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
370 lines
19 KiB
YAML
370 lines
19 KiB
YAML
# NOTE: Assertions have been autogenerated by utils/update_mir_test_checks.py
|
|
# RUN: llc -mtriple=amdgcn-amd-amdhsa -mcpu=hawaii -run-pass=instruction-select -verify-machineinstrs -o - %s | FileCheck %s -check-prefix=GCN
|
|
# RUN: llc -mtriple=amdgcn-amd-amdhsa -mcpu=fiji -run-pass=instruction-select -verify-machineinstrs -o - %s | FileCheck %s -check-prefix=VI
|
|
# RUN: llc -mtriple=amdgcn-amd-amdhsa -mcpu=gfx1100 -mattr=+real-true16 -run-pass=instruction-select -verify-machineinstrs -o - %s | FileCheck %s -check-prefixes=GFX11,GFX11-TRUE16
|
|
# RUN: llc -mtriple=amdgcn-amd-amdhsa -mcpu=gfx1100 -mattr=-real-true16 -run-pass=instruction-select -verify-machineinstrs -o - %s | FileCheck %s -check-prefixes=GFX11,GFX11-FAKE16
|
|
|
|
---
|
|
|
|
name: fptoui
|
|
legalized: true
|
|
regBankSelected: true
|
|
|
|
body: |
|
|
bb.0:
|
|
liveins: $sgpr0, $vgpr0, $vgpr3_vgpr4
|
|
|
|
; GCN-LABEL: name: fptoui
|
|
; GCN: liveins: $sgpr0, $vgpr0, $vgpr3_vgpr4
|
|
; GCN-NEXT: {{ $}}
|
|
; GCN-NEXT: [[COPY:%[0-9]+]]:sreg_32 = COPY $sgpr0
|
|
; GCN-NEXT: [[COPY1:%[0-9]+]]:vgpr_32 = COPY $vgpr0
|
|
; GCN-NEXT: [[COPY2:%[0-9]+]]:vreg_64 = COPY $vgpr3_vgpr4
|
|
; GCN-NEXT: [[V_CVT_U32_F32_e64_:%[0-9]+]]:vgpr_32 = nofpexcept V_CVT_U32_F32_e64 0, [[COPY]], 0, 0, implicit $mode, implicit $exec
|
|
; GCN-NEXT: [[V_CVT_U32_F32_e64_1:%[0-9]+]]:vgpr_32 = nofpexcept V_CVT_U32_F32_e64 0, [[COPY1]], 0, 0, implicit $mode, implicit $exec
|
|
; GCN-NEXT: FLAT_STORE_DWORD [[COPY2]], [[V_CVT_U32_F32_e64_]], 0, 0, implicit $exec, implicit $flat_scr :: (store (s32), addrspace 1)
|
|
; GCN-NEXT: FLAT_STORE_DWORD [[COPY2]], [[V_CVT_U32_F32_e64_1]], 0, 0, implicit $exec, implicit $flat_scr :: (store (s32), addrspace 1)
|
|
;
|
|
; VI-LABEL: name: fptoui
|
|
; VI: liveins: $sgpr0, $vgpr0, $vgpr3_vgpr4
|
|
; VI-NEXT: {{ $}}
|
|
; VI-NEXT: [[COPY:%[0-9]+]]:sreg_32 = COPY $sgpr0
|
|
; VI-NEXT: [[COPY1:%[0-9]+]]:vgpr_32 = COPY $vgpr0
|
|
; VI-NEXT: [[COPY2:%[0-9]+]]:vreg_64 = COPY $vgpr3_vgpr4
|
|
; VI-NEXT: [[V_CVT_U32_F32_e64_:%[0-9]+]]:vgpr_32 = nofpexcept V_CVT_U32_F32_e64 0, [[COPY]], 0, 0, implicit $mode, implicit $exec
|
|
; VI-NEXT: [[V_CVT_U32_F32_e64_1:%[0-9]+]]:vgpr_32 = nofpexcept V_CVT_U32_F32_e64 0, [[COPY1]], 0, 0, implicit $mode, implicit $exec
|
|
; VI-NEXT: FLAT_STORE_DWORD [[COPY2]], [[V_CVT_U32_F32_e64_]], 0, 0, implicit $exec, implicit $flat_scr :: (store (s32), addrspace 1)
|
|
; VI-NEXT: FLAT_STORE_DWORD [[COPY2]], [[V_CVT_U32_F32_e64_1]], 0, 0, implicit $exec, implicit $flat_scr :: (store (s32), addrspace 1)
|
|
;
|
|
; GFX11-LABEL: name: fptoui
|
|
; GFX11: liveins: $sgpr0, $vgpr0, $vgpr3_vgpr4
|
|
; GFX11-NEXT: {{ $}}
|
|
; GFX11-NEXT: [[COPY:%[0-9]+]]:sreg_32 = COPY $sgpr0
|
|
; GFX11-NEXT: [[COPY1:%[0-9]+]]:vgpr_32 = COPY $vgpr0
|
|
; GFX11-NEXT: [[COPY2:%[0-9]+]]:vreg_64 = COPY $vgpr3_vgpr4
|
|
; GFX11-NEXT: [[V_CVT_U32_F32_e64_:%[0-9]+]]:vgpr_32 = nofpexcept V_CVT_U32_F32_e64 0, [[COPY]], 0, 0, implicit $mode, implicit $exec
|
|
; GFX11-NEXT: [[V_CVT_U32_F32_e64_1:%[0-9]+]]:vgpr_32 = nofpexcept V_CVT_U32_F32_e64 0, [[COPY1]], 0, 0, implicit $mode, implicit $exec
|
|
; GFX11-NEXT: GLOBAL_STORE_DWORD [[COPY2]], [[V_CVT_U32_F32_e64_]], 0, 0, implicit $exec :: (store (s32), addrspace 1)
|
|
; GFX11-NEXT: GLOBAL_STORE_DWORD [[COPY2]], [[V_CVT_U32_F32_e64_1]], 0, 0, implicit $exec :: (store (s32), addrspace 1)
|
|
%0:sgpr(s32) = COPY $sgpr0
|
|
|
|
%1:vgpr(s32) = COPY $vgpr0
|
|
|
|
%2:vgpr(p1) = COPY $vgpr3_vgpr4
|
|
|
|
%3:vgpr(s32) = G_FPTOUI %0
|
|
|
|
%4:vgpr(s32) = G_FPTOUI %1
|
|
|
|
G_STORE %3, %2 :: (store (s32), addrspace 1)
|
|
G_STORE %4, %2 :: (store (s32), addrspace 1)
|
|
...
|
|
|
|
---
|
|
name: fptoui_s16_to_s32_vv
|
|
legalized: true
|
|
regBankSelected: true
|
|
tracksRegLiveness: true
|
|
|
|
body: |
|
|
bb.0:
|
|
liveins: $vgpr0
|
|
|
|
; GCN-LABEL: name: fptoui_s16_to_s32_vv
|
|
; GCN: liveins: $vgpr0
|
|
; GCN-NEXT: {{ $}}
|
|
; GCN-NEXT: [[COPY:%[0-9]+]]:vgpr_32 = COPY $vgpr0
|
|
; GCN-NEXT: [[V_CVT_F32_F16_e64_:%[0-9]+]]:vgpr_32 = nofpexcept V_CVT_F32_F16_e64 0, [[COPY]], 0, 0, implicit $mode, implicit $exec
|
|
; GCN-NEXT: [[V_CVT_U32_F32_e32_:%[0-9]+]]:vgpr_32 = nofpexcept V_CVT_U32_F32_e32 [[V_CVT_F32_F16_e64_]], implicit $mode, implicit $exec
|
|
; GCN-NEXT: $vgpr0 = COPY [[V_CVT_U32_F32_e32_]]
|
|
;
|
|
; VI-LABEL: name: fptoui_s16_to_s32_vv
|
|
; VI: liveins: $vgpr0
|
|
; VI-NEXT: {{ $}}
|
|
; VI-NEXT: [[COPY:%[0-9]+]]:vgpr_32 = COPY $vgpr0
|
|
; VI-NEXT: [[V_CVT_F32_F16_e64_:%[0-9]+]]:vgpr_32 = nofpexcept V_CVT_F32_F16_e64 0, [[COPY]], 0, 0, implicit $mode, implicit $exec
|
|
; VI-NEXT: [[V_CVT_U32_F32_e32_:%[0-9]+]]:vgpr_32 = nofpexcept V_CVT_U32_F32_e32 [[V_CVT_F32_F16_e64_]], implicit $mode, implicit $exec
|
|
; VI-NEXT: $vgpr0 = COPY [[V_CVT_U32_F32_e32_]]
|
|
;
|
|
; GFX11-TRUE16-LABEL: name: fptoui_s16_to_s32_vv
|
|
; GFX11-TRUE16: liveins: $vgpr0
|
|
; GFX11-TRUE16-NEXT: {{ $}}
|
|
; GFX11-TRUE16-NEXT: [[COPY:%[0-9]+]]:vgpr_32 = COPY $vgpr0
|
|
; GFX11-TRUE16-NEXT: [[COPY1:%[0-9]+]]:vgpr_16 = COPY [[COPY]].lo16
|
|
; GFX11-TRUE16-NEXT: [[V_CVT_F32_F16_t16_e64_:%[0-9]+]]:vgpr_32 = nofpexcept V_CVT_F32_F16_t16_e64 0, [[COPY1]], 0, 0, 0, implicit $mode, implicit $exec
|
|
; GFX11-TRUE16-NEXT: [[V_CVT_U32_F32_e32_:%[0-9]+]]:vgpr_32 = nofpexcept V_CVT_U32_F32_e32 [[V_CVT_F32_F16_t16_e64_]], implicit $mode, implicit $exec
|
|
; GFX11-TRUE16-NEXT: $vgpr0 = COPY [[V_CVT_U32_F32_e32_]]
|
|
;
|
|
; GFX11-FAKE16-LABEL: name: fptoui_s16_to_s32_vv
|
|
; GFX11-FAKE16: liveins: $vgpr0
|
|
; GFX11-FAKE16-NEXT: {{ $}}
|
|
; GFX11-FAKE16-NEXT: [[COPY:%[0-9]+]]:vgpr_32 = COPY $vgpr0
|
|
; GFX11-FAKE16-NEXT: [[V_CVT_F32_F16_fake16_e64_:%[0-9]+]]:vgpr_32 = nofpexcept V_CVT_F32_F16_fake16_e64 0, [[COPY]], 0, 0, implicit $mode, implicit $exec
|
|
; GFX11-FAKE16-NEXT: [[V_CVT_U32_F32_e32_:%[0-9]+]]:vgpr_32 = nofpexcept V_CVT_U32_F32_e32 [[V_CVT_F32_F16_fake16_e64_]], implicit $mode, implicit $exec
|
|
; GFX11-FAKE16-NEXT: $vgpr0 = COPY [[V_CVT_U32_F32_e32_]]
|
|
%0:vgpr(s32) = COPY $vgpr0
|
|
%1:vgpr(s16) = G_TRUNC %0
|
|
%2:vgpr(s32) = G_FPTOUI %1
|
|
$vgpr0 = COPY %2
|
|
...
|
|
|
|
---
|
|
name: fptoui_s16_to_s32_vs
|
|
legalized: true
|
|
regBankSelected: true
|
|
tracksRegLiveness: true
|
|
|
|
body: |
|
|
bb.0:
|
|
liveins: $sgpr0
|
|
|
|
; GCN-LABEL: name: fptoui_s16_to_s32_vs
|
|
; GCN: liveins: $sgpr0
|
|
; GCN-NEXT: {{ $}}
|
|
; GCN-NEXT: [[COPY:%[0-9]+]]:sreg_32 = COPY $sgpr0
|
|
; GCN-NEXT: [[V_CVT_F32_F16_e64_:%[0-9]+]]:vgpr_32 = nofpexcept V_CVT_F32_F16_e64 0, [[COPY]], 0, 0, implicit $mode, implicit $exec
|
|
; GCN-NEXT: [[V_CVT_U32_F32_e32_:%[0-9]+]]:vgpr_32 = nofpexcept V_CVT_U32_F32_e32 [[V_CVT_F32_F16_e64_]], implicit $mode, implicit $exec
|
|
; GCN-NEXT: $vgpr0 = COPY [[V_CVT_U32_F32_e32_]]
|
|
;
|
|
; VI-LABEL: name: fptoui_s16_to_s32_vs
|
|
; VI: liveins: $sgpr0
|
|
; VI-NEXT: {{ $}}
|
|
; VI-NEXT: [[COPY:%[0-9]+]]:sreg_32 = COPY $sgpr0
|
|
; VI-NEXT: [[V_CVT_F32_F16_e64_:%[0-9]+]]:vgpr_32 = nofpexcept V_CVT_F32_F16_e64 0, [[COPY]], 0, 0, implicit $mode, implicit $exec
|
|
; VI-NEXT: [[V_CVT_U32_F32_e32_:%[0-9]+]]:vgpr_32 = nofpexcept V_CVT_U32_F32_e32 [[V_CVT_F32_F16_e64_]], implicit $mode, implicit $exec
|
|
; VI-NEXT: $vgpr0 = COPY [[V_CVT_U32_F32_e32_]]
|
|
;
|
|
; GFX11-TRUE16-LABEL: name: fptoui_s16_to_s32_vs
|
|
; GFX11-TRUE16: liveins: $sgpr0
|
|
; GFX11-TRUE16-NEXT: {{ $}}
|
|
; GFX11-TRUE16-NEXT: [[COPY:%[0-9]+]]:sreg_32 = COPY $sgpr0
|
|
; GFX11-TRUE16-NEXT: [[V_CVT_F32_F16_t16_e64_:%[0-9]+]]:vgpr_32 = nofpexcept V_CVT_F32_F16_t16_e64 0, [[COPY]], 0, 0, 0, implicit $mode, implicit $exec
|
|
; GFX11-TRUE16-NEXT: [[V_CVT_U32_F32_e32_:%[0-9]+]]:vgpr_32 = nofpexcept V_CVT_U32_F32_e32 [[V_CVT_F32_F16_t16_e64_]], implicit $mode, implicit $exec
|
|
; GFX11-TRUE16-NEXT: $vgpr0 = COPY [[V_CVT_U32_F32_e32_]]
|
|
;
|
|
; GFX11-FAKE16-LABEL: name: fptoui_s16_to_s32_vs
|
|
; GFX11-FAKE16: liveins: $sgpr0
|
|
; GFX11-FAKE16-NEXT: {{ $}}
|
|
; GFX11-FAKE16-NEXT: [[COPY:%[0-9]+]]:sreg_32 = COPY $sgpr0
|
|
; GFX11-FAKE16-NEXT: [[V_CVT_F32_F16_fake16_e64_:%[0-9]+]]:vgpr_32 = nofpexcept V_CVT_F32_F16_fake16_e64 0, [[COPY]], 0, 0, implicit $mode, implicit $exec
|
|
; GFX11-FAKE16-NEXT: [[V_CVT_U32_F32_e32_:%[0-9]+]]:vgpr_32 = nofpexcept V_CVT_U32_F32_e32 [[V_CVT_F32_F16_fake16_e64_]], implicit $mode, implicit $exec
|
|
; GFX11-FAKE16-NEXT: $vgpr0 = COPY [[V_CVT_U32_F32_e32_]]
|
|
%0:sgpr(s32) = COPY $sgpr0
|
|
%1:sgpr(s16) = G_TRUNC %0
|
|
%2:vgpr(s32) = G_FPTOUI %1
|
|
$vgpr0 = COPY %2
|
|
...
|
|
|
|
---
|
|
name: fptoui_s16_to_s32_fneg_vv
|
|
legalized: true
|
|
regBankSelected: true
|
|
tracksRegLiveness: true
|
|
|
|
body: |
|
|
bb.0:
|
|
liveins: $vgpr0
|
|
|
|
; GCN-LABEL: name: fptoui_s16_to_s32_fneg_vv
|
|
; GCN: liveins: $vgpr0
|
|
; GCN-NEXT: {{ $}}
|
|
; GCN-NEXT: [[COPY:%[0-9]+]]:vgpr_32 = COPY $vgpr0
|
|
; GCN-NEXT: [[S_MOV_B32_:%[0-9]+]]:sreg_32 = S_MOV_B32 32768
|
|
; GCN-NEXT: [[V_XOR_B32_e64_:%[0-9]+]]:vgpr_32 = V_XOR_B32_e64 [[S_MOV_B32_]], [[COPY]], implicit $exec
|
|
; GCN-NEXT: [[V_CVT_F32_F16_e64_:%[0-9]+]]:vgpr_32 = nofpexcept V_CVT_F32_F16_e64 0, [[V_XOR_B32_e64_]], 0, 0, implicit $mode, implicit $exec
|
|
; GCN-NEXT: [[V_CVT_U32_F32_e32_:%[0-9]+]]:vgpr_32 = nofpexcept V_CVT_U32_F32_e32 [[V_CVT_F32_F16_e64_]], implicit $mode, implicit $exec
|
|
; GCN-NEXT: $vgpr0 = COPY [[V_CVT_U32_F32_e32_]]
|
|
;
|
|
; VI-LABEL: name: fptoui_s16_to_s32_fneg_vv
|
|
; VI: liveins: $vgpr0
|
|
; VI-NEXT: {{ $}}
|
|
; VI-NEXT: [[COPY:%[0-9]+]]:vgpr_32 = COPY $vgpr0
|
|
; VI-NEXT: [[S_MOV_B32_:%[0-9]+]]:sreg_32 = S_MOV_B32 32768
|
|
; VI-NEXT: [[V_XOR_B32_e64_:%[0-9]+]]:vgpr_32 = V_XOR_B32_e64 [[S_MOV_B32_]], [[COPY]], implicit $exec
|
|
; VI-NEXT: [[V_CVT_F32_F16_e64_:%[0-9]+]]:vgpr_32 = nofpexcept V_CVT_F32_F16_e64 0, [[V_XOR_B32_e64_]], 0, 0, implicit $mode, implicit $exec
|
|
; VI-NEXT: [[V_CVT_U32_F32_e32_:%[0-9]+]]:vgpr_32 = nofpexcept V_CVT_U32_F32_e32 [[V_CVT_F32_F16_e64_]], implicit $mode, implicit $exec
|
|
; VI-NEXT: $vgpr0 = COPY [[V_CVT_U32_F32_e32_]]
|
|
;
|
|
; GFX11-TRUE16-LABEL: name: fptoui_s16_to_s32_fneg_vv
|
|
; GFX11-TRUE16: liveins: $vgpr0
|
|
; GFX11-TRUE16-NEXT: {{ $}}
|
|
; GFX11-TRUE16-NEXT: [[COPY:%[0-9]+]]:vgpr_32 = COPY $vgpr0
|
|
; GFX11-TRUE16-NEXT: [[COPY1:%[0-9]+]]:vgpr_16 = COPY [[COPY]].lo16
|
|
; GFX11-TRUE16-NEXT: [[V_XOR_B16_t16_e64_:%[0-9]+]]:vgpr_16 = V_XOR_B16_t16_e64 0, 32768, 0, [[COPY1]], 0, implicit $exec
|
|
; GFX11-TRUE16-NEXT: [[V_CVT_F32_F16_t16_e64_:%[0-9]+]]:vgpr_32 = nofpexcept V_CVT_F32_F16_t16_e64 0, [[V_XOR_B16_t16_e64_]], 0, 0, 0, implicit $mode, implicit $exec
|
|
; GFX11-TRUE16-NEXT: [[V_CVT_U32_F32_e32_:%[0-9]+]]:vgpr_32 = nofpexcept V_CVT_U32_F32_e32 [[V_CVT_F32_F16_t16_e64_]], implicit $mode, implicit $exec
|
|
; GFX11-TRUE16-NEXT: $vgpr0 = COPY [[V_CVT_U32_F32_e32_]]
|
|
;
|
|
; GFX11-FAKE16-LABEL: name: fptoui_s16_to_s32_fneg_vv
|
|
; GFX11-FAKE16: liveins: $vgpr0
|
|
; GFX11-FAKE16-NEXT: {{ $}}
|
|
; GFX11-FAKE16-NEXT: [[COPY:%[0-9]+]]:vgpr_32 = COPY $vgpr0
|
|
; GFX11-FAKE16-NEXT: [[S_MOV_B32_:%[0-9]+]]:sreg_32 = S_MOV_B32 32768
|
|
; GFX11-FAKE16-NEXT: [[V_XOR_B32_e64_:%[0-9]+]]:vgpr_32 = V_XOR_B32_e64 [[S_MOV_B32_]], [[COPY]], implicit $exec
|
|
; GFX11-FAKE16-NEXT: [[V_CVT_F32_F16_fake16_e64_:%[0-9]+]]:vgpr_32 = nofpexcept V_CVT_F32_F16_fake16_e64 0, [[V_XOR_B32_e64_]], 0, 0, implicit $mode, implicit $exec
|
|
; GFX11-FAKE16-NEXT: [[V_CVT_U32_F32_e32_:%[0-9]+]]:vgpr_32 = nofpexcept V_CVT_U32_F32_e32 [[V_CVT_F32_F16_fake16_e64_]], implicit $mode, implicit $exec
|
|
; GFX11-FAKE16-NEXT: $vgpr0 = COPY [[V_CVT_U32_F32_e32_]]
|
|
%0:vgpr(s32) = COPY $vgpr0
|
|
%1:vgpr(s16) = G_TRUNC %0
|
|
%2:vgpr(s16) = G_FNEG %1
|
|
%3:vgpr(s32) = G_FPTOUI %2
|
|
$vgpr0 = COPY %3
|
|
...
|
|
|
|
---
|
|
name: fptoui_s16_to_s1_vv
|
|
legalized: true
|
|
regBankSelected: true
|
|
tracksRegLiveness: true
|
|
|
|
body: |
|
|
bb.0:
|
|
liveins: $vgpr0
|
|
|
|
; GCN-LABEL: name: fptoui_s16_to_s1_vv
|
|
; GCN: liveins: $vgpr0
|
|
; GCN-NEXT: {{ $}}
|
|
; GCN-NEXT: [[COPY:%[0-9]+]]:vgpr_32 = COPY $vgpr0
|
|
; GCN-NEXT: [[V_CVT_F32_F16_e64_:%[0-9]+]]:vgpr_32 = nofpexcept V_CVT_F32_F16_e64 0, [[COPY]], 0, 0, implicit $mode, implicit $exec
|
|
; GCN-NEXT: [[V_CVT_U32_F32_e32_:%[0-9]+]]:vgpr_32 = nofpexcept V_CVT_U32_F32_e32 [[V_CVT_F32_F16_e64_]], implicit $mode, implicit $exec
|
|
; GCN-NEXT: S_ENDPGM 0, implicit [[V_CVT_U32_F32_e32_]]
|
|
;
|
|
; VI-LABEL: name: fptoui_s16_to_s1_vv
|
|
; VI: liveins: $vgpr0
|
|
; VI-NEXT: {{ $}}
|
|
; VI-NEXT: [[COPY:%[0-9]+]]:vgpr_32 = COPY $vgpr0
|
|
; VI-NEXT: [[V_CVT_F32_F16_e64_:%[0-9]+]]:vgpr_32 = nofpexcept V_CVT_F32_F16_e64 0, [[COPY]], 0, 0, implicit $mode, implicit $exec
|
|
; VI-NEXT: [[V_CVT_U32_F32_e32_:%[0-9]+]]:vgpr_32 = nofpexcept V_CVT_U32_F32_e32 [[V_CVT_F32_F16_e64_]], implicit $mode, implicit $exec
|
|
; VI-NEXT: S_ENDPGM 0, implicit [[V_CVT_U32_F32_e32_]]
|
|
;
|
|
; GFX11-TRUE16-LABEL: name: fptoui_s16_to_s1_vv
|
|
; GFX11-TRUE16: liveins: $vgpr0
|
|
; GFX11-TRUE16-NEXT: {{ $}}
|
|
; GFX11-TRUE16-NEXT: [[COPY:%[0-9]+]]:vgpr_32 = COPY $vgpr0
|
|
; GFX11-TRUE16-NEXT: [[COPY1:%[0-9]+]]:vgpr_16 = COPY [[COPY]].lo16
|
|
; GFX11-TRUE16-NEXT: [[V_CVT_F32_F16_t16_e64_:%[0-9]+]]:vgpr_32 = nofpexcept V_CVT_F32_F16_t16_e64 0, [[COPY1]], 0, 0, 0, implicit $mode, implicit $exec
|
|
; GFX11-TRUE16-NEXT: [[V_CVT_U32_F32_e32_:%[0-9]+]]:vgpr_32 = nofpexcept V_CVT_U32_F32_e32 [[V_CVT_F32_F16_t16_e64_]], implicit $mode, implicit $exec
|
|
; GFX11-TRUE16-NEXT: [[COPY2:%[0-9]+]]:vgpr_16 = COPY [[V_CVT_U32_F32_e32_]].lo16
|
|
; GFX11-TRUE16-NEXT: S_ENDPGM 0, implicit [[COPY2]]
|
|
;
|
|
; GFX11-FAKE16-LABEL: name: fptoui_s16_to_s1_vv
|
|
; GFX11-FAKE16: liveins: $vgpr0
|
|
; GFX11-FAKE16-NEXT: {{ $}}
|
|
; GFX11-FAKE16-NEXT: [[COPY:%[0-9]+]]:vgpr_32 = COPY $vgpr0
|
|
; GFX11-FAKE16-NEXT: [[V_CVT_F32_F16_fake16_e64_:%[0-9]+]]:vgpr_32 = nofpexcept V_CVT_F32_F16_fake16_e64 0, [[COPY]], 0, 0, implicit $mode, implicit $exec
|
|
; GFX11-FAKE16-NEXT: [[V_CVT_U32_F32_e32_:%[0-9]+]]:vgpr_32 = nofpexcept V_CVT_U32_F32_e32 [[V_CVT_F32_F16_fake16_e64_]], implicit $mode, implicit $exec
|
|
; GFX11-FAKE16-NEXT: S_ENDPGM 0, implicit [[V_CVT_U32_F32_e32_]]
|
|
%0:vgpr(s32) = COPY $vgpr0
|
|
%1:vgpr(s16) = G_TRUNC %0
|
|
%2:vgpr(s32) = G_FPTOUI %1
|
|
%3:vgpr(s1) = G_TRUNC %2
|
|
S_ENDPGM 0, implicit %3
|
|
...
|
|
|
|
---
|
|
name: fptoui_s16_to_s1_vs
|
|
legalized: true
|
|
regBankSelected: true
|
|
tracksRegLiveness: true
|
|
|
|
body: |
|
|
bb.0:
|
|
liveins: $sgpr0
|
|
|
|
; GCN-LABEL: name: fptoui_s16_to_s1_vs
|
|
; GCN: liveins: $sgpr0
|
|
; GCN-NEXT: {{ $}}
|
|
; GCN-NEXT: [[COPY:%[0-9]+]]:sreg_32 = COPY $sgpr0
|
|
; GCN-NEXT: [[V_CVT_F32_F16_e64_:%[0-9]+]]:vgpr_32 = nofpexcept V_CVT_F32_F16_e64 0, [[COPY]], 0, 0, implicit $mode, implicit $exec
|
|
; GCN-NEXT: [[V_CVT_U32_F32_e32_:%[0-9]+]]:vgpr_32 = nofpexcept V_CVT_U32_F32_e32 [[V_CVT_F32_F16_e64_]], implicit $mode, implicit $exec
|
|
; GCN-NEXT: S_ENDPGM 0, implicit [[V_CVT_U32_F32_e32_]]
|
|
;
|
|
; VI-LABEL: name: fptoui_s16_to_s1_vs
|
|
; VI: liveins: $sgpr0
|
|
; VI-NEXT: {{ $}}
|
|
; VI-NEXT: [[COPY:%[0-9]+]]:sreg_32 = COPY $sgpr0
|
|
; VI-NEXT: [[V_CVT_F32_F16_e64_:%[0-9]+]]:vgpr_32 = nofpexcept V_CVT_F32_F16_e64 0, [[COPY]], 0, 0, implicit $mode, implicit $exec
|
|
; VI-NEXT: [[V_CVT_U32_F32_e32_:%[0-9]+]]:vgpr_32 = nofpexcept V_CVT_U32_F32_e32 [[V_CVT_F32_F16_e64_]], implicit $mode, implicit $exec
|
|
; VI-NEXT: S_ENDPGM 0, implicit [[V_CVT_U32_F32_e32_]]
|
|
;
|
|
; GFX11-TRUE16-LABEL: name: fptoui_s16_to_s1_vs
|
|
; GFX11-TRUE16: liveins: $sgpr0
|
|
; GFX11-TRUE16-NEXT: {{ $}}
|
|
; GFX11-TRUE16-NEXT: [[COPY:%[0-9]+]]:sreg_32 = COPY $sgpr0
|
|
; GFX11-TRUE16-NEXT: [[V_CVT_F32_F16_t16_e64_:%[0-9]+]]:vgpr_32 = nofpexcept V_CVT_F32_F16_t16_e64 0, [[COPY]], 0, 0, 0, implicit $mode, implicit $exec
|
|
; GFX11-TRUE16-NEXT: [[V_CVT_U32_F32_e32_:%[0-9]+]]:vgpr_32 = nofpexcept V_CVT_U32_F32_e32 [[V_CVT_F32_F16_t16_e64_]], implicit $mode, implicit $exec
|
|
; GFX11-TRUE16-NEXT: [[COPY1:%[0-9]+]]:vgpr_16 = COPY [[V_CVT_U32_F32_e32_]].lo16
|
|
; GFX11-TRUE16-NEXT: S_ENDPGM 0, implicit [[COPY1]]
|
|
;
|
|
; GFX11-FAKE16-LABEL: name: fptoui_s16_to_s1_vs
|
|
; GFX11-FAKE16: liveins: $sgpr0
|
|
; GFX11-FAKE16-NEXT: {{ $}}
|
|
; GFX11-FAKE16-NEXT: [[COPY:%[0-9]+]]:sreg_32 = COPY $sgpr0
|
|
; GFX11-FAKE16-NEXT: [[V_CVT_F32_F16_fake16_e64_:%[0-9]+]]:vgpr_32 = nofpexcept V_CVT_F32_F16_fake16_e64 0, [[COPY]], 0, 0, implicit $mode, implicit $exec
|
|
; GFX11-FAKE16-NEXT: [[V_CVT_U32_F32_e32_:%[0-9]+]]:vgpr_32 = nofpexcept V_CVT_U32_F32_e32 [[V_CVT_F32_F16_fake16_e64_]], implicit $mode, implicit $exec
|
|
; GFX11-FAKE16-NEXT: S_ENDPGM 0, implicit [[V_CVT_U32_F32_e32_]]
|
|
%0:sgpr(s32) = COPY $sgpr0
|
|
%1:sgpr(s16) = G_TRUNC %0
|
|
%2:vgpr(s32) = G_FPTOUI %1
|
|
%3:vgpr(s1) = G_TRUNC %2
|
|
S_ENDPGM 0, implicit %3
|
|
...
|
|
|
|
---
|
|
name: fptoui_s16_to_s1_fneg_vv
|
|
legalized: true
|
|
regBankSelected: true
|
|
tracksRegLiveness: true
|
|
|
|
body: |
|
|
bb.0:
|
|
liveins: $vgpr0
|
|
|
|
; GCN-LABEL: name: fptoui_s16_to_s1_fneg_vv
|
|
; GCN: liveins: $vgpr0
|
|
; GCN-NEXT: {{ $}}
|
|
; GCN-NEXT: [[COPY:%[0-9]+]]:vgpr_32 = COPY $vgpr0
|
|
; GCN-NEXT: [[S_MOV_B32_:%[0-9]+]]:sreg_32 = S_MOV_B32 32768
|
|
; GCN-NEXT: [[V_XOR_B32_e64_:%[0-9]+]]:vgpr_32 = V_XOR_B32_e64 [[S_MOV_B32_]], [[COPY]], implicit $exec
|
|
; GCN-NEXT: [[V_CVT_F32_F16_e64_:%[0-9]+]]:vgpr_32 = nofpexcept V_CVT_F32_F16_e64 0, [[V_XOR_B32_e64_]], 0, 0, implicit $mode, implicit $exec
|
|
; GCN-NEXT: [[V_CVT_U32_F32_e32_:%[0-9]+]]:vgpr_32 = nofpexcept V_CVT_U32_F32_e32 [[V_CVT_F32_F16_e64_]], implicit $mode, implicit $exec
|
|
; GCN-NEXT: S_ENDPGM 0, implicit [[V_CVT_U32_F32_e32_]]
|
|
;
|
|
; VI-LABEL: name: fptoui_s16_to_s1_fneg_vv
|
|
; VI: liveins: $vgpr0
|
|
; VI-NEXT: {{ $}}
|
|
; VI-NEXT: [[COPY:%[0-9]+]]:vgpr_32 = COPY $vgpr0
|
|
; VI-NEXT: [[S_MOV_B32_:%[0-9]+]]:sreg_32 = S_MOV_B32 32768
|
|
; VI-NEXT: [[V_XOR_B32_e64_:%[0-9]+]]:vgpr_32 = V_XOR_B32_e64 [[S_MOV_B32_]], [[COPY]], implicit $exec
|
|
; VI-NEXT: [[V_CVT_F32_F16_e64_:%[0-9]+]]:vgpr_32 = nofpexcept V_CVT_F32_F16_e64 0, [[V_XOR_B32_e64_]], 0, 0, implicit $mode, implicit $exec
|
|
; VI-NEXT: [[V_CVT_U32_F32_e32_:%[0-9]+]]:vgpr_32 = nofpexcept V_CVT_U32_F32_e32 [[V_CVT_F32_F16_e64_]], implicit $mode, implicit $exec
|
|
; VI-NEXT: S_ENDPGM 0, implicit [[V_CVT_U32_F32_e32_]]
|
|
;
|
|
; GFX11-TRUE16-LABEL: name: fptoui_s16_to_s1_fneg_vv
|
|
; GFX11-TRUE16: liveins: $vgpr0
|
|
; GFX11-TRUE16-NEXT: {{ $}}
|
|
; GFX11-TRUE16-NEXT: [[COPY:%[0-9]+]]:vgpr_32 = COPY $vgpr0
|
|
; GFX11-TRUE16-NEXT: [[COPY1:%[0-9]+]]:vgpr_16 = COPY [[COPY]].lo16
|
|
; GFX11-TRUE16-NEXT: [[V_XOR_B16_t16_e64_:%[0-9]+]]:vgpr_16 = V_XOR_B16_t16_e64 0, 32768, 0, [[COPY1]], 0, implicit $exec
|
|
; GFX11-TRUE16-NEXT: [[V_CVT_F32_F16_t16_e64_:%[0-9]+]]:vgpr_32 = nofpexcept V_CVT_F32_F16_t16_e64 0, [[V_XOR_B16_t16_e64_]], 0, 0, 0, implicit $mode, implicit $exec
|
|
; GFX11-TRUE16-NEXT: [[V_CVT_U32_F32_e32_:%[0-9]+]]:vgpr_32 = nofpexcept V_CVT_U32_F32_e32 [[V_CVT_F32_F16_t16_e64_]], implicit $mode, implicit $exec
|
|
; GFX11-TRUE16-NEXT: [[COPY2:%[0-9]+]]:vgpr_16 = COPY [[V_CVT_U32_F32_e32_]].lo16
|
|
; GFX11-TRUE16-NEXT: S_ENDPGM 0, implicit [[COPY2]]
|
|
;
|
|
; GFX11-FAKE16-LABEL: name: fptoui_s16_to_s1_fneg_vv
|
|
; GFX11-FAKE16: liveins: $vgpr0
|
|
; GFX11-FAKE16-NEXT: {{ $}}
|
|
; GFX11-FAKE16-NEXT: [[COPY:%[0-9]+]]:vgpr_32 = COPY $vgpr0
|
|
; GFX11-FAKE16-NEXT: [[S_MOV_B32_:%[0-9]+]]:sreg_32 = S_MOV_B32 32768
|
|
; GFX11-FAKE16-NEXT: [[V_XOR_B32_e64_:%[0-9]+]]:vgpr_32 = V_XOR_B32_e64 [[S_MOV_B32_]], [[COPY]], implicit $exec
|
|
; GFX11-FAKE16-NEXT: [[V_CVT_F32_F16_fake16_e64_:%[0-9]+]]:vgpr_32 = nofpexcept V_CVT_F32_F16_fake16_e64 0, [[V_XOR_B32_e64_]], 0, 0, implicit $mode, implicit $exec
|
|
; GFX11-FAKE16-NEXT: [[V_CVT_U32_F32_e32_:%[0-9]+]]:vgpr_32 = nofpexcept V_CVT_U32_F32_e32 [[V_CVT_F32_F16_fake16_e64_]], implicit $mode, implicit $exec
|
|
; GFX11-FAKE16-NEXT: S_ENDPGM 0, implicit [[V_CVT_U32_F32_e32_]]
|
|
%0:vgpr(s32) = COPY $vgpr0
|
|
%1:vgpr(s16) = G_TRUNC %0
|
|
%2:vgpr(s16) = G_FNEG %1
|
|
%3:vgpr(s32) = G_FPTOUI %2
|
|
%4:vgpr(s1) = G_TRUNC %3
|
|
S_ENDPGM 0, implicit %4
|
|
...
|