[CI][GEMM][FA] Remove default path with env vars configuration (#2882)

The performance of default path without env vars is similar to the
performance of default path with env vars.
This PR removes running default path with env vars in CI.

Signed-off-by: Whitney Tsang <[email protected]>

Author: whitneywhtsang

.github/workflows/triton-benchmarks.yml

@@ -157,21 +157,8 @@ jobs:
           source ../../scripts/capture-hw-details.sh
           python ../../scripts/build_report.py $REPORTS/matmul-performance-base.csv $REPORTS/gemm-triton-report.csv --benchmark gemm --compiler triton --param_cols "B,M,K,N" --tflops_col Triton-TFlops --hbm_col "Triton-GB/s" --tag $TAG
           python ../../scripts/build_report.py $REPORTS/matmul-performance-base.csv $REPORTS/gemm-xetla-report.csv --benchmark gemm --compiler xetla --param_cols "B,M,K,N" --tflops_col XeTLA-TFlops --hbm_col "XeTLA-GB/s" --tag $TAG
-
-      - name: Run Triton GEMM kernel benchmark - default path
-        if: ${{ steps.install.outcome == 'success' && !cancelled() && !contains(fromJson(inputs.skip_benchmarks || '[]'), 'gemm_benchmark.py_default') }}
-        run: |
-          cd benchmarks/triton_kernels_benchmark
-          # Default path:
-          TRITON_INTEL_ADVANCED_PATH=0 \
-          IGC_VISAOptions=" -enableBCR -nolocalra" \
-          IGC_DisableLoopUnroll=1 \
-          python gemm_benchmark.py --reports $REPORTS
-          mv $REPORTS/matmul-performance.csv $REPORTS/matmul-performance-default-path.csv
-
-          source ../../scripts/capture-hw-details.sh
-          TAG="${TAG}-dflt"
-          python ../../scripts/build_report.py $REPORTS/matmul-performance-default-path.csv $REPORTS/gemm-triton-default-report.csv --benchmark gemm --compiler triton --param_cols "B,M,K,N" --tflops_col Triton-TFlops --hbm_col "Triton-GB/s" --tag $TAG
+          # FIXME: Remove after Grafana migration
+          python ../../scripts/build_report.py $REPORTS/matmul-performance-base.csv $REPORTS/gemm-triton-report.csv --benchmark gemm --compiler triton --param_cols "B,M,K,N" --tflops_col Triton-TFlops --hbm_col "Triton-GB/s" --tag ${TAG}-dflt
 
       - name: Run Triton GEMM kernel benchmark - advanced path
         if: ${{ steps.install.outcome == 'success' && !cancelled() && !contains(fromJson(inputs.skip_benchmarks || '[]'), 'gemm_benchmark.py_advanced') }}
@@ -259,19 +246,8 @@ jobs:
           source ../../scripts/capture-hw-details.sh
           python ../../scripts/build_report.py $REPORTS/attn-performance.csv $REPORTS/attn-triton-report.csv --benchmark attn --compiler triton --param_cols "Z,H,N_CTX,D_HEAD,CAUSAL" --tflops_col Triton-TFlops --hbm_col "Triton-GB/s" --tag $TAG
           python ../../scripts/build_report.py $REPORTS/attn-performance.csv $REPORTS/attn-xetla-report.csv --benchmark attn --compiler xetla --param_cols "Z,H,N_CTX,D_HEAD,CAUSAL" --tflops_col XeTLA-TFlops --hbm_col "XeTLA-GB/s" --tag $TAG
-
-      - name: Run Triton FA kernel benchmark - default path
-        if: ${{ steps.install.outcome == 'success' && !cancelled() && !contains(fromJson(inputs.skip_benchmark || '[]'), 'flash_attention_fwd_benchmark.py_default') }}
-        run: |
-          cd benchmarks/triton_kernels_benchmark
-          TRITON_INTEL_ADVANCED_PATH=0 \
-          TRITON_INTEL_ENABLE_ADDRESS_PAYLOAD_OPT=1 \
-          IGC_VISAOptions=" -enableBCR" \
-          python flash_attention_fwd_benchmark.py --reports $REPORTS
-
-          TAG="${TAG}-dflt"
-          source ../../scripts/capture-hw-details.sh
-          python ../../scripts/build_report.py $REPORTS/attn-performance.csv $REPORTS/attn-triton-default-report.csv --benchmark attn --compiler triton --param_cols "Z,H,N_CTX,D_HEAD,CAUSAL" --tflops_col Triton-TFlops --hbm_col "Triton-GB/s" --tag $TAG
+          # FIXME: Remove after Grafana migration
+          python ../../scripts/build_report.py $REPORTS/attn-performance.csv $REPORTS/attn-triton-default-report.csv --benchmark attn --compiler triton --param_cols "Z,H,N_CTX,D_HEAD,CAUSAL" --tflops_col Triton-TFlops --hbm_col "Triton-GB/s" --tag ${TAG}-dflt
 
       - name: Run Triton FA kernel benchmark - advanced path
         if: ${{ steps.install.outcome == 'success' && !cancelled() && !contains(fromJson(inputs.skip_benchmarks || '[]'), 'flash_attention_fwd_benchmark.py_advanced') }}

scripts/test-triton.sh

@@ -263,10 +263,7 @@ run_benchmark_gemm() {
   python setup.py install
 
   echo "Default path:"
-  TRITON_INTEL_ADVANCED_PATH=0 \
-    IGC_VISAOptions=" -enableBCR -nolocalra" \
-    IGC_DisableLoopUnroll=1 \
-    python $TRITON_PROJ/benchmarks/triton_kernels_benchmark/gemm_benchmark.py
+  python $TRITON_PROJ/benchmarks/triton_kernels_benchmark/gemm_benchmark.py
 
   echo "Advanced path:"
   TRITON_INTEL_ADVANCED_PATH=1 \
@@ -283,10 +280,7 @@ run_benchmark_attention() {
   python setup.py install
 
   echo "Default path:"
-  TRITON_INTEL_ADVANCED_PATH=0 \
-    TRITON_INTEL_ENABLE_ADDRESS_PAYLOAD_OPT=1 \
-    IGC_VISAOptions=" -enableBCR" \
-    python $TRITON_PROJ/benchmarks/triton_kernels_benchmark/flash_attention_fwd_benchmark.py
+  python $TRITON_PROJ/benchmarks/triton_kernels_benchmark/flash_attention_fwd_benchmark.py
 
   echo "Advanced path:"
   TRITON_INTEL_ADVANCED_PATH=1 \