derek.holloway
7502bf64a4
CodeQL / Analyze (actions) (push) Canceled after 0s
CodeQL / Analyze (python) (push) Canceled after 0s
pre-commit / pre-commit (push) Canceled after 0s
Multi-Arch CI / setup (push) Canceled after 0s
Publish build_manylinux_x86_64 images / publish_build_manylinux_x86_64 (push) Canceled after 0s
Setup (Multi-Arch) / setup (push) Canceled after 0s
Multi-Arch CI / Linux::${{ fromJSON(needs.setup.outputs.linux_build_config || '{}').build_variant_label || 'skip' }} (push) Canceled after 0s
Publish TheRock Docker image / build-and-push-image (push) Canceled after 0s
Publish no_rocm_image_ubi10 image / publish_no_rocm_image_ubi10 (push) Canceled after 0s
Multi-Arch CI / Windows::${{ fromJSON(needs.setup.outputs.windows_build_config || '{}').build_variant_label || 'skip' }} (push) Canceled after 0s
Publish TheRock Docker image / build-and-push-image (push) Canceled after 0s
Multi-Arch CI / Manifest Diff (push) Canceled after 0s
TheRock Manifest Diff Report / Generate Manifest Diff Report (push) Canceled after 0s
Multi-Arch CI / CI Summary (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04 images / publish_no_rocm_image_ubuntu24_04 (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04_ocl_rt images / publish_no_rocm_image_ubuntu24_04_ocl_rt (push) Canceled after 0s
Publish TheRock Docker image / build-and-push-image (push) Canceled after 0s
Publish TheRock Docker image / build-and-push-image (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04 images / publish_no_rocm_image_ubuntu24_04_rocgdb (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04_rocgdb images / publish_no_rocm_image_ubuntu24_04_rocgdb (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04 images / publish_no_rocm_image_ubuntu24_04_ocl_rt (push) Canceled after 0s
Publish TheRock Docker image / build-and-push-image (push) Canceled after 0s
Unit Tests / Unit Tests :: ubuntu-24.04 (push) Canceled after 0s
Unit Tests / Unit Tests :: windows-2022 (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04 images / publish_no_rocm_image_ubuntu24_04_media (push) Canceled after 0s
Unit Tests / Unit Tests Summary (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04 images / publish_no_rocm_image_ubuntu24_04_openmpi (push) Canceled after 0s
52 lines
4.5 KiB
LLVM
52 lines
4.5 KiB
LLVM
; NOTE: Assertions have been autogenerated by utils/update_analyze_test_checks.py
|
|
; RUN: opt -passes="print<cost-model>" -cost-kind=all 2>&1 -disable-output -mtriple=thumbv8m.main < %s | FileCheck %s --check-prefix=V8M
|
|
; RUN: opt -passes="print<cost-model>" -cost-kind=all 2>&1 -disable-output -mtriple=armv8a-linux-gnueabihf < %s | FileCheck %s --check-prefix=NEON
|
|
; RUN: opt -passes="print<cost-model>" -cost-kind=all 2>&1 -disable-output -mtriple=armv8.1m.main -mattr=+mve < %s | FileCheck %s --check-prefix=MVE
|
|
|
|
target datalayout = "e-m:e-p:32:32-Fi8-i64:64-v128:64:128-a:0:32-n32-S64"
|
|
|
|
define i32 @reduce_i1(i32 %arg) {
|
|
; V8M-LABEL: 'reduce_i1'
|
|
; V8M-NEXT: Cost Model: Found costs of 1 for: %V1 = call i1 @llvm.vector.reduce.or.v1i1(<1 x i1> undef)
|
|
; V8M-NEXT: Cost Model: Found costs of RThru:3 CodeSize:2 Lat:2 SizeLat:2 for: %V2 = call i1 @llvm.vector.reduce.or.v2i1(<2 x i1> undef)
|
|
; V8M-NEXT: Cost Model: Found costs of RThru:5 CodeSize:2 Lat:2 SizeLat:2 for: %V4 = call i1 @llvm.vector.reduce.or.v4i1(<4 x i1> undef)
|
|
; V8M-NEXT: Cost Model: Found costs of RThru:9 CodeSize:2 Lat:2 SizeLat:2 for: %V8 = call i1 @llvm.vector.reduce.or.v8i1(<8 x i1> undef)
|
|
; V8M-NEXT: Cost Model: Found costs of RThru:17 CodeSize:2 Lat:2 SizeLat:2 for: %V16 = call i1 @llvm.vector.reduce.or.v16i1(<16 x i1> undef)
|
|
; V8M-NEXT: Cost Model: Found costs of RThru:33 CodeSize:2 Lat:2 SizeLat:2 for: %V32 = call i1 @llvm.vector.reduce.or.v32i1(<32 x i1> undef)
|
|
; V8M-NEXT: Cost Model: Found costs of RThru:65 CodeSize:2 Lat:2 SizeLat:2 for: %V64 = call i1 @llvm.vector.reduce.or.v64i1(<64 x i1> undef)
|
|
; V8M-NEXT: Cost Model: Found costs of RThru:129 CodeSize:2 Lat:2 SizeLat:2 for: %V128 = call i1 @llvm.vector.reduce.or.v128i1(<128 x i1> undef)
|
|
; V8M-NEXT: Cost Model: Found costs of 1 for: ret i32 undef
|
|
;
|
|
; NEON-LABEL: 'reduce_i1'
|
|
; NEON-NEXT: Cost Model: Found costs of 3 for: %V1 = call i1 @llvm.vector.reduce.or.v1i1(<1 x i1> undef)
|
|
; NEON-NEXT: Cost Model: Found costs of 2 for: %V2 = call i1 @llvm.vector.reduce.or.v2i1(<2 x i1> undef)
|
|
; NEON-NEXT: Cost Model: Found costs of 2 for: %V4 = call i1 @llvm.vector.reduce.or.v4i1(<4 x i1> undef)
|
|
; NEON-NEXT: Cost Model: Found costs of 2 for: %V8 = call i1 @llvm.vector.reduce.or.v8i1(<8 x i1> undef)
|
|
; NEON-NEXT: Cost Model: Found costs of 2 for: %V16 = call i1 @llvm.vector.reduce.or.v16i1(<16 x i1> undef)
|
|
; NEON-NEXT: Cost Model: Found costs of RThru:97 CodeSize:2 Lat:2 SizeLat:2 for: %V32 = call i1 @llvm.vector.reduce.or.v32i1(<32 x i1> undef)
|
|
; NEON-NEXT: Cost Model: Found costs of RThru:193 CodeSize:2 Lat:2 SizeLat:2 for: %V64 = call i1 @llvm.vector.reduce.or.v64i1(<64 x i1> undef)
|
|
; NEON-NEXT: Cost Model: Found costs of RThru:385 CodeSize:2 Lat:2 SizeLat:2 for: %V128 = call i1 @llvm.vector.reduce.or.v128i1(<128 x i1> undef)
|
|
; NEON-NEXT: Cost Model: Found costs of RThru:0 CodeSize:1 Lat:1 SizeLat:1 for: ret i32 undef
|
|
;
|
|
; MVE-LABEL: 'reduce_i1'
|
|
; MVE-NEXT: Cost Model: Found costs of 4 for: %V1 = call i1 @llvm.vector.reduce.or.v1i1(<1 x i1> undef)
|
|
; MVE-NEXT: Cost Model: Found costs of RThru:3 CodeSize:2 Lat:2 SizeLat:2 for: %V2 = call i1 @llvm.vector.reduce.or.v2i1(<2 x i1> undef)
|
|
; MVE-NEXT: Cost Model: Found costs of RThru:3 CodeSize:2 Lat:2 SizeLat:2 for: %V4 = call i1 @llvm.vector.reduce.or.v4i1(<4 x i1> undef)
|
|
; MVE-NEXT: Cost Model: Found costs of RThru:3 CodeSize:2 Lat:2 SizeLat:2 for: %V8 = call i1 @llvm.vector.reduce.or.v8i1(<8 x i1> undef)
|
|
; MVE-NEXT: Cost Model: Found costs of RThru:3 CodeSize:2 Lat:2 SizeLat:2 for: %V16 = call i1 @llvm.vector.reduce.or.v16i1(<16 x i1> undef)
|
|
; MVE-NEXT: Cost Model: Found costs of RThru:257 CodeSize:2 Lat:2 SizeLat:2 for: %V32 = call i1 @llvm.vector.reduce.or.v32i1(<32 x i1> undef)
|
|
; MVE-NEXT: Cost Model: Found costs of RThru:513 CodeSize:2 Lat:2 SizeLat:2 for: %V64 = call i1 @llvm.vector.reduce.or.v64i1(<64 x i1> undef)
|
|
; MVE-NEXT: Cost Model: Found costs of RThru:1025 CodeSize:2 Lat:2 SizeLat:2 for: %V128 = call i1 @llvm.vector.reduce.or.v128i1(<128 x i1> undef)
|
|
; MVE-NEXT: Cost Model: Found costs of RThru:0 CodeSize:1 Lat:1 SizeLat:1 for: ret i32 undef
|
|
;
|
|
%V1 = call i1 @llvm.vector.reduce.or.v1i1(<1 x i1> undef)
|
|
%V2 = call i1 @llvm.vector.reduce.or.v2i1(<2 x i1> undef)
|
|
%V4 = call i1 @llvm.vector.reduce.or.v4i1(<4 x i1> undef)
|
|
%V8 = call i1 @llvm.vector.reduce.or.v8i1(<8 x i1> undef)
|
|
%V16 = call i1 @llvm.vector.reduce.or.v16i1(<16 x i1> undef)
|
|
%V32 = call i1 @llvm.vector.reduce.or.v32i1(<32 x i1> undef)
|
|
%V64 = call i1 @llvm.vector.reduce.or.v64i1(<64 x i1> undef)
|
|
%V128 = call i1 @llvm.vector.reduce.or.v128i1(<128 x i1> undef)
|
|
ret i32 undef
|
|
}
|