[llvm] [RISCV][CostModel] Add cost for fabs/fsqrt of type bf16/f16 (PR #121685)

via llvm-commits llvm-commits at lists.llvm.org
Sun Jan 5 01:02:40 PST 2025


https://github.com/LiqinWeng created https://github.com/llvm/llvm-project/pull/121685

None

>From f85df2cfb5dc6f0ba07332c00b410e86a9589ed9 Mon Sep 17 00:00:00 2001
From: LiqinWeng <liqin.weng at spacemit.com>
Date: Sun, 5 Jan 2025 16:53:48 +0800
Subject: [PATCH] [RISCV][CostModel] Add cost for fabs/fsqrt of type bf16/f16

---
 .../Target/RISCV/RISCVTargetTransformInfo.cpp | 60 +++++++++---
 .../CostModel/RISCV/fp-min-max-abs.ll         | 50 ++++++----
 .../Analysis/CostModel/RISCV/fp-sqrt-pow.ll   | 98 ++++++++++++-------
 3 files changed, 140 insertions(+), 68 deletions(-)

diff --git a/llvm/lib/Target/RISCV/RISCVTargetTransformInfo.cpp b/llvm/lib/Target/RISCV/RISCVTargetTransformInfo.cpp
index 850d6244affa50..3595c615e1f618 100644
--- a/llvm/lib/Target/RISCV/RISCVTargetTransformInfo.cpp
+++ b/llvm/lib/Target/RISCV/RISCVTargetTransformInfo.cpp
@@ -1035,21 +1035,59 @@ RISCVTTIImpl::getIntrinsicInstrCost(const IntrinsicCostAttributes &ICA,
     }
     break;
   }
-  case Intrinsic::fabs:
+  case Intrinsic::fabs: {
+    auto LT = getTypeLegalizationCost(RetTy);
+    if (ST->hasVInstructions() && LT.second.isVector()) {
+      // lui a0, 8
+      // addi a0, a0, -1
+      // vsetvli a1, zero, e16, m1, ta, ma
+      // vand.vx v8, v8, a0
+      // f16 with zvfhmin and bf16 with zvfhbmin
+      if (LT.second.getVectorElementType() == MVT::bf16 ||
+          (LT.second.getVectorElementType() == MVT::f16 &&
+           !ST->hasVInstructionsF16()))
+        return LT.first * getRISCVInstructionCost(RISCV::VAND_VX, LT.second,
+                                                  CostKind) +
+               2;
+      else
+        return LT.first *
+               getRISCVInstructionCost(RISCV::VFSGNJX_VV, LT.second, CostKind);
+    }
+    break;
+  }
   case Intrinsic::sqrt: {
     auto LT = getTypeLegalizationCost(RetTy);
-    // TODO: add f16/bf16, bf16 with zvfbfmin && f16 with zvfhmin
+    auto NVT = LT.second;
     if (ST->hasVInstructions() && LT.second.isVector()) {
-      unsigned Op;
-      switch (ICA.getID()) {
-      case Intrinsic::fabs:
-        Op = RISCV::VFSGNJX_VV;
-        break;
-      case Intrinsic::sqrt:
-        Op = RISCV::VFSQRT_V;
-        break;
+      SmallVector<unsigned, 3> Opcodes;
+      // f16 with zvfhmin and bf16 with zvfbfmin and the type of nxv32[b]f16
+      // will be spilt.
+      if (LT.second.getVectorElementType() == MVT::bf16) {
+        if (LT.second == MVT::nxv32bf16) {
+          Opcodes = {RISCV::VFWCVTBF16_F_F_V, RISCV::VFWCVTBF16_F_F_V,
+                     RISCV::VFSQRT_V,         RISCV::VFSQRT_V,
+                     RISCV::VFNCVTBF16_F_F_W, RISCV::VFNCVTBF16_F_F_W};
+          NVT = MVT::nxv16f32;
+        } else {
+          Opcodes = {RISCV::VFWCVTBF16_F_F_V, RISCV::VFSQRT_V,
+                     RISCV::VFNCVTBF16_F_F_W};
+          NVT = TLI->getTypeToPromoteTo(ISD::FSQRT, NVT);
+        }
+      } else if (LT.second.getVectorElementType() == MVT::f16 &&
+                 !ST->hasVInstructionsF16()) {
+        if (LT.second == MVT::nxv32f16) {
+          Opcodes = {RISCV::VFWCVT_F_F_V, RISCV::VFWCVT_F_F_V,
+                     RISCV::VFSQRT_V,     RISCV::VFSQRT_V,
+                     RISCV::VFNCVT_F_F_W, RISCV::VFNCVT_F_F_W};
+          NVT = MVT::nxv16f32;
+        } else {
+          Opcodes = {RISCV::VFWCVT_F_F_V, RISCV::VFSQRT_V, RISCV::VFNCVT_F_F_W};
+          NVT = TLI->getTypeToPromoteTo(ISD::FSQRT, NVT);
+        }
+      } else {
+        Opcodes = {RISCV::VFSQRT_V};
       }
-      return LT.first * getRISCVInstructionCost(Op, LT.second, CostKind);
+      return LT.first * getRISCVInstructionCost(Opcodes, NVT, CostKind);
     }
     break;
   }
diff --git a/llvm/test/Analysis/CostModel/RISCV/fp-min-max-abs.ll b/llvm/test/Analysis/CostModel/RISCV/fp-min-max-abs.ll
index 9eb06a07f135fa..8ff70453194381 100644
--- a/llvm/test/Analysis/CostModel/RISCV/fp-min-max-abs.ll
+++ b/llvm/test/Analysis/CostModel/RISCV/fp-min-max-abs.ll
@@ -5,14 +5,14 @@
 define void @fabs() {
 ; CHECK-LABEL: 'fabs'
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %1 = call bfloat @llvm.fabs.bf16(bfloat undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %2 = call <2 x bfloat> @llvm.fabs.v2bf16(<2 x bfloat> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %3 = call <4 x bfloat> @llvm.fabs.v4bf16(<4 x bfloat> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %4 = call <8 x bfloat> @llvm.fabs.v8bf16(<8 x bfloat> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %5 = call <16 x bfloat> @llvm.fabs.v16bf16(<16 x bfloat> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %6 = call <vscale x 2 x bfloat> @llvm.fabs.nxv2bf16(<vscale x 2 x bfloat> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %7 = call <vscale x 4 x bfloat> @llvm.fabs.nxv4bf16(<vscale x 4 x bfloat> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %8 = call <vscale x 8 x bfloat> @llvm.fabs.nxv8bf16(<vscale x 8 x bfloat> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %9 = call <vscale x 16 x bfloat> @llvm.fabs.nxv16bf16(<vscale x 16 x bfloat> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %2 = call <2 x bfloat> @llvm.fabs.v2bf16(<2 x bfloat> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %3 = call <4 x bfloat> @llvm.fabs.v4bf16(<4 x bfloat> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %4 = call <8 x bfloat> @llvm.fabs.v8bf16(<8 x bfloat> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %5 = call <16 x bfloat> @llvm.fabs.v16bf16(<16 x bfloat> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %6 = call <vscale x 2 x bfloat> @llvm.fabs.nxv2bf16(<vscale x 2 x bfloat> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %7 = call <vscale x 4 x bfloat> @llvm.fabs.nxv4bf16(<vscale x 4 x bfloat> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %8 = call <vscale x 8 x bfloat> @llvm.fabs.nxv8bf16(<vscale x 8 x bfloat> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %9 = call <vscale x 16 x bfloat> @llvm.fabs.nxv16bf16(<vscale x 16 x bfloat> undef)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %10 = call float @llvm.fabs.f32(float undef)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %11 = call <2 x float> @llvm.fabs.v2f32(<2 x float> undef)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %12 = call <4 x float> @llvm.fabs.v4f32(<4 x float> undef)
@@ -66,17 +66,29 @@ define void @fabs() {
 }
 
 define void @fabs_f16() {
-; CHECK-LABEL: 'fabs_f16'
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %1 = call half @llvm.fabs.f16(half undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %2 = call <2 x half> @llvm.fabs.v2f16(<2 x half> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %3 = call <4 x half> @llvm.fabs.v4f16(<4 x half> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %4 = call <8 x half> @llvm.fabs.v8f16(<8 x half> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %5 = call <16 x half> @llvm.fabs.v16f16(<16 x half> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %6 = call <vscale x 2 x half> @llvm.fabs.nxv2f16(<vscale x 2 x half> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %7 = call <vscale x 4 x half> @llvm.fabs.nxv4f16(<vscale x 4 x half> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %8 = call <vscale x 8 x half> @llvm.fabs.nxv8f16(<vscale x 8 x half> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %9 = call <vscale x 16 x half> @llvm.fabs.nxv16f16(<vscale x 16 x half> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret void
+; ZVFH-LABEL: 'fabs_f16'
+; ZVFH-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %1 = call half @llvm.fabs.f16(half undef)
+; ZVFH-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %2 = call <2 x half> @llvm.fabs.v2f16(<2 x half> undef)
+; ZVFH-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %3 = call <4 x half> @llvm.fabs.v4f16(<4 x half> undef)
+; ZVFH-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %4 = call <8 x half> @llvm.fabs.v8f16(<8 x half> undef)
+; ZVFH-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %5 = call <16 x half> @llvm.fabs.v16f16(<16 x half> undef)
+; ZVFH-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %6 = call <vscale x 2 x half> @llvm.fabs.nxv2f16(<vscale x 2 x half> undef)
+; ZVFH-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %7 = call <vscale x 4 x half> @llvm.fabs.nxv4f16(<vscale x 4 x half> undef)
+; ZVFH-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %8 = call <vscale x 8 x half> @llvm.fabs.nxv8f16(<vscale x 8 x half> undef)
+; ZVFH-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %9 = call <vscale x 16 x half> @llvm.fabs.nxv16f16(<vscale x 16 x half> undef)
+; ZVFH-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret void
+;
+; ZVFHMIN-LABEL: 'fabs_f16'
+; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %1 = call half @llvm.fabs.f16(half undef)
+; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %2 = call <2 x half> @llvm.fabs.v2f16(<2 x half> undef)
+; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %3 = call <4 x half> @llvm.fabs.v4f16(<4 x half> undef)
+; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %4 = call <8 x half> @llvm.fabs.v8f16(<8 x half> undef)
+; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %5 = call <16 x half> @llvm.fabs.v16f16(<16 x half> undef)
+; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %6 = call <vscale x 2 x half> @llvm.fabs.nxv2f16(<vscale x 2 x half> undef)
+; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %7 = call <vscale x 4 x half> @llvm.fabs.nxv4f16(<vscale x 4 x half> undef)
+; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %8 = call <vscale x 8 x half> @llvm.fabs.nxv8f16(<vscale x 8 x half> undef)
+; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %9 = call <vscale x 16 x half> @llvm.fabs.nxv16f16(<vscale x 16 x half> undef)
+; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret void
 ;
   call half @llvm.fabs.f16(half undef)
   call <2 x half> @llvm.fabs.v2f16(<2 x half> undef)
diff --git a/llvm/test/Analysis/CostModel/RISCV/fp-sqrt-pow.ll b/llvm/test/Analysis/CostModel/RISCV/fp-sqrt-pow.ll
index 446627f6bf3c0e..85b0399671883e 100644
--- a/llvm/test/Analysis/CostModel/RISCV/fp-sqrt-pow.ll
+++ b/llvm/test/Analysis/CostModel/RISCV/fp-sqrt-pow.ll
@@ -5,33 +5,35 @@
 define void @sqrt() {
 ; CHECK-LABEL: 'sqrt'
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %1 = call bfloat @llvm.sqrt.bf16(bfloat undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %2 = call <2 x bfloat> @llvm.sqrt.v2bf16(<2 x bfloat> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %3 = call <4 x bfloat> @llvm.sqrt.v4bf16(<4 x bfloat> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %4 = call <8 x bfloat> @llvm.sqrt.v8bf16(<8 x bfloat> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %5 = call <16 x bfloat> @llvm.sqrt.v16bf16(<16 x bfloat> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %6 = call <vscale x 2 x bfloat> @llvm.sqrt.nxv2bf16(<vscale x 2 x bfloat> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %7 = call <vscale x 4 x bfloat> @llvm.sqrt.nxv4bf16(<vscale x 4 x bfloat> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %8 = call <vscale x 8 x bfloat> @llvm.sqrt.nxv8bf16(<vscale x 8 x bfloat> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %9 = call <vscale x 16 x bfloat> @llvm.sqrt.nxv16bf16(<vscale x 16 x bfloat> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %10 = call float @llvm.sqrt.f32(float undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %11 = call <2 x float> @llvm.sqrt.v2f32(<2 x float> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %12 = call <4 x float> @llvm.sqrt.v4f32(<4 x float> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %13 = call <8 x float> @llvm.sqrt.v8f32(<8 x float> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %14 = call <16 x float> @llvm.sqrt.v16f32(<16 x float> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %15 = call <vscale x 1 x float> @llvm.sqrt.nxv1f32(<vscale x 1 x float> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %16 = call <vscale x 2 x float> @llvm.sqrt.nxv2f32(<vscale x 2 x float> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %17 = call <vscale x 4 x float> @llvm.sqrt.nxv4f32(<vscale x 4 x float> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %18 = call <vscale x 8 x float> @llvm.sqrt.nxv8f32(<vscale x 8 x float> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %19 = call <vscale x 16 x float> @llvm.sqrt.nxv16f32(<vscale x 16 x float> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %20 = call double @llvm.sqrt.f64(double undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %21 = call <2 x double> @llvm.sqrt.v2f64(<2 x double> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %22 = call <4 x double> @llvm.sqrt.v4f64(<4 x double> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %23 = call <8 x double> @llvm.sqrt.v8f64(<8 x double> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %24 = call <16 x double> @llvm.sqrt.v16f64(<16 x double> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %25 = call <vscale x 1 x double> @llvm.sqrt.nxv1f64(<vscale x 1 x double> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %26 = call <vscale x 2 x double> @llvm.sqrt.nxv2f64(<vscale x 2 x double> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %27 = call <vscale x 4 x double> @llvm.sqrt.nxv4f64(<vscale x 4 x double> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %28 = call <vscale x 8 x double> @llvm.sqrt.nxv8f64(<vscale x 8 x double> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %2 = call <2 x bfloat> @llvm.sqrt.v2bf16(<2 x bfloat> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %3 = call <4 x bfloat> @llvm.sqrt.v4bf16(<4 x bfloat> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %4 = call <8 x bfloat> @llvm.sqrt.v8bf16(<8 x bfloat> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %5 = call <16 x bfloat> @llvm.sqrt.v16bf16(<16 x bfloat> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %6 = call <32 x bfloat> @llvm.sqrt.v32bf16(<32 x bfloat> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %7 = call <vscale x 2 x bfloat> @llvm.sqrt.nxv2bf16(<vscale x 2 x bfloat> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %8 = call <vscale x 4 x bfloat> @llvm.sqrt.nxv4bf16(<vscale x 4 x bfloat> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %9 = call <vscale x 8 x bfloat> @llvm.sqrt.nxv8bf16(<vscale x 8 x bfloat> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %10 = call <vscale x 16 x bfloat> @llvm.sqrt.nxv16bf16(<vscale x 16 x bfloat> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %11 = call float @llvm.sqrt.f32(float undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %12 = call <2 x float> @llvm.sqrt.v2f32(<2 x float> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %13 = call <4 x float> @llvm.sqrt.v4f32(<4 x float> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %14 = call <8 x float> @llvm.sqrt.v8f32(<8 x float> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %15 = call <16 x float> @llvm.sqrt.v16f32(<16 x float> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %16 = call <vscale x 1 x float> @llvm.sqrt.nxv1f32(<vscale x 1 x float> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %17 = call <vscale x 2 x float> @llvm.sqrt.nxv2f32(<vscale x 2 x float> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %18 = call <vscale x 4 x float> @llvm.sqrt.nxv4f32(<vscale x 4 x float> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %19 = call <vscale x 8 x float> @llvm.sqrt.nxv8f32(<vscale x 8 x float> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %20 = call <vscale x 16 x float> @llvm.sqrt.nxv16f32(<vscale x 16 x float> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 48 for instruction: %21 = call <vscale x 32 x bfloat> @llvm.sqrt.nxv32bf16(<vscale x 32 x bfloat> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %22 = call double @llvm.sqrt.f64(double undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %23 = call <2 x double> @llvm.sqrt.v2f64(<2 x double> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %24 = call <4 x double> @llvm.sqrt.v4f64(<4 x double> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %25 = call <8 x double> @llvm.sqrt.v8f64(<8 x double> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %26 = call <16 x double> @llvm.sqrt.v16f64(<16 x double> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %27 = call <vscale x 1 x double> @llvm.sqrt.nxv1f64(<vscale x 1 x double> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %28 = call <vscale x 2 x double> @llvm.sqrt.nxv2f64(<vscale x 2 x double> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %29 = call <vscale x 4 x double> @llvm.sqrt.nxv4f64(<vscale x 4 x double> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %30 = call <vscale x 8 x double> @llvm.sqrt.nxv8f64(<vscale x 8 x double> undef)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret void
 ;
   call bfloat @llvm.sqrt.bf16(bfloat undef)
@@ -39,6 +41,7 @@ define void @sqrt() {
   call <4 x bfloat> @llvm.sqrt.v4bf16(<4 x bfloat> undef)
   call <8 x bfloat> @llvm.sqrt.v8bf16(<8 x bfloat> undef)
   call <16 x bfloat> @llvm.sqrt.v16bf16(<16 x bfloat> undef)
+  call <32 x bfloat> @llvm.sqrt.v32bf16(<32 x bfloat> undef)
   call <vscale x 2 x bfloat> @llvm.sqrt.nxv2bf16(<vscale x 2 x bfloat> undef)
   call <vscale x 4 x bfloat> @llvm.sqrt.nxv4bf16(<vscale x 4 x bfloat> undef)
   call <vscale x 8 x bfloat> @llvm.sqrt.nxv8bf16(<vscale x 8 x bfloat> undef)
@@ -53,6 +56,7 @@ define void @sqrt() {
   call <vscale x 4 x float> @llvm.sqrt.nxv4f32(<vscale x 4 x float> undef)
   call <vscale x 8 x float> @llvm.sqrt.nxv8f32(<vscale x 8 x float> undef)
   call <vscale x 16 x float> @llvm.sqrt.nxv16f32(<vscale x 16 x float> undef)
+  call <vscale x 32 x bfloat> @llvm.sqrt.nxv32bf16(<vscale x 32 x bfloat> undef)
   call double @llvm.sqrt.f64(double undef)
   call <2 x double> @llvm.sqrt.v2f64(<2 x double> undef)
   call <4 x double> @llvm.sqrt.v4f64(<4 x double> undef)
@@ -66,27 +70,45 @@ define void @sqrt() {
 }
 
 define void @sqrt_f16() {
-; CHECK-LABEL: 'sqrt_f16'
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %1 = call half @llvm.sqrt.f16(half undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %2 = call <2 x half> @llvm.sqrt.v2f16(<2 x half> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %3 = call <4 x half> @llvm.sqrt.v4f16(<4 x half> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %4 = call <8 x half> @llvm.sqrt.v8f16(<8 x half> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %5 = call <16 x half> @llvm.sqrt.v16f16(<16 x half> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %6 = call <vscale x 2 x half> @llvm.sqrt.nxv2f16(<vscale x 2 x half> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %7 = call <vscale x 4 x half> @llvm.sqrt.nxv4f16(<vscale x 4 x half> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %8 = call <vscale x 8 x half> @llvm.sqrt.nxv8f16(<vscale x 8 x half> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %9 = call <vscale x 16 x half> @llvm.sqrt.nxv16f16(<vscale x 16 x half> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret void
+; ZVFH-LABEL: 'sqrt_f16'
+; ZVFH-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %1 = call half @llvm.sqrt.f16(half undef)
+; ZVFH-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %2 = call <2 x half> @llvm.sqrt.v2f16(<2 x half> undef)
+; ZVFH-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %3 = call <4 x half> @llvm.sqrt.v4f16(<4 x half> undef)
+; ZVFH-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %4 = call <8 x half> @llvm.sqrt.v8f16(<8 x half> undef)
+; ZVFH-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %5 = call <16 x half> @llvm.sqrt.v16f16(<16 x half> undef)
+; ZVFH-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %6 = call <32 x half> @llvm.sqrt.v32f16(<32 x half> undef)
+; ZVFH-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %7 = call <vscale x 2 x half> @llvm.sqrt.nxv2f16(<vscale x 2 x half> undef)
+; ZVFH-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %8 = call <vscale x 4 x half> @llvm.sqrt.nxv4f16(<vscale x 4 x half> undef)
+; ZVFH-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %9 = call <vscale x 8 x half> @llvm.sqrt.nxv8f16(<vscale x 8 x half> undef)
+; ZVFH-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %10 = call <vscale x 16 x half> @llvm.sqrt.nxv16f16(<vscale x 16 x half> undef)
+; ZVFH-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %11 = call <vscale x 32 x half> @llvm.sqrt.nxv32f16(<vscale x 32 x half> undef)
+; ZVFH-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret void
+;
+; ZVFHMIN-LABEL: 'sqrt_f16'
+; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %1 = call half @llvm.sqrt.f16(half undef)
+; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %2 = call <2 x half> @llvm.sqrt.v2f16(<2 x half> undef)
+; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %3 = call <4 x half> @llvm.sqrt.v4f16(<4 x half> undef)
+; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %4 = call <8 x half> @llvm.sqrt.v8f16(<8 x half> undef)
+; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %5 = call <16 x half> @llvm.sqrt.v16f16(<16 x half> undef)
+; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %6 = call <32 x half> @llvm.sqrt.v32f16(<32 x half> undef)
+; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %7 = call <vscale x 2 x half> @llvm.sqrt.nxv2f16(<vscale x 2 x half> undef)
+; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %8 = call <vscale x 4 x half> @llvm.sqrt.nxv4f16(<vscale x 4 x half> undef)
+; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %9 = call <vscale x 8 x half> @llvm.sqrt.nxv8f16(<vscale x 8 x half> undef)
+; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %10 = call <vscale x 16 x half> @llvm.sqrt.nxv16f16(<vscale x 16 x half> undef)
+; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 48 for instruction: %11 = call <vscale x 32 x half> @llvm.sqrt.nxv32f16(<vscale x 32 x half> undef)
+; ZVFHMIN-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret void
 ;
   call half @llvm.sqrt.f16(half undef)
   call <2 x half> @llvm.sqrt.v2f16(<2 x half> undef)
   call <4 x half> @llvm.sqrt.v4f16(<4 x half> undef)
   call <8 x half> @llvm.sqrt.v8f16(<8 x half> undef)
   call <16 x half> @llvm.sqrt.v16f16(<16 x half> undef)
+  call <32 x half> @llvm.sqrt.v32f16(<32 x half> undef)
   call <vscale x 2 x half> @llvm.sqrt.nxv2f16(<vscale x 2 x half> undef)
   call <vscale x 4 x half> @llvm.sqrt.nxv4f16(<vscale x 4 x half> undef)
   call <vscale x 8 x half> @llvm.sqrt.nxv8f16(<vscale x 8 x half> undef)
   call <vscale x 16 x half> @llvm.sqrt.nxv16f16(<vscale x 16 x half> undef)
+  call <vscale x 32 x half> @llvm.sqrt.nxv32f16(<vscale x 32 x half> undef)
   ret void
 }
 



More information about the llvm-commits mailing list