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
115 lines
5.7 KiB
YAML
115 lines
5.7 KiB
YAML
# NOTE: Assertions have been autogenerated by utils/update_mir_test_checks.py UTC_ARGS: --version 6
|
|
|
|
# RUN: llc -mtriple=amdgcn-amd-amdhsa -verify-machineinstrs -run-pass=si-lower-sgpr-spills -o - %s | FileCheck -check-prefix=SGPR-SPILL %s
|
|
# RUN: llc -mtriple=amdgcn-amd-amdhsa -O0 -start-before=si-lower-sgpr-spills -stop-after=regallocfast,1 -verify-machineinstrs %s -o - | FileCheck -check-prefix=WWM-REGALLOC %s
|
|
|
|
|
|
# When SGPR spills to a virtual VGPR lane occur in both a loop header and the latch,
|
|
# the IMPLICIT_DEF for the lane VGPR must be placed in the preheader (not the header).
|
|
# Establish that the virtual VGPR is live-in to the header and wwm regallocfast inserts
|
|
# a restore, preserving the latch writes.
|
|
|
|
---
|
|
name: sgpr_spill_loop_header_and_latch_implicit_def_in_preheader
|
|
tracksRegLiveness: true
|
|
frameInfo:
|
|
maxAlignment: 4
|
|
stack:
|
|
- { id: 0, type: spill-slot, size: 4, alignment: 4, stack-id: sgpr-spill }
|
|
machineFunctionInfo:
|
|
isEntryFunction: false
|
|
scratchRSrcReg: '$sgpr0_sgpr1_sgpr2_sgpr3'
|
|
stackPtrOffsetReg: '$sgpr32'
|
|
frameOffsetReg: '$sgpr33'
|
|
hasSpilledSGPRs: true
|
|
body: |
|
|
; SGPR-SPILL-LABEL: name: sgpr_spill_loop_header_and_latch_implicit_def_in_preheader
|
|
; SGPR-SPILL: bb.0:
|
|
; SGPR-SPILL-NEXT: successors: %bb.1(0x80000000)
|
|
; SGPR-SPILL-NEXT: liveins: $sgpr10, $sgpr11, $sgpr30_sgpr31
|
|
; SGPR-SPILL-NEXT: {{ $}}
|
|
; SGPR-SPILL-NEXT: [[DEF:%[0-9]+]]:vgpr_32 = IMPLICIT_DEF
|
|
; SGPR-SPILL-NEXT: S_BRANCH %bb.1
|
|
; SGPR-SPILL-NEXT: {{ $}}
|
|
; SGPR-SPILL-NEXT: bb.1:
|
|
; SGPR-SPILL-NEXT: successors: %bb.2(0x40000000), %bb.3(0x40000000)
|
|
; SGPR-SPILL-NEXT: liveins: $sgpr10, $sgpr11, $sgpr30_sgpr31
|
|
; SGPR-SPILL-NEXT: {{ $}}
|
|
; SGPR-SPILL-NEXT: $sgpr10 = SI_RESTORE_S32_FROM_VGPR [[DEF]], 0
|
|
; SGPR-SPILL-NEXT: S_CMP_EQ_U32 $sgpr11, 0, implicit-def $scc
|
|
; SGPR-SPILL-NEXT: [[DEF:%[0-9]+]]:vgpr_32 = SI_SPILL_S32_TO_VGPR killed $sgpr10, 0, [[DEF]]
|
|
; SGPR-SPILL-NEXT: S_CBRANCH_SCC1 %bb.2, implicit killed $scc
|
|
; SGPR-SPILL-NEXT: S_BRANCH %bb.3
|
|
; SGPR-SPILL-NEXT: {{ $}}
|
|
; SGPR-SPILL-NEXT: bb.2:
|
|
; SGPR-SPILL-NEXT: successors: %bb.1(0x80000000)
|
|
; SGPR-SPILL-NEXT: liveins: $sgpr10, $sgpr11, $sgpr30_sgpr31
|
|
; SGPR-SPILL-NEXT: {{ $}}
|
|
; SGPR-SPILL-NEXT: $sgpr10 = SI_RESTORE_S32_FROM_VGPR [[DEF]], 0
|
|
; SGPR-SPILL-NEXT: $sgpr10 = S_MOV_B32 1
|
|
; SGPR-SPILL-NEXT: [[DEF:%[0-9]+]]:vgpr_32 = SI_SPILL_S32_TO_VGPR killed $sgpr10, 0, [[DEF]]
|
|
; SGPR-SPILL-NEXT: S_BRANCH %bb.1
|
|
; SGPR-SPILL-NEXT: {{ $}}
|
|
; SGPR-SPILL-NEXT: bb.3:
|
|
; SGPR-SPILL-NEXT: liveins: $sgpr30_sgpr31
|
|
; SGPR-SPILL-NEXT: {{ $}}
|
|
; SGPR-SPILL-NEXT: S_SETPC_B64 $sgpr30_sgpr31
|
|
;
|
|
; WWM-REGALLOC-LABEL: name: sgpr_spill_loop_header_and_latch_implicit_def_in_preheader
|
|
; WWM-REGALLOC: bb.0:
|
|
; WWM-REGALLOC-NEXT: successors: %bb.1(0x80000000)
|
|
; WWM-REGALLOC-NEXT: liveins: $sgpr10, $sgpr11, $sgpr30_sgpr31
|
|
; WWM-REGALLOC-NEXT: {{ $}}
|
|
; WWM-REGALLOC-NEXT: renamable $vgpr63 = IMPLICIT_DEF
|
|
; WWM-REGALLOC-NEXT: S_BRANCH %bb.1
|
|
; WWM-REGALLOC-NEXT: {{ $}}
|
|
; WWM-REGALLOC-NEXT: bb.1:
|
|
; WWM-REGALLOC-NEXT: successors: %bb.2(0x40000000), %bb.3(0x40000000)
|
|
; WWM-REGALLOC-NEXT: liveins: $sgpr10, $sgpr11, $sgpr30_sgpr31
|
|
; WWM-REGALLOC-NEXT: {{ $}}
|
|
; WWM-REGALLOC-NEXT: $vgpr63 = SI_SPILL_WWM_V32_RESTORE %stack.1, $sgpr32, 0, implicit $exec :: (load (s32) from %stack.1, addrspace 5)
|
|
; WWM-REGALLOC-NEXT: $sgpr10 = SI_RESTORE_S32_FROM_VGPR $vgpr63, 0
|
|
; WWM-REGALLOC-NEXT: S_CMP_EQ_U32 $sgpr11, 0, implicit-def $scc
|
|
; WWM-REGALLOC-NEXT: $vgpr63 = SI_SPILL_S32_TO_VGPR $sgpr10, 0, $vgpr63
|
|
; WWM-REGALLOC-NEXT: SI_SPILL_WWM_V32_SAVE killed $vgpr63, %stack.1, $sgpr32, 0, implicit $exec :: (store (s32) into %stack.1, addrspace 5)
|
|
; WWM-REGALLOC-NEXT: S_CBRANCH_SCC1 %bb.2, implicit killed $scc
|
|
; WWM-REGALLOC-NEXT: S_BRANCH %bb.3
|
|
; WWM-REGALLOC-NEXT: {{ $}}
|
|
; WWM-REGALLOC-NEXT: bb.2:
|
|
; WWM-REGALLOC-NEXT: successors: %bb.1(0x80000000)
|
|
; WWM-REGALLOC-NEXT: liveins: $sgpr10, $sgpr11, $sgpr30_sgpr31
|
|
; WWM-REGALLOC-NEXT: {{ $}}
|
|
; WWM-REGALLOC-NEXT: $vgpr63 = SI_SPILL_WWM_V32_RESTORE %stack.1, $sgpr32, 0, implicit $exec :: (load (s32) from %stack.1, addrspace 5)
|
|
; WWM-REGALLOC-NEXT: dead $sgpr10 = SI_RESTORE_S32_FROM_VGPR $vgpr63, 0
|
|
; WWM-REGALLOC-NEXT: $sgpr10 = S_MOV_B32 1
|
|
; WWM-REGALLOC-NEXT: $vgpr63 = SI_SPILL_S32_TO_VGPR $sgpr10, 0, $vgpr63
|
|
; WWM-REGALLOC-NEXT: SI_SPILL_WWM_V32_SAVE killed $vgpr63, %stack.1, $sgpr32, 0, implicit $exec :: (store (s32) into %stack.1, addrspace 5)
|
|
; WWM-REGALLOC-NEXT: S_BRANCH %bb.1
|
|
; WWM-REGALLOC-NEXT: {{ $}}
|
|
; WWM-REGALLOC-NEXT: bb.3:
|
|
; WWM-REGALLOC-NEXT: liveins: $sgpr30_sgpr31
|
|
; WWM-REGALLOC-NEXT: {{ $}}
|
|
; WWM-REGALLOC-NEXT: S_SETPC_B64 killed $sgpr30_sgpr31
|
|
|
|
|
|
bb.0:
|
|
liveins: $sgpr10, $sgpr11, $sgpr30_sgpr31
|
|
S_BRANCH %bb.1
|
|
bb.1:
|
|
liveins: $sgpr10, $sgpr11, $sgpr30_sgpr31
|
|
renamable $sgpr10 = SI_SPILL_S32_RESTORE %stack.0, implicit $exec, implicit $sgpr0_sgpr1_sgpr2_sgpr3, implicit $sgpr32
|
|
S_CMP_EQ_U32 $sgpr11, 0, implicit-def $scc
|
|
SI_SPILL_S32_SAVE killed $sgpr10, %stack.0, implicit $exec, implicit $sgpr0_sgpr1_sgpr2_sgpr3, implicit $sgpr32
|
|
S_CBRANCH_SCC1 %bb.2, implicit killed $scc
|
|
S_BRANCH %bb.3
|
|
bb.2:
|
|
liveins: $sgpr10, $sgpr11, $sgpr30_sgpr31
|
|
renamable $sgpr10 = SI_SPILL_S32_RESTORE %stack.0, implicit $exec, implicit $sgpr0_sgpr1_sgpr2_sgpr3, implicit $sgpr32
|
|
$sgpr10 = S_MOV_B32 1
|
|
SI_SPILL_S32_SAVE killed $sgpr10, %stack.0, implicit $exec, implicit $sgpr0_sgpr1_sgpr2_sgpr3, implicit $sgpr32
|
|
S_BRANCH %bb.1
|
|
bb.3:
|
|
liveins: $sgpr30_sgpr31
|
|
S_SETPC_B64 $sgpr30_sgpr31
|
|
## NOTE: These prefixes are unused and the list is autogenerated. Do not add tests below this line:
|