AI/rocm-systems
2
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Replace hipLaunchKernel -> hipLaunchKernelGGL

Browse Source 
Change-Id: I4d99009e1199811d417becf1e1b934ec4d4e30be
This commit is contained in:
Maneesh Gupta
2018-10-17 12:01:44 +05:30
parent c9e4ed1149
commit 52e320f396
78 changed files with 246 additions and 264 deletions
Download Patch File Download Diff File Expand all files Collapse all files
samples/2_Cookbook/7_streams/Readme.md
+3 -3
View File
@@ -26,15 +26,15 @@ and create stream using `hipStreamCreate` as follows:
` for(int i=0;i<num_streams;i++) `
` hipStreamCreate(&streams[i]); `
and while kernel launch, we make the following changes in 5th parameter to hipLaunchKernel(having 0 as the default stream value):
and while kernel launch, we make the following changes in 5th parameter to hipLaunchKernelGGL(having 0 as the default stream value):
` hipLaunchKernel(matrixTranspose_static_shared, `
` hipLaunchKernelGGL(matrixTranspose_static_shared, `
dim3(WIDTH/THREADS_PER_BLOCK_X, WIDTH/THREADS_PER_BLOCK_Y),
dim3(THREADS_PER_BLOCK_X, THREADS_PER_BLOCK_Y),
0, streams[0],
gpuTransposeMatrix[0], data[0], width);
` hipLaunchKernel(matrixTranspose_dynamic_shared, `
` hipLaunchKernelGGL(matrixTranspose_dynamic_shared, `
dim3(WIDTH/THREADS_PER_BLOCK_X, WIDTH/THREADS_PER_BLOCK_Y),
dim3(THREADS_PER_BLOCK_X, THREADS_PER_BLOCK_Y),
sizeof(float)*WIDTH*WIDTH, streams[1],
samples/2_Cookbook/7_streams/stream.cpp
+4 -4
View File
@@ -30,7 +30,7 @@ THE SOFTWARE.
using namespace std;
__global__ void matrixTranspose_static_shared(hipLaunchParm lp, float* out, float* in,
__global__ void matrixTranspose_static_shared(float* out, float* in,
const int width) {
__shared__ float sharedMem[WIDTH * WIDTH];
@@ -44,7 +44,7 @@ __global__ void matrixTranspose_static_shared(hipLaunchParm lp, float* out, floa
out[y * width + x] = sharedMem[y * width + x];
}
__global__ void matrixTranspose_dynamic_shared(hipLaunchParm lp, float* out, float* in,
__global__ void matrixTranspose_dynamic_shared(float* out, float* in,
const int width) {
// declare dynamic shared memory
HIP_DYNAMIC_SHARED(float, sharedMem)
@@ -71,12 +71,12 @@ void MultipleStream(float** data, float* randArray, float** gpuTransposeMatrix,
hipMemcpyAsync(data[i], randArray, NUM * sizeof(float), hipMemcpyHostToDevice, streams[i]);
}
hipLaunchKernel(matrixTranspose_static_shared,
hipLaunchKernelGGL(matrixTranspose_static_shared,
dim3(WIDTH / THREADS_PER_BLOCK_X, WIDTH / THREADS_PER_BLOCK_Y),
dim3(THREADS_PER_BLOCK_X, THREADS_PER_BLOCK_Y), 0, streams[0],
gpuTransposeMatrix[0], data[0], width);
hipLaunchKernel(matrixTranspose_dynamic_shared,
hipLaunchKernelGGL(matrixTranspose_dynamic_shared,
dim3(WIDTH / THREADS_PER_BLOCK_X, WIDTH / THREADS_PER_BLOCK_Y),
dim3(THREADS_PER_BLOCK_X, THREADS_PER_BLOCK_Y), sizeof(float) * WIDTH * WIDTH,
streams[1], gpuTransposeMatrix[1], data[1], width);