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
105 lines
9.3 KiB
LLVM
105 lines
9.3 KiB
LLVM
; NOTE: Assertions have been autogenerated by utils/update_analyze_test_checks.py
|
|
; RUN: opt < %s -passes="print<cost-model>" 2>&1 -disable-output -mtriple=x86_64-unknown-linux-gnu -mattr=+pclmul | FileCheck %s --check-prefix=PCLMUL
|
|
; RUN: opt < %s -passes="print<cost-model>" 2>&1 -disable-output -mtriple=x86_64-unknown-linux-gnu -mattr=-pclmul | FileCheck %s --check-prefix=NO-PCLMUL
|
|
|
|
|
|
define void @clmul(i128 %a128, i128 %b128, i64 %a64, i64 %b64, i32 %a32, i32 %b32, i16 %a16, i16 %b16, i8 %a8, i8 %b8) {
|
|
; PCLMUL-LABEL: 'clmul'
|
|
; PCLMUL-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %call_i128 = call i128 @llvm.clmul.i128(i128 %a128, i128 %b128)
|
|
; PCLMUL-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %call_i64 = call i64 @llvm.clmul.i64(i64 %a64, i64 %b64)
|
|
; PCLMUL-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %call_i32 = call i32 @llvm.clmul.i32(i32 %a32, i32 %b32)
|
|
; PCLMUL-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %call_i16 = call i16 @llvm.clmul.i16(i16 %a16, i16 %b16)
|
|
; PCLMUL-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %call_i8 = call i8 @llvm.clmul.i8(i8 %a8, i8 %b8)
|
|
; PCLMUL-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret void
|
|
;
|
|
; NO-PCLMUL-LABEL: 'clmul'
|
|
; NO-PCLMUL-NEXT: Cost Model: Found an estimated cost of 768 for instruction: %call_i128 = call i128 @llvm.clmul.i128(i128 %a128, i128 %b128)
|
|
; NO-PCLMUL-NEXT: Cost Model: Found an estimated cost of 192 for instruction: %call_i64 = call i64 @llvm.clmul.i64(i64 %a64, i64 %b64)
|
|
; NO-PCLMUL-NEXT: Cost Model: Found an estimated cost of 96 for instruction: %call_i32 = call i32 @llvm.clmul.i32(i32 %a32, i32 %b32)
|
|
; NO-PCLMUL-NEXT: Cost Model: Found an estimated cost of 48 for instruction: %call_i16 = call i16 @llvm.clmul.i16(i16 %a16, i16 %b16)
|
|
; NO-PCLMUL-NEXT: Cost Model: Found an estimated cost of 24 for instruction: %call_i8 = call i8 @llvm.clmul.i8(i8 %a8, i8 %b8)
|
|
; NO-PCLMUL-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret void
|
|
;
|
|
%call_i128 = call i128 @llvm.clmul.i128(i128 %a128, i128 %b128)
|
|
%call_i64 = call i64 @llvm.clmul.i64(i64 %a64, i64 %b64)
|
|
%call_i32 = call i32 @llvm.clmul.i32(i32 %a32, i32 %b32)
|
|
%call_i16 = call i16 @llvm.clmul.i16(i16 %a16, i16 %b16)
|
|
%call_i8 = call i8 @llvm.clmul.i8(i8 %a8, i8 %b8)
|
|
ret void
|
|
}
|
|
|
|
define void @clmul_128(<1 x i128> %a128, <1 x i128> %b128, <2 x i64> %a64, <2 x i64> %b64, <4 x i32> %a32, <4 x i32> %b32, <8 x i16> %a16, <8 x i16> %b16, <16 x i8> %a8, <16 x i8> %b8) {
|
|
; PCLMUL-LABEL: 'clmul_128'
|
|
; PCLMUL-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %call_i128 = call <1 x i128> @llvm.clmul.v1i128(<1 x i128> %a128, <1 x i128> %b128)
|
|
; PCLMUL-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %call_i64 = call <2 x i64> @llvm.clmul.v2i64(<2 x i64> %a64, <2 x i64> %b64)
|
|
; PCLMUL-NEXT: Cost Model: Found an estimated cost of 2 for instruction: %call_i32 = call <4 x i32> @llvm.clmul.v4i32(<4 x i32> %a32, <4 x i32> %b32)
|
|
; PCLMUL-NEXT: Cost Model: Found an estimated cost of 48 for instruction: %call_i16 = call <8 x i16> @llvm.clmul.v8i16(<8 x i16> %a16, <8 x i16> %b16)
|
|
; PCLMUL-NEXT: Cost Model: Found an estimated cost of 40 for instruction: %call_i8 = call <16 x i8> @llvm.clmul.v16i8(<16 x i8> %a8, <16 x i8> %b8)
|
|
; PCLMUL-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret void
|
|
;
|
|
; NO-PCLMUL-LABEL: 'clmul_128'
|
|
; NO-PCLMUL-NEXT: Cost Model: Found an estimated cost of 768 for instruction: %call_i128 = call <1 x i128> @llvm.clmul.v1i128(<1 x i128> %a128, <1 x i128> %b128)
|
|
; NO-PCLMUL-NEXT: Cost Model: Found an estimated cost of 576 for instruction: %call_i64 = call <2 x i64> @llvm.clmul.v2i64(<2 x i64> %a64, <2 x i64> %b64)
|
|
; NO-PCLMUL-NEXT: Cost Model: Found an estimated cost of 160 for instruction: %call_i32 = call <4 x i32> @llvm.clmul.v4i32(<4 x i32> %a32, <4 x i32> %b32)
|
|
; NO-PCLMUL-NEXT: Cost Model: Found an estimated cost of 48 for instruction: %call_i16 = call <8 x i16> @llvm.clmul.v8i16(<8 x i16> %a16, <8 x i16> %b16)
|
|
; NO-PCLMUL-NEXT: Cost Model: Found an estimated cost of 40 for instruction: %call_i8 = call <16 x i8> @llvm.clmul.v16i8(<16 x i8> %a8, <16 x i8> %b8)
|
|
; NO-PCLMUL-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret void
|
|
;
|
|
%call_i128 = call <1 x i128> @llvm.clmul.v1i128(<1 x i128> %a128, <1 x i128> %b128)
|
|
%call_i64 = call <2 x i64> @llvm.clmul.v2i64(<2 x i64> %a64, <2 x i64> %b64)
|
|
%call_i32 = call <4 x i32> @llvm.clmul.v4i32(<4 x i32> %a32, <4 x i32> %b32)
|
|
%call_i16 = call <8 x i16> @llvm.clmul.v8i16(<8 x i16> %a16, <8 x i16> %b16)
|
|
%call_i8 = call <16 x i8> @llvm.clmul.v16i8(<16 x i8> %a8, <16 x i8> %b8)
|
|
ret void
|
|
}
|
|
|
|
define void @clmul_256(<2 x i128> %a128, <2 x i128> %b128, <4 x i64> %a64, <4 x i64> %b64, <8 x i32> %a32, <8 x i32> %b32, <16 x i16> %a16, <16 x i16> %b16, <32 x i8> %a8, <32 x i8> %b8) {
|
|
; PCLMUL-LABEL: 'clmul_256'
|
|
; PCLMUL-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %call_i128 = call <2 x i128> @llvm.clmul.v2i128(<2 x i128> %a128, <2 x i128> %b128)
|
|
; PCLMUL-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %call_i64 = call <4 x i64> @llvm.clmul.v4i64(<4 x i64> %a64, <4 x i64> %b64)
|
|
; PCLMUL-NEXT: Cost Model: Found an estimated cost of 4 for instruction: %call_i32 = call <8 x i32> @llvm.clmul.v8i32(<8 x i32> %a32, <8 x i32> %b32)
|
|
; PCLMUL-NEXT: Cost Model: Found an estimated cost of 96 for instruction: %call_i16 = call <16 x i16> @llvm.clmul.v16i16(<16 x i16> %a16, <16 x i16> %b16)
|
|
; PCLMUL-NEXT: Cost Model: Found an estimated cost of 80 for instruction: %call_i8 = call <32 x i8> @llvm.clmul.v32i8(<32 x i8> %a8, <32 x i8> %b8)
|
|
; PCLMUL-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret void
|
|
;
|
|
; NO-PCLMUL-LABEL: 'clmul_256'
|
|
; NO-PCLMUL-NEXT: Cost Model: Found an estimated cost of 1536 for instruction: %call_i128 = call <2 x i128> @llvm.clmul.v2i128(<2 x i128> %a128, <2 x i128> %b128)
|
|
; NO-PCLMUL-NEXT: Cost Model: Found an estimated cost of 1152 for instruction: %call_i64 = call <4 x i64> @llvm.clmul.v4i64(<4 x i64> %a64, <4 x i64> %b64)
|
|
; NO-PCLMUL-NEXT: Cost Model: Found an estimated cost of 320 for instruction: %call_i32 = call <8 x i32> @llvm.clmul.v8i32(<8 x i32> %a32, <8 x i32> %b32)
|
|
; NO-PCLMUL-NEXT: Cost Model: Found an estimated cost of 96 for instruction: %call_i16 = call <16 x i16> @llvm.clmul.v16i16(<16 x i16> %a16, <16 x i16> %b16)
|
|
; NO-PCLMUL-NEXT: Cost Model: Found an estimated cost of 80 for instruction: %call_i8 = call <32 x i8> @llvm.clmul.v32i8(<32 x i8> %a8, <32 x i8> %b8)
|
|
; NO-PCLMUL-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret void
|
|
;
|
|
%call_i128 = call <2 x i128> @llvm.clmul.v2i128(<2 x i128> %a128, <2 x i128> %b128)
|
|
%call_i64 = call <4 x i64> @llvm.clmul.v4i64(<4 x i64> %a64, <4 x i64> %b64)
|
|
%call_i32 = call <8 x i32> @llvm.clmul.v8i32(<8 x i32> %a32, <8 x i32> %b32)
|
|
%call_i16 = call <16 x i16> @llvm.clmul.v16i16(<16 x i16> %a16, <16 x i16> %b16)
|
|
%call_i8 = call <32 x i8> @llvm.clmul.v32i8(<32 x i8> %a8, <32 x i8> %b8)
|
|
ret void
|
|
}
|
|
|
|
define void @clmul_512(<4 x i128> %a128, <4 x i128> %b128, <8 x i64> %a64, <8 x i64> %b64, <16 x i32> %a32, <16 x i32> %b32, <32 x i16> %a16, <32 x i16> %b16, <64 x i8> %a8, <64 x i8> %b8) {
|
|
; PCLMUL-LABEL: 'clmul_512'
|
|
; PCLMUL-NEXT: Cost Model: Found an estimated cost of 16 for instruction: %call_i128 = call <4 x i128> @llvm.clmul.v4i128(<4 x i128> %a128, <4 x i128> %b128)
|
|
; PCLMUL-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %call_i64 = call <8 x i64> @llvm.clmul.v8i64(<8 x i64> %a64, <8 x i64> %b64)
|
|
; PCLMUL-NEXT: Cost Model: Found an estimated cost of 8 for instruction: %call_i32 = call <16 x i32> @llvm.clmul.v16i32(<16 x i32> %a32, <16 x i32> %b32)
|
|
; PCLMUL-NEXT: Cost Model: Found an estimated cost of 192 for instruction: %call_i16 = call <32 x i16> @llvm.clmul.v32i16(<32 x i16> %a16, <32 x i16> %b16)
|
|
; PCLMUL-NEXT: Cost Model: Found an estimated cost of 160 for instruction: %call_i8 = call <64 x i8> @llvm.clmul.v64i8(<64 x i8> %a8, <64 x i8> %b8)
|
|
; PCLMUL-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret void
|
|
;
|
|
; NO-PCLMUL-LABEL: 'clmul_512'
|
|
; NO-PCLMUL-NEXT: Cost Model: Found an estimated cost of 3072 for instruction: %call_i128 = call <4 x i128> @llvm.clmul.v4i128(<4 x i128> %a128, <4 x i128> %b128)
|
|
; NO-PCLMUL-NEXT: Cost Model: Found an estimated cost of 2304 for instruction: %call_i64 = call <8 x i64> @llvm.clmul.v8i64(<8 x i64> %a64, <8 x i64> %b64)
|
|
; NO-PCLMUL-NEXT: Cost Model: Found an estimated cost of 640 for instruction: %call_i32 = call <16 x i32> @llvm.clmul.v16i32(<16 x i32> %a32, <16 x i32> %b32)
|
|
; NO-PCLMUL-NEXT: Cost Model: Found an estimated cost of 192 for instruction: %call_i16 = call <32 x i16> @llvm.clmul.v32i16(<32 x i16> %a16, <32 x i16> %b16)
|
|
; NO-PCLMUL-NEXT: Cost Model: Found an estimated cost of 160 for instruction: %call_i8 = call <64 x i8> @llvm.clmul.v64i8(<64 x i8> %a8, <64 x i8> %b8)
|
|
; NO-PCLMUL-NEXT: Cost Model: Found an estimated cost of 0 for instruction: ret void
|
|
;
|
|
%call_i128 = call <4 x i128> @llvm.clmul.v4i128(<4 x i128> %a128, <4 x i128> %b128)
|
|
%call_i64 = call <8 x i64> @llvm.clmul.v8i64(<8 x i64> %a64, <8 x i64> %b64)
|
|
%call_i32 = call <16 x i32> @llvm.clmul.v16i32(<16 x i32> %a32, <16 x i32> %b32)
|
|
%call_i16 = call <32 x i16> @llvm.clmul.v32i16(<32 x i16> %a16, <32 x i16> %b16)
|
|
%call_i8 = call <64 x i8> @llvm.clmul.v64i8(<64 x i8> %a8, <64 x i8> %b8)
|
|
ret void
|
|
}
|