Disable the SLP vectorizer to prevent unnecessary wait (#1008)

* Disable the SLP vectorizer to prevent unnecessary wait

* Add comment to the reason of adding flag

* Fix wording

library/src/tensor_operation_instance/gpu/gemm/CMakeLists.txt

@@ -108,13 +108,15 @@ if (ENABLE_PIPELINE_V2_OPT)
        CK_EXPERIMENTAL_PIPELINE_V2_IGLP_OPT=1
     )
 
+    # TODO: The "-vectorize-slp=false" LLVM option is a workaround to prevent inefficient instruction scheduling
+    #       caused by the SLP Vectorizer. Remove this option after fix the SLP Vectorizer issue.
     # layout=NT
     set_source_files_properties(device_gemm_xdl_f16_f16_f16/km_kn_mn_default_pipeline_v2_opt_instance.cpp PROPERTIES
-       COMPILE_OPTIONS ";;"
+       COMPILE_OPTIONS ";-mllvm;-vectorize-slp=false"
        COMPILE_DEFINITIONS "${WAVES_PER_EU_DEFS};${IGLP_OPT_DEFS}")
     # layout=NN
     set_source_files_properties(device_gemm_xdl_f16_f16_f16/km_nk_mn_default_pipeline_v2_opt_instance.cpp PROPERTIES
-       COMPILE_OPTIONS ";;"
+       COMPILE_OPTIONS ";-mllvm;-vectorize-slp=false"
        COMPILE_DEFINITIONS "${WAVES_PER_EU_DEFS};${IGLP_OPT_DEFS}")
     # layout=TT
     set_source_files_properties(device_gemm_xdl_f16_f16_f16/mk_kn_mn_default_pipeline_v2_opt_instance.cpp PROPERTIES