update

example/01_gemm/gemm_xdl_bf16.cpp

@@ -13,6 +13,7 @@ using CShuffleDataType = ck::bhalf_t;
 
 using ALayout = Row;
 using BLayout = Col;
+// using BLayout = Row;
 using CLayout = Row;
 
 using AElementOp = PassThrough;

example/01_gemm/gemm_xdl_streamk.cpp

@@ -14,8 +14,8 @@ using CDataType        = ck::half_t;
 using F16 = ck::half_t;
 
 using ALayout = Row;
-using BLayout = Row;
-// using BLayout = Col;
+// using BLayout = Row;
+using BLayout = Col;
 using CLayout = Row;
 
 using AElementOp = PassThrough;

example/01_gemm/run_gemm_example_streamk_v2.inc

@@ -181,6 +181,11 @@ bool run_gemm(const ProblemType& problem_size, const ExecutionConfig& config)
     std::cout << "b_k_n: " << b_k_n.mDesc << std::endl;
     std::cout << "c_m_n: " << c_m_n_host_result.mDesc << std::endl;
 
+    //Added By Emin
+    
+
+    // Added By Emin
+
 #ifdef BUILD_INT4_EXAMPLE
     DeviceMem a_m_k_device_buf(sizeof(KernelADataType) * a_m_k.mDesc.GetElementSpaceSize());
     DeviceMem b_k_n_device_buf(sizeof(KernelBDataType) * b_k_n.mDesc.GetElementSpaceSize());

script/cmake-ck-dev.sh

@@ -15,8 +15,8 @@ else
 fi
 
 cmake                                                                                             \
--D CMAKE_PREFIX_PATH=/opt/rocm                                                                    \
--D CMAKE_CXX_COMPILER=/opt/rocm/bin/hipcc                                                         \
+-D CMAKE_PREFIX_PATH=/opt/rocm-6.2.1/                                                                 \
+-D CMAKE_CXX_COMPILER=/opt/rocm-6.2.1/bin/hipcc                                                        \
 -D CMAKE_CXX_FLAGS="-Xclang -mllvm -Xclang -enable-post-misched=0 -std=c++17 -O3 -ftemplate-backtrace-limit=0  -fPIE  -Wno-gnu-line-marker"     \
 -D CMAKE_BUILD_TYPE=Release                                                                       \
 -D BUILD_DEV=ON                                                                                   \
@@ -25,3 +25,7 @@ cmake
 -D USE_BITINT_EXTENSION_INT4=OFF                                                                  \
 $REST_ARGS                                                                                        \
 ${MY_PROJECT_SOURCE}
+
+
+# -D CMAKE_PREFIX_PATH=/opt/rocm                                                                    \
+# -D CMAKE_CXX_COMPILER=/opt/rocm/bin/hipcc     
\ No newline at end of file