Skip to content

[ET-VK] Modify conv 2d pw op shader and dispatch settings to linearly dispatch work accounting for linearity texture to improve performance. #7452

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 4 commits into from
Jan 3, 2025
Merged

[ET-VK] Modify conv 2d pw op shader and dispatch settings to linearly dispatch work accounting for linearity texture to improve performance. #7452

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
4 commits
Select commit Hold shift + click to select a range
713ad1d
[ET-VK] Modify conv 2d pw op shader and dispatch settings to linearly…
trivedivivek Dec 28, 2024
0e37dd2
Update on "[ET-VK] Modify conv 2d pw op shader and dispatch settings …
trivedivivek Jan 2, 2025
176892a
Update on "[ET-VK] Modify conv 2d pw op shader and dispatch settings …
trivedivivek Jan 3, 2025
5fc83ae
Update on "[ET-VK] Modify conv 2d pw op shader and dispatch settings …
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
7 changes: 6 additions & 1 deletion backends/vulkan/runtime/graph/ops/glsl/conv2d_pw.glsl
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -40,7 +40,12 @@ layout(local_size_x_id = 0, local_size_y_id = 1, local_size_z_id = 2) in;
* size is only 1x1, making it easier to re-use loaded texels from t_kernel.
*/
void main() {
const u16vec3 gpos = u16vec3(gl_GlobalInvocationID);
const uint16_t out_limits_y_scaled = uint16_t((out_limits.y + TILE_SIZE - 1) / TILE_SIZE);

const u16vec3 gpos = u16vec3(
gl_GlobalInvocationID.x / (out_limits_y_scaled * out_limits.z),
(gl_GlobalInvocationID.x / out_limits.z) % out_limits_y_scaled,
gl_GlobalInvocationID.x % out_limits.z);

// Output position for TILE_SIZE = 2
// +--------+--------+
Expand Down
4 changes: 4 additions & 0 deletions 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 @@ -372,6 +372,10 @@ void add_conv2d_node(

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

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

graph.execute_nodes().emplace_back(new DispatchNode(
graph,
shader,
Expand Down