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
75 lines
3.4 KiB
LLVM
75 lines
3.4 KiB
LLVM
; RUN: llc -mtriple aarch64-none-linux-gnu -mattr=+fp16fml < %s | FileCheck %s
|
|
|
|
declare <2 x float> @llvm.aarch64.neon.fmlal.v2f32.v4f16(<2 x float>, <4 x half>, <4 x half>)
|
|
declare <2 x float> @llvm.aarch64.neon.fmlsl.v2f32.v4f16(<2 x float>, <4 x half>, <4 x half>)
|
|
declare <2 x float> @llvm.aarch64.neon.fmlal2.v2f32.v4f16(<2 x float>, <4 x half>, <4 x half>)
|
|
declare <2 x float> @llvm.aarch64.neon.fmlsl2.v2f32.v4f16(<2 x float>, <4 x half>, <4 x half>)
|
|
declare <4 x float> @llvm.aarch64.neon.fmlal.v4f32.v8f16(<4 x float>, <8 x half>, <8 x half>)
|
|
declare <4 x float> @llvm.aarch64.neon.fmlsl.v4f32.v8f16(<4 x float>, <8 x half>, <8 x half>)
|
|
declare <4 x float> @llvm.aarch64.neon.fmlal2.v4f32.v8f16(<4 x float>, <8 x half>, <8 x half>)
|
|
declare <4 x float> @llvm.aarch64.neon.fmlsl2.v4f32.v8f16(<4 x float>, <8 x half>, <8 x half>)
|
|
|
|
define <2 x float> @test_vfmlal_low_u32(<2 x float> %a, <4 x half> %b, <4 x half> %c) #0 {
|
|
entry:
|
|
; CHECK-LABEL: test_vfmlal_low_u32:
|
|
; CHECK: fmlal v0.2s, v1.2h, v2.2h
|
|
%vfmlal_low2.i = call <2 x float> @llvm.aarch64.neon.fmlal.v2f32.v4f16(<2 x float> %a, <4 x half> %b, <4 x half> %c) #2
|
|
ret <2 x float> %vfmlal_low2.i
|
|
}
|
|
|
|
define <2 x float> @test_vfmlsl_low_u32(<2 x float> %a, <4 x half> %b, <4 x half> %c) #0 {
|
|
entry:
|
|
; CHECK-LABEL: test_vfmlsl_low_u32:
|
|
; CHECK: fmlsl v0.2s, v1.2h, v2.2h
|
|
%vfmlsl_low2.i = call <2 x float> @llvm.aarch64.neon.fmlsl.v2f32.v4f16(<2 x float> %a, <4 x half> %b, <4 x half> %c) #2
|
|
ret <2 x float> %vfmlsl_low2.i
|
|
}
|
|
|
|
define <2 x float> @test_vfmlal_high_u32(<2 x float> %a, <4 x half> %b, <4 x half> %c) #0 {
|
|
entry:
|
|
; CHECK-LABEL: test_vfmlal_high_u32:
|
|
; CHECK: fmlal2 v0.2s, v1.2h, v2.2h
|
|
%vfmlal_high2.i = call <2 x float> @llvm.aarch64.neon.fmlal2.v2f32.v4f16(<2 x float> %a, <4 x half> %b, <4 x half> %c) #2
|
|
ret <2 x float> %vfmlal_high2.i
|
|
}
|
|
|
|
define <2 x float> @test_vfmlsl_high_u32(<2 x float> %a, <4 x half> %b, <4 x half> %c) #0 {
|
|
entry:
|
|
; CHECK-LABEL: test_vfmlsl_high_u32:
|
|
; CHECK: fmlsl2 v0.2s, v1.2h, v2.2h
|
|
%vfmlsl_high2.i = call <2 x float> @llvm.aarch64.neon.fmlsl2.v2f32.v4f16(<2 x float> %a, <4 x half> %b, <4 x half> %c) #2
|
|
ret <2 x float> %vfmlsl_high2.i
|
|
}
|
|
|
|
define <4 x float> @test_vfmlalq_low_u32(<4 x float> %a, <8 x half> %b, <8 x half> %c) #0 {
|
|
entry:
|
|
; CHECK-LABEL: test_vfmlalq_low_u32:
|
|
; CHECK: fmlal v0.4s, v1.4h, v2.4h
|
|
%vfmlalq_low4.i = call <4 x float> @llvm.aarch64.neon.fmlal.v4f32.v8f16(<4 x float> %a, <8 x half> %b, <8 x half> %c) #2
|
|
ret <4 x float> %vfmlalq_low4.i
|
|
}
|
|
|
|
define <4 x float> @test_vfmlslq_low_u32(<4 x float> %a, <8 x half> %b, <8 x half> %c) #0 {
|
|
entry:
|
|
; CHECK-LABEL: test_vfmlslq_low_u32:
|
|
; CHECK: fmlsl v0.4s, v1.4h, v2.4h
|
|
%vfmlslq_low4.i = call <4 x float> @llvm.aarch64.neon.fmlsl.v4f32.v8f16(<4 x float> %a, <8 x half> %b, <8 x half> %c) #2
|
|
ret <4 x float> %vfmlslq_low4.i
|
|
}
|
|
|
|
define <4 x float> @test_vfmlalq_high_u32(<4 x float> %a, <8 x half> %b, <8 x half> %c) #0 {
|
|
entry:
|
|
; CHECK-LABEL: test_vfmlalq_high_u32:
|
|
; CHECK: fmlal2 v0.4s, v1.4h, v2.4h
|
|
%vfmlalq_high4.i = call <4 x float> @llvm.aarch64.neon.fmlal2.v4f32.v8f16(<4 x float> %a, <8 x half> %b, <8 x half> %c) #2
|
|
ret <4 x float> %vfmlalq_high4.i
|
|
}
|
|
|
|
define <4 x float> @test_vfmlslq_high_u32(<4 x float> %a, <8 x half> %b, <8 x half> %c) #0 {
|
|
entry:
|
|
; CHECK-LABEL: test_vfmlslq_high_u32:
|
|
; CHECK: fmlsl2 v0.4s, v1.4h, v2.4h
|
|
%vfmlslq_high4.i = call <4 x float> @llvm.aarch64.neon.fmlsl2.v4f32.v8f16(<4 x float> %a, <8 x half> %b, <8 x half> %c) #2
|
|
ret <4 x float> %vfmlslq_high4.i
|
|
}
|