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
102 lines
2.8 KiB
Plaintext
102 lines
2.8 KiB
Plaintext
# RUN: %stinkytofu-opt --arch gfx1250 %s --BuildUseDefChainPass --PeepholeOptimizationPass --print-output
|
|
#
|
|
# ADD+MUL Fusion: v_add + v_mul with constants → single v_fma.
|
|
# v_add v0,C1,v1 + v_mul v2,C2,v0 → v_fma v2,C2,v1,(C1*C2)
|
|
# Converted from PeepholeOptimizationPassTest.*ADDMUL*
|
|
|
|
#--- F32 basic: v_add v0,2.0,v1 + v_mul v2,3.0,v0 → v_fma v2,3.0,v1,6.0
|
|
# CHECK-LABEL: @add_mul_f32_basic
|
|
# CHECK: v_fma_f32
|
|
# CHECK-SAME: v2
|
|
# CHECK-SAME: 3
|
|
# CHECK-SAME: v1
|
|
# CHECK-SAME: 6
|
|
# CHECK-NOT: v_add_f32
|
|
# CHECK-NOT: v_mul_f32
|
|
|
|
st.func @add_mul_f32_basic() {
|
|
^entry:
|
|
v0 = "st.v_add_f32"(2.0, v1) { issueCycles = 1, latencyCycles = 5 }
|
|
v2 = "st.v_mul_f32"(3.0, v0) { issueCycles = 1, latencyCycles = 5 }
|
|
}
|
|
|
|
#--- F16 basic
|
|
# CHECK-LABEL: @add_mul_f16_basic
|
|
# CHECK: v_fma_f16
|
|
# CHECK-SAME: 3
|
|
# CHECK-SAME: v1
|
|
# CHECK-SAME: 6
|
|
# CHECK-NOT: v_add_f16
|
|
# CHECK-NOT: v_mul_f16
|
|
|
|
st.func @add_mul_f16_basic() {
|
|
^entry:
|
|
v0 = "st.v_add_f16"(2.0, v1) { issueCycles = 1, latencyCycles = 5 }
|
|
v2 = "st.v_mul_f16"(3.0, v0) { issueCycles = 1, latencyCycles = 5 }
|
|
}
|
|
|
|
#--- Negative: ADD result has multiple uses → no fusion
|
|
# CHECK-LABEL: @no_fusion_addmul_multiple_uses
|
|
# CHECK: v_add_f32
|
|
# CHECK: v_mul_f32
|
|
# CHECK: v_add_f32
|
|
|
|
st.func @no_fusion_addmul_multiple_uses() {
|
|
^entry:
|
|
v0 = "st.v_add_f32"(2.0, v1) { issueCycles = 1, latencyCycles = 5 }
|
|
v2 = "st.v_mul_f32"(3.0, v0) { issueCycles = 1, latencyCycles = 5 }
|
|
v3 = "st.v_add_f32"(v0, v4) { issueCycles = 1, latencyCycles = 5 }
|
|
}
|
|
|
|
#--- Negative: ADD has no constant → no fusion
|
|
# CHECK-LABEL: @no_fusion_addmul_add_no_const
|
|
# CHECK: v_add_f32
|
|
# CHECK: v_mul_f32
|
|
|
|
st.func @no_fusion_addmul_add_no_const() {
|
|
^entry:
|
|
v0 = "st.v_add_f32"(v1, v2) { issueCycles = 1, latencyCycles = 5 }
|
|
v3 = "st.v_mul_f32"(3.0, v0) { issueCycles = 1, latencyCycles = 5 }
|
|
}
|
|
|
|
#--- Negative: MUL has no constant → no fusion
|
|
# CHECK-LABEL: @no_fusion_addmul_mul_no_const
|
|
# CHECK: v_add_f32
|
|
# CHECK: v_mul_f32
|
|
|
|
st.func @no_fusion_addmul_mul_no_const() {
|
|
^entry:
|
|
v0 = "st.v_add_f32"(2.0, v1) { issueCycles = 1, latencyCycles = 5 }
|
|
v2 = "st.v_mul_f32"(v3, v0) { issueCycles = 1, latencyCycles = 5 }
|
|
}
|
|
|
|
#--- Negative constant: 3.0 * -2.0 = -6.0
|
|
# CHECK-LABEL: @add_mul_negative
|
|
# CHECK: v_fma_f32
|
|
# CHECK-SAME: 3
|
|
# CHECK-SAME: v1
|
|
# CHECK-SAME: -6
|
|
# CHECK-NOT: v_add_f32
|
|
# CHECK-NOT: v_mul_f32
|
|
|
|
st.func @add_mul_negative() {
|
|
^entry:
|
|
v0 = "st.v_add_f32"(-2.0, v1) { issueCycles = 1, latencyCycles = 5 }
|
|
v2 = "st.v_mul_f32"(3.0, v0) { issueCycles = 1, latencyCycles = 5 }
|
|
}
|
|
|
|
#--- Both negative: -3.0 * -2.0 = 6.0
|
|
# CHECK-LABEL: @add_mul_both_negative
|
|
# CHECK: v_fma_f32
|
|
# CHECK-SAME: -3
|
|
# CHECK-SAME: v1
|
|
# CHECK-SAME: 6
|
|
# CHECK-NOT: v_add_f32
|
|
# CHECK-NOT: v_mul_f32
|
|
|
|
st.func @add_mul_both_negative() {
|
|
^entry:
|
|
v0 = "st.v_add_f32"(-2.0, v1) { issueCycles = 1, latencyCycles = 5 }
|
|
v2 = "st.v_mul_f32"(-3.0, v0) { issueCycles = 1, latencyCycles = 5 }
|
|
}
|