refine code

Author: xiaolil1

CMakeLists.txt

@@ -317,11 +317,11 @@ if(BUILD_XPU)
       -fsycl
       --offload-compress
       -fsycl-targets=intel_gpu_pvc
-      -Xspirv-translator -spirv-ext=+SPV_INTEL_split_barrier,+SPV_INTEL_2d_block_io,+SPV_INTEL_subgroup_matrix_multiply_accumulate
+      -Xspirv-translator -spirv-ext=+SPV_INTEL_split_barrier #,+SPV_INTEL_2d_block_io,+SPV_INTEL_subgroup_matrix_multiply_accumulate
       -Xs
       -options "-cl-intel-enable-auto-large-GRF-mode -cl-poison-unsupported-fp64-kernels -cl-intel-greater-than-4GB-buffer-required"
     )
-    set(SYCL_COMPILE_FLAGS "-fsycl;-fhonor-nans;-fhonor-infinities;-fno-associative-math;-fno-approx-func;-fno-sycl-instrument-device-code;--offload-compress;-fsycl-targets=intel_gpu_pvc;-Xspirv-translator;-spirv-ext=+SPV_INTEL_split_barrier,+SPV_INTEL_2d_block_io,+SPV_INTEL_subgroup_matrix_multiply_accumulate;")
+    set(SYCL_COMPILE_FLAGS "-fsycl;-fhonor-nans;-fhonor-infinities;-fno-associative-math;-fno-approx-func;-fno-sycl-instrument-device-code;--offload-compress;-fsycl-targets=intel_gpu_pvc;")
 
     set_property(TARGET bitsandbytes PROPERTY CXX_STANDARD 20)
     target_compile_options(bitsandbytes PRIVATE ${SYCL_COMPILE_FLAGS})

run_case.sh

@@ -28,7 +28,7 @@
 
 
 #gdb -args python -m pytest -vs tests/test_xpu.py::TestXPU::test_gemm_4bit
-pytest -vs tests/test_xpu.py::TestXPU::test_gemm_4bit
-#python tests/test_xpu_db.py
+#pytest -vs tests/test_xpu.py::TestXPU::test_gemm_4bit
+python tests/test_xpu_db.py
 #gdb -args python tests/test_xpu_db.py
 #pytest tests/test_functional.py::TestQuantize4BitFunctional::test_gemv_4bit[dim=256-uint8-bf16-fc1-nf4-DQ_True-xpu]