Merge #148

148: Change Queue::submit to take &self r=kvark a=LaylConway

This small change allows multiple threads to submit buffers slightly more efficiently, by being able to do it concurrently without a lock on Queue.

In practice currently `Queue::submit` does a lot of locking internally for most of the function's duration, so it's only a small performance win right now, but it opens up the road for more improvements.

Co-authored-by: Layl <2385329-layl@users.noreply.gitlab.com>

Cargo.toml

@@ -36,6 +36,7 @@ rev = "b51053dc2dc3bbe9b2ba050fde42eeb6405fe092"
 
 [dependencies]
 arrayvec = "0.5"
+smallvec = "1"
 raw-window-handle = "0.3"
 
 [dev-dependencies]

examples/capture/main.rs

@@ -15,7 +15,7 @@ async fn run() {
     )
     .unwrap();
 
-    let (device, mut queue) = adapter.request_device(&wgpu::DeviceDescriptor {
+    let (device, queue) = adapter.request_device(&wgpu::DeviceDescriptor {
         extensions: wgpu::Extensions {
             anisotropic_filtering: false,
         },

examples/framework.rs

@@ -102,7 +102,7 @@ pub fn run<E: Example>(title: &str) {
     )
     .unwrap();
 
-    let (device, mut queue) = adapter.request_device(&wgpu::DeviceDescriptor {
+    let (device, queue) = adapter.request_device(&wgpu::DeviceDescriptor {
         extensions: wgpu::Extensions {
             anisotropic_filtering: false,
         },

examples/hello-compute/main.rs

@@ -24,7 +24,7 @@ async fn run() {
     )
     .unwrap();
 
-    let (device, mut queue) = adapter.request_device(&wgpu::DeviceDescriptor {
+    let (device, queue) = adapter.request_device(&wgpu::DeviceDescriptor {
         extensions: wgpu::Extensions {
             anisotropic_filtering: false,
         },

examples/hello-triangle/main.rs

@@ -44,7 +44,7 @@ fn main() {
     )
     .unwrap();
 
-    let (device, mut queue) = adapter.request_device(&wgpu::DeviceDescriptor {
+    let (device, queue) = adapter.request_device(&wgpu::DeviceDescriptor {
         extensions: wgpu::Extensions {
             anisotropic_filtering: false,
         },

src/lib.rs

@@ -5,6 +5,7 @@ use future::GpuFutureCompletion;
 pub use future::GpuFuture;
 
 use arrayvec::ArrayVec;
+use smallvec::SmallVec;
 
 use std::ffi::CString;
 use std::ops::Range;
@@ -261,7 +262,6 @@ pub struct ComputePass<'a> {
 #[derive(Debug)]
 pub struct Queue {
     id: wgc::id::QueueId,
-    temp_command_buffers: Vec<wgc::id::CommandBufferId>,
 }
 
 /// A resource that can be bound to a pipeline.
@@ -569,7 +569,6 @@ impl Adapter {
         };
         let queue = Queue {
             id: wgn::wgpu_device_get_queue(device.id),
-            temp_command_buffers: Vec::new(),
         };
         (device, queue)
     }
@@ -1382,14 +1381,14 @@ impl<'a> Drop for ComputePass<'a> {
 
 impl Queue {
     /// Submits a series of finished command buffers for execution.
-    pub fn submit(&mut self, command_buffers: &[CommandBuffer]) {
-        self.temp_command_buffers.clear();
-        self.temp_command_buffers
-            .extend(command_buffers.iter().map(|cb| cb.id));
+    pub fn submit(&self, command_buffers: &[CommandBuffer]) {
+        let temp_command_buffers = command_buffers.iter()
+            .map(|cb| cb.id)
+            .collect::<SmallVec<[_; 4]>>();
 
         wgn::wgpu_queue_submit(
             self.id,
-            self.temp_command_buffers.as_ptr(),
+            temp_command_buffers.as_ptr(),
             command_buffers.len(),
         );
     }