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
199 lines
7.1 KiB
LLVM
199 lines
7.1 KiB
LLVM
; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py UTC_ARGS: --version 5
|
|
; RUN: llc --mtriple=loongarch32 -mattr=+32s,+lasx --verify-machineinstrs < %s \
|
|
; RUN: | FileCheck %s --check-prefix=LA32
|
|
; RUN: llc --mtriple=loongarch64 -mattr=+lasx --verify-machineinstrs < %s \
|
|
; RUN: | FileCheck %s --check-prefix=LA64
|
|
|
|
declare <32 x i8> @llvm.bitreverse.v32i8(<32 x i8>)
|
|
|
|
define <32 x i8> @test_bitreverse_v32i8(<32 x i8> %a) nounwind {
|
|
; LA32-LABEL: test_bitreverse_v32i8:
|
|
; LA32: # %bb.0:
|
|
; LA32-NEXT: xvslli.b $xr1, $xr0, 4
|
|
; LA32-NEXT: xvsrli.b $xr0, $xr0, 4
|
|
; LA32-NEXT: xvor.v $xr0, $xr0, $xr1
|
|
; LA32-NEXT: xvandi.b $xr1, $xr0, 51
|
|
; LA32-NEXT: xvslli.b $xr1, $xr1, 2
|
|
; LA32-NEXT: xvsrli.b $xr0, $xr0, 2
|
|
; LA32-NEXT: xvandi.b $xr0, $xr0, 51
|
|
; LA32-NEXT: xvor.v $xr0, $xr0, $xr1
|
|
; LA32-NEXT: xvandi.b $xr1, $xr0, 85
|
|
; LA32-NEXT: xvslli.b $xr1, $xr1, 1
|
|
; LA32-NEXT: xvsrli.b $xr0, $xr0, 1
|
|
; LA32-NEXT: xvandi.b $xr0, $xr0, 85
|
|
; LA32-NEXT: xvor.v $xr0, $xr0, $xr1
|
|
; LA32-NEXT: ret
|
|
;
|
|
; LA64-LABEL: test_bitreverse_v32i8:
|
|
; LA64: # %bb.0:
|
|
; LA64-NEXT: xvpickve2gr.d $a0, $xr0, 2
|
|
; LA64-NEXT: bitrev.8b $a0, $a0
|
|
; LA64-NEXT: vinsgr2vr.d $vr2, $a0, 0
|
|
; LA64-NEXT: xvpickve2gr.d $a0, $xr0, 3
|
|
; LA64-NEXT: bitrev.8b $a0, $a0
|
|
; LA64-NEXT: vinsgr2vr.d $vr2, $a0, 1
|
|
; LA64-NEXT: xvpickve2gr.d $a0, $xr0, 0
|
|
; LA64-NEXT: bitrev.8b $a0, $a0
|
|
; LA64-NEXT: vinsgr2vr.d $vr1, $a0, 0
|
|
; LA64-NEXT: xvpickve2gr.d $a0, $xr0, 1
|
|
; LA64-NEXT: bitrev.8b $a0, $a0
|
|
; LA64-NEXT: vinsgr2vr.d $vr1, $a0, 1
|
|
; LA64-NEXT: xvpermi.q $xr1, $xr2, 2
|
|
; LA64-NEXT: xvori.b $xr0, $xr1, 0
|
|
; LA64-NEXT: ret
|
|
%b = call <32 x i8> @llvm.bitreverse.v32i8(<32 x i8> %a)
|
|
ret <32 x i8> %b
|
|
}
|
|
|
|
declare <16 x i16> @llvm.bitreverse.v16i16(<16 x i16>)
|
|
|
|
define <16 x i16> @test_bitreverse_v16i16(<16 x i16> %a) nounwind {
|
|
; LA32-LABEL: test_bitreverse_v16i16:
|
|
; LA32: # %bb.0:
|
|
; LA32-NEXT: xvpickve2gr.w $a0, $xr0, 5
|
|
; LA32-NEXT: bitrev.w $a0, $a0
|
|
; LA32-NEXT: vinsgr2vr.w $vr1, $a0, 0
|
|
; LA32-NEXT: xvpickve2gr.w $a0, $xr0, 4
|
|
; LA32-NEXT: bitrev.w $a0, $a0
|
|
; LA32-NEXT: vinsgr2vr.w $vr1, $a0, 1
|
|
; LA32-NEXT: xvpickve2gr.w $a0, $xr0, 7
|
|
; LA32-NEXT: bitrev.w $a0, $a0
|
|
; LA32-NEXT: vinsgr2vr.w $vr1, $a0, 2
|
|
; LA32-NEXT: xvpickve2gr.w $a0, $xr0, 6
|
|
; LA32-NEXT: bitrev.w $a0, $a0
|
|
; LA32-NEXT: vinsgr2vr.w $vr1, $a0, 3
|
|
; LA32-NEXT: xvpickve2gr.w $a0, $xr0, 1
|
|
; LA32-NEXT: bitrev.w $a0, $a0
|
|
; LA32-NEXT: vinsgr2vr.w $vr2, $a0, 0
|
|
; LA32-NEXT: xvpickve2gr.w $a0, $xr0, 0
|
|
; LA32-NEXT: bitrev.w $a0, $a0
|
|
; LA32-NEXT: vinsgr2vr.w $vr2, $a0, 1
|
|
; LA32-NEXT: xvpickve2gr.w $a0, $xr0, 3
|
|
; LA32-NEXT: bitrev.w $a0, $a0
|
|
; LA32-NEXT: vinsgr2vr.w $vr2, $a0, 2
|
|
; LA32-NEXT: xvpickve2gr.w $a0, $xr0, 2
|
|
; LA32-NEXT: bitrev.w $a0, $a0
|
|
; LA32-NEXT: vinsgr2vr.w $vr2, $a0, 3
|
|
; LA32-NEXT: xvpermi.q $xr2, $xr1, 2
|
|
; LA32-NEXT: xvshuf4i.h $xr0, $xr2, 27
|
|
; LA32-NEXT: ret
|
|
;
|
|
; LA64-LABEL: test_bitreverse_v16i16:
|
|
; LA64: # %bb.0:
|
|
; LA64-NEXT: xvpickve2gr.d $a0, $xr0, 2
|
|
; LA64-NEXT: bitrev.d $a0, $a0
|
|
; LA64-NEXT: vinsgr2vr.d $vr1, $a0, 0
|
|
; LA64-NEXT: xvpickve2gr.d $a0, $xr0, 3
|
|
; LA64-NEXT: bitrev.d $a0, $a0
|
|
; LA64-NEXT: vinsgr2vr.d $vr1, $a0, 1
|
|
; LA64-NEXT: xvpickve2gr.d $a0, $xr0, 0
|
|
; LA64-NEXT: bitrev.d $a0, $a0
|
|
; LA64-NEXT: vinsgr2vr.d $vr2, $a0, 0
|
|
; LA64-NEXT: xvpickve2gr.d $a0, $xr0, 1
|
|
; LA64-NEXT: bitrev.d $a0, $a0
|
|
; LA64-NEXT: vinsgr2vr.d $vr2, $a0, 1
|
|
; LA64-NEXT: xvpermi.q $xr2, $xr1, 2
|
|
; LA64-NEXT: xvshuf4i.h $xr0, $xr2, 27
|
|
; LA64-NEXT: ret
|
|
%b = call <16 x i16> @llvm.bitreverse.v16i16(<16 x i16> %a)
|
|
ret <16 x i16> %b
|
|
}
|
|
|
|
declare <8 x i32> @llvm.bitreverse.v8i32(<8 x i32>)
|
|
|
|
define <8 x i32> @test_bitreverse_v8i32(<8 x i32> %a) nounwind {
|
|
; LA32-LABEL: test_bitreverse_v8i32:
|
|
; LA32: # %bb.0:
|
|
; LA32-NEXT: xvpickve2gr.w $a0, $xr0, 4
|
|
; LA32-NEXT: bitrev.w $a0, $a0
|
|
; LA32-NEXT: vinsgr2vr.w $vr2, $a0, 0
|
|
; LA32-NEXT: xvpickve2gr.w $a0, $xr0, 5
|
|
; LA32-NEXT: bitrev.w $a0, $a0
|
|
; LA32-NEXT: vinsgr2vr.w $vr2, $a0, 1
|
|
; LA32-NEXT: xvpickve2gr.w $a0, $xr0, 6
|
|
; LA32-NEXT: bitrev.w $a0, $a0
|
|
; LA32-NEXT: vinsgr2vr.w $vr2, $a0, 2
|
|
; LA32-NEXT: xvpickve2gr.w $a0, $xr0, 7
|
|
; LA32-NEXT: bitrev.w $a0, $a0
|
|
; LA32-NEXT: vinsgr2vr.w $vr2, $a0, 3
|
|
; LA32-NEXT: xvpickve2gr.w $a0, $xr0, 0
|
|
; LA32-NEXT: bitrev.w $a0, $a0
|
|
; LA32-NEXT: vinsgr2vr.w $vr1, $a0, 0
|
|
; LA32-NEXT: xvpickve2gr.w $a0, $xr0, 1
|
|
; LA32-NEXT: bitrev.w $a0, $a0
|
|
; LA32-NEXT: vinsgr2vr.w $vr1, $a0, 1
|
|
; LA32-NEXT: xvpickve2gr.w $a0, $xr0, 2
|
|
; LA32-NEXT: bitrev.w $a0, $a0
|
|
; LA32-NEXT: vinsgr2vr.w $vr1, $a0, 2
|
|
; LA32-NEXT: xvpickve2gr.w $a0, $xr0, 3
|
|
; LA32-NEXT: bitrev.w $a0, $a0
|
|
; LA32-NEXT: vinsgr2vr.w $vr1, $a0, 3
|
|
; LA32-NEXT: xvpermi.q $xr1, $xr2, 2
|
|
; LA32-NEXT: xvori.b $xr0, $xr1, 0
|
|
; LA32-NEXT: ret
|
|
;
|
|
; LA64-LABEL: test_bitreverse_v8i32:
|
|
; LA64: # %bb.0:
|
|
; LA64-NEXT: xvpickve2gr.d $a0, $xr0, 2
|
|
; LA64-NEXT: bitrev.d $a0, $a0
|
|
; LA64-NEXT: vinsgr2vr.d $vr1, $a0, 0
|
|
; LA64-NEXT: xvpickve2gr.d $a0, $xr0, 3
|
|
; LA64-NEXT: bitrev.d $a0, $a0
|
|
; LA64-NEXT: vinsgr2vr.d $vr1, $a0, 1
|
|
; LA64-NEXT: xvpickve2gr.d $a0, $xr0, 0
|
|
; LA64-NEXT: bitrev.d $a0, $a0
|
|
; LA64-NEXT: vinsgr2vr.d $vr2, $a0, 0
|
|
; LA64-NEXT: xvpickve2gr.d $a0, $xr0, 1
|
|
; LA64-NEXT: bitrev.d $a0, $a0
|
|
; LA64-NEXT: vinsgr2vr.d $vr2, $a0, 1
|
|
; LA64-NEXT: xvpermi.q $xr2, $xr1, 2
|
|
; LA64-NEXT: xvshuf4i.w $xr0, $xr2, 177
|
|
; LA64-NEXT: ret
|
|
%b = call <8 x i32> @llvm.bitreverse.v8i32(<8 x i32> %a)
|
|
ret <8 x i32> %b
|
|
}
|
|
|
|
declare <4 x i64> @llvm.bitreverse.v4i64(<4 x i64>)
|
|
|
|
define <4 x i64> @test_bitreverse_v4i64(<4 x i64> %a) nounwind {
|
|
; LA32-LABEL: test_bitreverse_v4i64:
|
|
; LA32: # %bb.0:
|
|
; LA32-NEXT: pcalau12i $a0, %pc_hi20(.LCPI3_0)
|
|
; LA32-NEXT: xvld $xr1, $a0, %pc_lo12(.LCPI3_0)
|
|
; LA32-NEXT: xvshuf.b $xr0, $xr0, $xr0, $xr1
|
|
; LA32-NEXT: xvslli.b $xr1, $xr0, 4
|
|
; LA32-NEXT: xvsrli.b $xr0, $xr0, 4
|
|
; LA32-NEXT: xvor.v $xr0, $xr0, $xr1
|
|
; LA32-NEXT: xvandi.b $xr1, $xr0, 51
|
|
; LA32-NEXT: xvslli.b $xr1, $xr1, 2
|
|
; LA32-NEXT: xvsrli.b $xr0, $xr0, 2
|
|
; LA32-NEXT: xvandi.b $xr0, $xr0, 51
|
|
; LA32-NEXT: xvor.v $xr0, $xr0, $xr1
|
|
; LA32-NEXT: xvandi.b $xr1, $xr0, 85
|
|
; LA32-NEXT: xvslli.b $xr1, $xr1, 1
|
|
; LA32-NEXT: xvsrli.b $xr0, $xr0, 1
|
|
; LA32-NEXT: xvandi.b $xr0, $xr0, 85
|
|
; LA32-NEXT: xvor.v $xr0, $xr0, $xr1
|
|
; LA32-NEXT: ret
|
|
;
|
|
; LA64-LABEL: test_bitreverse_v4i64:
|
|
; LA64: # %bb.0:
|
|
; LA64-NEXT: xvpickve2gr.d $a0, $xr0, 2
|
|
; LA64-NEXT: bitrev.d $a0, $a0
|
|
; LA64-NEXT: vinsgr2vr.d $vr2, $a0, 0
|
|
; LA64-NEXT: xvpickve2gr.d $a0, $xr0, 3
|
|
; LA64-NEXT: bitrev.d $a0, $a0
|
|
; LA64-NEXT: vinsgr2vr.d $vr2, $a0, 1
|
|
; LA64-NEXT: xvpickve2gr.d $a0, $xr0, 0
|
|
; LA64-NEXT: bitrev.d $a0, $a0
|
|
; LA64-NEXT: vinsgr2vr.d $vr1, $a0, 0
|
|
; LA64-NEXT: xvpickve2gr.d $a0, $xr0, 1
|
|
; LA64-NEXT: bitrev.d $a0, $a0
|
|
; LA64-NEXT: vinsgr2vr.d $vr1, $a0, 1
|
|
; LA64-NEXT: xvpermi.q $xr1, $xr2, 2
|
|
; LA64-NEXT: xvori.b $xr0, $xr1, 0
|
|
; LA64-NEXT: ret
|
|
%b = call <4 x i64> @llvm.bitreverse.v4i64(<4 x i64> %a)
|
|
ret <4 x i64> %b
|
|
}
|