WebNN: Implement MLBuffer dispatch

Adds support to execute MLGraphs using MLBuffers. Allows the WebNN developer to directly bind MLBuffers as input/outputs to graphs for execution, which keeps MLBuffer data on-device after execution completes. In future CLs, dispatch can be further optimized. * Moves out per graph resources required by Dispatch(). * MLGraphBuilder.build no longer pre-allocates I/O. webmachinelearning/webnn#482 Bug: 1472888 Change-Id: I7400704cf60c149c47c20f22c50d5f12bff89cf9 Cq-Include-Trybots: luci.chromium.try:win11-blink-rel
web-platform-tests · Apr 17, 2024 · eb6ad9c · eb6ad9c
1 parent b8a81e1
commit eb6ad9c
Show file tree

Hide file tree

Showing 3 changed files with 435 additions and 1 deletion.
diff --git a/webnn/conformance_tests/buffer.https.any.js b/webnn/conformance_tests/buffer.https.any.js
@@ -14,3 +14,5 @@ testDestroyWebNNBuffer('destroyTwice');
 testReadWebNNBuffer('read');
 
 testWriteWebNNBuffer('write');
+
+testDispatchWebNNBuffer('dispatch');
diff --git a/webnn/conformance_tests/gpu/buffer.https.any.js b/webnn/conformance_tests/gpu/buffer.https.any.js
@@ -14,3 +14,5 @@ testDestroyWebNNBuffer('destroyTwice', 'gpu');
 testReadWebNNBuffer('read', 'gpu');
 
 testWriteWebNNBuffer('write', 'gpu');
+
+testDispatchWebNNBuffer('dispatch', 'gpu');