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
38 lines
1.9 KiB
Plaintext
38 lines
1.9 KiB
Plaintext
# RUN: %stinkytofu-opt --arch gfx1250 %s --StinkyDAGSchedulerPass --print-output
|
|
#
|
|
# AsmDirective preservation: .set directives act as non-movable
|
|
# side-effect boundaries in the DAG scheduler. Their relative order
|
|
# among instructions must be strictly preserved.
|
|
#
|
|
# Input: ds_load A, ds_load B, .set symA, ds_load C, ds_load D, .set symB, WMMA0, WMMA1
|
|
# Expect: ds_loads may reorder within each region, but .set directives
|
|
# stay at their original boundary positions.
|
|
#
|
|
# Region 1 (before symA): two ds_loads
|
|
# CHECK: ds_load_b128
|
|
# CHECK: ds_load_b128
|
|
# Directive boundary
|
|
# CHECK: asm_directive{{.*}}symA
|
|
# Region 2 (between symA and symB): two ds_loads
|
|
# CHECK: ds_load_b128
|
|
# CHECK: ds_load_b128
|
|
# Directive boundary
|
|
# CHECK: asm_directive{{.*}}symB
|
|
# Region 3 (after symB): WMMAs
|
|
# CHECK: v_wmma_f32_16x16x16_bf16
|
|
# CHECK: v_wmma_f32_16x16x16_bf16
|
|
# No directives should appear after the WMMAs
|
|
# CHECK-NOT: asm_directive
|
|
|
|
st.func @test() {
|
|
^entry:
|
|
v[8:11] = "st.ds_load_b128"(v100) { issueCycles = 1, latencyCycles = 56, mod.ds = { na = 1, offset = 0, gds = false }, mod.memtoken = { tokens = [0] } }
|
|
v[12:15] = "st.ds_load_b128"(v100) { issueCycles = 1, latencyCycles = 56, mod.ds = { na = 1, offset = 32, gds = false }, mod.memtoken = { tokens = [0] } }
|
|
"st.asm_directive"(".set", "symA")
|
|
v[16:19] = "st.ds_load_b128"(v100) { issueCycles = 1, latencyCycles = 56, mod.ds = { na = 1, offset = 8192, gds = false }, mod.memtoken = { tokens = [0] } }
|
|
v[20:23] = "st.ds_load_b128"(v100) { issueCycles = 1, latencyCycles = 56, mod.ds = { na = 1, offset = 8224, gds = false }, mod.memtoken = { tokens = [0] } }
|
|
"st.asm_directive"(".set", "symB")
|
|
v[0:7] = "st.v_wmma_f32_16x16x16_bf16"(v[8:15], v[16:23], v[0:7]) { mod.mfma = { reuseA = false, reuseB = false } }
|
|
v[0:7] = "st.v_wmma_f32_16x16x16_bf16"(v[16:23], v[20:23], v[0:7]) { mod.mfma = { reuseA = false, reuseB = false } }
|
|
}
|