Skip to content

[webgpu][dawn API optimization] reduce number of calls to wgpuDeviceGetQueue #24313

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
fs-eire merged 1 commit into from
Apr 4, 2025
Merged

[webgpu][dawn API optimization] reduce number of calls to wgpuDeviceGetQueue #24313

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
6 changes: 4 additions & 2 deletions onnxruntime/core/providers/webgpu/webgpu_context.cc
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -136,6 +136,8 @@ void WebGpuContext::Initialize(const WebGpuBufferCacheConfig& buffer_cache_confi

LOGS_DEFAULT(VERBOSE) << "WebGPU EP Context is created for: Instance=" << instance_.Get() << ", Device=" << device_.Get() << ".";

// cache device queue
device_queue_ = device_.GetQueue();
// cache adapter info
ORT_ENFORCE(Device().GetAdapterInfo(&adapter_info_));
// cache device limits
Expand Down Expand Up @@ -404,7 +406,7 @@ Status WebGpuContext::Run(ComputeContext& context, const ProgramBase& program) {
}

uniform_buffer = buffer_mgr_->Create(uniform_buffer_total_size, wgpu::BufferUsage::CopyDst | wgpu::BufferUsage::Uniform);
device_.GetQueue().WriteBuffer(uniform_buffer, 0, uniform_data_buffer.data(), uniform_buffer_total_size);
device_queue_.WriteBuffer(uniform_buffer, 0, uniform_data_buffer.data(), uniform_buffer_total_size);
}

const auto& compute_pass_encoder = GetComputePassEncoder();
Expand Down Expand Up @@ -696,7 +698,7 @@ void WebGpuContext::Flush() {
}

auto command_buffer = current_command_encoder_.Finish();
Device().GetQueue().Submit(1, &command_buffer);
device_queue_.Submit(1, &command_buffer);
BufferManager().RefreshPendingBuffers();
current_command_encoder_ = nullptr;
num_pending_dispatches_ = 0;
Expand Down
1 change: 1 addition & 0 deletions onnxruntime/core/providers/webgpu/webgpu_context.h
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -207,6 +207,7 @@ class WebGpuContext final {

webgpu::ValidationMode validation_mode_;

wgpu::Queue device_queue_;
wgpu::AdapterInfo adapter_info_;
wgpu::Limits device_limits_;
std::unordered_set<wgpu::FeatureName> device_features_;
Expand Down
Loading