webgpu: Fix HTML event loop integration#34631
Conversation
Signed-off-by: sagudev <16504129+sagudev@users.noreply.github.com>
Signed-off-by: sagudev <16504129+sagudev@users.noreply.github.com>
| @@ -1621,6 +1621,9 @@ impl ScriptThread { | |||
|
|
|||
| // TODO(#31871): Update the rendering: consolidate all reflow calls into one here? | |||
There was a problem hiding this comment.
Note: unrelated but this TODO can be removed now. Was missed previously.
Signed-off-by: sagudev <16504129+sagudev@users.noreply.github.com>
|
For the worker part(do we have offscreen canvas working with webgpu yet?) I have filed #34637 |
IIRC all offscreen support is unfinished/disabled by default, so I didn't really bother with it. Workers are in same state too IIRC, but I might be wrong. |
{"fail_fast": false, "matrix": [{"name": "WebGPU CTS", "workflow": "linux", "wpt_layout": "2020", "profile": "production", "unit_tests": false, "bencher": false, "wpt_args": "_webgpu"}]}
|
I just checked and firefox does update the wr image (so update the rendering of context) as part of reflow/layout. So inverse of what we did here. cc @gterzian |
Funny timing of your comment as I just posted #37703, which consolidates all the canvas flushes into one place. I think it doesn't matter too much whether this happens in |
|
I actually think this should be done in reflow (step 22 of update the rendering). And yeah, I want to consolidate that too. |
rebase of #33673, now possible because of latest changes to update rendering stuff in servo.
./mach build -ddoes not report any errors./mach test-tidydoes not report any errors