[llvm] 40363d5 - [RISCV] Add cost model tests for fp rounding ops for bf16. NFC
Luke Lau via llvm-commits
llvm-commits at lists.llvm.org
Mon Oct 28 07:59:15 PDT 2024
Author: Luke Lau
Date: 2024-10-28T14:59:06Z
New Revision: 40363d506db8ab3382339dc4463372dcfcf931b7
URL: https://github.com/llvm/llvm-project/commit/40363d506db8ab3382339dc4463372dcfcf931b7
DIFF: https://github.com/llvm/llvm-project/commit/40363d506db8ab3382339dc4463372dcfcf931b7.diff
LOG: [RISCV] Add cost model tests for fp rounding ops for bf16. NFC
Added:
Modified:
llvm/test/Analysis/CostModel/RISCV/fround.ll
Removed:
################################################################################
diff --git a/llvm/test/Analysis/CostModel/RISCV/fround.ll b/llvm/test/Analysis/CostModel/RISCV/fround.ll
index b4740f223eca3a..c6826760a45bee 100644
--- a/llvm/test/Analysis/CostModel/RISCV/fround.ll
+++ b/llvm/test/Analysis/CostModel/RISCV/fround.ll
@@ -1,30 +1,50 @@
; NOTE: Assertions have been autogenerated by utils/update_analyze_test_checks.py
-; RUN: opt < %s -passes="print<cost-model>" 2>&1 -disable-output -S -mtriple=riscv64 -mattr=+v,+f,+d,+zvfh | FileCheck %s --check-prefixes=CHECK,ZVFH
-; RUN: opt < %s -passes="print<cost-model>" 2>&1 -disable-output -S -mtriple=riscv64 -mattr=+v,+f,+d,+zvfhmin | FileCheck %s --check-prefixes=CHECK,ZVFHMIN
+; RUN: opt < %s -passes="print<cost-model>" 2>&1 -disable-output -S -mtriple=riscv64 -mattr=+v,+f,+d,+zvfh,+zvfbfmin | FileCheck %s --check-prefixes=CHECK,ZVFH
+; RUN: opt < %s -passes="print<cost-model>" 2>&1 -disable-output -S -mtriple=riscv64 -mattr=+v,+f,+d,+zvfhmin,+zvfbfmin | FileCheck %s --check-prefixes=CHECK,ZVFHMIN
define void @floor() {
; CHECK-LABEL: 'floor'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %1 = call float @llvm.floor.f32(float undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %2 = call <2 x float> @llvm.floor.v2f32(<2 x float> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %3 = call <4 x float> @llvm.floor.v4f32(<4 x float> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %4 = call <8 x float> @llvm.floor.v8f32(<8 x float> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %5 = call <16 x float> @llvm.floor.v16f32(<16 x float> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %6 = call <vscale x 1 x float> @llvm.floor.nxv1f32(<vscale x 1 x float> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %7 = call <vscale x 2 x float> @llvm.floor.nxv2f32(<vscale x 2 x float> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %8 = call <vscale x 4 x float> @llvm.floor.nxv4f32(<vscale x 4 x float> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %9 = call <vscale x 8 x float> @llvm.floor.nxv8f32(<vscale x 8 x float> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %10 = call <vscale x 16 x float> @llvm.floor.nxv16f32(<vscale x 16 x float> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %11 = call double @llvm.floor.f64(double undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %12 = call <2 x double> @llvm.floor.v2f64(<2 x double> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %13 = call <4 x double> @llvm.floor.v4f64(<4 x double> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %14 = call <8 x double> @llvm.floor.v8f64(<8 x double> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %15 = call <16 x double> @llvm.floor.v16f64(<16 x double> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %16 = call <vscale x 1 x double> @llvm.floor.nxv1f64(<vscale x 1 x double> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %17 = call <vscale x 2 x double> @llvm.floor.nxv2f64(<vscale x 2 x double> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %18 = call <vscale x 4 x double> @llvm.floor.nxv4f64(<vscale x 4 x double> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %19 = call <vscale x 8 x double> @llvm.floor.nxv8f64(<vscale x 8 x double> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %1 = call bfloat @llvm.floor.bf16(bfloat undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %2 = call <2 x bfloat> @llvm.floor.v2bf16(<2 x bfloat> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %3 = call <4 x bfloat> @llvm.floor.v4bf16(<4 x bfloat> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %4 = call <8 x bfloat> @llvm.floor.v8bf16(<8 x bfloat> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %5 = call <16 x bfloat> @llvm.floor.v16bf16(<16 x bfloat> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %6 = call <vscale x 1 x bfloat> @llvm.floor.nxv1bf16(<vscale x 1 x bfloat> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %7 = call <vscale x 2 x bfloat> @llvm.floor.nxv2bf16(<vscale x 2 x bfloat> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %8 = call <vscale x 4 x bfloat> @llvm.floor.nxv4bf16(<vscale x 4 x bfloat> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %9 = call <vscale x 8 x bfloat> @llvm.floor.nxv8bf16(<vscale x 8 x bfloat> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %10 = call <vscale x 16 x bfloat> @llvm.floor.nxv16bf16(<vscale x 16 x bfloat> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %11 = call float @llvm.floor.f32(float undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %12 = call <2 x float> @llvm.floor.v2f32(<2 x float> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %13 = call <4 x float> @llvm.floor.v4f32(<4 x float> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %14 = call <8 x float> @llvm.floor.v8f32(<8 x float> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %15 = call <16 x float> @llvm.floor.v16f32(<16 x float> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %16 = call <vscale x 1 x float> @llvm.floor.nxv1f32(<vscale x 1 x float> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %17 = call <vscale x 2 x float> @llvm.floor.nxv2f32(<vscale x 2 x float> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %18 = call <vscale x 4 x float> @llvm.floor.nxv4f32(<vscale x 4 x float> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %19 = call <vscale x 8 x float> @llvm.floor.nxv8f32(<vscale x 8 x float> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %20 = call <vscale x 16 x float> @llvm.floor.nxv16f32(<vscale x 16 x float> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %21 = call double @llvm.floor.f64(double undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %22 = call <2 x double> @llvm.floor.v2f64(<2 x double> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %23 = call <4 x double> @llvm.floor.v4f64(<4 x double> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %24 = call <8 x double> @llvm.floor.v8f64(<8 x double> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %25 = call <16 x double> @llvm.floor.v16f64(<16 x double> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %26 = call <vscale x 1 x double> @llvm.floor.nxv1f64(<vscale x 1 x double> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %27 = call <vscale x 2 x double> @llvm.floor.nxv2f64(<vscale x 2 x double> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %28 = call <vscale x 4 x double> @llvm.floor.nxv4f64(<vscale x 4 x double> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %29 = call <vscale x 8 x double> @llvm.floor.nxv8f64(<vscale x 8 x double> undef)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret void
;
+ call bfloat @llvm.floor.bf16(bfloat undef)
+ call <2 x bfloat> @llvm.floor.v2bf16(<2 x bfloat> undef)
+ call <4 x bfloat> @llvm.floor.v4bf16(<4 x bfloat> undef)
+ call <8 x bfloat> @llvm.floor.v8bf16(<8 x bfloat> undef)
+ call <16 x bfloat> @llvm.floor.v16bf16(<16 x bfloat> undef)
+ call <vscale x 1 x bfloat> @llvm.floor.nvx1bf16(<vscale x 1 x bfloat> undef)
+ call <vscale x 2 x bfloat> @llvm.floor.nvx2bf16(<vscale x 2 x bfloat> undef)
+ call <vscale x 4 x bfloat> @llvm.floor.nvx4bf16(<vscale x 4 x bfloat> undef)
+ call <vscale x 8 x bfloat> @llvm.floor.nvx8bf16(<vscale x 8 x bfloat> undef)
+ call <vscale x 16 x bfloat> @llvm.floor.nvx16bf16(<vscale x 16 x bfloat> undef)
call float @llvm.floor.f32(float undef)
call <2 x float> @llvm.floor.v2f32(<2 x float> undef)
call <4 x float> @llvm.floor.v4f32(<4 x float> undef)
@@ -89,27 +109,47 @@ define void @floor_fp16() {
define void @ceil() {
; CHECK-LABEL: 'ceil'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %1 = call float @llvm.ceil.f32(float undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %2 = call <2 x float> @llvm.ceil.v2f32(<2 x float> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %3 = call <4 x float> @llvm.ceil.v4f32(<4 x float> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %4 = call <8 x float> @llvm.ceil.v8f32(<8 x float> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %5 = call <16 x float> @llvm.ceil.v16f32(<16 x float> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %6 = call <vscale x 1 x float> @llvm.ceil.nxv1f32(<vscale x 1 x float> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %7 = call <vscale x 2 x float> @llvm.ceil.nxv2f32(<vscale x 2 x float> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %8 = call <vscale x 4 x float> @llvm.ceil.nxv4f32(<vscale x 4 x float> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %9 = call <vscale x 8 x float> @llvm.ceil.nxv8f32(<vscale x 8 x float> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %10 = call <vscale x 16 x float> @llvm.ceil.nxv16f32(<vscale x 16 x float> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %11 = call double @llvm.ceil.f64(double undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %12 = call <2 x double> @llvm.ceil.v2f64(<2 x double> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %13 = call <4 x double> @llvm.ceil.v4f64(<4 x double> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %14 = call <8 x double> @llvm.ceil.v8f64(<8 x double> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %15 = call <16 x double> @llvm.ceil.v16f64(<16 x double> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %16 = call <vscale x 1 x double> @llvm.ceil.nxv1f64(<vscale x 1 x double> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %17 = call <vscale x 2 x double> @llvm.ceil.nxv2f64(<vscale x 2 x double> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %18 = call <vscale x 4 x double> @llvm.ceil.nxv4f64(<vscale x 4 x double> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %19 = call <vscale x 8 x double> @llvm.ceil.nxv8f64(<vscale x 8 x double> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %1 = call bfloat @llvm.ceil.bf16(bfloat undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %2 = call <2 x bfloat> @llvm.ceil.v2bf16(<2 x bfloat> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %3 = call <4 x bfloat> @llvm.ceil.v4bf16(<4 x bfloat> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %4 = call <8 x bfloat> @llvm.ceil.v8bf16(<8 x bfloat> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %5 = call <16 x bfloat> @llvm.ceil.v16bf16(<16 x bfloat> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %6 = call <vscale x 1 x bfloat> @llvm.ceil.nxv1bf16(<vscale x 1 x bfloat> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %7 = call <vscale x 2 x bfloat> @llvm.ceil.nxv2bf16(<vscale x 2 x bfloat> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %8 = call <vscale x 4 x bfloat> @llvm.ceil.nxv4bf16(<vscale x 4 x bfloat> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %9 = call <vscale x 8 x bfloat> @llvm.ceil.nxv8bf16(<vscale x 8 x bfloat> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %10 = call <vscale x 16 x bfloat> @llvm.ceil.nxv16bf16(<vscale x 16 x bfloat> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %11 = call float @llvm.ceil.f32(float undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %12 = call <2 x float> @llvm.ceil.v2f32(<2 x float> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %13 = call <4 x float> @llvm.ceil.v4f32(<4 x float> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %14 = call <8 x float> @llvm.ceil.v8f32(<8 x float> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %15 = call <16 x float> @llvm.ceil.v16f32(<16 x float> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %16 = call <vscale x 1 x float> @llvm.ceil.nxv1f32(<vscale x 1 x float> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %17 = call <vscale x 2 x float> @llvm.ceil.nxv2f32(<vscale x 2 x float> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %18 = call <vscale x 4 x float> @llvm.ceil.nxv4f32(<vscale x 4 x float> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %19 = call <vscale x 8 x float> @llvm.ceil.nxv8f32(<vscale x 8 x float> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %20 = call <vscale x 16 x float> @llvm.ceil.nxv16f32(<vscale x 16 x float> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %21 = call double @llvm.ceil.f64(double undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %22 = call <2 x double> @llvm.ceil.v2f64(<2 x double> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %23 = call <4 x double> @llvm.ceil.v4f64(<4 x double> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %24 = call <8 x double> @llvm.ceil.v8f64(<8 x double> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %25 = call <16 x double> @llvm.ceil.v16f64(<16 x double> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %26 = call <vscale x 1 x double> @llvm.ceil.nxv1f64(<vscale x 1 x double> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %27 = call <vscale x 2 x double> @llvm.ceil.nxv2f64(<vscale x 2 x double> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %28 = call <vscale x 4 x double> @llvm.ceil.nxv4f64(<vscale x 4 x double> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %29 = call <vscale x 8 x double> @llvm.ceil.nxv8f64(<vscale x 8 x double> undef)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret void
;
+ call bfloat @llvm.ceil.bf16(bfloat undef)
+ call <2 x bfloat> @llvm.ceil.v2bf16(<2 x bfloat> undef)
+ call <4 x bfloat> @llvm.ceil.v4bf16(<4 x bfloat> undef)
+ call <8 x bfloat> @llvm.ceil.v8bf16(<8 x bfloat> undef)
+ call <16 x bfloat> @llvm.ceil.v16bf16(<16 x bfloat> undef)
+ call <vscale x 1 x bfloat> @llvm.ceil.nvx1bf16(<vscale x 1 x bfloat> undef)
+ call <vscale x 2 x bfloat> @llvm.ceil.nvx2bf16(<vscale x 2 x bfloat> undef)
+ call <vscale x 4 x bfloat> @llvm.ceil.nvx4bf16(<vscale x 4 x bfloat> undef)
+ call <vscale x 8 x bfloat> @llvm.ceil.nvx8bf16(<vscale x 8 x bfloat> undef)
+ call <vscale x 16 x bfloat> @llvm.ceil.nvx16bf16(<vscale x 16 x bfloat> undef)
call float @llvm.ceil.f32(float undef)
call <2 x float> @llvm.ceil.v2f32(<2 x float> undef)
call <4 x float> @llvm.ceil.v4f32(<4 x float> undef)
@@ -174,27 +214,47 @@ define void @ceil_fp16() {
define void @trunc() {
; CHECK-LABEL: 'trunc'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %1 = call float @llvm.trunc.f32(float undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %2 = call <2 x float> @llvm.trunc.v2f32(<2 x float> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %3 = call <4 x float> @llvm.trunc.v4f32(<4 x float> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %4 = call <8 x float> @llvm.trunc.v8f32(<8 x float> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %5 = call <16 x float> @llvm.trunc.v16f32(<16 x float> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %6 = call <vscale x 1 x float> @llvm.trunc.nxv1f32(<vscale x 1 x float> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %7 = call <vscale x 2 x float> @llvm.trunc.nxv2f32(<vscale x 2 x float> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %8 = call <vscale x 4 x float> @llvm.trunc.nxv4f32(<vscale x 4 x float> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %9 = call <vscale x 8 x float> @llvm.trunc.nxv8f32(<vscale x 8 x float> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %10 = call <vscale x 16 x float> @llvm.trunc.nxv16f32(<vscale x 16 x float> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %11 = call double @llvm.trunc.f64(double undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %12 = call <2 x double> @llvm.trunc.v2f64(<2 x double> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %13 = call <4 x double> @llvm.trunc.v4f64(<4 x double> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %14 = call <8 x double> @llvm.trunc.v8f64(<8 x double> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %15 = call <16 x double> @llvm.trunc.v16f64(<16 x double> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %16 = call <vscale x 1 x double> @llvm.trunc.nxv1f64(<vscale x 1 x double> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %17 = call <vscale x 2 x double> @llvm.trunc.nxv2f64(<vscale x 2 x double> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %18 = call <vscale x 4 x double> @llvm.trunc.nxv4f64(<vscale x 4 x double> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %19 = call <vscale x 8 x double> @llvm.trunc.nxv8f64(<vscale x 8 x double> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %1 = call bfloat @llvm.trunc.bf16(bfloat undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %2 = call <2 x bfloat> @llvm.trunc.v2bf16(<2 x bfloat> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %3 = call <4 x bfloat> @llvm.trunc.v4bf16(<4 x bfloat> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %4 = call <8 x bfloat> @llvm.trunc.v8bf16(<8 x bfloat> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %5 = call <16 x bfloat> @llvm.trunc.v16bf16(<16 x bfloat> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %6 = call <vscale x 1 x bfloat> @llvm.trunc.nxv1bf16(<vscale x 1 x bfloat> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %7 = call <vscale x 2 x bfloat> @llvm.trunc.nxv2bf16(<vscale x 2 x bfloat> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %8 = call <vscale x 4 x bfloat> @llvm.trunc.nxv4bf16(<vscale x 4 x bfloat> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %9 = call <vscale x 8 x bfloat> @llvm.trunc.nxv8bf16(<vscale x 8 x bfloat> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %10 = call <vscale x 16 x bfloat> @llvm.trunc.nxv16bf16(<vscale x 16 x bfloat> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %11 = call float @llvm.trunc.f32(float undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %12 = call <2 x float> @llvm.trunc.v2f32(<2 x float> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %13 = call <4 x float> @llvm.trunc.v4f32(<4 x float> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %14 = call <8 x float> @llvm.trunc.v8f32(<8 x float> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %15 = call <16 x float> @llvm.trunc.v16f32(<16 x float> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %16 = call <vscale x 1 x float> @llvm.trunc.nxv1f32(<vscale x 1 x float> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %17 = call <vscale x 2 x float> @llvm.trunc.nxv2f32(<vscale x 2 x float> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %18 = call <vscale x 4 x float> @llvm.trunc.nxv4f32(<vscale x 4 x float> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %19 = call <vscale x 8 x float> @llvm.trunc.nxv8f32(<vscale x 8 x float> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %20 = call <vscale x 16 x float> @llvm.trunc.nxv16f32(<vscale x 16 x float> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %21 = call double @llvm.trunc.f64(double undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %22 = call <2 x double> @llvm.trunc.v2f64(<2 x double> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %23 = call <4 x double> @llvm.trunc.v4f64(<4 x double> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %24 = call <8 x double> @llvm.trunc.v8f64(<8 x double> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %25 = call <16 x double> @llvm.trunc.v16f64(<16 x double> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %26 = call <vscale x 1 x double> @llvm.trunc.nxv1f64(<vscale x 1 x double> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %27 = call <vscale x 2 x double> @llvm.trunc.nxv2f64(<vscale x 2 x double> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %28 = call <vscale x 4 x double> @llvm.trunc.nxv4f64(<vscale x 4 x double> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %29 = call <vscale x 8 x double> @llvm.trunc.nxv8f64(<vscale x 8 x double> undef)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret void
;
+ call bfloat @llvm.trunc.bf16(bfloat undef)
+ call <2 x bfloat> @llvm.trunc.v2bf16(<2 x bfloat> undef)
+ call <4 x bfloat> @llvm.trunc.v4bf16(<4 x bfloat> undef)
+ call <8 x bfloat> @llvm.trunc.v8bf16(<8 x bfloat> undef)
+ call <16 x bfloat> @llvm.trunc.v16bf16(<16 x bfloat> undef)
+ call <vscale x 1 x bfloat> @llvm.trunc.nvx1bf16(<vscale x 1 x bfloat> undef)
+ call <vscale x 2 x bfloat> @llvm.trunc.nvx2bf16(<vscale x 2 x bfloat> undef)
+ call <vscale x 4 x bfloat> @llvm.trunc.nvx4bf16(<vscale x 4 x bfloat> undef)
+ call <vscale x 8 x bfloat> @llvm.trunc.nvx8bf16(<vscale x 8 x bfloat> undef)
+ call <vscale x 16 x bfloat> @llvm.trunc.nvx16bf16(<vscale x 16 x bfloat> undef)
call float @llvm.trunc.f32(float undef)
call <2 x float> @llvm.trunc.v2f32(<2 x float> undef)
call <4 x float> @llvm.trunc.v4f32(<4 x float> undef)
@@ -259,27 +319,47 @@ define void @trunc_fp16() {
define void @rint() {
; CHECK-LABEL: 'rint'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %1 = call float @llvm.rint.f32(float undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %2 = call <2 x float> @llvm.rint.v2f32(<2 x float> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %3 = call <4 x float> @llvm.rint.v4f32(<4 x float> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %4 = call <8 x float> @llvm.rint.v8f32(<8 x float> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %5 = call <16 x float> @llvm.rint.v16f32(<16 x float> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %6 = call <vscale x 1 x float> @llvm.rint.nxv1f32(<vscale x 1 x float> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %7 = call <vscale x 2 x float> @llvm.rint.nxv2f32(<vscale x 2 x float> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %8 = call <vscale x 4 x float> @llvm.rint.nxv4f32(<vscale x 4 x float> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %9 = call <vscale x 8 x float> @llvm.rint.nxv8f32(<vscale x 8 x float> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %10 = call <vscale x 16 x float> @llvm.rint.nxv16f32(<vscale x 16 x float> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %11 = call double @llvm.rint.f64(double undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %12 = call <2 x double> @llvm.rint.v2f64(<2 x double> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %13 = call <4 x double> @llvm.rint.v4f64(<4 x double> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %14 = call <8 x double> @llvm.rint.v8f64(<8 x double> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %15 = call <16 x double> @llvm.rint.v16f64(<16 x double> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %16 = call <vscale x 1 x double> @llvm.rint.nxv1f64(<vscale x 1 x double> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %17 = call <vscale x 2 x double> @llvm.rint.nxv2f64(<vscale x 2 x double> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %18 = call <vscale x 4 x double> @llvm.rint.nxv4f64(<vscale x 4 x double> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %19 = call <vscale x 8 x double> @llvm.rint.nxv8f64(<vscale x 8 x double> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %1 = call bfloat @llvm.rint.bf16(bfloat undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %2 = call <2 x bfloat> @llvm.rint.v2bf16(<2 x bfloat> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %3 = call <4 x bfloat> @llvm.rint.v4bf16(<4 x bfloat> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %4 = call <8 x bfloat> @llvm.rint.v8bf16(<8 x bfloat> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %5 = call <16 x bfloat> @llvm.rint.v16bf16(<16 x bfloat> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %6 = call <vscale x 1 x bfloat> @llvm.rint.nxv1bf16(<vscale x 1 x bfloat> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %7 = call <vscale x 2 x bfloat> @llvm.rint.nxv2bf16(<vscale x 2 x bfloat> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %8 = call <vscale x 4 x bfloat> @llvm.rint.nxv4bf16(<vscale x 4 x bfloat> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %9 = call <vscale x 8 x bfloat> @llvm.rint.nxv8bf16(<vscale x 8 x bfloat> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %10 = call <vscale x 16 x bfloat> @llvm.rint.nxv16bf16(<vscale x 16 x bfloat> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %11 = call float @llvm.rint.f32(float undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %12 = call <2 x float> @llvm.rint.v2f32(<2 x float> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %13 = call <4 x float> @llvm.rint.v4f32(<4 x float> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %14 = call <8 x float> @llvm.rint.v8f32(<8 x float> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %15 = call <16 x float> @llvm.rint.v16f32(<16 x float> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %16 = call <vscale x 1 x float> @llvm.rint.nxv1f32(<vscale x 1 x float> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %17 = call <vscale x 2 x float> @llvm.rint.nxv2f32(<vscale x 2 x float> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %18 = call <vscale x 4 x float> @llvm.rint.nxv4f32(<vscale x 4 x float> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %19 = call <vscale x 8 x float> @llvm.rint.nxv8f32(<vscale x 8 x float> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %20 = call <vscale x 16 x float> @llvm.rint.nxv16f32(<vscale x 16 x float> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %21 = call double @llvm.rint.f64(double undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %22 = call <2 x double> @llvm.rint.v2f64(<2 x double> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %23 = call <4 x double> @llvm.rint.v4f64(<4 x double> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %24 = call <8 x double> @llvm.rint.v8f64(<8 x double> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %25 = call <16 x double> @llvm.rint.v16f64(<16 x double> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %26 = call <vscale x 1 x double> @llvm.rint.nxv1f64(<vscale x 1 x double> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %27 = call <vscale x 2 x double> @llvm.rint.nxv2f64(<vscale x 2 x double> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %28 = call <vscale x 4 x double> @llvm.rint.nxv4f64(<vscale x 4 x double> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %29 = call <vscale x 8 x double> @llvm.rint.nxv8f64(<vscale x 8 x double> undef)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret void
;
+ call bfloat @llvm.rint.bf16(bfloat undef)
+ call <2 x bfloat> @llvm.rint.v2bf16(<2 x bfloat> undef)
+ call <4 x bfloat> @llvm.rint.v4bf16(<4 x bfloat> undef)
+ call <8 x bfloat> @llvm.rint.v8bf16(<8 x bfloat> undef)
+ call <16 x bfloat> @llvm.rint.v16bf16(<16 x bfloat> undef)
+ call <vscale x 1 x bfloat> @llvm.rint.nvx1bf16(<vscale x 1 x bfloat> undef)
+ call <vscale x 2 x bfloat> @llvm.rint.nvx2bf16(<vscale x 2 x bfloat> undef)
+ call <vscale x 4 x bfloat> @llvm.rint.nvx4bf16(<vscale x 4 x bfloat> undef)
+ call <vscale x 8 x bfloat> @llvm.rint.nvx8bf16(<vscale x 8 x bfloat> undef)
+ call <vscale x 16 x bfloat> @llvm.rint.nvx16bf16(<vscale x 16 x bfloat> undef)
call float @llvm.rint.f32(float undef)
call <2 x float> @llvm.rint.v2f32(<2 x float> undef)
call <4 x float> @llvm.rint.v4f32(<4 x float> undef)
@@ -344,27 +424,47 @@ define void @rint_fp16() {
define void @lrint() {
; CHECK-LABEL: 'lrint'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %1 = call i64 @llvm.lrint.i64.f32(float undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %2 = call <2 x i64> @llvm.lrint.v2i64.v2f32(<2 x float> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %3 = call <4 x i64> @llvm.lrint.v4i64.v4f32(<4 x float> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %4 = call <8 x i64> @llvm.lrint.v8i64.v8f32(<8 x float> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %5 = call <16 x i64> @llvm.lrint.v16i64.v16f32(<16 x float> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %6 = call <vscale x 1 x i64> @llvm.lrint.nxv1i64.nxv1f32(<vscale x 1 x float> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %7 = call <vscale x 2 x i64> @llvm.lrint.nxv2i64.nxv2f32(<vscale x 2 x float> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %8 = call <vscale x 4 x i64> @llvm.lrint.nxv4i64.nxv4f32(<vscale x 4 x float> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %9 = call <vscale x 8 x i64> @llvm.lrint.nxv8i64.nxv8f32(<vscale x 8 x float> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %10 = call <vscale x 16 x i64> @llvm.lrint.nxv16i64.nxv16f32(<vscale x 16 x float> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %11 = call i64 @llvm.lrint.i64.f64(double undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %12 = call <2 x i64> @llvm.lrint.v2i64.v2f64(<2 x double> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %13 = call <4 x i64> @llvm.lrint.v4i64.v4f64(<4 x double> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %14 = call <8 x i64> @llvm.lrint.v8i64.v8f64(<8 x double> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %15 = call <16 x i64> @llvm.lrint.v16i64.v16f64(<16 x double> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %16 = call <vscale x 1 x i64> @llvm.lrint.nxv1i64.nxv1f64(<vscale x 1 x double> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %17 = call <vscale x 2 x i64> @llvm.lrint.nxv2i64.nxv2f64(<vscale x 2 x double> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %18 = call <vscale x 4 x i64> @llvm.lrint.nxv4i64.nxv4f64(<vscale x 4 x double> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %19 = call <vscale x 8 x i64> @llvm.lrint.nxv8i64.nxv8f64(<vscale x 8 x double> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %1 = call i64 @llvm.lrint.i64.bf16(bfloat undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %2 = call <2 x i64> @llvm.lrint.v2i64.v2bf16(<2 x bfloat> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %3 = call <4 x i64> @llvm.lrint.v4i64.v4bf16(<4 x bfloat> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %4 = call <8 x i64> @llvm.lrint.v8i64.v8bf16(<8 x bfloat> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %5 = call <16 x i64> @llvm.lrint.v16i64.v16bf16(<16 x bfloat> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %6 = call <vscale x 1 x i64> @llvm.lrint.nxv1i64.nxv1bf16(<vscale x 1 x bfloat> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %7 = call <vscale x 2 x i64> @llvm.lrint.nxv2i64.nxv2bf16(<vscale x 2 x bfloat> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %8 = call <vscale x 4 x i64> @llvm.lrint.nxv4i64.nxv4bf16(<vscale x 4 x bfloat> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %9 = call <vscale x 8 x i64> @llvm.lrint.nxv8i64.nxv8bf16(<vscale x 8 x bfloat> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %10 = call <vscale x 16 x i64> @llvm.lrint.nxv16i64.nxv16bf16(<vscale x 16 x bfloat> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %11 = call i64 @llvm.lrint.i64.f32(float undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %12 = call <2 x i64> @llvm.lrint.v2i64.v2f32(<2 x float> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %13 = call <4 x i64> @llvm.lrint.v4i64.v4f32(<4 x float> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %14 = call <8 x i64> @llvm.lrint.v8i64.v8f32(<8 x float> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %15 = call <16 x i64> @llvm.lrint.v16i64.v16f32(<16 x float> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %16 = call <vscale x 1 x i64> @llvm.lrint.nxv1i64.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 i64> @llvm.lrint.nxv2i64.nxv2f32(<vscale x 2 x float> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %18 = call <vscale x 4 x i64> @llvm.lrint.nxv4i64.nxv4f32(<vscale x 4 x float> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %19 = call <vscale x 8 x i64> @llvm.lrint.nxv8i64.nxv8f32(<vscale x 8 x float> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %20 = call <vscale x 16 x i64> @llvm.lrint.nxv16i64.nxv16f32(<vscale x 16 x float> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %21 = call i64 @llvm.lrint.i64.f64(double undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %22 = call <2 x i64> @llvm.lrint.v2i64.v2f64(<2 x double> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %23 = call <4 x i64> @llvm.lrint.v4i64.v4f64(<4 x double> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %24 = call <8 x i64> @llvm.lrint.v8i64.v8f64(<8 x double> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %25 = call <16 x i64> @llvm.lrint.v16i64.v16f64(<16 x double> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %26 = call <vscale x 1 x i64> @llvm.lrint.nxv1i64.nxv1f64(<vscale x 1 x double> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %27 = call <vscale x 2 x i64> @llvm.lrint.nxv2i64.nxv2f64(<vscale x 2 x double> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %28 = call <vscale x 4 x i64> @llvm.lrint.nxv4i64.nxv4f64(<vscale x 4 x double> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %29 = call <vscale x 8 x i64> @llvm.lrint.nxv8i64.nxv8f64(<vscale x 8 x double> undef)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret void
;
+ call i64 @llvm.lrint.i64.bf16(bfloat undef)
+ call <2 x i64> @llvm.lrint.v2i64.v2bf16(<2 x bfloat> undef)
+ call <4 x i64> @llvm.lrint.v4i64.v4bf16(<4 x bfloat> undef)
+ call <8 x i64> @llvm.lrint.v8i64.v8bf16(<8 x bfloat> undef)
+ call <16 x i64> @llvm.lrint.v16i64.v16bf16(<16 x bfloat> undef)
+ call <vscale x 1 x i64> @llvm.lrint.nvx1i64.nvx1bf16(<vscale x 1 x bfloat> undef)
+ call <vscale x 2 x i64> @llvm.lrint.nvx2i64.nvx2bf16(<vscale x 2 x bfloat> undef)
+ call <vscale x 4 x i64> @llvm.lrint.nvx4i64.nvx4bf16(<vscale x 4 x bfloat> undef)
+ call <vscale x 8 x i64> @llvm.lrint.nvx8i64.nvx8bf16(<vscale x 8 x bfloat> undef)
+ call <vscale x 16 x i64> @llvm.lrint.nvx16i64.nvx16bf16(<vscale x 16 x bfloat> undef)
call i64 @llvm.lrint.i64.f32(float undef)
call <2 x i64> @llvm.lrint.v2i64.v2f32(<2 x float> undef)
call <4 x i64> @llvm.lrint.v4i64.v4f32(<4 x float> undef)
@@ -416,27 +516,47 @@ define void @lrint_fp16() {
define void @llrint() {
; CHECK-LABEL: 'llrint'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %1 = call i64 @llvm.llrint.i64.f32(float undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %2 = call <2 x i64> @llvm.llrint.v2i64.v2f32(<2 x float> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %3 = call <4 x i64> @llvm.llrint.v4i64.v4f32(<4 x float> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %4 = call <8 x i64> @llvm.llrint.v8i64.v8f32(<8 x float> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %5 = call <16 x i64> @llvm.llrint.v16i64.v16f32(<16 x float> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %6 = call <vscale x 1 x i64> @llvm.llrint.nxv1i64.nxv1f32(<vscale x 1 x float> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %7 = call <vscale x 2 x i64> @llvm.llrint.nxv2i64.nxv2f32(<vscale x 2 x float> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %8 = call <vscale x 4 x i64> @llvm.llrint.nxv4i64.nxv4f32(<vscale x 4 x float> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %9 = call <vscale x 8 x i64> @llvm.llrint.nxv8i64.nxv8f32(<vscale x 8 x float> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %10 = call <vscale x 16 x i64> @llvm.llrint.nxv16i64.nxv16f32(<vscale x 16 x float> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %11 = call i64 @llvm.llrint.i64.f64(double undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %12 = call <2 x i64> @llvm.llrint.v2i64.v2f64(<2 x double> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %13 = call <4 x i64> @llvm.llrint.v4i64.v4f64(<4 x double> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %14 = call <8 x i64> @llvm.llrint.v8i64.v8f64(<8 x double> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %15 = call <16 x i64> @llvm.llrint.v16i64.v16f64(<16 x double> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %16 = call <vscale x 1 x i64> @llvm.llrint.nxv1i64.nxv1f64(<vscale x 1 x double> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %17 = call <vscale x 2 x i64> @llvm.llrint.nxv2i64.nxv2f64(<vscale x 2 x double> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %18 = call <vscale x 4 x i64> @llvm.llrint.nxv4i64.nxv4f64(<vscale x 4 x double> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %19 = call <vscale x 8 x i64> @llvm.llrint.nxv8i64.nxv8f64(<vscale x 8 x double> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %1 = call i64 @llvm.llrint.i64.bf16(bfloat undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %2 = call <2 x i64> @llvm.llrint.v2i64.v2bf16(<2 x bfloat> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %3 = call <4 x i64> @llvm.llrint.v4i64.v4bf16(<4 x bfloat> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %4 = call <8 x i64> @llvm.llrint.v8i64.v8bf16(<8 x bfloat> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %5 = call <16 x i64> @llvm.llrint.v16i64.v16bf16(<16 x bfloat> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %6 = call <vscale x 1 x i64> @llvm.llrint.nxv1i64.nxv1bf16(<vscale x 1 x bfloat> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %7 = call <vscale x 2 x i64> @llvm.llrint.nxv2i64.nxv2bf16(<vscale x 2 x bfloat> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %8 = call <vscale x 4 x i64> @llvm.llrint.nxv4i64.nxv4bf16(<vscale x 4 x bfloat> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %9 = call <vscale x 8 x i64> @llvm.llrint.nxv8i64.nxv8bf16(<vscale x 8 x bfloat> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %10 = call <vscale x 16 x i64> @llvm.llrint.nxv16i64.nxv16bf16(<vscale x 16 x bfloat> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %11 = call i64 @llvm.llrint.i64.f32(float undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %12 = call <2 x i64> @llvm.llrint.v2i64.v2f32(<2 x float> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %13 = call <4 x i64> @llvm.llrint.v4i64.v4f32(<4 x float> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %14 = call <8 x i64> @llvm.llrint.v8i64.v8f32(<8 x float> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %15 = call <16 x i64> @llvm.llrint.v16i64.v16f32(<16 x float> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %16 = call <vscale x 1 x i64> @llvm.llrint.nxv1i64.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 i64> @llvm.llrint.nxv2i64.nxv2f32(<vscale x 2 x float> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %18 = call <vscale x 4 x i64> @llvm.llrint.nxv4i64.nxv4f32(<vscale x 4 x float> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %19 = call <vscale x 8 x i64> @llvm.llrint.nxv8i64.nxv8f32(<vscale x 8 x float> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %20 = call <vscale x 16 x i64> @llvm.llrint.nxv16i64.nxv16f32(<vscale x 16 x float> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %21 = call i64 @llvm.llrint.i64.f64(double undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %22 = call <2 x i64> @llvm.llrint.v2i64.v2f64(<2 x double> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %23 = call <4 x i64> @llvm.llrint.v4i64.v4f64(<4 x double> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %24 = call <8 x i64> @llvm.llrint.v8i64.v8f64(<8 x double> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %25 = call <16 x i64> @llvm.llrint.v16i64.v16f64(<16 x double> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %26 = call <vscale x 1 x i64> @llvm.llrint.nxv1i64.nxv1f64(<vscale x 1 x double> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %27 = call <vscale x 2 x i64> @llvm.llrint.nxv2i64.nxv2f64(<vscale x 2 x double> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %28 = call <vscale x 4 x i64> @llvm.llrint.nxv4i64.nxv4f64(<vscale x 4 x double> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %29 = call <vscale x 8 x i64> @llvm.llrint.nxv8i64.nxv8f64(<vscale x 8 x double> undef)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret void
;
+ call i64 @llvm.llrint.i64.bf16(bfloat undef)
+ call <2 x i64> @llvm.llrint.v2i64.v2bf16(<2 x bfloat> undef)
+ call <4 x i64> @llvm.llrint.v4i64.v4bf16(<4 x bfloat> undef)
+ call <8 x i64> @llvm.llrint.v8i64.v8bf16(<8 x bfloat> undef)
+ call <16 x i64> @llvm.llrint.v16i64.v16bf16(<16 x bfloat> undef)
+ call <vscale x 1 x i64> @llvm.llrint.nvx1i64.nvx1bf16(<vscale x 1 x bfloat> undef)
+ call <vscale x 2 x i64> @llvm.llrint.nvx2i64.nvx2bf16(<vscale x 2 x bfloat> undef)
+ call <vscale x 4 x i64> @llvm.llrint.nvx4i64.nvx4bf16(<vscale x 4 x bfloat> undef)
+ call <vscale x 8 x i64> @llvm.llrint.nvx8i64.nvx8bf16(<vscale x 8 x bfloat> undef)
+ call <vscale x 16 x i64> @llvm.llrint.nvx16i64.nvx16bf16(<vscale x 16 x bfloat> undef)
call i64 @llvm.llrint.i64.f32(float undef)
call <2 x i64> @llvm.llrint.v2i64.v2f32(<2 x float> undef)
call <4 x i64> @llvm.llrint.v4i64.v4f32(<4 x float> undef)
@@ -488,27 +608,47 @@ define void @llrint_fp16() {
define void @nearbyint() {
; CHECK-LABEL: 'nearbyint'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %1 = call float @llvm.nearbyint.f32(float undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %2 = call <2 x float> @llvm.nearbyint.v2f32(<2 x float> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %3 = call <4 x float> @llvm.nearbyint.v4f32(<4 x float> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %4 = call <8 x float> @llvm.nearbyint.v8f32(<8 x float> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %5 = call <16 x float> @llvm.nearbyint.v16f32(<16 x float> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %6 = call <vscale x 1 x float> @llvm.nearbyint.nxv1f32(<vscale x 1 x float> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %7 = call <vscale x 2 x float> @llvm.nearbyint.nxv2f32(<vscale x 2 x float> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %8 = call <vscale x 4 x float> @llvm.nearbyint.nxv4f32(<vscale x 4 x float> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %9 = call <vscale x 8 x float> @llvm.nearbyint.nxv8f32(<vscale x 8 x float> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %10 = call <vscale x 16 x float> @llvm.nearbyint.nxv16f32(<vscale x 16 x float> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %11 = call double @llvm.nearbyint.f64(double undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %12 = call <2 x double> @llvm.nearbyint.v2f64(<2 x double> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %13 = call <4 x double> @llvm.nearbyint.v4f64(<4 x double> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %14 = call <8 x double> @llvm.nearbyint.v8f64(<8 x double> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %15 = call <16 x double> @llvm.nearbyint.v16f64(<16 x double> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %16 = call <vscale x 1 x double> @llvm.nearbyint.nxv1f64(<vscale x 1 x double> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %17 = call <vscale x 2 x double> @llvm.nearbyint.nxv2f64(<vscale x 2 x double> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %18 = call <vscale x 4 x double> @llvm.nearbyint.nxv4f64(<vscale x 4 x double> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %19 = call <vscale x 8 x double> @llvm.nearbyint.nxv8f64(<vscale x 8 x double> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %1 = call bfloat @llvm.nearbyint.bf16(bfloat undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %2 = call <2 x bfloat> @llvm.nearbyint.v2bf16(<2 x bfloat> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %3 = call <4 x bfloat> @llvm.nearbyint.v4bf16(<4 x bfloat> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %4 = call <8 x bfloat> @llvm.nearbyint.v8bf16(<8 x bfloat> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %5 = call <16 x bfloat> @llvm.nearbyint.v16bf16(<16 x bfloat> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %6 = call <vscale x 1 x bfloat> @llvm.nearbyint.nxv1bf16(<vscale x 1 x bfloat> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %7 = call <vscale x 2 x bfloat> @llvm.nearbyint.nxv2bf16(<vscale x 2 x bfloat> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %8 = call <vscale x 4 x bfloat> @llvm.nearbyint.nxv4bf16(<vscale x 4 x bfloat> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %9 = call <vscale x 8 x bfloat> @llvm.nearbyint.nxv8bf16(<vscale x 8 x bfloat> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %10 = call <vscale x 16 x bfloat> @llvm.nearbyint.nxv16bf16(<vscale x 16 x bfloat> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %11 = call float @llvm.nearbyint.f32(float undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %12 = call <2 x float> @llvm.nearbyint.v2f32(<2 x float> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %13 = call <4 x float> @llvm.nearbyint.v4f32(<4 x float> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %14 = call <8 x float> @llvm.nearbyint.v8f32(<8 x float> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %15 = call <16 x float> @llvm.nearbyint.v16f32(<16 x float> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %16 = call <vscale x 1 x float> @llvm.nearbyint.nxv1f32(<vscale x 1 x float> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %17 = call <vscale x 2 x float> @llvm.nearbyint.nxv2f32(<vscale x 2 x float> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %18 = call <vscale x 4 x float> @llvm.nearbyint.nxv4f32(<vscale x 4 x float> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %19 = call <vscale x 8 x float> @llvm.nearbyint.nxv8f32(<vscale x 8 x float> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %20 = call <vscale x 16 x float> @llvm.nearbyint.nxv16f32(<vscale x 16 x float> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %21 = call double @llvm.nearbyint.f64(double undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %22 = call <2 x double> @llvm.nearbyint.v2f64(<2 x double> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %23 = call <4 x double> @llvm.nearbyint.v4f64(<4 x double> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %24 = call <8 x double> @llvm.nearbyint.v8f64(<8 x double> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %25 = call <16 x double> @llvm.nearbyint.v16f64(<16 x double> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %26 = call <vscale x 1 x double> @llvm.nearbyint.nxv1f64(<vscale x 1 x double> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %27 = call <vscale x 2 x double> @llvm.nearbyint.nxv2f64(<vscale x 2 x double> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %28 = call <vscale x 4 x double> @llvm.nearbyint.nxv4f64(<vscale x 4 x double> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %29 = call <vscale x 8 x double> @llvm.nearbyint.nxv8f64(<vscale x 8 x double> undef)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret void
;
+ call bfloat @llvm.nearbyint.bf16(bfloat undef)
+ call <2 x bfloat> @llvm.nearbyint.v2bf16(<2 x bfloat> undef)
+ call <4 x bfloat> @llvm.nearbyint.v4bf16(<4 x bfloat> undef)
+ call <8 x bfloat> @llvm.nearbyint.v8bf16(<8 x bfloat> undef)
+ call <16 x bfloat> @llvm.nearbyint.v16bf16(<16 x bfloat> undef)
+ call <vscale x 1 x bfloat> @llvm.nearbyint.nvx1bf16(<vscale x 1 x bfloat> undef)
+ call <vscale x 2 x bfloat> @llvm.nearbyint.nvx2bf16(<vscale x 2 x bfloat> undef)
+ call <vscale x 4 x bfloat> @llvm.nearbyint.nvx4bf16(<vscale x 4 x bfloat> undef)
+ call <vscale x 8 x bfloat> @llvm.nearbyint.nvx8bf16(<vscale x 8 x bfloat> undef)
+ call <vscale x 16 x bfloat> @llvm.nearbyint.nvx16bf16(<vscale x 16 x bfloat> undef)
call float @llvm.nearbyint.f32(float undef)
call <2 x float> @llvm.nearbyint.v2f32(<2 x float> undef)
call <4 x float> @llvm.nearbyint.v4f32(<4 x float> undef)
@@ -573,27 +713,47 @@ define void @nearbyint_fp16() {
define void @round() {
; CHECK-LABEL: 'round'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %1 = call float @llvm.round.f32(float undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %2 = call <2 x float> @llvm.round.v2f32(<2 x float> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %3 = call <4 x float> @llvm.round.v4f32(<4 x float> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %4 = call <8 x float> @llvm.round.v8f32(<8 x float> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %5 = call <16 x float> @llvm.round.v16f32(<16 x float> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %6 = call <vscale x 1 x float> @llvm.round.nxv1f32(<vscale x 1 x float> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %7 = call <vscale x 2 x float> @llvm.round.nxv2f32(<vscale x 2 x float> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %8 = call <vscale x 4 x float> @llvm.round.nxv4f32(<vscale x 4 x float> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %9 = call <vscale x 8 x float> @llvm.round.nxv8f32(<vscale x 8 x float> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %10 = call <vscale x 16 x float> @llvm.round.nxv16f32(<vscale x 16 x float> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %11 = call double @llvm.round.f64(double undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %12 = call <2 x double> @llvm.round.v2f64(<2 x double> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %13 = call <4 x double> @llvm.round.v4f64(<4 x double> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %14 = call <8 x double> @llvm.round.v8f64(<8 x double> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %15 = call <16 x double> @llvm.round.v16f64(<16 x double> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %16 = call <vscale x 1 x double> @llvm.round.nxv1f64(<vscale x 1 x double> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %17 = call <vscale x 2 x double> @llvm.round.nxv2f64(<vscale x 2 x double> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %18 = call <vscale x 4 x double> @llvm.round.nxv4f64(<vscale x 4 x double> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %19 = call <vscale x 8 x double> @llvm.round.nxv8f64(<vscale x 8 x double> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %1 = call bfloat @llvm.round.bf16(bfloat undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %2 = call <2 x bfloat> @llvm.round.v2bf16(<2 x bfloat> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %3 = call <4 x bfloat> @llvm.round.v4bf16(<4 x bfloat> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %4 = call <8 x bfloat> @llvm.round.v8bf16(<8 x bfloat> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %5 = call <16 x bfloat> @llvm.round.v16bf16(<16 x bfloat> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %6 = call <vscale x 1 x bfloat> @llvm.round.nxv1bf16(<vscale x 1 x bfloat> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %7 = call <vscale x 2 x bfloat> @llvm.round.nxv2bf16(<vscale x 2 x bfloat> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %8 = call <vscale x 4 x bfloat> @llvm.round.nxv4bf16(<vscale x 4 x bfloat> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %9 = call <vscale x 8 x bfloat> @llvm.round.nxv8bf16(<vscale x 8 x bfloat> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %10 = call <vscale x 16 x bfloat> @llvm.round.nxv16bf16(<vscale x 16 x bfloat> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %11 = call float @llvm.round.f32(float undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %12 = call <2 x float> @llvm.round.v2f32(<2 x float> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %13 = call <4 x float> @llvm.round.v4f32(<4 x float> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %14 = call <8 x float> @llvm.round.v8f32(<8 x float> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %15 = call <16 x float> @llvm.round.v16f32(<16 x float> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %16 = call <vscale x 1 x float> @llvm.round.nxv1f32(<vscale x 1 x float> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %17 = call <vscale x 2 x float> @llvm.round.nxv2f32(<vscale x 2 x float> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %18 = call <vscale x 4 x float> @llvm.round.nxv4f32(<vscale x 4 x float> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %19 = call <vscale x 8 x float> @llvm.round.nxv8f32(<vscale x 8 x float> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %20 = call <vscale x 16 x float> @llvm.round.nxv16f32(<vscale x 16 x float> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %21 = call double @llvm.round.f64(double undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %22 = call <2 x double> @llvm.round.v2f64(<2 x double> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %23 = call <4 x double> @llvm.round.v4f64(<4 x double> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %24 = call <8 x double> @llvm.round.v8f64(<8 x double> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %25 = call <16 x double> @llvm.round.v16f64(<16 x double> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %26 = call <vscale x 1 x double> @llvm.round.nxv1f64(<vscale x 1 x double> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %27 = call <vscale x 2 x double> @llvm.round.nxv2f64(<vscale x 2 x double> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %28 = call <vscale x 4 x double> @llvm.round.nxv4f64(<vscale x 4 x double> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %29 = call <vscale x 8 x double> @llvm.round.nxv8f64(<vscale x 8 x double> undef)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret void
;
+ call bfloat @llvm.round.bf16(bfloat undef)
+ call <2 x bfloat> @llvm.round.v2bf16(<2 x bfloat> undef)
+ call <4 x bfloat> @llvm.round.v4bf16(<4 x bfloat> undef)
+ call <8 x bfloat> @llvm.round.v8bf16(<8 x bfloat> undef)
+ call <16 x bfloat> @llvm.round.v16bf16(<16 x bfloat> undef)
+ call <vscale x 1 x bfloat> @llvm.round.nvx1bf16(<vscale x 1 x bfloat> undef)
+ call <vscale x 2 x bfloat> @llvm.round.nvx2bf16(<vscale x 2 x bfloat> undef)
+ call <vscale x 4 x bfloat> @llvm.round.nvx4bf16(<vscale x 4 x bfloat> undef)
+ call <vscale x 8 x bfloat> @llvm.round.nvx8bf16(<vscale x 8 x bfloat> undef)
+ call <vscale x 16 x bfloat> @llvm.round.nvx16bf16(<vscale x 16 x bfloat> undef)
call float @llvm.round.f32(float undef)
call <2 x float> @llvm.round.v2f32(<2 x float> undef)
call <4 x float> @llvm.round.v4f32(<4 x float> undef)
@@ -658,27 +818,47 @@ define void @round_fp16() {
define void @roundeven() {
; CHECK-LABEL: 'roundeven'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %1 = call float @llvm.roundeven.f32(float undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %2 = call <2 x float> @llvm.roundeven.v2f32(<2 x float> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %3 = call <4 x float> @llvm.roundeven.v4f32(<4 x float> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %4 = call <8 x float> @llvm.roundeven.v8f32(<8 x float> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %5 = call <16 x float> @llvm.roundeven.v16f32(<16 x float> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %6 = call <vscale x 1 x float> @llvm.roundeven.nxv1f32(<vscale x 1 x float> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %7 = call <vscale x 2 x float> @llvm.roundeven.nxv2f32(<vscale x 2 x float> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %8 = call <vscale x 4 x float> @llvm.roundeven.nxv4f32(<vscale x 4 x float> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %9 = call <vscale x 8 x float> @llvm.roundeven.nxv8f32(<vscale x 8 x float> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %10 = call <vscale x 16 x float> @llvm.roundeven.nxv16f32(<vscale x 16 x float> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %11 = call double @llvm.roundeven.f64(double undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %12 = call <2 x double> @llvm.roundeven.v2f64(<2 x double> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %13 = call <4 x double> @llvm.roundeven.v4f64(<4 x double> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %14 = call <8 x double> @llvm.roundeven.v8f64(<8 x double> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %15 = call <16 x double> @llvm.roundeven.v16f64(<16 x double> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %16 = call <vscale x 1 x double> @llvm.roundeven.nxv1f64(<vscale x 1 x double> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %17 = call <vscale x 2 x double> @llvm.roundeven.nxv2f64(<vscale x 2 x double> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %18 = call <vscale x 4 x double> @llvm.roundeven.nxv4f64(<vscale x 4 x double> undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %19 = call <vscale x 8 x double> @llvm.roundeven.nxv8f64(<vscale x 8 x double> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %1 = call bfloat @llvm.roundeven.bf16(bfloat undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %2 = call <2 x bfloat> @llvm.roundeven.v2bf16(<2 x bfloat> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %3 = call <4 x bfloat> @llvm.roundeven.v4bf16(<4 x bfloat> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %4 = call <8 x bfloat> @llvm.roundeven.v8bf16(<8 x bfloat> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %5 = call <16 x bfloat> @llvm.roundeven.v16bf16(<16 x bfloat> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %6 = call <vscale x 1 x bfloat> @llvm.roundeven.nxv1bf16(<vscale x 1 x bfloat> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %7 = call <vscale x 2 x bfloat> @llvm.roundeven.nxv2bf16(<vscale x 2 x bfloat> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %8 = call <vscale x 4 x bfloat> @llvm.roundeven.nxv4bf16(<vscale x 4 x bfloat> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %9 = call <vscale x 8 x bfloat> @llvm.roundeven.nxv8bf16(<vscale x 8 x bfloat> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %10 = call <vscale x 16 x bfloat> @llvm.roundeven.nxv16bf16(<vscale x 16 x bfloat> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %11 = call float @llvm.roundeven.f32(float undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %12 = call <2 x float> @llvm.roundeven.v2f32(<2 x float> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %13 = call <4 x float> @llvm.roundeven.v4f32(<4 x float> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %14 = call <8 x float> @llvm.roundeven.v8f32(<8 x float> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %15 = call <16 x float> @llvm.roundeven.v16f32(<16 x float> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %16 = call <vscale x 1 x float> @llvm.roundeven.nxv1f32(<vscale x 1 x float> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %17 = call <vscale x 2 x float> @llvm.roundeven.nxv2f32(<vscale x 2 x float> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %18 = call <vscale x 4 x float> @llvm.roundeven.nxv4f32(<vscale x 4 x float> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %19 = call <vscale x 8 x float> @llvm.roundeven.nxv8f32(<vscale x 8 x float> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %20 = call <vscale x 16 x float> @llvm.roundeven.nxv16f32(<vscale x 16 x float> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %21 = call double @llvm.roundeven.f64(double undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %22 = call <2 x double> @llvm.roundeven.v2f64(<2 x double> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %23 = call <4 x double> @llvm.roundeven.v4f64(<4 x double> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %24 = call <8 x double> @llvm.roundeven.v8f64(<8 x double> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %25 = call <16 x double> @llvm.roundeven.v16f64(<16 x double> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %26 = call <vscale x 1 x double> @llvm.roundeven.nxv1f64(<vscale x 1 x double> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %27 = call <vscale x 2 x double> @llvm.roundeven.nxv2f64(<vscale x 2 x double> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %28 = call <vscale x 4 x double> @llvm.roundeven.nxv4f64(<vscale x 4 x double> undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %29 = call <vscale x 8 x double> @llvm.roundeven.nxv8f64(<vscale x 8 x double> undef)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret void
;
+ call bfloat @llvm.roundeven.bf16(bfloat undef)
+ call <2 x bfloat> @llvm.roundeven.v2bf16(<2 x bfloat> undef)
+ call <4 x bfloat> @llvm.roundeven.v4bf16(<4 x bfloat> undef)
+ call <8 x bfloat> @llvm.roundeven.v8bf16(<8 x bfloat> undef)
+ call <16 x bfloat> @llvm.roundeven.v16bf16(<16 x bfloat> undef)
+ call <vscale x 1 x bfloat> @llvm.roundeven.nvx1bf16(<vscale x 1 x bfloat> undef)
+ call <vscale x 2 x bfloat> @llvm.roundeven.nvx2bf16(<vscale x 2 x bfloat> undef)
+ call <vscale x 4 x bfloat> @llvm.roundeven.nvx4bf16(<vscale x 4 x bfloat> undef)
+ call <vscale x 8 x bfloat> @llvm.roundeven.nvx8bf16(<vscale x 8 x bfloat> undef)
+ call <vscale x 16 x bfloat> @llvm.roundeven.nvx16bf16(<vscale x 16 x bfloat> undef)
call float @llvm.roundeven.f32(float undef)
call <2 x float> @llvm.roundeven.v2f32(<2 x float> undef)
call <4 x float> @llvm.roundeven.v4f32(<4 x float> undef)
@@ -743,25 +923,43 @@ define void @roundeven_fp16() {
define void @vp_ceil() {
; CHECK-LABEL: 'vp_ceil'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %1 = call <2 x float> @llvm.vp.ceil.v2f32(<2 x float> undef, <2 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %2 = call <4 x float> @llvm.vp.ceil.v4f32(<4 x float> undef, <4 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %3 = call <8 x float> @llvm.vp.ceil.v8f32(<8 x float> undef, <8 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %4 = call <16 x float> @llvm.vp.ceil.v16f32(<16 x float> undef, <16 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %5 = call <vscale x 1 x float> @llvm.vp.ceil.nxv1f32(<vscale x 1 x float> undef, <vscale x 1 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %6 = call <vscale x 2 x float> @llvm.vp.ceil.nxv2f32(<vscale x 2 x float> undef, <vscale x 2 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %7 = call <vscale x 4 x float> @llvm.vp.ceil.nxv4f32(<vscale x 4 x float> undef, <vscale x 4 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %8 = call <vscale x 8 x float> @llvm.vp.ceil.nxv8f32(<vscale x 8 x float> undef, <vscale x 8 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %9 = call <vscale x 16 x float> @llvm.vp.ceil.nxv16f32(<vscale x 16 x float> undef, <vscale x 16 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %10 = call <2 x double> @llvm.vp.ceil.v2f64(<2 x double> undef, <2 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %11 = call <4 x double> @llvm.vp.ceil.v4f64(<4 x double> undef, <4 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %12 = call <8 x double> @llvm.vp.ceil.v8f64(<8 x double> undef, <8 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %13 = call <16 x double> @llvm.vp.ceil.v16f64(<16 x double> undef, <16 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %14 = call <vscale x 1 x double> @llvm.vp.ceil.nxv1f64(<vscale x 1 x double> undef, <vscale x 1 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %15 = call <vscale x 2 x double> @llvm.vp.ceil.nxv2f64(<vscale x 2 x double> undef, <vscale x 2 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %16 = call <vscale x 4 x double> @llvm.vp.ceil.nxv4f64(<vscale x 4 x double> undef, <vscale x 4 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %17 = call <vscale x 8 x double> @llvm.vp.ceil.nxv8f64(<vscale x 8 x double> undef, <vscale x 8 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %1 = call <2 x bfloat> @llvm.vp.ceil.v2bf16(<2 x bfloat> undef, <2 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %2 = call <4 x bfloat> @llvm.vp.ceil.v4bf16(<4 x bfloat> undef, <4 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %3 = call <8 x bfloat> @llvm.vp.ceil.v8bf16(<8 x bfloat> undef, <8 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %4 = call <16 x bfloat> @llvm.vp.ceil.v16bf16(<16 x bfloat> undef, <16 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %5 = call <vscale x 1 x bfloat> @llvm.vp.ceil.nxv1bf16(<vscale x 1 x bfloat> undef, <vscale x 1 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %6 = call <vscale x 2 x bfloat> @llvm.vp.ceil.nxv2bf16(<vscale x 2 x bfloat> undef, <vscale x 2 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %7 = call <vscale x 4 x bfloat> @llvm.vp.ceil.nxv4bf16(<vscale x 4 x bfloat> undef, <vscale x 4 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %8 = call <vscale x 8 x bfloat> @llvm.vp.ceil.nxv8bf16(<vscale x 8 x bfloat> undef, <vscale x 8 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %9 = call <vscale x 16 x bfloat> @llvm.vp.ceil.nxv16bf16(<vscale x 16 x bfloat> undef, <vscale x 16 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %10 = call <2 x float> @llvm.vp.ceil.v2f32(<2 x float> undef, <2 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %11 = call <4 x float> @llvm.vp.ceil.v4f32(<4 x float> undef, <4 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %12 = call <8 x float> @llvm.vp.ceil.v8f32(<8 x float> undef, <8 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %13 = call <16 x float> @llvm.vp.ceil.v16f32(<16 x float> undef, <16 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %14 = call <vscale x 1 x float> @llvm.vp.ceil.nxv1f32(<vscale x 1 x float> undef, <vscale x 1 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %15 = call <vscale x 2 x float> @llvm.vp.ceil.nxv2f32(<vscale x 2 x float> undef, <vscale x 2 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %16 = call <vscale x 4 x float> @llvm.vp.ceil.nxv4f32(<vscale x 4 x float> undef, <vscale x 4 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %17 = call <vscale x 8 x float> @llvm.vp.ceil.nxv8f32(<vscale x 8 x float> undef, <vscale x 8 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %18 = call <vscale x 16 x float> @llvm.vp.ceil.nxv16f32(<vscale x 16 x float> undef, <vscale x 16 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %19 = call <2 x double> @llvm.vp.ceil.v2f64(<2 x double> undef, <2 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %20 = call <4 x double> @llvm.vp.ceil.v4f64(<4 x double> undef, <4 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %21 = call <8 x double> @llvm.vp.ceil.v8f64(<8 x double> undef, <8 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %22 = call <16 x double> @llvm.vp.ceil.v16f64(<16 x double> undef, <16 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %23 = call <vscale x 1 x double> @llvm.vp.ceil.nxv1f64(<vscale x 1 x double> undef, <vscale x 1 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %24 = call <vscale x 2 x double> @llvm.vp.ceil.nxv2f64(<vscale x 2 x double> undef, <vscale x 2 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %25 = call <vscale x 4 x double> @llvm.vp.ceil.nxv4f64(<vscale x 4 x double> undef, <vscale x 4 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %26 = call <vscale x 8 x double> @llvm.vp.ceil.nxv8f64(<vscale x 8 x double> undef, <vscale x 8 x i1> undef, i32 undef)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret void
;
+ call <2 x bfloat> @llvm.vp.ceil.v2bf16(<2 x bfloat> undef, <2 x i1> undef, i32 undef)
+ call <4 x bfloat> @llvm.vp.ceil.v4bf16(<4 x bfloat> undef, <4 x i1> undef, i32 undef)
+ call <8 x bfloat> @llvm.vp.ceil.v8bf16(<8 x bfloat> undef, <8 x i1> undef, i32 undef)
+ call <16 x bfloat> @llvm.vp.ceil.v16bf16(<16 x bfloat> undef, <16 x i1> undef, i32 undef)
+ call <vscale x 1 x bfloat> @llvm.vp.ceil.nvx1bf16(<vscale x 1 x bfloat> undef, <vscale x 1 x i1> undef, i32 undef)
+ call <vscale x 2 x bfloat> @llvm.vp.ceil.nvx2bf16(<vscale x 2 x bfloat> undef, <vscale x 2 x i1> undef, i32 undef)
+ call <vscale x 4 x bfloat> @llvm.vp.ceil.nvx4bf16(<vscale x 4 x bfloat> undef, <vscale x 4 x i1> undef, i32 undef)
+ call <vscale x 8 x bfloat> @llvm.vp.ceil.nvx8bf16(<vscale x 8 x bfloat> undef, <vscale x 8 x i1> undef, i32 undef)
+ call <vscale x 16 x bfloat> @llvm.vp.ceil.nvx16bf16(<vscale x 16 x bfloat> undef, <vscale x 16 x i1> undef, i32 undef)
call <2 x float> @llvm.vp.ceil.v2f32(<2 x float> undef, <2 x i1> undef, i32 undef)
call <4 x float> @llvm.vp.ceil.v4f32(<4 x float> undef, <4 x i1> undef, i32 undef)
call <8 x float> @llvm.vp.ceil.v8f32(<8 x float> undef, <8 x i1> undef, i32 undef)
@@ -821,25 +1019,43 @@ define void @vp_ceil_f16() {
define void @vp_floor() {
; CHECK-LABEL: 'vp_floor'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %1 = call <2 x float> @llvm.vp.floor.v2f32(<2 x float> undef, <2 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %2 = call <4 x float> @llvm.vp.floor.v4f32(<4 x float> undef, <4 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %3 = call <8 x float> @llvm.vp.floor.v8f32(<8 x float> undef, <8 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %4 = call <16 x float> @llvm.vp.floor.v16f32(<16 x float> undef, <16 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %5 = call <vscale x 1 x float> @llvm.vp.floor.nxv1f32(<vscale x 1 x float> undef, <vscale x 1 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %6 = call <vscale x 2 x float> @llvm.vp.floor.nxv2f32(<vscale x 2 x float> undef, <vscale x 2 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %7 = call <vscale x 4 x float> @llvm.vp.floor.nxv4f32(<vscale x 4 x float> undef, <vscale x 4 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %8 = call <vscale x 8 x float> @llvm.vp.floor.nxv8f32(<vscale x 8 x float> undef, <vscale x 8 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %9 = call <vscale x 16 x float> @llvm.vp.floor.nxv16f32(<vscale x 16 x float> undef, <vscale x 16 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %10 = call <2 x double> @llvm.vp.floor.v2f64(<2 x double> undef, <2 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %11 = call <4 x double> @llvm.vp.floor.v4f64(<4 x double> undef, <4 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %12 = call <8 x double> @llvm.vp.floor.v8f64(<8 x double> undef, <8 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %13 = call <16 x double> @llvm.vp.floor.v16f64(<16 x double> undef, <16 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %14 = call <vscale x 1 x double> @llvm.vp.floor.nxv1f64(<vscale x 1 x double> undef, <vscale x 1 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %15 = call <vscale x 2 x double> @llvm.vp.floor.nxv2f64(<vscale x 2 x double> undef, <vscale x 2 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %16 = call <vscale x 4 x double> @llvm.vp.floor.nxv4f64(<vscale x 4 x double> undef, <vscale x 4 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %17 = call <vscale x 8 x double> @llvm.vp.floor.nxv8f64(<vscale x 8 x double> undef, <vscale x 8 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %1 = call <2 x bfloat> @llvm.vp.floor.v2bf16(<2 x bfloat> undef, <2 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %2 = call <4 x bfloat> @llvm.vp.floor.v4bf16(<4 x bfloat> undef, <4 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %3 = call <8 x bfloat> @llvm.vp.floor.v8bf16(<8 x bfloat> undef, <8 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %4 = call <16 x bfloat> @llvm.vp.floor.v16bf16(<16 x bfloat> undef, <16 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %5 = call <vscale x 1 x bfloat> @llvm.vp.floor.nxv1bf16(<vscale x 1 x bfloat> undef, <vscale x 1 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %6 = call <vscale x 2 x bfloat> @llvm.vp.floor.nxv2bf16(<vscale x 2 x bfloat> undef, <vscale x 2 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %7 = call <vscale x 4 x bfloat> @llvm.vp.floor.nxv4bf16(<vscale x 4 x bfloat> undef, <vscale x 4 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %8 = call <vscale x 8 x bfloat> @llvm.vp.floor.nxv8bf16(<vscale x 8 x bfloat> undef, <vscale x 8 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %9 = call <vscale x 16 x bfloat> @llvm.vp.floor.nxv16bf16(<vscale x 16 x bfloat> undef, <vscale x 16 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %10 = call <2 x float> @llvm.vp.floor.v2f32(<2 x float> undef, <2 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %11 = call <4 x float> @llvm.vp.floor.v4f32(<4 x float> undef, <4 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %12 = call <8 x float> @llvm.vp.floor.v8f32(<8 x float> undef, <8 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %13 = call <16 x float> @llvm.vp.floor.v16f32(<16 x float> undef, <16 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %14 = call <vscale x 1 x float> @llvm.vp.floor.nxv1f32(<vscale x 1 x float> undef, <vscale x 1 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %15 = call <vscale x 2 x float> @llvm.vp.floor.nxv2f32(<vscale x 2 x float> undef, <vscale x 2 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %16 = call <vscale x 4 x float> @llvm.vp.floor.nxv4f32(<vscale x 4 x float> undef, <vscale x 4 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %17 = call <vscale x 8 x float> @llvm.vp.floor.nxv8f32(<vscale x 8 x float> undef, <vscale x 8 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %18 = call <vscale x 16 x float> @llvm.vp.floor.nxv16f32(<vscale x 16 x float> undef, <vscale x 16 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %19 = call <2 x double> @llvm.vp.floor.v2f64(<2 x double> undef, <2 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %20 = call <4 x double> @llvm.vp.floor.v4f64(<4 x double> undef, <4 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %21 = call <8 x double> @llvm.vp.floor.v8f64(<8 x double> undef, <8 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %22 = call <16 x double> @llvm.vp.floor.v16f64(<16 x double> undef, <16 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %23 = call <vscale x 1 x double> @llvm.vp.floor.nxv1f64(<vscale x 1 x double> undef, <vscale x 1 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %24 = call <vscale x 2 x double> @llvm.vp.floor.nxv2f64(<vscale x 2 x double> undef, <vscale x 2 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %25 = call <vscale x 4 x double> @llvm.vp.floor.nxv4f64(<vscale x 4 x double> undef, <vscale x 4 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %26 = call <vscale x 8 x double> @llvm.vp.floor.nxv8f64(<vscale x 8 x double> undef, <vscale x 8 x i1> undef, i32 undef)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret void
;
+ call <2 x bfloat> @llvm.vp.floor.v2bf16(<2 x bfloat> undef, <2 x i1> undef, i32 undef)
+ call <4 x bfloat> @llvm.vp.floor.v4bf16(<4 x bfloat> undef, <4 x i1> undef, i32 undef)
+ call <8 x bfloat> @llvm.vp.floor.v8bf16(<8 x bfloat> undef, <8 x i1> undef, i32 undef)
+ call <16 x bfloat> @llvm.vp.floor.v16bf16(<16 x bfloat> undef, <16 x i1> undef, i32 undef)
+ call <vscale x 1 x bfloat> @llvm.vp.floor.nvx1bf16(<vscale x 1 x bfloat> undef, <vscale x 1 x i1> undef, i32 undef)
+ call <vscale x 2 x bfloat> @llvm.vp.floor.nvx2bf16(<vscale x 2 x bfloat> undef, <vscale x 2 x i1> undef, i32 undef)
+ call <vscale x 4 x bfloat> @llvm.vp.floor.nvx4bf16(<vscale x 4 x bfloat> undef, <vscale x 4 x i1> undef, i32 undef)
+ call <vscale x 8 x bfloat> @llvm.vp.floor.nvx8bf16(<vscale x 8 x bfloat> undef, <vscale x 8 x i1> undef, i32 undef)
+ call <vscale x 16 x bfloat> @llvm.vp.floor.nvx16bf16(<vscale x 16 x bfloat> undef, <vscale x 16 x i1> undef, i32 undef)
call <2 x float> @llvm.vp.floor.v2f32(<2 x float> undef, <2 x i1> undef, i32 undef)
call <4 x float> @llvm.vp.floor.v4f32(<4 x float> undef, <4 x i1> undef, i32 undef)
call <8 x float> @llvm.vp.floor.v8f32(<8 x float> undef, <8 x i1> undef, i32 undef)
@@ -899,25 +1115,43 @@ define void @vp_floor_f16() {
define void @vp_round() {
; CHECK-LABEL: 'vp_round'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %1 = call <2 x float> @llvm.vp.round.v2f32(<2 x float> undef, <2 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %2 = call <4 x float> @llvm.vp.round.v4f32(<4 x float> undef, <4 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %3 = call <8 x float> @llvm.vp.round.v8f32(<8 x float> undef, <8 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %4 = call <16 x float> @llvm.vp.round.v16f32(<16 x float> undef, <16 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %5 = call <vscale x 1 x float> @llvm.vp.round.nxv1f32(<vscale x 1 x float> undef, <vscale x 1 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %6 = call <vscale x 2 x float> @llvm.vp.round.nxv2f32(<vscale x 2 x float> undef, <vscale x 2 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %7 = call <vscale x 4 x float> @llvm.vp.round.nxv4f32(<vscale x 4 x float> undef, <vscale x 4 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %8 = call <vscale x 8 x float> @llvm.vp.round.nxv8f32(<vscale x 8 x float> undef, <vscale x 8 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %9 = call <vscale x 16 x float> @llvm.vp.round.nxv16f32(<vscale x 16 x float> undef, <vscale x 16 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %10 = call <2 x double> @llvm.vp.round.v2f64(<2 x double> undef, <2 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %11 = call <4 x double> @llvm.vp.round.v4f64(<4 x double> undef, <4 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %12 = call <8 x double> @llvm.vp.round.v8f64(<8 x double> undef, <8 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %13 = call <16 x double> @llvm.vp.round.v16f64(<16 x double> undef, <16 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %14 = call <vscale x 1 x double> @llvm.vp.round.nxv1f64(<vscale x 1 x double> undef, <vscale x 1 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %15 = call <vscale x 2 x double> @llvm.vp.round.nxv2f64(<vscale x 2 x double> undef, <vscale x 2 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %16 = call <vscale x 4 x double> @llvm.vp.round.nxv4f64(<vscale x 4 x double> undef, <vscale x 4 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %17 = call <vscale x 8 x double> @llvm.vp.round.nxv8f64(<vscale x 8 x double> undef, <vscale x 8 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %1 = call <2 x bfloat> @llvm.vp.round.v2bf16(<2 x bfloat> undef, <2 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %2 = call <4 x bfloat> @llvm.vp.round.v4bf16(<4 x bfloat> undef, <4 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %3 = call <8 x bfloat> @llvm.vp.round.v8bf16(<8 x bfloat> undef, <8 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %4 = call <16 x bfloat> @llvm.vp.round.v16bf16(<16 x bfloat> undef, <16 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %5 = call <vscale x 1 x bfloat> @llvm.vp.round.nxv1bf16(<vscale x 1 x bfloat> undef, <vscale x 1 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %6 = call <vscale x 2 x bfloat> @llvm.vp.round.nxv2bf16(<vscale x 2 x bfloat> undef, <vscale x 2 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %7 = call <vscale x 4 x bfloat> @llvm.vp.round.nxv4bf16(<vscale x 4 x bfloat> undef, <vscale x 4 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %8 = call <vscale x 8 x bfloat> @llvm.vp.round.nxv8bf16(<vscale x 8 x bfloat> undef, <vscale x 8 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %9 = call <vscale x 16 x bfloat> @llvm.vp.round.nxv16bf16(<vscale x 16 x bfloat> undef, <vscale x 16 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %10 = call <2 x float> @llvm.vp.round.v2f32(<2 x float> undef, <2 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %11 = call <4 x float> @llvm.vp.round.v4f32(<4 x float> undef, <4 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %12 = call <8 x float> @llvm.vp.round.v8f32(<8 x float> undef, <8 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %13 = call <16 x float> @llvm.vp.round.v16f32(<16 x float> undef, <16 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %14 = call <vscale x 1 x float> @llvm.vp.round.nxv1f32(<vscale x 1 x float> undef, <vscale x 1 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %15 = call <vscale x 2 x float> @llvm.vp.round.nxv2f32(<vscale x 2 x float> undef, <vscale x 2 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %16 = call <vscale x 4 x float> @llvm.vp.round.nxv4f32(<vscale x 4 x float> undef, <vscale x 4 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %17 = call <vscale x 8 x float> @llvm.vp.round.nxv8f32(<vscale x 8 x float> undef, <vscale x 8 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %18 = call <vscale x 16 x float> @llvm.vp.round.nxv16f32(<vscale x 16 x float> undef, <vscale x 16 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %19 = call <2 x double> @llvm.vp.round.v2f64(<2 x double> undef, <2 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %20 = call <4 x double> @llvm.vp.round.v4f64(<4 x double> undef, <4 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %21 = call <8 x double> @llvm.vp.round.v8f64(<8 x double> undef, <8 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %22 = call <16 x double> @llvm.vp.round.v16f64(<16 x double> undef, <16 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %23 = call <vscale x 1 x double> @llvm.vp.round.nxv1f64(<vscale x 1 x double> undef, <vscale x 1 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %24 = call <vscale x 2 x double> @llvm.vp.round.nxv2f64(<vscale x 2 x double> undef, <vscale x 2 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %25 = call <vscale x 4 x double> @llvm.vp.round.nxv4f64(<vscale x 4 x double> undef, <vscale x 4 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %26 = call <vscale x 8 x double> @llvm.vp.round.nxv8f64(<vscale x 8 x double> undef, <vscale x 8 x i1> undef, i32 undef)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret void
;
+ call <2 x bfloat> @llvm.vp.round.v2bf16(<2 x bfloat> undef, <2 x i1> undef, i32 undef)
+ call <4 x bfloat> @llvm.vp.round.v4bf16(<4 x bfloat> undef, <4 x i1> undef, i32 undef)
+ call <8 x bfloat> @llvm.vp.round.v8bf16(<8 x bfloat> undef, <8 x i1> undef, i32 undef)
+ call <16 x bfloat> @llvm.vp.round.v16bf16(<16 x bfloat> undef, <16 x i1> undef, i32 undef)
+ call <vscale x 1 x bfloat> @llvm.vp.round.nvx1bf16(<vscale x 1 x bfloat> undef, <vscale x 1 x i1> undef, i32 undef)
+ call <vscale x 2 x bfloat> @llvm.vp.round.nvx2bf16(<vscale x 2 x bfloat> undef, <vscale x 2 x i1> undef, i32 undef)
+ call <vscale x 4 x bfloat> @llvm.vp.round.nvx4bf16(<vscale x 4 x bfloat> undef, <vscale x 4 x i1> undef, i32 undef)
+ call <vscale x 8 x bfloat> @llvm.vp.round.nvx8bf16(<vscale x 8 x bfloat> undef, <vscale x 8 x i1> undef, i32 undef)
+ call <vscale x 16 x bfloat> @llvm.vp.round.nvx16bf16(<vscale x 16 x bfloat> undef, <vscale x 16 x i1> undef, i32 undef)
call <2 x float> @llvm.vp.round.v2f32(<2 x float> undef, <2 x i1> undef, i32 undef)
call <4 x float> @llvm.vp.round.v4f32(<4 x float> undef, <4 x i1> undef, i32 undef)
call <8 x float> @llvm.vp.round.v8f32(<8 x float> undef, <8 x i1> undef, i32 undef)
@@ -977,25 +1211,43 @@ define void @vp_round_f16() {
define void @vp_roundeven() {
; CHECK-LABEL: 'vp_roundeven'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %1 = call <2 x float> @llvm.vp.roundeven.v2f32(<2 x float> undef, <2 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %2 = call <4 x float> @llvm.vp.roundeven.v4f32(<4 x float> undef, <4 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %3 = call <8 x float> @llvm.vp.roundeven.v8f32(<8 x float> undef, <8 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %4 = call <16 x float> @llvm.vp.roundeven.v16f32(<16 x float> undef, <16 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %5 = call <vscale x 1 x float> @llvm.vp.roundeven.nxv1f32(<vscale x 1 x float> undef, <vscale x 1 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %6 = call <vscale x 2 x float> @llvm.vp.roundeven.nxv2f32(<vscale x 2 x float> undef, <vscale x 2 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %7 = call <vscale x 4 x float> @llvm.vp.roundeven.nxv4f32(<vscale x 4 x float> undef, <vscale x 4 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %8 = call <vscale x 8 x float> @llvm.vp.roundeven.nxv8f32(<vscale x 8 x float> undef, <vscale x 8 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %9 = call <vscale x 16 x float> @llvm.vp.roundeven.nxv16f32(<vscale x 16 x float> undef, <vscale x 16 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %10 = call <2 x double> @llvm.vp.roundeven.v2f64(<2 x double> undef, <2 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %11 = call <4 x double> @llvm.vp.roundeven.v4f64(<4 x double> undef, <4 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %12 = call <8 x double> @llvm.vp.roundeven.v8f64(<8 x double> undef, <8 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %13 = call <16 x double> @llvm.vp.roundeven.v16f64(<16 x double> undef, <16 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %14 = call <vscale x 1 x double> @llvm.vp.roundeven.nxv1f64(<vscale x 1 x double> undef, <vscale x 1 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %15 = call <vscale x 2 x double> @llvm.vp.roundeven.nxv2f64(<vscale x 2 x double> undef, <vscale x 2 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %16 = call <vscale x 4 x double> @llvm.vp.roundeven.nxv4f64(<vscale x 4 x double> undef, <vscale x 4 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %17 = call <vscale x 8 x double> @llvm.vp.roundeven.nxv8f64(<vscale x 8 x double> undef, <vscale x 8 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %1 = call <2 x bfloat> @llvm.vp.roundeven.v2bf16(<2 x bfloat> undef, <2 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %2 = call <4 x bfloat> @llvm.vp.roundeven.v4bf16(<4 x bfloat> undef, <4 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %3 = call <8 x bfloat> @llvm.vp.roundeven.v8bf16(<8 x bfloat> undef, <8 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %4 = call <16 x bfloat> @llvm.vp.roundeven.v16bf16(<16 x bfloat> undef, <16 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %5 = call <vscale x 1 x bfloat> @llvm.vp.roundeven.nxv1bf16(<vscale x 1 x bfloat> undef, <vscale x 1 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %6 = call <vscale x 2 x bfloat> @llvm.vp.roundeven.nxv2bf16(<vscale x 2 x bfloat> undef, <vscale x 2 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %7 = call <vscale x 4 x bfloat> @llvm.vp.roundeven.nxv4bf16(<vscale x 4 x bfloat> undef, <vscale x 4 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %8 = call <vscale x 8 x bfloat> @llvm.vp.roundeven.nxv8bf16(<vscale x 8 x bfloat> undef, <vscale x 8 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %9 = call <vscale x 16 x bfloat> @llvm.vp.roundeven.nxv16bf16(<vscale x 16 x bfloat> undef, <vscale x 16 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %10 = call <2 x float> @llvm.vp.roundeven.v2f32(<2 x float> undef, <2 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %11 = call <4 x float> @llvm.vp.roundeven.v4f32(<4 x float> undef, <4 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %12 = call <8 x float> @llvm.vp.roundeven.v8f32(<8 x float> undef, <8 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %13 = call <16 x float> @llvm.vp.roundeven.v16f32(<16 x float> undef, <16 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %14 = call <vscale x 1 x float> @llvm.vp.roundeven.nxv1f32(<vscale x 1 x float> undef, <vscale x 1 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %15 = call <vscale x 2 x float> @llvm.vp.roundeven.nxv2f32(<vscale x 2 x float> undef, <vscale x 2 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %16 = call <vscale x 4 x float> @llvm.vp.roundeven.nxv4f32(<vscale x 4 x float> undef, <vscale x 4 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %17 = call <vscale x 8 x float> @llvm.vp.roundeven.nxv8f32(<vscale x 8 x float> undef, <vscale x 8 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %18 = call <vscale x 16 x float> @llvm.vp.roundeven.nxv16f32(<vscale x 16 x float> undef, <vscale x 16 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %19 = call <2 x double> @llvm.vp.roundeven.v2f64(<2 x double> undef, <2 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %20 = call <4 x double> @llvm.vp.roundeven.v4f64(<4 x double> undef, <4 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %21 = call <8 x double> @llvm.vp.roundeven.v8f64(<8 x double> undef, <8 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %22 = call <16 x double> @llvm.vp.roundeven.v16f64(<16 x double> undef, <16 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %23 = call <vscale x 1 x double> @llvm.vp.roundeven.nxv1f64(<vscale x 1 x double> undef, <vscale x 1 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %24 = call <vscale x 2 x double> @llvm.vp.roundeven.nxv2f64(<vscale x 2 x double> undef, <vscale x 2 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %25 = call <vscale x 4 x double> @llvm.vp.roundeven.nxv4f64(<vscale x 4 x double> undef, <vscale x 4 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %26 = call <vscale x 8 x double> @llvm.vp.roundeven.nxv8f64(<vscale x 8 x double> undef, <vscale x 8 x i1> undef, i32 undef)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret void
;
+ call <2 x bfloat> @llvm.vp.roundeven.v2bf16(<2 x bfloat> undef, <2 x i1> undef, i32 undef)
+ call <4 x bfloat> @llvm.vp.roundeven.v4bf16(<4 x bfloat> undef, <4 x i1> undef, i32 undef)
+ call <8 x bfloat> @llvm.vp.roundeven.v8bf16(<8 x bfloat> undef, <8 x i1> undef, i32 undef)
+ call <16 x bfloat> @llvm.vp.roundeven.v16bf16(<16 x bfloat> undef, <16 x i1> undef, i32 undef)
+ call <vscale x 1 x bfloat> @llvm.vp.roundeven.nvx1bf16(<vscale x 1 x bfloat> undef, <vscale x 1 x i1> undef, i32 undef)
+ call <vscale x 2 x bfloat> @llvm.vp.roundeven.nvx2bf16(<vscale x 2 x bfloat> undef, <vscale x 2 x i1> undef, i32 undef)
+ call <vscale x 4 x bfloat> @llvm.vp.roundeven.nvx4bf16(<vscale x 4 x bfloat> undef, <vscale x 4 x i1> undef, i32 undef)
+ call <vscale x 8 x bfloat> @llvm.vp.roundeven.nvx8bf16(<vscale x 8 x bfloat> undef, <vscale x 8 x i1> undef, i32 undef)
+ call <vscale x 16 x bfloat> @llvm.vp.roundeven.nvx16bf16(<vscale x 16 x bfloat> undef, <vscale x 16 x i1> undef, i32 undef)
call <2 x float> @llvm.vp.roundeven.v2f32(<2 x float> undef, <2 x i1> undef, i32 undef)
call <4 x float> @llvm.vp.roundeven.v4f32(<4 x float> undef, <4 x i1> undef, i32 undef)
call <8 x float> @llvm.vp.roundeven.v8f32(<8 x float> undef, <8 x i1> undef, i32 undef)
@@ -1055,25 +1307,43 @@ define void @vp_roundeven_f16() {
define void @vp_roundtozero() {
; CHECK-LABEL: 'vp_roundtozero'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %1 = call <2 x float> @llvm.vp.roundtozero.v2f32(<2 x float> undef, <2 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %2 = call <4 x float> @llvm.vp.roundtozero.v4f32(<4 x float> undef, <4 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %3 = call <8 x float> @llvm.vp.roundtozero.v8f32(<8 x float> undef, <8 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %4 = call <16 x float> @llvm.vp.roundtozero.v16f32(<16 x float> undef, <16 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %5 = call <vscale x 1 x float> @llvm.vp.roundtozero.nxv1f32(<vscale x 1 x float> undef, <vscale x 1 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %6 = call <vscale x 2 x float> @llvm.vp.roundtozero.nxv2f32(<vscale x 2 x float> undef, <vscale x 2 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %7 = call <vscale x 4 x float> @llvm.vp.roundtozero.nxv4f32(<vscale x 4 x float> undef, <vscale x 4 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %8 = call <vscale x 8 x float> @llvm.vp.roundtozero.nxv8f32(<vscale x 8 x float> undef, <vscale x 8 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %9 = call <vscale x 16 x float> @llvm.vp.roundtozero.nxv16f32(<vscale x 16 x float> undef, <vscale x 16 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %10 = call <2 x double> @llvm.vp.roundtozero.v2f64(<2 x double> undef, <2 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %11 = call <4 x double> @llvm.vp.roundtozero.v4f64(<4 x double> undef, <4 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %12 = call <8 x double> @llvm.vp.roundtozero.v8f64(<8 x double> undef, <8 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %13 = call <16 x double> @llvm.vp.roundtozero.v16f64(<16 x double> undef, <16 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %14 = call <vscale x 1 x double> @llvm.vp.roundtozero.nxv1f64(<vscale x 1 x double> undef, <vscale x 1 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %15 = call <vscale x 2 x double> @llvm.vp.roundtozero.nxv2f64(<vscale x 2 x double> undef, <vscale x 2 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %16 = call <vscale x 4 x double> @llvm.vp.roundtozero.nxv4f64(<vscale x 4 x double> undef, <vscale x 4 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %17 = call <vscale x 8 x double> @llvm.vp.roundtozero.nxv8f64(<vscale x 8 x double> undef, <vscale x 8 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %1 = call <2 x bfloat> @llvm.vp.roundtozero.v2bf16(<2 x bfloat> undef, <2 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %2 = call <4 x bfloat> @llvm.vp.roundtozero.v4bf16(<4 x bfloat> undef, <4 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %3 = call <8 x bfloat> @llvm.vp.roundtozero.v8bf16(<8 x bfloat> undef, <8 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %4 = call <16 x bfloat> @llvm.vp.roundtozero.v16bf16(<16 x bfloat> undef, <16 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %5 = call <vscale x 1 x bfloat> @llvm.vp.roundtozero.nxv1bf16(<vscale x 1 x bfloat> undef, <vscale x 1 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %6 = call <vscale x 2 x bfloat> @llvm.vp.roundtozero.nxv2bf16(<vscale x 2 x bfloat> undef, <vscale x 2 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %7 = call <vscale x 4 x bfloat> @llvm.vp.roundtozero.nxv4bf16(<vscale x 4 x bfloat> undef, <vscale x 4 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %8 = call <vscale x 8 x bfloat> @llvm.vp.roundtozero.nxv8bf16(<vscale x 8 x bfloat> undef, <vscale x 8 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %9 = call <vscale x 16 x bfloat> @llvm.vp.roundtozero.nxv16bf16(<vscale x 16 x bfloat> undef, <vscale x 16 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %10 = call <2 x float> @llvm.vp.roundtozero.v2f32(<2 x float> undef, <2 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %11 = call <4 x float> @llvm.vp.roundtozero.v4f32(<4 x float> undef, <4 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %12 = call <8 x float> @llvm.vp.roundtozero.v8f32(<8 x float> undef, <8 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %13 = call <16 x float> @llvm.vp.roundtozero.v16f32(<16 x float> undef, <16 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %14 = call <vscale x 1 x float> @llvm.vp.roundtozero.nxv1f32(<vscale x 1 x float> undef, <vscale x 1 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %15 = call <vscale x 2 x float> @llvm.vp.roundtozero.nxv2f32(<vscale x 2 x float> undef, <vscale x 2 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %16 = call <vscale x 4 x float> @llvm.vp.roundtozero.nxv4f32(<vscale x 4 x float> undef, <vscale x 4 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %17 = call <vscale x 8 x float> @llvm.vp.roundtozero.nxv8f32(<vscale x 8 x float> undef, <vscale x 8 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %18 = call <vscale x 16 x float> @llvm.vp.roundtozero.nxv16f32(<vscale x 16 x float> undef, <vscale x 16 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %19 = call <2 x double> @llvm.vp.roundtozero.v2f64(<2 x double> undef, <2 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %20 = call <4 x double> @llvm.vp.roundtozero.v4f64(<4 x double> undef, <4 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %21 = call <8 x double> @llvm.vp.roundtozero.v8f64(<8 x double> undef, <8 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %22 = call <16 x double> @llvm.vp.roundtozero.v16f64(<16 x double> undef, <16 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %23 = call <vscale x 1 x double> @llvm.vp.roundtozero.nxv1f64(<vscale x 1 x double> undef, <vscale x 1 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %24 = call <vscale x 2 x double> @llvm.vp.roundtozero.nxv2f64(<vscale x 2 x double> undef, <vscale x 2 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %25 = call <vscale x 4 x double> @llvm.vp.roundtozero.nxv4f64(<vscale x 4 x double> undef, <vscale x 4 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %26 = call <vscale x 8 x double> @llvm.vp.roundtozero.nxv8f64(<vscale x 8 x double> undef, <vscale x 8 x i1> undef, i32 undef)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret void
;
+ call <2 x bfloat> @llvm.vp.roundtozero.v2bf16(<2 x bfloat> undef, <2 x i1> undef, i32 undef)
+ call <4 x bfloat> @llvm.vp.roundtozero.v4bf16(<4 x bfloat> undef, <4 x i1> undef, i32 undef)
+ call <8 x bfloat> @llvm.vp.roundtozero.v8bf16(<8 x bfloat> undef, <8 x i1> undef, i32 undef)
+ call <16 x bfloat> @llvm.vp.roundtozero.v16bf16(<16 x bfloat> undef, <16 x i1> undef, i32 undef)
+ call <vscale x 1 x bfloat> @llvm.vp.roundtozero.nvx1bf16(<vscale x 1 x bfloat> undef, <vscale x 1 x i1> undef, i32 undef)
+ call <vscale x 2 x bfloat> @llvm.vp.roundtozero.nvx2bf16(<vscale x 2 x bfloat> undef, <vscale x 2 x i1> undef, i32 undef)
+ call <vscale x 4 x bfloat> @llvm.vp.roundtozero.nvx4bf16(<vscale x 4 x bfloat> undef, <vscale x 4 x i1> undef, i32 undef)
+ call <vscale x 8 x bfloat> @llvm.vp.roundtozero.nvx8bf16(<vscale x 8 x bfloat> undef, <vscale x 8 x i1> undef, i32 undef)
+ call <vscale x 16 x bfloat> @llvm.vp.roundtozero.nvx16bf16(<vscale x 16 x bfloat> undef, <vscale x 16 x i1> undef, i32 undef)
call <2 x float> @llvm.vp.roundtozero.v2f32(<2 x float> undef, <2 x i1> undef, i32 undef)
call <4 x float> @llvm.vp.roundtozero.v4f32(<4 x float> undef, <4 x i1> undef, i32 undef)
call <8 x float> @llvm.vp.roundtozero.v8f32(<8 x float> undef, <8 x i1> undef, i32 undef)
@@ -1133,25 +1403,43 @@ define void @vp_roundtozero_f16() {
define void @vp_rint() {
; CHECK-LABEL: 'vp_rint'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %1 = call <2 x float> @llvm.vp.rint.v2f32(<2 x float> undef, <2 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %2 = call <4 x float> @llvm.vp.rint.v4f32(<4 x float> undef, <4 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %3 = call <8 x float> @llvm.vp.rint.v8f32(<8 x float> undef, <8 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %4 = call <16 x float> @llvm.vp.rint.v16f32(<16 x float> undef, <16 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %5 = call <vscale x 1 x float> @llvm.vp.rint.nxv1f32(<vscale x 1 x float> undef, <vscale x 1 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %6 = call <vscale x 2 x float> @llvm.vp.rint.nxv2f32(<vscale x 2 x float> undef, <vscale x 2 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %7 = call <vscale x 4 x float> @llvm.vp.rint.nxv4f32(<vscale x 4 x float> undef, <vscale x 4 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %8 = call <vscale x 8 x float> @llvm.vp.rint.nxv8f32(<vscale x 8 x float> undef, <vscale x 8 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %9 = call <vscale x 16 x float> @llvm.vp.rint.nxv16f32(<vscale x 16 x float> undef, <vscale x 16 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %10 = call <2 x double> @llvm.vp.rint.v2f64(<2 x double> undef, <2 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %11 = call <4 x double> @llvm.vp.rint.v4f64(<4 x double> undef, <4 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %12 = call <8 x double> @llvm.vp.rint.v8f64(<8 x double> undef, <8 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %13 = call <16 x double> @llvm.vp.rint.v16f64(<16 x double> undef, <16 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %14 = call <vscale x 1 x double> @llvm.vp.rint.nxv1f64(<vscale x 1 x double> undef, <vscale x 1 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %15 = call <vscale x 2 x double> @llvm.vp.rint.nxv2f64(<vscale x 2 x double> undef, <vscale x 2 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %16 = call <vscale x 4 x double> @llvm.vp.rint.nxv4f64(<vscale x 4 x double> undef, <vscale x 4 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %17 = call <vscale x 8 x double> @llvm.vp.rint.nxv8f64(<vscale x 8 x double> undef, <vscale x 8 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %1 = call <2 x bfloat> @llvm.vp.rint.v2bf16(<2 x bfloat> undef, <2 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %2 = call <4 x bfloat> @llvm.vp.rint.v4bf16(<4 x bfloat> undef, <4 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %3 = call <8 x bfloat> @llvm.vp.rint.v8bf16(<8 x bfloat> undef, <8 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %4 = call <16 x bfloat> @llvm.vp.rint.v16bf16(<16 x bfloat> undef, <16 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %5 = call <vscale x 1 x bfloat> @llvm.vp.rint.nxv1bf16(<vscale x 1 x bfloat> undef, <vscale x 1 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %6 = call <vscale x 2 x bfloat> @llvm.vp.rint.nxv2bf16(<vscale x 2 x bfloat> undef, <vscale x 2 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %7 = call <vscale x 4 x bfloat> @llvm.vp.rint.nxv4bf16(<vscale x 4 x bfloat> undef, <vscale x 4 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %8 = call <vscale x 8 x bfloat> @llvm.vp.rint.nxv8bf16(<vscale x 8 x bfloat> undef, <vscale x 8 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %9 = call <vscale x 16 x bfloat> @llvm.vp.rint.nxv16bf16(<vscale x 16 x bfloat> undef, <vscale x 16 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %10 = call <2 x float> @llvm.vp.rint.v2f32(<2 x float> undef, <2 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %11 = call <4 x float> @llvm.vp.rint.v4f32(<4 x float> undef, <4 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %12 = call <8 x float> @llvm.vp.rint.v8f32(<8 x float> undef, <8 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %13 = call <16 x float> @llvm.vp.rint.v16f32(<16 x float> undef, <16 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %14 = call <vscale x 1 x float> @llvm.vp.rint.nxv1f32(<vscale x 1 x float> undef, <vscale x 1 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %15 = call <vscale x 2 x float> @llvm.vp.rint.nxv2f32(<vscale x 2 x float> undef, <vscale x 2 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %16 = call <vscale x 4 x float> @llvm.vp.rint.nxv4f32(<vscale x 4 x float> undef, <vscale x 4 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %17 = call <vscale x 8 x float> @llvm.vp.rint.nxv8f32(<vscale x 8 x float> undef, <vscale x 8 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %18 = call <vscale x 16 x float> @llvm.vp.rint.nxv16f32(<vscale x 16 x float> undef, <vscale x 16 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %19 = call <2 x double> @llvm.vp.rint.v2f64(<2 x double> undef, <2 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %20 = call <4 x double> @llvm.vp.rint.v4f64(<4 x double> undef, <4 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %21 = call <8 x double> @llvm.vp.rint.v8f64(<8 x double> undef, <8 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %22 = call <16 x double> @llvm.vp.rint.v16f64(<16 x double> undef, <16 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %23 = call <vscale x 1 x double> @llvm.vp.rint.nxv1f64(<vscale x 1 x double> undef, <vscale x 1 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %24 = call <vscale x 2 x double> @llvm.vp.rint.nxv2f64(<vscale x 2 x double> undef, <vscale x 2 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %25 = call <vscale x 4 x double> @llvm.vp.rint.nxv4f64(<vscale x 4 x double> undef, <vscale x 4 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 5 for instruction: %26 = call <vscale x 8 x double> @llvm.vp.rint.nxv8f64(<vscale x 8 x double> undef, <vscale x 8 x i1> undef, i32 undef)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret void
;
+ call <2 x bfloat> @llvm.vp.rint.v2bf16(<2 x bfloat> undef, <2 x i1> undef, i32 undef)
+ call <4 x bfloat> @llvm.vp.rint.v4bf16(<4 x bfloat> undef, <4 x i1> undef, i32 undef)
+ call <8 x bfloat> @llvm.vp.rint.v8bf16(<8 x bfloat> undef, <8 x i1> undef, i32 undef)
+ call <16 x bfloat> @llvm.vp.rint.v16bf16(<16 x bfloat> undef, <16 x i1> undef, i32 undef)
+ call <vscale x 1 x bfloat> @llvm.vp.rint.nvx1bf16(<vscale x 1 x bfloat> undef, <vscale x 1 x i1> undef, i32 undef)
+ call <vscale x 2 x bfloat> @llvm.vp.rint.nvx2bf16(<vscale x 2 x bfloat> undef, <vscale x 2 x i1> undef, i32 undef)
+ call <vscale x 4 x bfloat> @llvm.vp.rint.nvx4bf16(<vscale x 4 x bfloat> undef, <vscale x 4 x i1> undef, i32 undef)
+ call <vscale x 8 x bfloat> @llvm.vp.rint.nvx8bf16(<vscale x 8 x bfloat> undef, <vscale x 8 x i1> undef, i32 undef)
+ call <vscale x 16 x bfloat> @llvm.vp.rint.nvx16bf16(<vscale x 16 x bfloat> undef, <vscale x 16 x i1> undef, i32 undef)
call <2 x float> @llvm.vp.rint.v2f32(<2 x float> undef, <2 x i1> undef, i32 undef)
call <4 x float> @llvm.vp.rint.v4f32(<4 x float> undef, <4 x i1> undef, i32 undef)
call <8 x float> @llvm.vp.rint.v8f32(<8 x float> undef, <8 x i1> undef, i32 undef)
@@ -1211,25 +1499,43 @@ define void @vp_rint_f16() {
define void @vp_nearbyint() {
; CHECK-LABEL: 'vp_nearbyint'
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %1 = call <2 x float> @llvm.vp.nearbyint.v2f32(<2 x float> undef, <2 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %2 = call <4 x float> @llvm.vp.nearbyint.v4f32(<4 x float> undef, <4 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %3 = call <8 x float> @llvm.vp.nearbyint.v8f32(<8 x float> undef, <8 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %4 = call <16 x float> @llvm.vp.nearbyint.v16f32(<16 x float> undef, <16 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %5 = call <vscale x 1 x float> @llvm.vp.nearbyint.nxv1f32(<vscale x 1 x float> undef, <vscale x 1 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %6 = call <vscale x 2 x float> @llvm.vp.nearbyint.nxv2f32(<vscale x 2 x float> undef, <vscale x 2 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %7 = call <vscale x 4 x float> @llvm.vp.nearbyint.nxv4f32(<vscale x 4 x float> undef, <vscale x 4 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %8 = call <vscale x 8 x float> @llvm.vp.nearbyint.nxv8f32(<vscale x 8 x float> undef, <vscale x 8 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %9 = call <vscale x 16 x float> @llvm.vp.nearbyint.nxv16f32(<vscale x 16 x float> undef, <vscale x 16 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %10 = call <2 x double> @llvm.vp.nearbyint.v2f64(<2 x double> undef, <2 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %11 = call <4 x double> @llvm.vp.nearbyint.v4f64(<4 x double> undef, <4 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %12 = call <8 x double> @llvm.vp.nearbyint.v8f64(<8 x double> undef, <8 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %13 = call <16 x double> @llvm.vp.nearbyint.v16f64(<16 x double> undef, <16 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %14 = call <vscale x 1 x double> @llvm.vp.nearbyint.nxv1f64(<vscale x 1 x double> undef, <vscale x 1 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %15 = call <vscale x 2 x double> @llvm.vp.nearbyint.nxv2f64(<vscale x 2 x double> undef, <vscale x 2 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %16 = call <vscale x 4 x double> @llvm.vp.nearbyint.nxv4f64(<vscale x 4 x double> undef, <vscale x 4 x i1> undef, i32 undef)
-; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %17 = call <vscale x 8 x double> @llvm.vp.nearbyint.nxv8f64(<vscale x 8 x double> undef, <vscale x 8 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %1 = call <2 x bfloat> @llvm.vp.nearbyint.v2bf16(<2 x bfloat> undef, <2 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %2 = call <4 x bfloat> @llvm.vp.nearbyint.v4bf16(<4 x bfloat> undef, <4 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %3 = call <8 x bfloat> @llvm.vp.nearbyint.v8bf16(<8 x bfloat> undef, <8 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %4 = call <16 x bfloat> @llvm.vp.nearbyint.v16bf16(<16 x bfloat> undef, <16 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %5 = call <vscale x 1 x bfloat> @llvm.vp.nearbyint.nxv1bf16(<vscale x 1 x bfloat> undef, <vscale x 1 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %6 = call <vscale x 2 x bfloat> @llvm.vp.nearbyint.nxv2bf16(<vscale x 2 x bfloat> undef, <vscale x 2 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %7 = call <vscale x 4 x bfloat> @llvm.vp.nearbyint.nxv4bf16(<vscale x 4 x bfloat> undef, <vscale x 4 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %8 = call <vscale x 8 x bfloat> @llvm.vp.nearbyint.nxv8bf16(<vscale x 8 x bfloat> undef, <vscale x 8 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 1 for instruction: %9 = call <vscale x 16 x bfloat> @llvm.vp.nearbyint.nxv16bf16(<vscale x 16 x bfloat> undef, <vscale x 16 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %10 = call <2 x float> @llvm.vp.nearbyint.v2f32(<2 x float> undef, <2 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %11 = call <4 x float> @llvm.vp.nearbyint.v4f32(<4 x float> undef, <4 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %12 = call <8 x float> @llvm.vp.nearbyint.v8f32(<8 x float> undef, <8 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %13 = call <16 x float> @llvm.vp.nearbyint.v16f32(<16 x float> undef, <16 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %14 = call <vscale x 1 x float> @llvm.vp.nearbyint.nxv1f32(<vscale x 1 x float> undef, <vscale x 1 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %15 = call <vscale x 2 x float> @llvm.vp.nearbyint.nxv2f32(<vscale x 2 x float> undef, <vscale x 2 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %16 = call <vscale x 4 x float> @llvm.vp.nearbyint.nxv4f32(<vscale x 4 x float> undef, <vscale x 4 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %17 = call <vscale x 8 x float> @llvm.vp.nearbyint.nxv8f32(<vscale x 8 x float> undef, <vscale x 8 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %18 = call <vscale x 16 x float> @llvm.vp.nearbyint.nxv16f32(<vscale x 16 x float> undef, <vscale x 16 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %19 = call <2 x double> @llvm.vp.nearbyint.v2f64(<2 x double> undef, <2 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %20 = call <4 x double> @llvm.vp.nearbyint.v4f64(<4 x double> undef, <4 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %21 = call <8 x double> @llvm.vp.nearbyint.v8f64(<8 x double> undef, <8 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %22 = call <16 x double> @llvm.vp.nearbyint.v16f64(<16 x double> undef, <16 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %23 = call <vscale x 1 x double> @llvm.vp.nearbyint.nxv1f64(<vscale x 1 x double> undef, <vscale x 1 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %24 = call <vscale x 2 x double> @llvm.vp.nearbyint.nxv2f64(<vscale x 2 x double> undef, <vscale x 2 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %25 = call <vscale x 4 x double> @llvm.vp.nearbyint.nxv4f64(<vscale x 4 x double> undef, <vscale x 4 x i1> undef, i32 undef)
+; CHECK-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %26 = call <vscale x 8 x double> @llvm.vp.nearbyint.nxv8f64(<vscale x 8 x double> undef, <vscale x 8 x i1> undef, i32 undef)
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret void
;
+ call <2 x bfloat> @llvm.vp.nearbyint.v2bf16(<2 x bfloat> undef, <2 x i1> undef, i32 undef)
+ call <4 x bfloat> @llvm.vp.nearbyint.v4bf16(<4 x bfloat> undef, <4 x i1> undef, i32 undef)
+ call <8 x bfloat> @llvm.vp.nearbyint.v8bf16(<8 x bfloat> undef, <8 x i1> undef, i32 undef)
+ call <16 x bfloat> @llvm.vp.nearbyint.v16bf16(<16 x bfloat> undef, <16 x i1> undef, i32 undef)
+ call <vscale x 1 x bfloat> @llvm.vp.nearbyint.nvx1bf16(<vscale x 1 x bfloat> undef, <vscale x 1 x i1> undef, i32 undef)
+ call <vscale x 2 x bfloat> @llvm.vp.nearbyint.nvx2bf16(<vscale x 2 x bfloat> undef, <vscale x 2 x i1> undef, i32 undef)
+ call <vscale x 4 x bfloat> @llvm.vp.nearbyint.nvx4bf16(<vscale x 4 x bfloat> undef, <vscale x 4 x i1> undef, i32 undef)
+ call <vscale x 8 x bfloat> @llvm.vp.nearbyint.nvx8bf16(<vscale x 8 x bfloat> undef, <vscale x 8 x i1> undef, i32 undef)
+ call <vscale x 16 x bfloat> @llvm.vp.nearbyint.nvx16bf16(<vscale x 16 x bfloat> undef, <vscale x 16 x i1> undef, i32 undef)
call <2 x float> @llvm.vp.nearbyint.v2f32(<2 x float> undef, <2 x i1> undef, i32 undef)
call <4 x float> @llvm.vp.nearbyint.v4f32(<4 x float> undef, <4 x i1> undef, i32 undef)
call <8 x float> @llvm.vp.nearbyint.v8f32(<8 x float> undef, <8 x i1> undef, i32 undef)
More information about the llvm-commits
mailing list