[llvm] [VP] Check if VP ops with functional intrinsics are speculatable (PR #69504)
Luke Lau via llvm-commits
llvm-commits at lists.llvm.org
Wed Oct 18 12:50:27 PDT 2023
https://github.com/lukel97 created https://github.com/llvm/llvm-project/pull/69504
Noticed whilst working on #69494. VP intrinsics whose functional equivalent is
an intrinsic were being marked as their lanes being non-speculatable, even if
the underlying intrinsic was speculatable.
This meant that
%1 = call <4 x i32> @llvm.vp.umax(<4 x i32> %x, <4 x i32> %y, <4 x i1> %mask, i32 %evl)
would be expanded out to
%.splatinsert = insertelement <4 x i32> poison, i32 %evl, i64 0
%.splat = shufflevector <4 x i32> %.splatinsert, <4 x i32> poison, <4 x i32> zeroinitializer
%1 = icmp ult <4 x i32> <i32 0, i32 1, i32 2, i32 3>, %.splat
%2 = and <4 x i1> %1, %mask
%3 = call <4 x i32> @llvm.umax.v4i32(<4 x i32> %x, <4 x i32> %y)
instead of
%1 = call <4 x i32> @llvm.umax.v4i32(<4 x i32> %x, <4 x i32> %y)
The cause of this was isSafeToSpeculativelyExecuteWithOpcode checking the
function attributes for the VP instruction itself, not the functional
intrinsic. Since isSafeToSpeculativelyExecuteWithOpcode expects an already
materialized instruction, we can't use it directly for the intrinsic case. So
this fixes it by manually checking the function attributes on the intrinsic.
>From adfce648fbfa0184ee9f2e9aa39655b2ce2a8c0a Mon Sep 17 00:00:00 2001
From: Luke Lau <luke at igalia.com>
Date: Wed, 18 Oct 2023 15:37:34 -0400
Subject: [PATCH] [VP] Check if VP ops with functional intrinsics are
speculatable
Noticed whilst working on #69494. VP intrinsics whose functional equivalent is
an intrinsic were being marked as their lanes being non-speculatable, even if
the underlying intrinsic was speculatable.
This meant that
%1 = call <4 x i32> @llvm.vp.umax(<4 x i32> %x, <4 x i32> %y, <4 x i1> %mask, i32 %evl)
would be expanded out to
%.splatinsert = insertelement <4 x i32> poison, i32 %evl, i64 0
%.splat = shufflevector <4 x i32> %.splatinsert, <4 x i32> poison, <4 x i32> zeroinitializer
%1 = icmp ult <4 x i32> <i32 0, i32 1, i32 2, i32 3>, %.splat
%2 = and <4 x i1> %1, %mask
%3 = call <4 x i32> @llvm.umax.v4i32(<4 x i32> %x, <4 x i32> %y)
instead of
%1 = call <4 x i32> @llvm.umax.v4i32(<4 x i32> %x, <4 x i32> %y)
The cause of this was isSafeToSpeculativelyExecuteWithOpcode checking the
function attributes for the VP instruction itself, not the functional
intrinsic. Since isSafeToSpeculativelyExecuteWithOpcode expects an already
materialized instruction, we can't use it directly for the intrinsic case. So
this fixes it by manually checking the function attributes on the intrinsic.
---
llvm/lib/CodeGen/ExpandVectorPredication.cpp | 9 +-
llvm/test/CodeGen/Generic/expand-vp.ll | 92 +++++++++++++-------
2 files changed, 68 insertions(+), 33 deletions(-)
diff --git a/llvm/lib/CodeGen/ExpandVectorPredication.cpp b/llvm/lib/CodeGen/ExpandVectorPredication.cpp
index 2d4da33a566dc15..6c873a9aee27f7f 100644
--- a/llvm/lib/CodeGen/ExpandVectorPredication.cpp
+++ b/llvm/lib/CodeGen/ExpandVectorPredication.cpp
@@ -123,9 +123,12 @@ static bool maySpeculateLanes(VPIntrinsic &VPI) {
if (isa<VPReductionIntrinsic>(VPI))
return false;
// Fallback to whether the intrinsic is speculatable.
- std::optional<unsigned> OpcOpt = VPI.getFunctionalOpcode();
- unsigned FunctionalOpc = OpcOpt.value_or((unsigned)Instruction::Call);
- return isSafeToSpeculativelyExecuteWithOpcode(FunctionalOpc, &VPI);
+ if (auto IntrID = VPI.getFunctionalIntrinsicID())
+ return Intrinsic::getAttributes(VPI.getContext(), *IntrID)
+ .hasFnAttr(Attribute::AttrKind::Speculatable);
+ if (auto Opc = VPI.getFunctionalOpcode())
+ return isSafeToSpeculativelyExecuteWithOpcode(*Opc, &VPI);
+ return false;
}
//// } Helpers
diff --git a/llvm/test/CodeGen/Generic/expand-vp.ll b/llvm/test/CodeGen/Generic/expand-vp.ll
index 509f86a64d9ce7c..40d183273b86d67 100644
--- a/llvm/test/CodeGen/Generic/expand-vp.ll
+++ b/llvm/test/CodeGen/Generic/expand-vp.ll
@@ -18,6 +18,10 @@ declare <8 x i32> @llvm.vp.sdiv.v8i32(<8 x i32>, <8 x i32>, <8 x i1>, i32)
declare <8 x i32> @llvm.vp.srem.v8i32(<8 x i32>, <8 x i32>, <8 x i1>, i32)
declare <8 x i32> @llvm.vp.udiv.v8i32(<8 x i32>, <8 x i32>, <8 x i1>, i32)
declare <8 x i32> @llvm.vp.urem.v8i32(<8 x i32>, <8 x i32>, <8 x i1>, i32)
+declare <8 x i32> @llvm.vp.smax.v8i32(<8 x i32>, <8 x i32>, <8 x i1>, i32)
+declare <8 x i32> @llvm.vp.smin.v8i32(<8 x i32>, <8 x i32>, <8 x i1>, i32)
+declare <8 x i32> @llvm.vp.umax.v8i32(<8 x i32>, <8 x i32>, <8 x i1>, i32)
+declare <8 x i32> @llvm.vp.umin.v8i32(<8 x i32>, <8 x i32>, <8 x i1>, i32)
; Bit arith
declare <8 x i32> @llvm.vp.and.v8i32(<8 x i32>, <8 x i32>, <8 x i1>, i32)
declare <8 x i32> @llvm.vp.xor.v8i32(<8 x i32>, <8 x i32>, <8 x i1>, i32)
@@ -52,12 +56,16 @@ define void @test_vp_int_v8(<8 x i32> %i0, <8 x i32> %i1, <8 x i32> %i2, <8 x i3
%r4 = call <8 x i32> @llvm.vp.srem.v8i32(<8 x i32> %i0, <8 x i32> %i1, <8 x i1> %m, i32 %n)
%r5 = call <8 x i32> @llvm.vp.udiv.v8i32(<8 x i32> %i0, <8 x i32> %i1, <8 x i1> %m, i32 %n)
%r6 = call <8 x i32> @llvm.vp.urem.v8i32(<8 x i32> %i0, <8 x i32> %i1, <8 x i1> %m, i32 %n)
- %r7 = call <8 x i32> @llvm.vp.and.v8i32(<8 x i32> %i0, <8 x i32> %i1, <8 x i1> %m, i32 %n)
- %r8 = call <8 x i32> @llvm.vp.or.v8i32(<8 x i32> %i0, <8 x i32> %i1, <8 x i1> %m, i32 %n)
- %r9 = call <8 x i32> @llvm.vp.xor.v8i32(<8 x i32> %i0, <8 x i32> %i1, <8 x i1> %m, i32 %n)
- %rA = call <8 x i32> @llvm.vp.ashr.v8i32(<8 x i32> %i0, <8 x i32> %i1, <8 x i1> %m, i32 %n)
- %rB = call <8 x i32> @llvm.vp.lshr.v8i32(<8 x i32> %i0, <8 x i32> %i1, <8 x i1> %m, i32 %n)
- %rC = call <8 x i32> @llvm.vp.shl.v8i32(<8 x i32> %i0, <8 x i32> %i1, <8 x i1> %m, i32 %n)
+ %r7 = call <8 x i32> @llvm.vp.smax.v8i32(<8 x i32> %i0, <8 x i32> %i1, <8 x i1> %m, i32 %n)
+ %r8 = call <8 x i32> @llvm.vp.smin.v8i32(<8 x i32> %i0, <8 x i32> %i1, <8 x i1> %m, i32 %n)
+ %r9 = call <8 x i32> @llvm.vp.umax.v8i32(<8 x i32> %i0, <8 x i32> %i1, <8 x i1> %m, i32 %n)
+ %rA = call <8 x i32> @llvm.vp.umin.v8i32(<8 x i32> %i0, <8 x i32> %i1, <8 x i1> %m, i32 %n)
+ %rB = call <8 x i32> @llvm.vp.and.v8i32(<8 x i32> %i0, <8 x i32> %i1, <8 x i1> %m, i32 %n)
+ %rC = call <8 x i32> @llvm.vp.or.v8i32(<8 x i32> %i0, <8 x i32> %i1, <8 x i1> %m, i32 %n)
+ %rD = call <8 x i32> @llvm.vp.xor.v8i32(<8 x i32> %i0, <8 x i32> %i1, <8 x i1> %m, i32 %n)
+ %rE = call <8 x i32> @llvm.vp.ashr.v8i32(<8 x i32> %i0, <8 x i32> %i1, <8 x i1> %m, i32 %n)
+ %rF = call <8 x i32> @llvm.vp.lshr.v8i32(<8 x i32> %i0, <8 x i32> %i1, <8 x i1> %m, i32 %n)
+ %r10 = call <8 x i32> @llvm.vp.shl.v8i32(<8 x i32> %i0, <8 x i32> %i1, <8 x i1> %m, i32 %n)
ret void
}
@@ -70,6 +78,10 @@ declare <vscale x 4 x i32> @llvm.vp.sdiv.nxv4i32(<vscale x 4 x i32>, <vscale x 4
declare <vscale x 4 x i32> @llvm.vp.srem.nxv4i32(<vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i1>, i32)
declare <vscale x 4 x i32> @llvm.vp.udiv.nxv4i32(<vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i1>, i32)
declare <vscale x 4 x i32> @llvm.vp.urem.nxv4i32(<vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i1>, i32)
+declare <vscale x 4 x i32> @llvm.vp.smax.nxv4i32(<vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i1>, i32)
+declare <vscale x 4 x i32> @llvm.vp.smin.nxv4i32(<vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i1>, i32)
+declare <vscale x 4 x i32> @llvm.vp.umax.nxv4i32(<vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i1>, i32)
+declare <vscale x 4 x i32> @llvm.vp.umin.nxv4i32(<vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i1>, i32)
; Bit arith
declare <vscale x 4 x i32> @llvm.vp.and.nxv4i32(<vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i1>, i32)
declare <vscale x 4 x i32> @llvm.vp.xor.nxv4i32(<vscale x 4 x i32>, <vscale x 4 x i32>, <vscale x 4 x i1>, i32)
@@ -87,12 +99,16 @@ define void @test_vp_int_vscale(<vscale x 4 x i32> %i0, <vscale x 4 x i32> %i1,
%r4 = call <vscale x 4 x i32> @llvm.vp.srem.nxv4i32(<vscale x 4 x i32> %i0, <vscale x 4 x i32> %i1, <vscale x 4 x i1> %m, i32 %n)
%r5 = call <vscale x 4 x i32> @llvm.vp.udiv.nxv4i32(<vscale x 4 x i32> %i0, <vscale x 4 x i32> %i1, <vscale x 4 x i1> %m, i32 %n)
%r6 = call <vscale x 4 x i32> @llvm.vp.urem.nxv4i32(<vscale x 4 x i32> %i0, <vscale x 4 x i32> %i1, <vscale x 4 x i1> %m, i32 %n)
- %r7 = call <vscale x 4 x i32> @llvm.vp.and.nxv4i32(<vscale x 4 x i32> %i0, <vscale x 4 x i32> %i1, <vscale x 4 x i1> %m, i32 %n)
- %r8 = call <vscale x 4 x i32> @llvm.vp.or.nxv4i32(<vscale x 4 x i32> %i0, <vscale x 4 x i32> %i1, <vscale x 4 x i1> %m, i32 %n)
- %r9 = call <vscale x 4 x i32> @llvm.vp.xor.nxv4i32(<vscale x 4 x i32> %i0, <vscale x 4 x i32> %i1, <vscale x 4 x i1> %m, i32 %n)
- %rA = call <vscale x 4 x i32> @llvm.vp.ashr.nxv4i32(<vscale x 4 x i32> %i0, <vscale x 4 x i32> %i1, <vscale x 4 x i1> %m, i32 %n)
- %rB = call <vscale x 4 x i32> @llvm.vp.lshr.nxv4i32(<vscale x 4 x i32> %i0, <vscale x 4 x i32> %i1, <vscale x 4 x i1> %m, i32 %n)
- %rC = call <vscale x 4 x i32> @llvm.vp.shl.nxv4i32(<vscale x 4 x i32> %i0, <vscale x 4 x i32> %i1, <vscale x 4 x i1> %m, i32 %n)
+ %r7 = call <vscale x 4 x i32> @llvm.vp.smax.nxv4i32(<vscale x 4 x i32> %i0, <vscale x 4 x i32> %i1, <vscale x 4 x i1> %m, i32 %n)
+ %r8 = call <vscale x 4 x i32> @llvm.vp.smin.nxv4i32(<vscale x 4 x i32> %i0, <vscale x 4 x i32> %i1, <vscale x 4 x i1> %m, i32 %n)
+ %r9 = call <vscale x 4 x i32> @llvm.vp.umax.nxv4i32(<vscale x 4 x i32> %i0, <vscale x 4 x i32> %i1, <vscale x 4 x i1> %m, i32 %n)
+ %rA = call <vscale x 4 x i32> @llvm.vp.umin.nxv4i32(<vscale x 4 x i32> %i0, <vscale x 4 x i32> %i1, <vscale x 4 x i1> %m, i32 %n)
+ %rB = call <vscale x 4 x i32> @llvm.vp.and.nxv4i32(<vscale x 4 x i32> %i0, <vscale x 4 x i32> %i1, <vscale x 4 x i1> %m, i32 %n)
+ %rC = call <vscale x 4 x i32> @llvm.vp.or.nxv4i32(<vscale x 4 x i32> %i0, <vscale x 4 x i32> %i1, <vscale x 4 x i1> %m, i32 %n)
+ %rD = call <vscale x 4 x i32> @llvm.vp.xor.nxv4i32(<vscale x 4 x i32> %i0, <vscale x 4 x i32> %i1, <vscale x 4 x i1> %m, i32 %n)
+ %rE = call <vscale x 4 x i32> @llvm.vp.ashr.nxv4i32(<vscale x 4 x i32> %i0, <vscale x 4 x i32> %i1, <vscale x 4 x i1> %m, i32 %n)
+ %rF = call <vscale x 4 x i32> @llvm.vp.lshr.nxv4i32(<vscale x 4 x i32> %i0, <vscale x 4 x i32> %i1, <vscale x 4 x i1> %m, i32 %n)
+ %r10 = call <vscale x 4 x i32> @llvm.vp.shl.nxv4i32(<vscale x 4 x i32> %i0, <vscale x 4 x i32> %i1, <vscale x 4 x i1> %m, i32 %n)
ret void
}
@@ -166,6 +182,10 @@ define void @test_vp_cmp_v8(<8 x i32> %i0, <8 x i32> %i1, <8 x float> %f0, <8 x
; ALL-CONVERT: %{{.+}} = udiv <8 x i32> %i0, %{{.+}}
; ALL-CONVERT-NOT: %{{.+}} = urem <8 x i32> %i0, %i1
; ALL-CONVERT: %{{.+}} = urem <8 x i32> %i0, %{{.+}}
+; ALL-CONVERT-NEXT: %{{.+}} = call <8 x i32> @llvm.smax.v8i32(<8 x i32> %i0, <8 x i32> %i1)
+; ALL-CONVERT-NEXT: %{{.+}} = call <8 x i32> @llvm.smin.v8i32(<8 x i32> %i0, <8 x i32> %i1)
+; ALL-CONVERT-NEXT: %{{.+}} = call <8 x i32> @llvm.umax.v8i32(<8 x i32> %i0, <8 x i32> %i1)
+; ALL-CONVERT-NEXT: %{{.+}} = call <8 x i32> @llvm.umin.v8i32(<8 x i32> %i0, <8 x i32> %i1)
; ALL-CONVERT-NEXT: %{{.+}} = and <8 x i32> %i0, %i1
; ALL-CONVERT-NEXT: %{{.+}} = or <8 x i32> %i0, %i1
; ALL-CONVERT-NEXT: %{{.+}} = xor <8 x i32> %i0, %i1
@@ -263,12 +283,16 @@ define void @test_vp_cmp_v8(<8 x i32> %i0, <8 x i32> %i1, <8 x float> %f0, <8 x
; LEGAL_LEGAL-NEXT: %r4 = call <8 x i32> @llvm.vp.srem.v8i32(<8 x i32> %i0, <8 x i32> %i1, <8 x i1> %m, i32 %n)
; LEGAL_LEGAL-NEXT: %r5 = call <8 x i32> @llvm.vp.udiv.v8i32(<8 x i32> %i0, <8 x i32> %i1, <8 x i1> %m, i32 %n)
; LEGAL_LEGAL-NEXT: %r6 = call <8 x i32> @llvm.vp.urem.v8i32(<8 x i32> %i0, <8 x i32> %i1, <8 x i1> %m, i32 %n)
-; LEGAL_LEGAL-NEXT: %r7 = call <8 x i32> @llvm.vp.and.v8i32(<8 x i32> %i0, <8 x i32> %i1, <8 x i1> %m, i32 %n)
-; LEGAL_LEGAL-NEXT: %r8 = call <8 x i32> @llvm.vp.or.v8i32(<8 x i32> %i0, <8 x i32> %i1, <8 x i1> %m, i32 %n)
-; LEGAL_LEGAL-NEXT: %r9 = call <8 x i32> @llvm.vp.xor.v8i32(<8 x i32> %i0, <8 x i32> %i1, <8 x i1> %m, i32 %n)
-; LEGAL_LEGAL-NEXT: %rA = call <8 x i32> @llvm.vp.ashr.v8i32(<8 x i32> %i0, <8 x i32> %i1, <8 x i1> %m, i32 %n)
-; LEGAL_LEGAL-NEXT: %rB = call <8 x i32> @llvm.vp.lshr.v8i32(<8 x i32> %i0, <8 x i32> %i1, <8 x i1> %m, i32 %n)
-; LEGAL_LEGAL-NEXT: %rC = call <8 x i32> @llvm.vp.shl.v8i32(<8 x i32> %i0, <8 x i32> %i1, <8 x i1> %m, i32 %n)
+; LEGAL_LEGAL-NEXT: %r7 = call <8 x i32> @llvm.vp.smax.v8i32(<8 x i32> %i0, <8 x i32> %i1, <8 x i1> %m, i32 %n)
+; LEGAL_LEGAL-NEXT: %r8 = call <8 x i32> @llvm.vp.smin.v8i32(<8 x i32> %i0, <8 x i32> %i1, <8 x i1> %m, i32 %n)
+; LEGAL_LEGAL-NEXT: %r9 = call <8 x i32> @llvm.vp.umax.v8i32(<8 x i32> %i0, <8 x i32> %i1, <8 x i1> %m, i32 %n)
+; LEGAL_LEGAL-NEXT: %rA = call <8 x i32> @llvm.vp.umin.v8i32(<8 x i32> %i0, <8 x i32> %i1, <8 x i1> %m, i32 %n)
+; LEGAL_LEGAL-NEXT: %rB = call <8 x i32> @llvm.vp.and.v8i32(<8 x i32> %i0, <8 x i32> %i1, <8 x i1> %m, i32 %n)
+; LEGAL_LEGAL-NEXT: %rC = call <8 x i32> @llvm.vp.or.v8i32(<8 x i32> %i0, <8 x i32> %i1, <8 x i1> %m, i32 %n)
+; LEGAL_LEGAL-NEXT: %rD = call <8 x i32> @llvm.vp.xor.v8i32(<8 x i32> %i0, <8 x i32> %i1, <8 x i1> %m, i32 %n)
+; LEGAL_LEGAL-NEXT: %rE = call <8 x i32> @llvm.vp.ashr.v8i32(<8 x i32> %i0, <8 x i32> %i1, <8 x i1> %m, i32 %n)
+; LEGAL_LEGAL-NEXT: %rF = call <8 x i32> @llvm.vp.lshr.v8i32(<8 x i32> %i0, <8 x i32> %i1, <8 x i1> %m, i32 %n)
+; LEGAL_LEGAL-NEXT: %r10 = call <8 x i32> @llvm.vp.shl.v8i32(<8 x i32> %i0, <8 x i32> %i1, <8 x i1> %m, i32 %n)
; LEGAL_LEGAL-NEXT: ret void
; LEGAL_LEGAL:define void @test_vp_int_vscale(<vscale x 4 x i32> %i0, <vscale x 4 x i32> %i1, <vscale x 4 x i32> %i2, <vscale x 4 x i32> %f3, <vscale x 4 x i1> %m, i32 %n) {
@@ -279,12 +303,16 @@ define void @test_vp_cmp_v8(<8 x i32> %i0, <8 x i32> %i1, <8 x float> %f0, <8 x
; LEGAL_LEGAL-NEXT: %r4 = call <vscale x 4 x i32> @llvm.vp.srem.nxv4i32(<vscale x 4 x i32> %i0, <vscale x 4 x i32> %i1, <vscale x 4 x i1> %m, i32 %n)
; LEGAL_LEGAL-NEXT: %r5 = call <vscale x 4 x i32> @llvm.vp.udiv.nxv4i32(<vscale x 4 x i32> %i0, <vscale x 4 x i32> %i1, <vscale x 4 x i1> %m, i32 %n)
; LEGAL_LEGAL-NEXT: %r6 = call <vscale x 4 x i32> @llvm.vp.urem.nxv4i32(<vscale x 4 x i32> %i0, <vscale x 4 x i32> %i1, <vscale x 4 x i1> %m, i32 %n)
-; LEGAL_LEGAL-NEXT: %r7 = call <vscale x 4 x i32> @llvm.vp.and.nxv4i32(<vscale x 4 x i32> %i0, <vscale x 4 x i32> %i1, <vscale x 4 x i1> %m, i32 %n)
-; LEGAL_LEGAL-NEXT: %r8 = call <vscale x 4 x i32> @llvm.vp.or.nxv4i32(<vscale x 4 x i32> %i0, <vscale x 4 x i32> %i1, <vscale x 4 x i1> %m, i32 %n)
-; LEGAL_LEGAL-NEXT: %r9 = call <vscale x 4 x i32> @llvm.vp.xor.nxv4i32(<vscale x 4 x i32> %i0, <vscale x 4 x i32> %i1, <vscale x 4 x i1> %m, i32 %n)
-; LEGAL_LEGAL-NEXT: %rA = call <vscale x 4 x i32> @llvm.vp.ashr.nxv4i32(<vscale x 4 x i32> %i0, <vscale x 4 x i32> %i1, <vscale x 4 x i1> %m, i32 %n)
-; LEGAL_LEGAL-NEXT: %rB = call <vscale x 4 x i32> @llvm.vp.lshr.nxv4i32(<vscale x 4 x i32> %i0, <vscale x 4 x i32> %i1, <vscale x 4 x i1> %m, i32 %n)
-; LEGAL_LEGAL-NEXT: %rC = call <vscale x 4 x i32> @llvm.vp.shl.nxv4i32(<vscale x 4 x i32> %i0, <vscale x 4 x i32> %i1, <vscale x 4 x i1> %m, i32 %n)
+; LEGAL_LEGAL-NEXT: %r7 = call <vscale x 4 x i32> @llvm.vp.smax.nxv4i32(<vscale x 4 x i32> %i0, <vscale x 4 x i32> %i1, <vscale x 4 x i1> %m, i32 %n)
+; LEGAL_LEGAL-NEXT: %r8 = call <vscale x 4 x i32> @llvm.vp.smin.nxv4i32(<vscale x 4 x i32> %i0, <vscale x 4 x i32> %i1, <vscale x 4 x i1> %m, i32 %n)
+; LEGAL_LEGAL-NEXT: %r9 = call <vscale x 4 x i32> @llvm.vp.umax.nxv4i32(<vscale x 4 x i32> %i0, <vscale x 4 x i32> %i1, <vscale x 4 x i1> %m, i32 %n)
+; LEGAL_LEGAL-NEXT: %rA = call <vscale x 4 x i32> @llvm.vp.umin.nxv4i32(<vscale x 4 x i32> %i0, <vscale x 4 x i32> %i1, <vscale x 4 x i1> %m, i32 %n)
+; LEGAL_LEGAL-NEXT: %rB = call <vscale x 4 x i32> @llvm.vp.and.nxv4i32(<vscale x 4 x i32> %i0, <vscale x 4 x i32> %i1, <vscale x 4 x i1> %m, i32 %n)
+; LEGAL_LEGAL-NEXT: %rC = call <vscale x 4 x i32> @llvm.vp.or.nxv4i32(<vscale x 4 x i32> %i0, <vscale x 4 x i32> %i1, <vscale x 4 x i1> %m, i32 %n)
+; LEGAL_LEGAL-NEXT: %rD = call <vscale x 4 x i32> @llvm.vp.xor.nxv4i32(<vscale x 4 x i32> %i0, <vscale x 4 x i32> %i1, <vscale x 4 x i1> %m, i32 %n)
+; LEGAL_LEGAL-NEXT: %rE = call <vscale x 4 x i32> @llvm.vp.ashr.nxv4i32(<vscale x 4 x i32> %i0, <vscale x 4 x i32> %i1, <vscale x 4 x i1> %m, i32 %n)
+; LEGAL_LEGAL-NEXT: %rF = call <vscale x 4 x i32> @llvm.vp.lshr.nxv4i32(<vscale x 4 x i32> %i0, <vscale x 4 x i32> %i1, <vscale x 4 x i1> %m, i32 %n)
+; LEGAL_LEGAL-NEXT: %r10 = call <vscale x 4 x i32> @llvm.vp.shl.nxv4i32(<vscale x 4 x i32> %i0, <vscale x 4 x i32> %i1, <vscale x 4 x i1> %m, i32 %n)
; LEGAL_LEGAL-NEXT: ret void
; LEGAL_LEGAL: define void @test_vp_reduce_int_v4(i32 %start, <4 x i32> %vi, <4 x i1> %m, i32 %n) {
@@ -342,12 +370,16 @@ define void @test_vp_cmp_v8(<8 x i32> %i0, <8 x i32> %i1, <8 x float> %f0, <8 x
; DISCARD_LEGAL-NOT: %r4 = call <8 x i32> @llvm.vp.srem.v8i32(<8 x i32> %i0, <8 x i32> %i1, <8 x i1> %m, i32 8)
; DISCARD_LEGAL-NOT: %r5 = call <8 x i32> @llvm.vp.udiv.v8i32(<8 x i32> %i0, <8 x i32> %i1, <8 x i1> %m, i32 8)
; DISCARD_LEGAL-NOT: %r6 = call <8 x i32> @llvm.vp.urem.v8i32(<8 x i32> %i0, <8 x i32> %i1, <8 x i1> %m, i32 8)
-; DISCARD_LEGAL: %r7 = call <8 x i32> @llvm.vp.and.v8i32(<8 x i32> %i0, <8 x i32> %i1, <8 x i1> %m, i32 8)
-; DISCARD_LEGAL-NEXT: %r8 = call <8 x i32> @llvm.vp.or.v8i32(<8 x i32> %i0, <8 x i32> %i1, <8 x i1> %m, i32 8)
-; DISCARD_LEGAL-NEXT: %r9 = call <8 x i32> @llvm.vp.xor.v8i32(<8 x i32> %i0, <8 x i32> %i1, <8 x i1> %m, i32 8)
-; DISCARD_LEGAL-NEXT: %rA = call <8 x i32> @llvm.vp.ashr.v8i32(<8 x i32> %i0, <8 x i32> %i1, <8 x i1> %m, i32 8)
-; DISCARD_LEGAL-NEXT: %rB = call <8 x i32> @llvm.vp.lshr.v8i32(<8 x i32> %i0, <8 x i32> %i1, <8 x i1> %m, i32 8)
-; DISCARD_LEGAL-NEXT: %rC = call <8 x i32> @llvm.vp.shl.v8i32(<8 x i32> %i0, <8 x i32> %i1, <8 x i1> %m, i32 8)
+; DISCARD_LEGAL: %r7 = call <8 x i32> @llvm.vp.smax.v8i32(<8 x i32> %i0, <8 x i32> %i1, <8 x i1> %m, i32 8)
+; DISCARD_LEGAL: %r8 = call <8 x i32> @llvm.vp.smin.v8i32(<8 x i32> %i0, <8 x i32> %i1, <8 x i1> %m, i32 8)
+; DISCARD_LEGAL: %r9 = call <8 x i32> @llvm.vp.umax.v8i32(<8 x i32> %i0, <8 x i32> %i1, <8 x i1> %m, i32 8)
+; DISCARD_LEGAL: %rA = call <8 x i32> @llvm.vp.umin.v8i32(<8 x i32> %i0, <8 x i32> %i1, <8 x i1> %m, i32 8)
+; DISCARD_LEGAL-NEXT: %rB = call <8 x i32> @llvm.vp.and.v8i32(<8 x i32> %i0, <8 x i32> %i1, <8 x i1> %m, i32 8)
+; DISCARD_LEGAL-NEXT: %rC = call <8 x i32> @llvm.vp.or.v8i32(<8 x i32> %i0, <8 x i32> %i1, <8 x i1> %m, i32 8)
+; DISCARD_LEGAL-NEXT: %rD = call <8 x i32> @llvm.vp.xor.v8i32(<8 x i32> %i0, <8 x i32> %i1, <8 x i1> %m, i32 8)
+; DISCARD_LEGAL-NEXT: %rE = call <8 x i32> @llvm.vp.ashr.v8i32(<8 x i32> %i0, <8 x i32> %i1, <8 x i1> %m, i32 8)
+; DISCARD_LEGAL-NEXT: %rF = call <8 x i32> @llvm.vp.lshr.v8i32(<8 x i32> %i0, <8 x i32> %i1, <8 x i1> %m, i32 8)
+; DISCARD_LEGAL-NEXT: %r10 = call <8 x i32> @llvm.vp.shl.v8i32(<8 x i32> %i0, <8 x i32> %i1, <8 x i1> %m, i32 8)
; DISCARD_LEGAL-NEXT: ret void
; TODO compute vscale only once and use caching.
More information about the llvm-commits
mailing list