[llvm] [RISCV][TTI] Add vp.fneg intrinsic cost with functionalOP (PR #114378)
via llvm-commits
llvm-commits at lists.llvm.org
Sun Nov 3 21:02:23 PST 2024
https://github.com/LiqinWeng updated https://github.com/llvm/llvm-project/pull/114378
>From db372706eb94ca16a129302bd9f49a949ddfff9b Mon Sep 17 00:00:00 2001
From: LiqinWeng <liqin.weng at spacemit.com>
Date: Fri, 1 Nov 2024 11:10:24 +0800
Subject: [PATCH 1/2] [RISCV][TTI] Add vp.fneg intrinsic cost with functionalOP
---
.../Target/RISCV/RISCVTargetTransformInfo.cpp | 3 +-
.../CostModel/RISCV/rvv-intrinsics.ll | 63 +++++++++++++++++++
2 files changed, 65 insertions(+), 1 deletion(-)
diff --git a/llvm/lib/Target/RISCV/RISCVTargetTransformInfo.cpp b/llvm/lib/Target/RISCV/RISCVTargetTransformInfo.cpp
index b84c94f718b287..c92e85cd462834 100644
--- a/llvm/lib/Target/RISCV/RISCVTargetTransformInfo.cpp
+++ b/llvm/lib/Target/RISCV/RISCVTargetTransformInfo.cpp
@@ -1123,7 +1123,8 @@ RISCVTTIImpl::getIntrinsicInstrCost(const IntrinsicCostAttributes &ICA,
case Intrinsic::vp_fsub:
case Intrinsic::vp_fmul:
case Intrinsic::vp_fdiv:
- case Intrinsic::vp_frem: {
+ case Intrinsic::vp_frem:
+ case Intrinsic::vp_fneg: {
std::optional<unsigned> FOp =
VPIntrinsic::getFunctionalOpcodeForVP(ICA.getID());
assert(FOp.has_value());
diff --git a/llvm/test/Analysis/CostModel/RISCV/rvv-intrinsics.ll b/llvm/test/Analysis/CostModel/RISCV/rvv-intrinsics.ll
index bb98508f239c1b..d061cc9bcdc843 100644
--- a/llvm/test/Analysis/CostModel/RISCV/rvv-intrinsics.ll
+++ b/llvm/test/Analysis/CostModel/RISCV/rvv-intrinsics.ll
@@ -1501,6 +1501,68 @@ define void @vp_fadd(){
}
+define void @vp_fneg() {
+; CHECK-LABEL: 'vp_fneg'
+; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %t0 = call <2 x float> @llvm.vp.fneg.v2f32(<2 x float> undef, <2 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %t2 = call <4 x float> @llvm.vp.fneg.v4f32(<4 x float> undef, <4 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %t4 = call <8 x float> @llvm.vp.fneg.v8f32(<8 x float> undef, <8 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %t6 = call <16 x float> @llvm.vp.fneg.v16f32(<16 x float> undef, <16 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %t8 = call <2 x double> @llvm.vp.fneg.v2f64(<2 x double> undef, <2 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %t10 = call <4 x double> @llvm.vp.fneg.v4f64(<4 x double> undef, <4 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %t12 = call <8 x double> @llvm.vp.fneg.v8f64(<8 x double> undef, <8 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 16 for instruction: %t14 = call <16 x double> @llvm.vp.fneg.v16f64(<16 x double> undef, <16 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %t17 = call <vscale x 2 x float> @llvm.vp.fneg.nxv2f32(<vscale x 2 x float> undef, <vscale x 2 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %t19 = call <vscale x 4 x float> @llvm.vp.fneg.nxv4f32(<vscale x 4 x float> undef, <vscale x 4 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %t21 = call <vscale x 8 x float> @llvm.vp.fneg.nxv8f32(<vscale x 8 x float> undef, <vscale x 8 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 16 for instruction: %t23 = call <vscale x 16 x float> @llvm.vp.fneg.nxv16f32(<vscale x 16 x float> undef, <vscale x 16 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %t25 = call <vscale x 2 x double> @llvm.vp.fneg.nxv2f64(<vscale x 2 x double> undef, <vscale x 2 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %t27 = call <vscale x 4 x double> @llvm.vp.fneg.nxv4f64(<vscale x 4 x double> undef, <vscale x 4 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 16 for instruction: %t29 = call <vscale x 8 x double> @llvm.vp.fneg.nxv8f64(<vscale x 8 x double> undef, <vscale x 8 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 32 for instruction: %t31 = call <vscale x 16 x double> @llvm.vp.fneg.nxv16f64(<vscale x 16 x double> undef, <vscale x 16 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret void
+;
+; TYPEBASED-LABEL: 'vp_fneg'
+; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %t0 = call <2 x float> @llvm.vp.fneg.v2f32(<2 x float> undef, <2 x i1> undef, i32 undef)
+; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %t2 = call <4 x float> @llvm.vp.fneg.v4f32(<4 x float> undef, <4 x i1> undef, i32 undef)
+; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %t4 = call <8 x float> @llvm.vp.fneg.v8f32(<8 x float> undef, <8 x i1> undef, i32 undef)
+; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %t6 = call <16 x float> @llvm.vp.fneg.v16f32(<16 x float> undef, <16 x i1> undef, i32 undef)
+; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %t8 = call <2 x double> @llvm.vp.fneg.v2f64(<2 x double> undef, <2 x i1> undef, i32 undef)
+; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %t10 = call <4 x double> @llvm.vp.fneg.v4f64(<4 x double> undef, <4 x i1> undef, i32 undef)
+; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %t12 = call <8 x double> @llvm.vp.fneg.v8f64(<8 x double> undef, <8 x i1> undef, i32 undef)
+; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 16 for instruction: %t14 = call <16 x double> @llvm.vp.fneg.v16f64(<16 x double> undef, <16 x i1> undef, i32 undef)
+; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %t17 = call <vscale x 2 x float> @llvm.vp.fneg.nxv2f32(<vscale x 2 x float> undef, <vscale x 2 x i1> undef, i32 undef)
+; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %t19 = call <vscale x 4 x float> @llvm.vp.fneg.nxv4f32(<vscale x 4 x float> undef, <vscale x 4 x i1> undef, i32 undef)
+; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %t21 = call <vscale x 8 x float> @llvm.vp.fneg.nxv8f32(<vscale x 8 x float> undef, <vscale x 8 x i1> undef, i32 undef)
+; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 16 for instruction: %t23 = call <vscale x 16 x float> @llvm.vp.fneg.nxv16f32(<vscale x 16 x float> undef, <vscale x 16 x i1> undef, i32 undef)
+; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %t25 = call <vscale x 2 x double> @llvm.vp.fneg.nxv2f64(<vscale x 2 x double> undef, <vscale x 2 x i1> undef, i32 undef)
+; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %t27 = call <vscale x 4 x double> @llvm.vp.fneg.nxv4f64(<vscale x 4 x double> undef, <vscale x 4 x i1> undef, i32 undef)
+; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 16 for instruction: %t29 = call <vscale x 8 x double> @llvm.vp.fneg.nxv8f64(<vscale x 8 x double> undef, <vscale x 8 x i1> undef, i32 undef)
+; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 32 for instruction: %t31 = call <vscale x 16 x double> @llvm.vp.fneg.nxv16f64(<vscale x 16 x double> undef, <vscale x 16 x i1> undef, i32 undef)
+; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret void
+;
+ %t0 = call <2 x float> @llvm.vp.fneg.v2f32(<2 x float> undef, <2 x i1> undef, i32 undef)
+ %t2 = call <4 x float> @llvm.vp.fneg.v4f32(<4 x float> undef, <4 x i1> undef, i32 undef)
+ %t4 = call <8 x float> @llvm.vp.fneg.v8f32(<8 x float> undef, <8 x i1> undef, i32 undef)
+ %t6 = call <16 x float> @llvm.vp.fneg.v16f32(<16 x float> undef, <16 x i1> undef, i32 undef)
+
+ %t8 = call <2 x double> @llvm.vp.fneg.v2f64(<2 x double> undef, <2 x i1> undef, i32 undef)
+ %t10 = call <4 x double> @llvm.vp.fneg.v4f64(<4 x double> undef, <4 x i1> undef, i32 undef)
+ %t12 = call <8 x double> @llvm.vp.fneg.v8f64(<8 x double> undef, <8 x i1> undef, i32 undef)
+ %t14 = call <16 x double> @llvm.vp.fneg.v16f64(<16 x double> undef, <16 x i1> undef, i32 undef)
+
+ %t17 = call <vscale x 2 x float> @llvm.vp.fneg.nxv2f32(<vscale x 2 x float> undef, <vscale x 2 x i1> undef, i32 undef)
+ %t19 = call <vscale x 4 x float> @llvm.vp.fneg.nxv4f32(<vscale x 4 x float> undef, <vscale x 4 x i1> undef, i32 undef)
+ %t21 = call <vscale x 8 x float> @llvm.vp.fneg.nxv8f32(<vscale x 8 x float> undef, <vscale x 8 x i1> undef, i32 undef)
+ %t23 = call <vscale x 16 x float> @llvm.vp.fneg.nxv16f32(<vscale x 16 x float> undef, <vscale x 16 x i1> undef, i32 undef)
+
+ %t25 = call <vscale x 2 x double> @llvm.vp.fneg.nxv2f64(<vscale x 2 x double> undef, <vscale x 2 x i1> undef, i32 undef)
+ %t27 = call <vscale x 4 x double> @llvm.vp.fneg.nxv4f64(<vscale x 4 x double> undef, <vscale x 4 x i1> undef, i32 undef)
+ %t29 = call <vscale x 8 x double> @llvm.vp.fneg.nxv8f64(<vscale x 8 x double> undef, <vscale x 8 x i1> undef, i32 undef)
+ %t31 = call <vscale x 16 x double> @llvm.vp.fneg.nxv16f64(<vscale x 16 x double> undef, <vscale x 16 x i1> undef, i32 undef)
+
+ ret void
+}
+
declare <2 x i8> @llvm.vp.add.v2i8(<2 x i8>, <2 x i8>, <2 x i1>, i32)
declare <4 x i8> @llvm.vp.add.v4i8(<4 x i8>, <4 x i8>, <4 x i1>, i32)
declare <8 x i8> @llvm.vp.add.v8i8(<8 x i8>, <8 x i8>, <8 x i1>, i32)
@@ -1772,3 +1834,4 @@ declare <vscale x 1 x i64> @llvm.vp.fshl.nxv1i64(<vscale x 1 x i64>, <vscale x 1
declare <vscale x 2 x i64> @llvm.vp.fshl.nxv2i64(<vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i64>, <vscale x 2 x i1>, i32)
declare <vscale x 4 x i64> @llvm.vp.fshl.nxv4i64(<vscale x 4 x i64>, <vscale x 4 x i64>, <vscale x 4 x i64>, <vscale x 4 x i1>, i32)
declare <vscale x 8 x i64> @llvm.vp.fshl.nxv8i64(<vscale x 8 x i64>, <vscale x 8 x i64>, <vscale x 8 x i64>, <vscale x 8 x i1>, i32)
+
>From 87a78b1ee37ad079d22dfd623bd210df397fbd27 Mon Sep 17 00:00:00 2001
From: "Liqin.Weng" <liqin.weng at spacemit.com>
Date: Mon, 4 Nov 2024 13:01:48 +0800
Subject: [PATCH 2/2] fix the comments
---
.../test/Analysis/CostModel/RISCV/arith-fp.ll | 76 ++++++++++++++++---
.../CostModel/RISCV/rvv-intrinsics.ll | 62 ---------------
2 files changed, 65 insertions(+), 73 deletions(-)
diff --git a/llvm/test/Analysis/CostModel/RISCV/arith-fp.ll b/llvm/test/Analysis/CostModel/RISCV/arith-fp.ll
index b6b49982a73267..75cb3aa7bdb083 100644
--- a/llvm/test/Analysis/CostModel/RISCV/arith-fp.ll
+++ b/llvm/test/Analysis/CostModel/RISCV/arith-fp.ll
@@ -1,6 +1,6 @@
; NOTE: Assertions have been autogenerated by utils/update_analyze_test_checks.py
-; RUN: opt -passes="print<cost-model>" 2>&1 -disable-output -mtriple=riscv64 -mattr=+v,+f,+d,+zfh,+zvfh,+zvfbfmin < %s | FileCheck %s --check-prefixes=CHECK,ZVFH
-; RUN: opt -passes="print<cost-model>" 2>&1 -disable-output -mtriple=riscv64 -mattr=+v,+f,+d,+zfh,+zvfhmin,+zvfbfmin < %s | FileCheck %s --check-prefixes=CHECK,ZVFHMIN
+; RUN: opt -passes="print<cost-model>" 2>&1 -disable-output -mtriple=riscv64 -mattr=+v,+f,+d,+zfh,+zvfh,+zvfbfmin --type-based-intrinsic-cost=true < %s | FileCheck %s --check-prefixes=CHECK,ZVFH
+; RUN: opt -passes="print<cost-model>" 2>&1 -disable-output -mtriple=riscv64 -mattr=+v,+f,+d,+zfh,+zvfhmin,+zvfbfmin --type-based-intrinsic-cost=true < %s | FileCheck %s --check-prefixes=CHECK,ZVFHMIN
; Check that we don't crash querying costs when vectors are not enabled.
; RUN: opt -passes="print<cost-model>" 2>&1 -disable-output -mtriple=riscv64
@@ -661,6 +661,30 @@ define void @fneg() {
; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %NXV2F64 = fneg <vscale x 2 x double> undef
; CHECK-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %NXV4F64 = fneg <vscale x 4 x double> undef
; CHECK-NEXT: Cost Model: Found an estimated cost of 16 for instruction: %NXV8F64 = fneg <vscale x 8 x double> undef
+; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %V1BF16_VP = call <2 x bfloat> @llvm.vp.fneg.v2bf16(<2 x bfloat> undef, <2 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %V2BF16_VP = call <4 x bfloat> @llvm.vp.fneg.v4bf16(<4 x bfloat> undef, <4 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %V4BF16_VP = call <8 x bfloat> @llvm.vp.fneg.v8bf16(<8 x bfloat> undef, <8 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V8BF16_VP = call <16 x bfloat> @llvm.vp.fneg.v16bf16(<16 x bfloat> undef, <16 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %V1F32_VP = call <2 x float> @llvm.vp.fneg.v2f32(<2 x float> undef, <2 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %V2F32_VP = call <4 x float> @llvm.vp.fneg.v4f32(<4 x float> undef, <4 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V4F32_VP = call <8 x float> @llvm.vp.fneg.v8f32(<8 x float> undef, <8 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %V8F32_VP = call <16 x float> @llvm.vp.fneg.v16f32(<16 x float> undef, <16 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %V1F64_VP = call <2 x double> @llvm.vp.fneg.v2f64(<2 x double> undef, <2 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V2F64_VP = call <4 x double> @llvm.vp.fneg.v4f64(<4 x double> undef, <4 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %V4F64_VP = call <8 x double> @llvm.vp.fneg.v8f64(<8 x double> undef, <8 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 16 for instruction: %V8F64_VP = call <16 x double> @llvm.vp.fneg.v16f64(<16 x double> undef, <16 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %NXV1BF16_VP = call <vscale x 2 x bfloat> @llvm.vp.fneg.nxv2bf16(<vscale x 2 x bfloat> undef, <vscale x 2 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %NXV2BF16_VP = call <vscale x 4 x bfloat> @llvm.vp.fneg.nxv4bf16(<vscale x 4 x bfloat> undef, <vscale x 4 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %NXV4BF16_VP = call <vscale x 8 x bfloat> @llvm.vp.fneg.nxv8bf16(<vscale x 8 x bfloat> undef, <vscale x 8 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %NXV8BF16_VP = call <vscale x 16 x bfloat> @llvm.vp.fneg.nxv16bf16(<vscale x 16 x bfloat> undef, <vscale x 16 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %NXV1BF32_VP = call <vscale x 2 x float> @llvm.vp.fneg.nxv2f32(<vscale x 2 x float> undef, <vscale x 2 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %NXV2BF32_VP = call <vscale x 4 x float> @llvm.vp.fneg.nxv4f32(<vscale x 4 x float> undef, <vscale x 4 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %NXV4BF32_VP = call <vscale x 8 x float> @llvm.vp.fneg.nxv8f32(<vscale x 8 x float> undef, <vscale x 8 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 16 for instruction: %NXV8BF32_VP = call <vscale x 16 x float> @llvm.vp.fneg.nxv16f32(<vscale x 16 x float> undef, <vscale x 16 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %NXV1BF64_VP = call <vscale x 2 x double> @llvm.vp.fneg.nxv2f64(<vscale x 2 x double> undef, <vscale x 2 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %NXV2BF64_VP = call <vscale x 4 x double> @llvm.vp.fneg.nxv4f64(<vscale x 4 x double> undef, <vscale x 4 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 16 for instruction: %NXV4BF64_VP = call <vscale x 8 x double> @llvm.vp.fneg.nxv8f64(<vscale x 8 x double> undef, <vscale x 8 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 32 for instruction: %NXV8BF64_VP = call <vscale x 16 x double> @llvm.vp.fneg.nxv16f64(<vscale x 16 x double> undef, <vscale x 16 x i1> undef, i32 undef)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret void
;
%BF16 = fneg half undef
@@ -701,6 +725,36 @@ define void @fneg() {
%NXV4F64 = fneg <vscale x 4 x double> undef
%NXV8F64 = fneg <vscale x 8 x double> undef
+ %V1BF16_VP = call <2 x bfloat> @llvm.vp.fneg.v2f16(<2 x bfloat> undef, <2 x i1> undef, i32 undef)
+ %V2BF16_VP = call <4 x bfloat> @llvm.vp.fneg.v4f16(<4 x bfloat> undef, <4 x i1> undef, i32 undef)
+ %V4BF16_VP = call <8 x bfloat> @llvm.vp.fneg.v8f16(<8 x bfloat> undef, <8 x i1> undef, i32 undef)
+ %V8BF16_VP = call <16 x bfloat> @llvm.vp.fneg.v16f16(<16 x bfloat> undef, <16 x i1> undef, i32 undef)
+
+ %V1F32_VP = call <2 x float> @llvm.vp.fneg.v2f32(<2 x float> undef, <2 x i1> undef, i32 undef)
+ %V2F32_VP = call <4 x float> @llvm.vp.fneg.v4f32(<4 x float> undef, <4 x i1> undef, i32 undef)
+ %V4F32_VP = call <8 x float> @llvm.vp.fneg.v8f32(<8 x float> undef, <8 x i1> undef, i32 undef)
+ %V8F32_VP = call <16 x float> @llvm.vp.fneg.v16f32(<16 x float> undef, <16 x i1> undef, i32 undef)
+
+ %V1F64_VP = call <2 x double> @llvm.vp.fneg.v2f64(<2 x double> undef, <2 x i1> undef, i32 undef)
+ %V2F64_VP = call <4 x double> @llvm.vp.fneg.v4f64(<4 x double> undef, <4 x i1> undef, i32 undef)
+ %V4F64_VP = call <8 x double> @llvm.vp.fneg.v8f64(<8 x double> undef, <8 x i1> undef, i32 undef)
+ %V8F64_VP = call <16 x double> @llvm.vp.fneg.v16f64(<16 x double> undef, <16 x i1> undef, i32 undef)
+
+ %NXV1BF16_VP = call <vscale x 2 x bfloat> @llvm.vp.fneg.nxv2f16(<vscale x 2 x bfloat> undef, <vscale x 2 x i1> undef, i32 undef)
+ %NXV2BF16_VP = call <vscale x 4 x bfloat> @llvm.vp.fneg.nxv4f16(<vscale x 4 x bfloat> undef, <vscale x 4 x i1> undef, i32 undef)
+ %NXV4BF16_VP = call <vscale x 8 x bfloat> @llvm.vp.fneg.nxv8f16(<vscale x 8 x bfloat> undef, <vscale x 8 x i1> undef, i32 undef)
+ %NXV8BF16_VP = call <vscale x 16 x bfloat> @llvm.vp.fneg.nxv16f16(<vscale x 16 x bfloat> undef, <vscale x 16 x i1> undef, i32 undef)
+
+ %NXV1BF32_VP = call <vscale x 2 x float> @llvm.vp.fneg.nxv2f32(<vscale x 2 x float> undef, <vscale x 2 x i1> undef, i32 undef)
+ %NXV2BF32_VP = call <vscale x 4 x float> @llvm.vp.fneg.nxv4f32(<vscale x 4 x float> undef, <vscale x 4 x i1> undef, i32 undef)
+ %NXV4BF32_VP = call <vscale x 8 x float> @llvm.vp.fneg.nxv8f32(<vscale x 8 x float> undef, <vscale x 8 x i1> undef, i32 undef)
+ %NXV8BF32_VP = call <vscale x 16 x float> @llvm.vp.fneg.nxv16f32(<vscale x 16 x float> undef, <vscale x 16 x i1> undef, i32 undef)
+
+ %NXV1BF64_VP = call <vscale x 2 x double> @llvm.vp.fneg.nxv2f64(<vscale x 2 x double> undef, <vscale x 2 x i1> undef, i32 undef)
+ %NXV2BF64_VP = call <vscale x 4 x double> @llvm.vp.fneg.nxv4f64(<vscale x 4 x double> undef, <vscale x 4 x i1> undef, i32 undef)
+ %NXV4BF64_VP = call <vscale x 8 x double> @llvm.vp.fneg.nxv8f64(<vscale x 8 x double> undef, <vscale x 8 x i1> undef, i32 undef)
+ %NXV8BF64_VP = call <vscale x 16 x double> @llvm.vp.fneg.nxv16f64(<vscale x 16 x double> undef, <vscale x 16 x i1> undef, i32 undef)
+
ret void
}
@@ -746,10 +800,10 @@ define void @fcopysign() {
; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %F32 = call float @llvm.copysign.f32(float undef, float undef)
; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %F64 = call double @llvm.copysign.f64(double undef, double undef)
; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V1BF16 = call <1 x bfloat> @llvm.copysign.v1bf16(<1 x bfloat> undef, <1 x bfloat> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V2BF16 = call <2 x bfloat> @llvm.copysign.v2bf16(<2 x bfloat> undef, <2 x bfloat> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %V4BF16 = call <4 x bfloat> @llvm.copysign.v4bf16(<4 x bfloat> undef, <4 x bfloat> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 16 for instruction: %V8BF16 = call <8 x bfloat> @llvm.copysign.v8bf16(<8 x bfloat> undef, <8 x bfloat> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 47 for instruction: %V16BF16 = call <16 x bfloat> @llvm.copysign.v16bf16(<16 x bfloat> undef, <16 x bfloat> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %V2BF16 = call <2 x bfloat> @llvm.copysign.v2bf16(<2 x bfloat> undef, <2 x bfloat> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 22 for instruction: %V4BF16 = call <4 x bfloat> @llvm.copysign.v4bf16(<4 x bfloat> undef, <4 x bfloat> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 46 for instruction: %V8BF16 = call <8 x bfloat> @llvm.copysign.v8bf16(<8 x bfloat> undef, <8 x bfloat> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 109 for instruction: %V16BF16 = call <16 x bfloat> @llvm.copysign.v16bf16(<16 x bfloat> undef, <16 x bfloat> undef)
; CHECK-NEXT: Cost Model: Invalid cost for instruction: %NXV1BF16 = call <vscale x 1 x bfloat> @llvm.copysign.nxv1bf16(<vscale x 1 x bfloat> undef, <vscale x 1 x bfloat> undef)
; CHECK-NEXT: Cost Model: Invalid cost for instruction: %NXV2BF16 = call <vscale x 2 x bfloat> @llvm.copysign.nxv2bf16(<vscale x 2 x bfloat> undef, <vscale x 2 x bfloat> undef)
; CHECK-NEXT: Cost Model: Invalid cost for instruction: %NXV4BF16 = call <vscale x 4 x bfloat> @llvm.copysign.nxv4bf16(<vscale x 4 x bfloat> undef, <vscale x 4 x bfloat> undef)
@@ -836,11 +890,11 @@ define void @fcopysign_f16() {
; ZVFHMIN-LABEL: 'fcopysign_f16'
; ZVFHMIN-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %F16 = call half @llvm.copysign.f16(half undef, half undef)
; ZVFHMIN-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V1F16 = call <1 x half> @llvm.copysign.v1f16(<1 x half> undef, <1 x half> undef)
-; ZVFHMIN-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V2F16 = call <2 x half> @llvm.copysign.v2f16(<2 x half> undef, <2 x half> undef)
-; ZVFHMIN-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %V4F16 = call <4 x half> @llvm.copysign.v4f16(<4 x half> undef, <4 x half> undef)
-; ZVFHMIN-NEXT: Cost Model: Found an estimated cost of 16 for instruction: %V8F16 = call <8 x half> @llvm.copysign.v8f16(<8 x half> undef, <8 x half> undef)
-; ZVFHMIN-NEXT: Cost Model: Found an estimated cost of 47 for instruction: %V16F16 = call <16 x half> @llvm.copysign.v16f16(<16 x half> undef, <16 x half> undef)
-; ZVFHMIN-NEXT: Cost Model: Found an estimated cost of 95 for instruction: %V32F16 = call <32 x half> @llvm.copysign.v32f16(<32 x half> undef, <32 x half> undef)
+; ZVFHMIN-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %V2F16 = call <2 x half> @llvm.copysign.v2f16(<2 x half> undef, <2 x half> undef)
+; ZVFHMIN-NEXT: Cost Model: Found an estimated cost of 22 for instruction: %V4F16 = call <4 x half> @llvm.copysign.v4f16(<4 x half> undef, <4 x half> undef)
+; ZVFHMIN-NEXT: Cost Model: Found an estimated cost of 46 for instruction: %V8F16 = call <8 x half> @llvm.copysign.v8f16(<8 x half> undef, <8 x half> undef)
+; ZVFHMIN-NEXT: Cost Model: Found an estimated cost of 109 for instruction: %V16F16 = call <16 x half> @llvm.copysign.v16f16(<16 x half> undef, <16 x half> undef)
+; ZVFHMIN-NEXT: Cost Model: Found an estimated cost of 221 for instruction: %V32F16 = call <32 x half> @llvm.copysign.v32f16(<32 x half> undef, <32 x half> undef)
; ZVFHMIN-NEXT: Cost Model: Invalid cost for instruction: %NXV1F16 = call <vscale x 1 x half> @llvm.copysign.nxv1f16(<vscale x 1 x half> undef, <vscale x 1 x half> undef)
; ZVFHMIN-NEXT: Cost Model: Invalid cost for instruction: %NXV2F16 = call <vscale x 2 x half> @llvm.copysign.nxv2f16(<vscale x 2 x half> undef, <vscale x 2 x half> undef)
; ZVFHMIN-NEXT: Cost Model: Invalid cost for instruction: %NXV4F16 = call <vscale x 4 x half> @llvm.copysign.nxv4f16(<vscale x 4 x half> undef, <vscale x 4 x half> undef)
diff --git a/llvm/test/Analysis/CostModel/RISCV/rvv-intrinsics.ll b/llvm/test/Analysis/CostModel/RISCV/rvv-intrinsics.ll
index d061cc9bcdc843..a8af70682b6df3 100644
--- a/llvm/test/Analysis/CostModel/RISCV/rvv-intrinsics.ll
+++ b/llvm/test/Analysis/CostModel/RISCV/rvv-intrinsics.ll
@@ -1501,68 +1501,6 @@ define void @vp_fadd(){
}
-define void @vp_fneg() {
-; CHECK-LABEL: 'vp_fneg'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %t0 = call <2 x float> @llvm.vp.fneg.v2f32(<2 x float> undef, <2 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %t2 = call <4 x float> @llvm.vp.fneg.v4f32(<4 x float> undef, <4 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %t4 = call <8 x float> @llvm.vp.fneg.v8f32(<8 x float> undef, <8 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %t6 = call <16 x float> @llvm.vp.fneg.v16f32(<16 x float> undef, <16 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %t8 = call <2 x double> @llvm.vp.fneg.v2f64(<2 x double> undef, <2 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %t10 = call <4 x double> @llvm.vp.fneg.v4f64(<4 x double> undef, <4 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %t12 = call <8 x double> @llvm.vp.fneg.v8f64(<8 x double> undef, <8 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 16 for instruction: %t14 = call <16 x double> @llvm.vp.fneg.v16f64(<16 x double> undef, <16 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %t17 = call <vscale x 2 x float> @llvm.vp.fneg.nxv2f32(<vscale x 2 x float> undef, <vscale x 2 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %t19 = call <vscale x 4 x float> @llvm.vp.fneg.nxv4f32(<vscale x 4 x float> undef, <vscale x 4 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %t21 = call <vscale x 8 x float> @llvm.vp.fneg.nxv8f32(<vscale x 8 x float> undef, <vscale x 8 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 16 for instruction: %t23 = call <vscale x 16 x float> @llvm.vp.fneg.nxv16f32(<vscale x 16 x float> undef, <vscale x 16 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %t25 = call <vscale x 2 x double> @llvm.vp.fneg.nxv2f64(<vscale x 2 x double> undef, <vscale x 2 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %t27 = call <vscale x 4 x double> @llvm.vp.fneg.nxv4f64(<vscale x 4 x double> undef, <vscale x 4 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 16 for instruction: %t29 = call <vscale x 8 x double> @llvm.vp.fneg.nxv8f64(<vscale x 8 x double> undef, <vscale x 8 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 32 for instruction: %t31 = call <vscale x 16 x double> @llvm.vp.fneg.nxv16f64(<vscale x 16 x double> undef, <vscale x 16 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret void
-;
-; TYPEBASED-LABEL: 'vp_fneg'
-; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %t0 = call <2 x float> @llvm.vp.fneg.v2f32(<2 x float> undef, <2 x i1> undef, i32 undef)
-; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %t2 = call <4 x float> @llvm.vp.fneg.v4f32(<4 x float> undef, <4 x i1> undef, i32 undef)
-; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %t4 = call <8 x float> @llvm.vp.fneg.v8f32(<8 x float> undef, <8 x i1> undef, i32 undef)
-; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %t6 = call <16 x float> @llvm.vp.fneg.v16f32(<16 x float> undef, <16 x i1> undef, i32 undef)
-; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %t8 = call <2 x double> @llvm.vp.fneg.v2f64(<2 x double> undef, <2 x i1> undef, i32 undef)
-; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %t10 = call <4 x double> @llvm.vp.fneg.v4f64(<4 x double> undef, <4 x i1> undef, i32 undef)
-; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %t12 = call <8 x double> @llvm.vp.fneg.v8f64(<8 x double> undef, <8 x i1> undef, i32 undef)
-; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 16 for instruction: %t14 = call <16 x double> @llvm.vp.fneg.v16f64(<16 x double> undef, <16 x i1> undef, i32 undef)
-; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %t17 = call <vscale x 2 x float> @llvm.vp.fneg.nxv2f32(<vscale x 2 x float> undef, <vscale x 2 x i1> undef, i32 undef)
-; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %t19 = call <vscale x 4 x float> @llvm.vp.fneg.nxv4f32(<vscale x 4 x float> undef, <vscale x 4 x i1> undef, i32 undef)
-; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %t21 = call <vscale x 8 x float> @llvm.vp.fneg.nxv8f32(<vscale x 8 x float> undef, <vscale x 8 x i1> undef, i32 undef)
-; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 16 for instruction: %t23 = call <vscale x 16 x float> @llvm.vp.fneg.nxv16f32(<vscale x 16 x float> undef, <vscale x 16 x i1> undef, i32 undef)
-; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %t25 = call <vscale x 2 x double> @llvm.vp.fneg.nxv2f64(<vscale x 2 x double> undef, <vscale x 2 x i1> undef, i32 undef)
-; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %t27 = call <vscale x 4 x double> @llvm.vp.fneg.nxv4f64(<vscale x 4 x double> undef, <vscale x 4 x i1> undef, i32 undef)
-; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 16 for instruction: %t29 = call <vscale x 8 x double> @llvm.vp.fneg.nxv8f64(<vscale x 8 x double> undef, <vscale x 8 x i1> undef, i32 undef)
-; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 32 for instruction: %t31 = call <vscale x 16 x double> @llvm.vp.fneg.nxv16f64(<vscale x 16 x double> undef, <vscale x 16 x i1> undef, i32 undef)
-; TYPEBASED-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret void
-;
- %t0 = call <2 x float> @llvm.vp.fneg.v2f32(<2 x float> undef, <2 x i1> undef, i32 undef)
- %t2 = call <4 x float> @llvm.vp.fneg.v4f32(<4 x float> undef, <4 x i1> undef, i32 undef)
- %t4 = call <8 x float> @llvm.vp.fneg.v8f32(<8 x float> undef, <8 x i1> undef, i32 undef)
- %t6 = call <16 x float> @llvm.vp.fneg.v16f32(<16 x float> undef, <16 x i1> undef, i32 undef)
-
- %t8 = call <2 x double> @llvm.vp.fneg.v2f64(<2 x double> undef, <2 x i1> undef, i32 undef)
- %t10 = call <4 x double> @llvm.vp.fneg.v4f64(<4 x double> undef, <4 x i1> undef, i32 undef)
- %t12 = call <8 x double> @llvm.vp.fneg.v8f64(<8 x double> undef, <8 x i1> undef, i32 undef)
- %t14 = call <16 x double> @llvm.vp.fneg.v16f64(<16 x double> undef, <16 x i1> undef, i32 undef)
-
- %t17 = call <vscale x 2 x float> @llvm.vp.fneg.nxv2f32(<vscale x 2 x float> undef, <vscale x 2 x i1> undef, i32 undef)
- %t19 = call <vscale x 4 x float> @llvm.vp.fneg.nxv4f32(<vscale x 4 x float> undef, <vscale x 4 x i1> undef, i32 undef)
- %t21 = call <vscale x 8 x float> @llvm.vp.fneg.nxv8f32(<vscale x 8 x float> undef, <vscale x 8 x i1> undef, i32 undef)
- %t23 = call <vscale x 16 x float> @llvm.vp.fneg.nxv16f32(<vscale x 16 x float> undef, <vscale x 16 x i1> undef, i32 undef)
-
- %t25 = call <vscale x 2 x double> @llvm.vp.fneg.nxv2f64(<vscale x 2 x double> undef, <vscale x 2 x i1> undef, i32 undef)
- %t27 = call <vscale x 4 x double> @llvm.vp.fneg.nxv4f64(<vscale x 4 x double> undef, <vscale x 4 x i1> undef, i32 undef)
- %t29 = call <vscale x 8 x double> @llvm.vp.fneg.nxv8f64(<vscale x 8 x double> undef, <vscale x 8 x i1> undef, i32 undef)
- %t31 = call <vscale x 16 x double> @llvm.vp.fneg.nxv16f64(<vscale x 16 x double> undef, <vscale x 16 x i1> undef, i32 undef)
-
- ret void
-}
-
declare <2 x i8> @llvm.vp.add.v2i8(<2 x i8>, <2 x i8>, <2 x i1>, i32)
declare <4 x i8> @llvm.vp.add.v4i8(<4 x i8>, <4 x i8>, <4 x i1>, i32)
declare <8 x i8> @llvm.vp.add.v8i8(<8 x i8>, <8 x i8>, <8 x i1>, i32)
More information about the llvm-commits
mailing list