-
Notifications
You must be signed in to change notification settings - Fork 27
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.
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
Update to wgpu 0.20 #67
Conversation
231c333
to
5552967
Compare
I think this might also need to deal with the new timestamp related features flag? |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
@DJMcNab is right, the feature requirements for timers have changed. Need to go through all the places that configure this in the project
Due to a specification change write_timestamp is no longer supported on WebGPU.
wgpu::CommandEncoder::write_timestamp requires now the new wgpu::Features::TIMESTAMP_QUERY_INSIDE_ENCODERS feature which is available on all native backends but not on WebGPU.
5552967
to
09ec2cf
Compare
Unfortunately, I need to leave now unexpectedly to drive 250km and won't be back for some hours, so I won't be able to get back to this until about 24 hours from now at the earliest, so if anyone else can pick it up and run with it, they're welcome to it! |
Bump version to 0.17 to prepare for new release.
09ec2cf
to
281057f
Compare
I pushed an update but forgot to mention it ... this at least passes tests but there may be a better way to do parts of it? |
@Wumpf Just checking in on this if you have the time. We'd like to update Vello to wgpu 0.20 for the shader compilation improvements and this is needed for that. |
(But I know everyone is busy with the run up to RustNL and other things!) |
I'll try to have a close look and release a new version until Monday but don't want to promise anything ;-) |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
handling of the new feature split is still not quite correct, this needs more work
I can take it from here
I'm on it, but looks like I stumbled over a bug; managed to violate an internal invariant 🤔 Edit: Figured it out: Has to do with the fact that you can how have encoder scopes without queries containing timestamp_write scopes with queries containing pass scopes without queries. Before this wasn't possible. |
Had to change way results are reported to allow enabled timers to be nested inside of disabled
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
oof this was quite a chunk of work, but I think everything is in order now. @waywardmonkeys would be great if you could check on the changes I did before I put out a new release
I am on a trip this weekend without my laptop. (First time without a computer in over a decade!) I will try to visually review it via phone tonight. thanks for doing all of this! |
@Wumpf This seems to work once I fix up the code in the |
No description provided.