Skip to content

Commit 1b986b4

Browse files
RKSimonRKSimon
committed
[CostModel][X86] Add add/fadd reduction tests for SLM
1 parent cca01c0 commit 1b986b4

File tree

2 files changed

+256
-0
lines changed

2 files changed

+256
-0
lines changed

‎llvm/test/Analysis/CostModel/X86/reduce-add.ll

Lines changed: 37 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -8,6 +8,8 @@
88
; RUN: opt < %s -cost-model -mtriple=x86_64-apple-darwin -analyze -mattr=+avx512f,+avx512bw | FileCheck %s --check-prefixes=CHECK,AVX512,AVX512BW
99
; RUN: opt < %s -cost-model -mtriple=x86_64-apple-darwin -analyze -mattr=+avx512f,+avx512dq | FileCheck %s --check-prefixes=CHECK,AVX512,AVX512DQ
1010

11+
; RUN: opt < %s -cost-model -mtriple=x86_64-apple-darwin -analyze -mcpu=slm | FileCheck %s --check-prefixes=SLM
12+
1113
define i32 @reduce_i64(i32 %arg) {
1214
; SSE-LABEL: 'reduce_i64'
1315
; SSE-NEXT: Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.add.v1i64(<1 x i64> undef)
@@ -32,6 +34,14 @@ define i32 @reduce_i64(i32 %arg) {
3234
; AVX512-NEXT: Cost Model: Found an estimated cost of 7 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.add.v8i64(<8 x i64> undef)
3335
; AVX512-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.add.v16i64(<16 x i64> undef)
3436
; AVX512-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
37+
;
38+
; SLM-LABEL: 'reduce_i64'
39+
; SLM-NEXT: Cost Model: Found an estimated cost of 0 for instruction: %V1 = call i64 @llvm.experimental.vector.reduce.add.v1i64(<1 x i64> undef)
40+
; SLM-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %V2 = call i64 @llvm.experimental.vector.reduce.add.v2i64(<2 x i64> undef)
41+
; SLM-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V4 = call i64 @llvm.experimental.vector.reduce.add.v4i64(<4 x i64> undef)
42+
; SLM-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %V8 = call i64 @llvm.experimental.vector.reduce.add.v8i64(<8 x i64> undef)
43+
; SLM-NEXT: Cost Model: Found an estimated cost of 16 for instruction: %V16 = call i64 @llvm.experimental.vector.reduce.add.v16i64(<16 x i64> undef)
44+
; SLM-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
3545
;
3646
%V1 = call i64 @llvm.experimental.vector.reduce.add.v1i64(<1 x i64> undef)
3747
%V2 = call i64 @llvm.experimental.vector.reduce.add.v2i64(<2 x i64> undef)
@@ -65,6 +75,14 @@ define i32 @reduce_i32(i32 %arg) {
6575
; AVX512-NEXT: Cost Model: Found an estimated cost of 9 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.add.v16i32(<16 x i32> undef)
6676
; AVX512-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.add.v32i32(<32 x i32> undef)
6777
; AVX512-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
78+
;
79+
; SLM-LABEL: 'reduce_i32'
80+
; SLM-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %V2 = call i32 @llvm.experimental.vector.reduce.add.v2i32(<2 x i32> undef)
81+
; SLM-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V4 = call i32 @llvm.experimental.vector.reduce.add.v4i32(<4 x i32> undef)
82+
; SLM-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %V8 = call i32 @llvm.experimental.vector.reduce.add.v8i32(<8 x i32> undef)
83+
; SLM-NEXT: Cost Model: Found an estimated cost of 12 for instruction: %V16 = call i32 @llvm.experimental.vector.reduce.add.v16i32(<16 x i32> undef)
84+
; SLM-NEXT: Cost Model: Found an estimated cost of 24 for instruction: %V32 = call i32 @llvm.experimental.vector.reduce.add.v32i32(<32 x i32> undef)
85+
; SLM-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
6886
;
6987
%V2 = call i32 @llvm.experimental.vector.reduce.add.v2i32(<2 x i32> undef)
7088
%V4 = call i32 @llvm.experimental.vector.reduce.add.v4i32(<4 x i32> undef)
@@ -119,6 +137,15 @@ define i32 @reduce_i16(i32 %arg) {
119137
; AVX512DQ-NEXT: Cost Model: Found an estimated cost of 10 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.add.v32i16(<32 x i16> undef)
120138
; AVX512DQ-NEXT: Cost Model: Found an estimated cost of 20 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.add.v64i16(<64 x i16> undef)
121139
; AVX512DQ-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
140+
;
141+
; SLM-LABEL: 'reduce_i16'
142+
; SLM-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %V2 = call i16 @llvm.experimental.vector.reduce.add.v2i16(<2 x i16> undef)
143+
; SLM-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V4 = call i16 @llvm.experimental.vector.reduce.add.v4i16(<4 x i16> undef)
144+
; SLM-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %V8 = call i16 @llvm.experimental.vector.reduce.add.v8i16(<8 x i16> undef)
145+
; SLM-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %V16 = call i16 @llvm.experimental.vector.reduce.add.v16i16(<16 x i16> undef)
146+
; SLM-NEXT: Cost Model: Found an estimated cost of 16 for instruction: %V32 = call i16 @llvm.experimental.vector.reduce.add.v32i16(<32 x i16> undef)
147+
; SLM-NEXT: Cost Model: Found an estimated cost of 32 for instruction: %V64 = call i16 @llvm.experimental.vector.reduce.add.v64i16(<64 x i16> undef)
148+
; SLM-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
122149
;
123150
%V2 = call i16 @llvm.experimental.vector.reduce.add.v2i16(<2 x i16> undef)
124151
%V4 = call i16 @llvm.experimental.vector.reduce.add.v4i16(<4 x i16> undef)
@@ -179,6 +206,16 @@ define i32 @reduce_i8(i32 %arg) {
179206
; AVX512DQ-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.add.v64i8(<64 x i8> undef)
180207
; AVX512DQ-NEXT: Cost Model: Found an estimated cost of 16 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.add.v128i8(<128 x i8> undef)
181208
; AVX512DQ-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
209+
;
210+
; SLM-LABEL: 'reduce_i8'
211+
; SLM-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %V2 = call i8 @llvm.experimental.vector.reduce.add.v2i8(<2 x i8> undef)
212+
; SLM-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %V4 = call i8 @llvm.experimental.vector.reduce.add.v4i8(<4 x i8> undef)
213+
; SLM-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %V8 = call i8 @llvm.experimental.vector.reduce.add.v8i8(<8 x i8> undef)
214+
; SLM-NEXT: Cost Model: Found an estimated cost of 3 for instruction: %V16 = call i8 @llvm.experimental.vector.reduce.add.v16i8(<16 x i8> undef)
215+
; SLM-NEXT: Cost Model: Found an estimated cost of 6 for instruction: %V32 = call i8 @llvm.experimental.vector.reduce.add.v32i8(<32 x i8> undef)
216+
; SLM-NEXT: Cost Model: Found an estimated cost of 12 for instruction: %V64 = call i8 @llvm.experimental.vector.reduce.add.v64i8(<64 x i8> undef)
217+
; SLM-NEXT: Cost Model: Found an estimated cost of 24 for instruction: %V128 = call i8 @llvm.experimental.vector.reduce.add.v128i8(<128 x i8> undef)
218+
; SLM-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
182219
;
183220
%V2 = call i8 @llvm.experimental.vector.reduce.add.v2i8(<2 x i8> undef)
184221
%V4 = call i8 @llvm.experimental.vector.reduce.add.v4i8(<4 x i8> undef)

0 commit comments

Comments
 (0)