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
51 lines
1.9 KiB
ArmAsm
51 lines
1.9 KiB
ArmAsm
// Test that matrix_a_reuse is stripped on both halves of a K-split.
|
|
//
|
|
// matrix_a_reuse is a HW data-reuse hint asserting that the A matrix
|
|
// is identical to the previous WMMA's A. After a K-split, A is sliced
|
|
// into halves -- the data layout assumption no longer holds, so
|
|
// preserving the hint would make the hardware reuse stale data. The
|
|
// splitter strips the modifier on both halves (no perf hint, but
|
|
// correct semantics).
|
|
|
|
// RUN: %clang -target amdgcn-amd-amdhsa -mcpu=gfx1250 -nostdlib %s -o %t.elf
|
|
|
|
// RUN: hotswap-rewrite %t.elf \
|
|
// RUN: amdgcn-amd-amdhsa--gfx1250 amdgcn-amd-amdhsa--gfx1250 \
|
|
// RUN: --output %t.out.elf \
|
|
// RUN: | %FileCheck --check-prefix=API %s
|
|
// API: RESULT: SUCCESS
|
|
|
|
// RUN: %llvm-objdump -d %t.out.elf | %FileCheck --check-prefix=DISASM %s
|
|
|
|
.amdgcn_target "amdgcn-amd-amdhsa--gfx1250"
|
|
|
|
// DISASM-LABEL: <kernel>:
|
|
// DISASM-NOT: v_wmma_f32_16x16x128_fp8_bf8
|
|
// DISASM: s_branch
|
|
// DISASM: s_endpgm
|
|
|
|
// COM: Both halves end at `//` immediately after the operand list (no
|
|
// COM: matrix_a_reuse modifier suffix on either).
|
|
// DISASM: v_wmma_f32_16x16x64_fp8_bf8 v[40:47], v[0:7], v[8:15], v[40:47]{{[[:space:]]*\/\/}}
|
|
// DISASM-NEXT: v_wmma_f32_16x16x64_fp8_bf8 v[40:47], v[8:15], v[16:23], v[40:47]{{[[:space:]]*\/\/}}
|
|
// DISASM-NEXT: s_branch
|
|
|
|
// COM: Sanity: matrix_a_reuse must NOT appear anywhere on the K=64
|
|
// COM: replacement instructions for this kernel.
|
|
// DISASM-NOT: v_wmma_f32_16x16x64_fp8_bf8 v[40:47]{{.*}}matrix_a_reuse
|
|
.globl kernel
|
|
.p2align 8
|
|
.type kernel,@function
|
|
kernel:
|
|
v_wmma_f32_16x16x128_fp8_bf8 v[40:47], v[0:15], v[8:23], v[40:47] matrix_a_reuse
|
|
s_endpgm
|
|
.size kernel, .-kernel
|
|
|
|
// Idempotency.
|
|
// RUN: hotswap-rewrite %t.out.elf \
|
|
// RUN: amdgcn-amd-amdhsa--gfx1250 amdgcn-amd-amdhsa--gfx1250 \
|
|
// RUN: --output %t.out2.elf \
|
|
// RUN: | %FileCheck --check-prefix=API2 %s
|
|
// API2: RESULT: SUCCESS
|
|
// RUN: cmp %t.out.elf %t.out2.elf
|