Skip to content

[ET-VK] Changing texture access pattern for conv2d dw ops to improve performance. #7477

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

Merged
merged 3 commits into from
Jan 3, 2025
Merged

[ET-VK] Changing texture access pattern for conv2d dw ops to improve performance. #7477

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
969ebaf
[ET-VK] Changing texture access pattern for conv2d dw ops to improve …
trivedivivek Jan 2, 2025
de83341
Update on "[ET-VK] Changing texture access pattern for conv2d dw ops …
trivedivivek Jan 3, 2025
09373ca
Update on "[ET-VK] Changing texture access pattern for conv2d dw ops …
trivedivivek Jan 3, 2025
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
5 changes: 4 additions & 1 deletion backends/vulkan/runtime/graph/ops/glsl/conv2d_dw.glsl
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -35,7 +35,10 @@ layout(local_size_x_id = 0, local_size_y_id = 1, local_size_z_id = 2) in;
* output at a single output location.
*/
void main() {
const ivec3 pos = ivec3(gl_GlobalInvocationID);
const ivec3 pos = ivec3(
gl_GlobalInvocationID.x % out_limits.x,
(gl_GlobalInvocationID.x / out_limits.x) % out_limits.y,
gl_GlobalInvocationID.x / (out_limits.x * out_limits.y));

if (any(greaterThanEqual(pos, out_limits))) {
return;
Expand Down
5 changes: 4 additions & 1 deletion backends/vulkan/runtime/graph/ops/glsl/conv2d_dw_output_tile.glsl
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -39,7 +39,10 @@ layout(local_size_x_id = 0, local_size_y_id = 1, local_size_z_id = 2) in;
* output at a single output location.
*/
void main() {
const u16vec3 pos = u16vec3(gl_GlobalInvocationID);
const u16vec3 pos = u16vec3(
gl_GlobalInvocationID.x % out_limits.x,
(gl_GlobalInvocationID.x / out_limits.x) % out_limits.y,
gl_GlobalInvocationID.x / (out_limits.x * out_limits.y));

if (any(greaterThanEqual(pos, out_limits))) {
return;
Expand Down
2 changes: 1 addition & 1 deletion backends/vulkan/runtime/graph/ops/impl/Convolution.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -372,7 +372,7 @@ void add_conv2d_node(

utils::uvec3 wg_size = create_conv2d_global_wg_size(graph, method, out);

if (method == Conv2dMethod::Pointwise) {
if (method == Conv2dMethod::Pointwise || method == Conv2dMethod::Depthwise) {
wg_size = {wg_size[0] * wg_size[1] * wg_size[2], 1, 1};
}

Expand Down