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
203 lines
6.4 KiB
YAML
203 lines
6.4 KiB
YAML
# RUN: llc -march=hexagon -mcpu=hexagonv68 -mattr=+hvxv68,+hvx-length128b \
|
|
# RUN: -run-pass hexagon-qfp-optimizer -disable-qfp-opt-mul=false %s -o - | FileCheck %s --check-prefix=MUL-ENABLED
|
|
# RUN: llc -march=hexagon -mcpu=hexagonv68 -mattr=+hvxv68,+hvx-length128b \
|
|
# RUN: -run-pass hexagon-qfp-optimizer %s -o - | FileCheck %s --check-prefix=DEFAULT
|
|
# MUL-ENABLED-LABEL: name: qfpAdd32
|
|
# MUL-ENABLED: V6_vconv_sf_qf32
|
|
# MUL-ENABLED-NEXT: V6_vadd_qf32_mix
|
|
# MUL-ENABLED-NEXT: V6_vconv_sf_qf32
|
|
# MUL-ENABLED-NEXT: V6_vS32Ub_ai
|
|
# MUL-ENABLED-NEXT: V6_vadd_qf32
|
|
# DEFAULT-LABEL: name: qfpAdd32
|
|
# DEFAULT: V6_vconv_sf_qf32
|
|
# DEFAULT-NEXT: V6_vadd_qf32_mix
|
|
# DEFAULT-NEXT: V6_vconv_sf_qf32
|
|
# DEFAULT-NEXT: V6_vS32Ub_ai
|
|
# DEFAULT-NEXT: V6_vadd_qf32
|
|
---
|
|
name: qfpAdd32
|
|
tracksRegLiveness: true
|
|
body: |
|
|
bb.0:
|
|
liveins: $r0, $r1, $r2, $r3
|
|
%0:intregs = COPY $r0
|
|
%1:intregs = COPY $r1
|
|
%2:intregs = COPY $r2
|
|
%3:intregs = COPY $r3
|
|
%4:hvxvr = V6_vL32Ub_ai %0:intregs, 0
|
|
%5:hvxvr = V6_vL32Ub_ai %1:intregs, 0
|
|
%6:hvxvr = V6_vadd_sf %4:hvxvr, %5:hvxvr
|
|
%7:hvxvr = V6_vconv_sf_qf32 %6:hvxvr
|
|
%8:hvxvr = V6_vadd_sf %5:hvxvr, %7:hvxvr
|
|
%9:hvxvr = V6_vconv_sf_qf32 %8:hvxvr
|
|
V6_vS32Ub_ai %2:intregs, 0, %9:hvxvr
|
|
%10:hvxvr = V6_vadd_sf %7:hvxvr, %9:hvxvr
|
|
%11:hvxvr = V6_vconv_sf_qf32 %10:hvxvr
|
|
V6_vS32Ub_ai %3:intregs, 0, %11:hvxvr
|
|
...
|
|
# MUL-ENABLED-LABEL: name: qfpAdd16
|
|
# MUL-ENABLED: V6_vconv_hf_qf16
|
|
# MUL-ENABLED-NEXT: V6_vadd_qf16_mix
|
|
# MUL-ENABLED-NEXT: V6_vconv_hf_qf16
|
|
# MUL-ENABLED-NEXT: V6_vS32Ub_ai
|
|
# MUL-ENABLED-NEXT: V6_vadd_qf16
|
|
# DEFAULT-LABEL: name: qfpAdd16
|
|
# DEFAULT: V6_vconv_hf_qf16
|
|
# DEFAULT-NEXT: V6_vadd_qf16_mix
|
|
# DEFAULT-NEXT: V6_vconv_hf_qf16
|
|
# DEFAULT-NEXT: V6_vS32Ub_ai
|
|
# DEFAULT-NEXT: V6_vadd_qf16
|
|
---
|
|
name: qfpAdd16
|
|
tracksRegLiveness: true
|
|
body: |
|
|
bb.0:
|
|
liveins: $r0, $r1, $r2, $r3
|
|
%0:intregs = COPY $r0
|
|
%1:intregs = COPY $r1
|
|
%2:intregs = COPY $r2
|
|
%3:intregs = COPY $r3
|
|
%4:hvxvr = V6_vL32Ub_ai %0:intregs, 0
|
|
%5:hvxvr = V6_vL32Ub_ai %1:intregs, 0
|
|
%6:hvxvr = V6_vadd_hf %4:hvxvr, %5:hvxvr
|
|
%7:hvxvr = V6_vconv_hf_qf16 %6:hvxvr
|
|
%8:hvxvr = V6_vadd_hf %5:hvxvr, %7:hvxvr
|
|
%9:hvxvr = V6_vconv_hf_qf16 %8:hvxvr
|
|
V6_vS32Ub_ai %2:intregs, 0, %9:hvxvr
|
|
%10:hvxvr = V6_vadd_hf %7:hvxvr, %9:hvxvr
|
|
%11:hvxvr = V6_vconv_hf_qf16 %10:hvxvr
|
|
V6_vS32Ub_ai %3:intregs, 0, %11:hvxvr
|
|
...
|
|
# MUL-ENABLED-LABEL: name: qfpSub32
|
|
# MUL-ENABLED: V6_vconv_sf_qf32
|
|
# MUL-ENABLED-NEXT: V6_vsub_qf32_mix
|
|
# MUL-ENABLED-NEXT: V6_vconv_sf_qf32
|
|
# MUL-ENABLED-NEXT: V6_vS32Ub_ai
|
|
# MUL-ENABLED-NEXT: V6_vsub_qf32
|
|
# DEFAULT-LABEL: name: qfpSub32
|
|
# DEFAULT: V6_vconv_sf_qf32
|
|
# DEFAULT-NEXT: V6_vsub_qf32_mix
|
|
# DEFAULT-NEXT: V6_vconv_sf_qf32
|
|
# DEFAULT-NEXT: V6_vS32Ub_ai
|
|
# DEFAULT-NEXT: V6_vsub_qf32
|
|
---
|
|
name: qfpSub32
|
|
tracksRegLiveness: true
|
|
body: |
|
|
bb.0:
|
|
liveins: $r0, $r1, $r2, $r3
|
|
%0:intregs = COPY $r0
|
|
%1:intregs = COPY $r1
|
|
%2:intregs = COPY $r2
|
|
%3:intregs = COPY $r3
|
|
%4:hvxvr = V6_vL32Ub_ai %0:intregs, 0
|
|
%5:hvxvr = V6_vL32Ub_ai %1:intregs, 0
|
|
%6:hvxvr = V6_vsub_sf %4:hvxvr, %5:hvxvr
|
|
%7:hvxvr = V6_vconv_sf_qf32 %6:hvxvr
|
|
%8:hvxvr = V6_vsub_sf %7:hvxvr, %5:hvxvr
|
|
%9:hvxvr = V6_vconv_sf_qf32 %8:hvxvr
|
|
V6_vS32Ub_ai %2:intregs, 0, %9:hvxvr
|
|
%10:hvxvr = V6_vsub_sf %7:hvxvr, %9:hvxvr
|
|
%11:hvxvr = V6_vconv_sf_qf32 %10:hvxvr
|
|
V6_vS32Ub_ai %3:intregs, 0, %11:hvxvr
|
|
...
|
|
# MUL-ENABLED-LABEL: name: qfpSub16
|
|
# MUL-ENABLED: V6_vconv_hf_qf16
|
|
# MUL-ENABLED-NEXT: V6_vsub_qf16_mix
|
|
# MUL-ENABLED-NEXT: V6_vconv_hf_qf16
|
|
# MUL-ENABLED-NEXT: V6_vS32Ub_ai
|
|
# MUL-ENABLED-NEXT: V6_vsub_qf16
|
|
# DEFAULT-LABEL: name: qfpSub16
|
|
# DEFAULT: V6_vconv_hf_qf16
|
|
# DEFAULT-NEXT: V6_vsub_qf16_mix
|
|
# DEFAULT-NEXT: V6_vconv_hf_qf16
|
|
# DEFAULT-NEXT: V6_vS32Ub_ai
|
|
# DEFAULT-NEXT: V6_vsub_qf16
|
|
---
|
|
name: qfpSub16
|
|
tracksRegLiveness: true
|
|
body: |
|
|
bb.0:
|
|
liveins: $r0, $r1, $r2, $r3
|
|
%0:intregs = COPY $r0
|
|
%1:intregs = COPY $r1
|
|
%2:intregs = COPY $r2
|
|
%3:intregs = COPY $r3
|
|
%4:hvxvr = V6_vL32Ub_ai %0:intregs, 0
|
|
%5:hvxvr = V6_vL32Ub_ai %1:intregs, 0
|
|
%6:hvxvr = V6_vsub_hf %4:hvxvr, %5:hvxvr
|
|
%7:hvxvr = V6_vconv_hf_qf16 %6:hvxvr
|
|
%8:hvxvr = V6_vsub_hf %7:hvxvr, %5:hvxvr
|
|
%9:hvxvr = V6_vconv_hf_qf16 %8:hvxvr
|
|
V6_vS32Ub_ai %2:intregs, 0, %9:hvxvr
|
|
%10:hvxvr = V6_vsub_hf %7:hvxvr, %9:hvxvr
|
|
%11:hvxvr = V6_vconv_hf_qf16 %10:hvxvr
|
|
V6_vS32Ub_ai %3:intregs, 0, %11:hvxvr
|
|
...
|
|
# MUL-ENABLED-LABEL: name: qfpMul32
|
|
# MUL-ENABLED: V6_vmpy_qf32_sf
|
|
# MUL-ENABLED-NEXT: V6_vconv_sf_qf32
|
|
# MUL-ENABLED-NEXT: V6_vmpy_qf32_sf
|
|
# MUL-ENABLED-NEXT: V6_vconv_sf_qf32
|
|
# MUL-ENABLED-NEXT: V6_vmpy_qf32
|
|
# MUL-ENABLED-NEXT: V6_vS32Ub_ai
|
|
# DEFAULT-LABEL: name: qfpMul32
|
|
# DEFAULT: V6_vmpy_qf32_sf
|
|
# DEFAULT-NEXT: V6_vconv_sf_qf32
|
|
# DEFAULT-NEXT: V6_vmpy_qf32_sf
|
|
# DEFAULT-NEXT: V6_vconv_sf_qf32
|
|
# DEFAULT-NEXT: V6_vmpy_qf32_sf
|
|
# DEFAULT-NEXT: V6_vS32Ub_ai
|
|
---
|
|
name: qfpMul32
|
|
tracksRegLiveness: true
|
|
body: |
|
|
bb.0:
|
|
liveins: $r0, $r1, $r2, $r3
|
|
%0:intregs = COPY $r0
|
|
%1:intregs = COPY $r1
|
|
%2:intregs = COPY $r2
|
|
%3:intregs = COPY $r3
|
|
%4:hvxvr = V6_vL32Ub_ai %0:intregs, 0
|
|
%5:hvxvr = V6_vL32Ub_ai %1:intregs, 0
|
|
%6:hvxvr = V6_vL32Ub_ai %2:intregs, 0
|
|
%7:hvxvr = V6_vmpy_qf32_sf %4:hvxvr, %5:hvxvr
|
|
%8:hvxvr = V6_vconv_sf_qf32 %7:hvxvr
|
|
%9:hvxvr = V6_vmpy_qf32_sf %5:hvxvr, %6:hvxvr
|
|
%10:hvxvr = V6_vconv_sf_qf32 %9:hvxvr
|
|
%11:hvxvr = V6_vmpy_qf32_sf %8:hvxvr, %10:hvxvr
|
|
V6_vS32Ub_ai %3:intregs, 0, %11:hvxvr
|
|
...
|
|
# MUL-ENABLED-LABEL: name: qfpMul16
|
|
# MUL-ENABLED: V6_vconv_hf_qf16
|
|
# MUL-ENABLED-NEXT: V6_vmpy_qf16_mix_hf
|
|
# MUL-ENABLED-NEXT: V6_vconv_hf_qf16
|
|
# MUL-ENABLED-NEXT: V6_vS32Ub_ai
|
|
# MUL-ENABLED-NEXT: V6_vmpy_qf16
|
|
# DEFAULT-LABEL: name: qfpMul16
|
|
# DEFAULT: V6_vconv_hf_qf16
|
|
# DEFAULT-NEXT: V6_vmpy_qf16_hf
|
|
# DEFAULT-NEXT: V6_vconv_hf_qf16
|
|
# DEFAULT-NEXT: V6_vS32Ub_ai
|
|
# DEFAULT-NEXT: V6_vmpy_qf16_hf
|
|
---
|
|
name: qfpMul16
|
|
tracksRegLiveness: true
|
|
body: |
|
|
bb.0:
|
|
liveins: $r0, $r1, $r2, $r3
|
|
%0:intregs = COPY $r0
|
|
%1:intregs = COPY $r1
|
|
%2:intregs = COPY $r2
|
|
%3:intregs = COPY $r3
|
|
%4:hvxvr = V6_vL32Ub_ai %0:intregs, 0
|
|
%5:hvxvr = V6_vL32Ub_ai %1:intregs, 0
|
|
%6:hvxvr = V6_vmpy_qf16_hf %4:hvxvr, %5:hvxvr
|
|
%7:hvxvr = V6_vconv_hf_qf16 %6:hvxvr
|
|
%8:hvxvr = V6_vmpy_qf16_hf %5:hvxvr, %7:hvxvr
|
|
%9:hvxvr = V6_vconv_hf_qf16 %8:hvxvr
|
|
V6_vS32Ub_ai %2:intregs, 0, %9:hvxvr
|
|
%10:hvxvr = V6_vmpy_qf16_hf %7:hvxvr, %9:hvxvr
|
|
%11:hvxvr = V6_vconv_hf_qf16 %10:hvxvr
|
|
V6_vS32Ub_ai %3:intregs, 0, %11:hvxvr
|