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
76 lines
2.4 KiB
ArmAsm
76 lines
2.4 KiB
ArmAsm
// A far crossing split needs a forward gateway. Source-window growth may move
|
|
// the callable-entry VGPR-MSB setter only as the first copied instruction, so
|
|
// every entry still establishes the original mode before either WMMA half.
|
|
|
|
// RUN: %clang -target amdgcn-amd-amdhsa -mcpu=gfx1250 -nostdlib %s -o %t.elf
|
|
// RUN: env AMD_COMGR_EMIT_VERBOSE_LOGS=1 hotswap-rewrite %t.elf \
|
|
// RUN: amdgcn-amd-amdhsa--gfx1250 amdgcn-amd-amdhsa--gfx1250 \
|
|
// RUN: --output %t.out.elf 2>&1 | %FileCheck --check-prefix=API %s
|
|
// API: WMMA split: patched v_wmma_f32_16x16x128_fp8_fp8
|
|
// API-NOT: error:
|
|
// API: RESULT: SUCCESS
|
|
|
|
// RUN: %llvm-objdump -d %t.out.elf | %FileCheck --check-prefix=DISASM \
|
|
// RUN: --implicit-check-not=s_add_pc_i64 %s
|
|
// DISASM-LABEL: <test_wmma_vgpr_msb_far>:
|
|
// DISASM-NEXT: s_branch
|
|
// DISASM-NEXT: s_nop 0
|
|
// DISASM-NEXT: s_nop 0
|
|
// DISASM-NEXT: s_endpgm
|
|
// DISASM: s_set_vgpr_msb 0
|
|
// DISASM-NEXT: v_wmma_f32_16x16x64_fp8_fp8 v[162:169], v[250:257], v[114:121], 0
|
|
// DISASM-NEXT: s_set_vgpr_msb 1
|
|
// DISASM-NEXT: v_wmma_f32_16x16x64_fp8_fp8 v[162:169], v[2:9]{{.*v\[258:265\].*}}, v[122:129], v[162:169]
|
|
// DISASM-NEXT: s_set_vgpr_msb 0x100
|
|
|
|
// RUN: hotswap-rewrite %t.out.elf \
|
|
// RUN: amdgcn-amd-amdhsa--gfx1250 amdgcn-amd-amdhsa--gfx1250 \
|
|
// RUN: --check-idempotent | %FileCheck --check-prefix=IDEM %s
|
|
// IDEM: IDEMPOTENT: YES
|
|
|
|
.amdgcn_target "amdgcn-amd-amdhsa--gfx1250"
|
|
.text
|
|
.globl test_wmma_vgpr_msb_far
|
|
.p2align 8
|
|
.type test_wmma_vgpr_msb_far,@function
|
|
test_wmma_vgpr_msb_far:
|
|
s_set_vgpr_msb 0
|
|
v_wmma_f32_16x16x128_fp8_fp8 v[162:169], v[250:265], v[114:129], 0
|
|
s_endpgm
|
|
.Ltest_wmma_vgpr_msb_far_end:
|
|
.size test_wmma_vgpr_msb_far, .Ltest_wmma_vgpr_msb_far_end-test_wmma_vgpr_msb_far
|
|
|
|
// Safe external gateway space after the no-fallthrough terminator.
|
|
.rept 8
|
|
s_nop 0
|
|
.endr
|
|
|
|
// Keep the appended pool beyond s_branch reach from the split site.
|
|
.rept 40000
|
|
s_mov_b32 s3, s4
|
|
.endr
|
|
|
|
.rodata
|
|
.p2align 8
|
|
.amdhsa_kernel test_wmma_vgpr_msb_far
|
|
.amdhsa_next_free_vgpr 266
|
|
.amdhsa_next_free_sgpr 0
|
|
.end_amdhsa_kernel
|
|
|
|
.amdgpu_metadata
|
|
amdhsa.version:
|
|
- 3
|
|
- 0
|
|
amdhsa.kernels:
|
|
- .name: test_wmma_vgpr_msb_far
|
|
.symbol: test_wmma_vgpr_msb_far.kd
|
|
.sgpr_count: 0
|
|
.vgpr_count: 266
|
|
.kernarg_segment_size: 0
|
|
.group_segment_fixed_size: 0
|
|
.private_segment_fixed_size: 0
|
|
.kernarg_segment_align: 8
|
|
.wavefront_size: 64
|
|
.max_flat_workgroup_size: 256
|
|
.end_amdgpu_metadata
|