Fix typos

Author: William-An

src/cuda/GPU_Microbenchmark/hw_def/common/gpuConfig.h

@@ -146,7 +146,7 @@ inline void gpuAssert(cudaError_t code, const char *file, int line,
 
 inline cudaDeviceProp deviceProp;
 
-inline unsigned intilizeDeviceProp(unsigned deviceID, int argc, char *argv[])
+inline unsigned initializeDeviceProp(unsigned deviceID, int argc, char *argv[])
 {
 #ifdef TUNER
 

src/cuda/GPU_Microbenchmark/ubench/atomics/Atomic_add_bw/atomic_add_bw.cu

@@ -62,7 +62,7 @@ __global__ void atomic_bw(uint64_t *startClk, uint64_t *stopClk, T *data1,
 int main(int argc, char *argv[])
 {
 
-  intilizeDeviceProp(0, argc, argv);
+  initializeDeviceProp(0, argc, argv);
 
   unsigned ARRAY_SIZE = config.TOTAL_THREADS + (REPEAT_TIMES * config.WARP_SIZE);
   uint64_t *startClk = (uint64_t *)malloc(config.TOTAL_THREADS * sizeof(uint64_t));

src/cuda/GPU_Microbenchmark/ubench/atomics/Atomic_add_bw_conflict/atomic_add_bw_conflict.cu

@@ -66,7 +66,7 @@ __global__ void atomic_bw(uint32_t *startClk, uint32_t *stopClk, T *data1,
 int main(int argc, char *argv[])
 {
 
-  intilizeDeviceProp(0, argc, argv);
+  initializeDeviceProp(0, argc, argv);
 
   uint32_t *startClk = (uint32_t *)malloc(config.TOTAL_THREADS * sizeof(uint32_t));
   uint32_t *stopClk = (uint32_t *)malloc(config.TOTAL_THREADS * sizeof(uint32_t));

src/cuda/GPU_Microbenchmark/ubench/atomics/Atomic_add_lat/atomic_add_lat.cu

@@ -51,7 +51,7 @@ __global__ void atmoic_latency(uint32_t *startClk, uint32_t *stopClk, T *data1,
 int main(int argc, char *argv[])
 {
 
-  intilizeDeviceProp(0, argc, argv);
+  initializeDeviceProp(0, argc, argv);
 #ifdef TUNER
   config.THREADS_PER_BLOCK = 1;
   config.THREADS_PER_SM = 1;

src/cuda/GPU_Microbenchmark/ubench/core/MaxFlops_double/MaxFlops_double.cu

@@ -3,7 +3,7 @@
 int main(int argc, char *argv[])
 {
 
-  intilizeDeviceProp(0, argc, argv);
+  initializeDeviceProp(0, argc, argv);
 
   dpu_max_flops();
 

src/cuda/GPU_Microbenchmark/ubench/core/MaxFlops_float/MaxFlops_float.cu

@@ -3,7 +3,7 @@
 int main(int argc, char *argv[])
 {
 
-  intilizeDeviceProp(0, argc, argv);
+  initializeDeviceProp(0, argc, argv);
 
   fpu_max_flops();
 

src/cuda/GPU_Microbenchmark/ubench/core/MaxFlops_gmma/MaxFlops_gmma.cu

@@ -4,7 +4,7 @@
 
 int main(int argc, char *argv[])
 {
-  intilizeDeviceProp(0, argc, argv);
+  initializeDeviceProp(0, argc, argv);
 
   // Run comprehensive sweep over all valid MMA operations
   run_all_wgmma_maxflops_tests();

src/cuda/GPU_Microbenchmark/ubench/core/MaxFlops_gmma/MaxFlops_gmma_test.cu

@@ -5,7 +5,7 @@ void run_f32bf16bf16_64x256x16_test();
 
 int main()
 {
-  // Simple a test kernel to check if the library is working
+  // A simple test kernel to check if the library is working
   try {
     using TileShape = decltype(make_shape(cute::Int<64>{}, cute::Int<256>{}, cute::Int<16>{}));
     // Repeat once for the test

src/cuda/GPU_Microbenchmark/ubench/core/MaxFlops_half/MaxFlops_half.cu

@@ -3,7 +3,7 @@
 int main(int argc, char *argv[])
 {
 
-  intilizeDeviceProp(0, argc, argv);
+  initializeDeviceProp(0, argc, argv);
 
   fpu16_max_flops();
 

src/cuda/GPU_Microbenchmark/ubench/core/MaxIops_int32/MaxFlops_int32.h

@@ -59,7 +59,7 @@ __global__ void max_flops(uint32_t *startClk, uint32_t *stopClk, T *data1,
 
 float max_int32_flops(int argc, char* argv[]) {
 
-  intilizeDeviceProp(0,argc,argv); 
+  initializeDeviceProp(0,argc,argv); 
   #ifdef TUNER
 
   config.BLOCKS_NUM = 1;