[llvm] [ConstantFolding] Fold scalable get_active_lane_masks (PR #156659)
Matthew Devereau via llvm-commits
llvm-commits at lists.llvm.org
Thu Sep 4 04:59:19 PDT 2025
https://github.com/MDevereau updated https://github.com/llvm/llvm-project/pull/156659
>From 49833a23854be79658ed29b71743655bd970f6a6 Mon Sep 17 00:00:00 2001
From: Matthew Devereau <matthew.devereau at arm.com>
Date: Wed, 3 Sep 2025 12:42:57 +0000
Subject: [PATCH 1/2] [ConstantFolding] Fold scalable get_active_lane_masks
Scalable get_active_lane_mask intrinsics with a range of 0 can be lowered to
zeroinitializer. This helps remove no-op scalable masked stores and loads.
When the second operand is 0, this cannot be done (see #152140)
---
llvm/lib/Analysis/ConstantFolding.cpp | 7 ++++
.../ConstProp/active-lane-mask.ll | 34 +++++++++++++++++++
2 files changed, 41 insertions(+)
diff --git a/llvm/lib/Analysis/ConstantFolding.cpp b/llvm/lib/Analysis/ConstantFolding.cpp
index 2148431c1acce..8d849fd230648 100644
--- a/llvm/lib/Analysis/ConstantFolding.cpp
+++ b/llvm/lib/Analysis/ConstantFolding.cpp
@@ -4238,6 +4238,13 @@ static Constant *ConstantFoldScalableVectorCall(
return ConstantInt::getFalse(SVTy);
}
+ case Intrinsic::get_active_lane_mask: {
+ auto Op0 = cast<ConstantInt>(Operands[0])->getValue();
+ auto Op1 = cast<ConstantInt>(Operands[1])->getValue();
+ if ((Op0.uge(Op1) && (!Op1.isZero())))
+ return ConstantVector::getNullValue(SVTy);
+ break;
+ }
default:
break;
}
diff --git a/llvm/test/Transforms/InstSimplify/ConstProp/active-lane-mask.ll b/llvm/test/Transforms/InstSimplify/ConstProp/active-lane-mask.ll
index a904e697cc975..c4a465aa17f36 100644
--- a/llvm/test/Transforms/InstSimplify/ConstProp/active-lane-mask.ll
+++ b/llvm/test/Transforms/InstSimplify/ConstProp/active-lane-mask.ll
@@ -307,6 +307,40 @@ entry:
ret <4 x float> %var33
}
+define <vscale x 4 x i1> @nxv4i1_12_12() {
+; CHECK-LABEL: @nxv4i1_12_12(
+; CHECK-NEXT: entry:
+; CHECK-NEXT: ret <vscale x 4 x i1> zeroinitializer
+;
+entry:
+ %mask = call <vscale x 4 x i1> @llvm.get.active.lane.mask.nxv4i1.i32(i32 12, i32 12)
+ ret <vscale x 4 x i1> %mask
+}
+
+define <vscale x 4 x i1> @nxv4i1_8_4() {
+; CHECK-LABEL: @nxv4i1_8_4(
+; CHECK-NEXT: entry:
+; CHECK-NEXT: ret <vscale x 4 x i1> zeroinitializer
+;
+entry:
+ %mask = call <vscale x 4 x i1> @llvm.get.active.lane.mask.nxv4i1.i32(i32 8, i32 4)
+ ret <vscale x 4 x i1> %mask
+}
+
+define <vscale x 16 x i1> @nxv16i1_0_0() {
+; CHECK-LABEL: @nxv16i1_0_0(
+; CHECK-NEXT: entry:
+; CHECK-NEXT: [[MASK:%.*]] = call <vscale x 16 x i1> @llvm.get.active.lane.mask.nxv16i1.i64(i64 0, i64 0)
+; CHECK-NEXT: ret <vscale x 16 x i1> [[MASK]]
+;
+entry:
+ %mask = call <vscale x 16 x i1> @llvm.get.active.lane.mask.nxv16i1.i64(i64 0, i64 0)
+ ret <vscale x 16 x i1> %mask
+}
+
declare <4 x i1> @llvm.get.active.lane.mask.v4i1.i32(i32, i32)
declare <8 x i1> @llvm.get.active.lane.mask.v8i1.i32(i32, i32)
declare <16 x i1> @llvm.get.active.lane.mask.v16i1.i32(i32, i32)
+
+declare <vscale x 4 x i1> @llvm.get.active.lane.mask.nxv4i1.i32(i32, i32)
+declare <vscale x 16 x i1> @llvm.get.active.lane.mask.nxv4i1.i64(i64, i64)
>From 042b0292ec02d8c2ecb5ed81829b48c6967ee16f Mon Sep 17 00:00:00 2001
From: Matthew Devereau <matthew.devereau at arm.com>
Date: Thu, 4 Sep 2025 11:57:14 +0000
Subject: [PATCH 2/2] Remove Op1.isZero() condition
---
llvm/lib/Analysis/ConstantFolding.cpp | 2 +-
.../test/Transforms/InstSimplify/ConstProp/active-lane-mask.ll | 3 +--
2 files changed, 2 insertions(+), 3 deletions(-)
diff --git a/llvm/lib/Analysis/ConstantFolding.cpp b/llvm/lib/Analysis/ConstantFolding.cpp
index 8d849fd230648..67e6be5b70cb6 100644
--- a/llvm/lib/Analysis/ConstantFolding.cpp
+++ b/llvm/lib/Analysis/ConstantFolding.cpp
@@ -4241,7 +4241,7 @@ static Constant *ConstantFoldScalableVectorCall(
case Intrinsic::get_active_lane_mask: {
auto Op0 = cast<ConstantInt>(Operands[0])->getValue();
auto Op1 = cast<ConstantInt>(Operands[1])->getValue();
- if ((Op0.uge(Op1) && (!Op1.isZero())))
+ if (Op0.uge(Op1))
return ConstantVector::getNullValue(SVTy);
break;
}
diff --git a/llvm/test/Transforms/InstSimplify/ConstProp/active-lane-mask.ll b/llvm/test/Transforms/InstSimplify/ConstProp/active-lane-mask.ll
index c4a465aa17f36..ed26deb58eae4 100644
--- a/llvm/test/Transforms/InstSimplify/ConstProp/active-lane-mask.ll
+++ b/llvm/test/Transforms/InstSimplify/ConstProp/active-lane-mask.ll
@@ -330,8 +330,7 @@ entry:
define <vscale x 16 x i1> @nxv16i1_0_0() {
; CHECK-LABEL: @nxv16i1_0_0(
; CHECK-NEXT: entry:
-; CHECK-NEXT: [[MASK:%.*]] = call <vscale x 16 x i1> @llvm.get.active.lane.mask.nxv16i1.i64(i64 0, i64 0)
-; CHECK-NEXT: ret <vscale x 16 x i1> [[MASK]]
+; CHECK-NEXT: ret <vscale x 16 x i1> zeroinitializer
;
entry:
%mask = call <vscale x 16 x i1> @llvm.get.active.lane.mask.nxv16i1.i64(i64 0, i64 0)
More information about the llvm-commits
mailing list