64 %n) {
; CHECK-LABEL: @invariant_store_red_exit_is_phi( ; CHECK-LABEL: @invariant_store_red_exit_is_phi(
; CHECK: vector.ph: ; CHECK: vector.ph:
; CHECK: %[[ACTIVE_LANE_MASK_ENTRY:.*]] = call <vscale x 4 x i1> @llvm.get.active.lane.mask.nxv4i1.i64(i64 0, i64 %n) ; CHECK: %[[ACTIVE_LANE_MASK_ENTRY:.*]] = call <vscale x 4 x i1> @llvm.get.active.lane.mask.nxv4i1.i64(i64 0, i64 %n)
; CHECK: vector.body: ; CHECK: vector.body:
; CHECK: %[[ACTIVE_LANE_MASK:.*]] = phi <vscale x 4 x i1> [ %[[ACTIVE_LANE_MASK_ENTRY]], %vector.ph ], [ %[[ACTIVE_LANE_MASK_NEXT:.*]], %vector.body ] ; CHECK: %[[ACTIVE_LANE_MASK:.*]] = phi <vscale x 4 x i1> [ %[[ACTIVE_LANE_MASK_ENTRY]], %vector.ph ], [ %[[ACTIVE_LANE_MASK_NEXT:.*]], %vector.body ]
Show All 34 Lines

llvm/test/Transforms/LoopVectorize/AArch64/sve-tail-folding-cost.ll

; RUN: opt -S -loop-vectorize -prefer-predicate-over-epilogue=predicate-dont-vectorize <%s | FileCheck %s ; RUN: opt -S -loop-vectorize -prefer-predicate-over-epilogue=predicate-else-scalar-epilogue <%s | FileCheck %s
target triple = "aarch64-unknown-linux-gnu" target triple = "aarch64-unknown-linux-gnu"
; The uniform load of %d in the following loop triggers the special ; The uniform load of %d in the following loop triggers the special
; branch costing code in LoopVectorizationCostModel::getInstructionCost. ; branch costing code in LoopVectorizationCostModel::getInstructionCost.
; However, this should only affect the fixed-width cost because for ; However, this should only affect the fixed-width cost because for
; NEON it needs to scalarize the load, whereas for SVE it can use a predicated load. ; NEON it needs to scalarize the load, whereas for SVE it can use a predicated load.
; Because of how the LoopVectorizer annotates the load to need scalarization with ; Because of how the LoopVectorizer annotates the load to need scalarization with
Show All 26 Lines

llvm/test/Transforms/LoopVectorize/AArch64/sve-tail-folding-unroll.ll

; RUN: opt -S -loop-vectorize -prefer-predicate-over-epilogue=predicate-dont-vectorize -force-vector-interleave=4 -force-vector-width=4 < %s | FileCheck %s
; RUN: opt -S -loop-vectorize -prefer-predicate-over-epilogue=predicate-else-scalar-epilogue -force-vector-interleave=4 -force-vector-width=4 < %s | FileCheck %s ; RUN: opt -S -loop-vectorize -prefer-predicate-over-epilogue=predicate-else-scalar-epilogue -force-vector-interleave=4 -force-vector-width=4 < %s | FileCheck %s
target triple = "aarch64-unknown-linux-gnu" target triple = "aarch64-unknown-linux-gnu"
define void @simple_memset(i32 %val, i32* %ptr, i64 %n) #0 { define void @simple_memset(i32 %val, i32* %ptr, i64 %n) #0 {
; CHECK-LABEL: @simple_memset( ; CHECK-LABEL: @simple_memset(
; CHECK-NEXT: entry: ; CHECK-NEXT: entry:
▲ Show 20 Lines • Show All 288 Lines • Show Last 20 Lines

llvm/test/Transforms/LoopVectorize/AArch64/sve-tail-folding.ll

; RUN: opt -S -hints-allow-reordering=false -loop-vectorize -prefer-predicate-over-epilogue=predicate-dont-vectorize -prefer-inloop-reductions < %s | FileCheck %s
; RUN: opt -S -hints-allow-reordering=false -loop-vectorize -prefer-predicate-over-epilogue=predicate-else-scalar-epilogue -prefer-inloop-reductions < %s | FileCheck %s ; RUN: opt -S -hints-allow-reordering=false -loop-vectorize -prefer-predicate-over-epilogue=predicate-else-scalar-epilogue -prefer-inloop-reductions < %s | FileCheck %s
target triple = "aarch64-unknown-linux-gnu" target triple = "aarch64-unknown-linux-gnu"
define void @simple_memset(i32 %val, i32* %ptr, i64 %n) #0 { define void @simple_memset(i32 %val, i32* %ptr, i64 %n) #0 {
; CHECK-LABEL: @simple_memset( ; CHECK-LABEL: @simple_memset(
; CHECK-NEXT: entry: ; CHECK-NEXT: entry:
▲ Show 20 Lines • Show All 782 Lines • Show Last 20 Lines