blob: ca53d85e556a346c0b9c5ae292e5b0f63811d039 [file] [log] [blame]
; NOTE: Assertions have been autogenerated by utils/update_analyze_test_checks.py
; RUN: opt -cost-model -analyze -mtriple=thumbv8.1m.main-none-eabi -mattr=+mve < %s | FileCheck %s --check-prefixes=CHECK,CHECK-MVE
; RUN: opt -cost-model -analyze -mtriple=thumbv8.1m.main-none-eabi -mattr=+mve.fp < %s | FileCheck %s --check-prefixes=CHECK,CHECK-MVEFP
define void @icmp() {
; CHECK-LABEL: 'icmp'
; CHECK-NEXT: Cost Model: Found an estimated cost of 12 for instruction: %v2i8 = icmp slt <2 x i8> undef, undef
; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v4i8 = icmp slt <4 x i8> undef, undef
; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v8i8 = icmp slt <8 x i8> undef, undef
; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v16i8 = icmp slt <16 x i8> undef, undef
; CHECK-NEXT: Cost Model: Found an estimated cost of 1028 for instruction: %v32i8 = icmp slt <32 x i8> undef, undef
; CHECK-NEXT: Cost Model: Found an estimated cost of 12 for instruction: %v2i16 = icmp slt <2 x i16> undef, undef
; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v4i16 = icmp slt <4 x i16> undef, undef
; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v8i16 = icmp slt <8 x i16> undef, undef
; CHECK-NEXT: Cost Model: Found an estimated cost of 260 for instruction: %v16i16 = icmp slt <16 x i16> undef, undef
; CHECK-NEXT: Cost Model: Found an estimated cost of 12 for instruction: %v2i32 = icmp slt <2 x i32> undef, undef
; CHECK-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v4i32 = icmp slt <4 x i32> undef, undef
; CHECK-NEXT: Cost Model: Found an estimated cost of 68 for instruction: %v8i32 = icmp slt <8 x i32> undef, undef
; CHECK-NEXT: Cost Model: Found an estimated cost of 264 for instruction: %v16i32 = icmp slt <16 x i32> undef, undef
; CHECK-NEXT: Cost Model: Found an estimated cost of 12 for instruction: %v2i64 = icmp slt <2 x i64> undef, undef
; CHECK-NEXT: Cost Model: Found an estimated cost of 40 for instruction: %v4i64 = icmp slt <4 x i64> undef, undef
; CHECK-NEXT: Cost Model: Found an estimated cost of 144 for instruction: %v8i64 = icmp slt <8 x i64> undef, undef
; CHECK-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret void
;
%v2i8 = icmp slt <2 x i8> undef, undef
%v4i8 = icmp slt <4 x i8> undef, undef
%v8i8 = icmp slt <8 x i8> undef, undef
%v16i8 = icmp slt <16 x i8> undef, undef
%v32i8 = icmp slt <32 x i8> undef, undef
%v2i16 = icmp slt <2 x i16> undef, undef
%v4i16 = icmp slt <4 x i16> undef, undef
%v8i16 = icmp slt <8 x i16> undef, undef
%v16i16 = icmp slt <16 x i16> undef, undef
%v2i32 = icmp slt <2 x i32> undef, undef
%v4i32 = icmp slt <4 x i32> undef, undef
%v8i32 = icmp slt <8 x i32> undef, undef
%v16i32 = icmp slt <16 x i32> undef, undef
%v2i64 = icmp slt <2 x i64> undef, undef
%v4i64 = icmp slt <4 x i64> undef, undef
%v8i64 = icmp slt <8 x i64> undef, undef
ret void
}
define void @fcmp() {
; CHECK-MVE-LABEL: 'fcmp'
; CHECK-MVE-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %v2f16 = fcmp olt <2 x half> undef, undef
; CHECK-MVE-NEXT: Cost Model: Found an estimated cost of 36 for instruction: %v4f16 = fcmp olt <4 x half> undef, undef
; CHECK-MVE-NEXT: Cost Model: Found an estimated cost of 136 for instruction: %v8f16 = fcmp olt <8 x half> undef, undef
; CHECK-MVE-NEXT: Cost Model: Found an estimated cost of 528 for instruction: %v16f16 = fcmp olt <16 x half> undef, undef
; CHECK-MVE-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %v2f32 = fcmp olt <2 x float> undef, undef
; CHECK-MVE-NEXT: Cost Model: Found an estimated cost of 36 for instruction: %v4f32 = fcmp olt <4 x float> undef, undef
; CHECK-MVE-NEXT: Cost Model: Found an estimated cost of 136 for instruction: %v8f32 = fcmp olt <8 x float> undef, undef
; CHECK-MVE-NEXT: Cost Model: Found an estimated cost of 528 for instruction: %v16f32 = fcmp olt <16 x float> undef, undef
; CHECK-MVE-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %v2f64 = fcmp olt <2 x double> undef, undef
; CHECK-MVE-NEXT: Cost Model: Found an estimated cost of 36 for instruction: %v4f64 = fcmp olt <4 x double> undef, undef
; CHECK-MVE-NEXT: Cost Model: Found an estimated cost of 136 for instruction: %v8f64 = fcmp olt <8 x double> undef, undef
; CHECK-MVE-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret void
;
; CHECK-MVEFP-LABEL: 'fcmp'
; CHECK-MVEFP-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v2f16 = fcmp olt <2 x half> undef, undef
; CHECK-MVEFP-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v4f16 = fcmp olt <4 x half> undef, undef
; CHECK-MVEFP-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v8f16 = fcmp olt <8 x half> undef, undef
; CHECK-MVEFP-NEXT: Cost Model: Found an estimated cost of 260 for instruction: %v16f16 = fcmp olt <16 x half> undef, undef
; CHECK-MVEFP-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v2f32 = fcmp olt <2 x float> undef, undef
; CHECK-MVEFP-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %v4f32 = fcmp olt <4 x float> undef, undef
; CHECK-MVEFP-NEXT: Cost Model: Found an estimated cost of 68 for instruction: %v8f32 = fcmp olt <8 x float> undef, undef
; CHECK-MVEFP-NEXT: Cost Model: Found an estimated cost of 264 for instruction: %v16f32 = fcmp olt <16 x float> undef, undef
; CHECK-MVEFP-NEXT: Cost Model: Found an estimated cost of 12 for instruction: %v2f64 = fcmp olt <2 x double> undef, undef
; CHECK-MVEFP-NEXT: Cost Model: Found an estimated cost of 40 for instruction: %v4f64 = fcmp olt <4 x double> undef, undef
; CHECK-MVEFP-NEXT: Cost Model: Found an estimated cost of 144 for instruction: %v8f64 = fcmp olt <8 x double> undef, undef
; CHECK-MVEFP-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret void
;
%v2f16 = fcmp olt <2 x half> undef, undef
%v4f16 = fcmp olt <4 x half> undef, undef
%v8f16 = fcmp olt <8 x half> undef, undef
%v16f16 = fcmp olt <16 x half> undef, undef
%v2f32 = fcmp olt <2 x float> undef, undef
%v4f32 = fcmp olt <4 x float> undef, undef
%v8f32 = fcmp olt <8 x float> undef, undef
%v16f32 = fcmp olt <16 x float> undef, undef
%v2f64 = fcmp olt <2 x double> undef, undef
%v4f64 = fcmp olt <4 x double> undef, undef
%v8f64 = fcmp olt <8 x double> undef, undef
ret void
}