Skip to content

Change device pointers to host to enable concurrent copy for multi-stream variant #1

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
Baxi-codes wants to merge 1 commit into
base: main
Choose a base branch
from
Open

Change device pointers to host to enable concurrent copy for multi-stream variant #1

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
2 changes: 2 additions & 0 deletions src/conv_kernels.cu
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -202,6 +202,8 @@ void conv2d_variant3(const float* input, const float* kernel, float* output,
// - Degree of overlap visible on the NSYS timeline (copies concurrent with kernels).
// - Any change in kernel performance (avoid starving compute with too many small chunks).
//
// NOTE: const float* input, const float* kernel are host pointers in this task
//
// =============================================================================


Expand Down
2 changes: 1 addition & 1 deletion src/main.cu
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -157,7 +157,7 @@ int main(int argc, char** argv) {
conv2d_variant3(device_input, device_kernel, device_output,
batch_size, height, width, kernel_size, stream);
} else if (impl_name == "variant4") {
conv2d_variant4(device_input, device_kernel, device_output,
conv2d_variant4(host_images.data(), host_kernel.data(), device_output,
batch_size, height, width, kernel_size, stream);
} else {
std::fprintf(stderr, "Unknown implementation: %s\n", impl_name.c_str());
Expand Down