-
-
Notifications
You must be signed in to change notification settings - Fork 307
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
WebGPU 0.8
#737
WebGPU 0.8
#737
Conversation
…inside an if (it's not allowed)
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.
dst_factor: wgpu::BlendFactor::DstColor, | ||
pub const LIGHTEST: wgpu::BlendComponent = wgpu::BlendComponent { | ||
src_factor: wgpu::BlendFactor::One, | ||
dst_factor: wgpu::BlendFactor::One, |
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.
Thanks for checking these blend factors! Would you mind sharing what was wrong with these and how you arrived at these fixes?
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.
Webgpu validation layer has complained that if the operation is Max
/ Min
, the only option that makes sense for src / dst is One
@@ -588,7 +588,7 @@ where | |||
.extent(wgpu::Extent3d { | |||
width, | |||
height, | |||
depth: array_layers, | |||
depth_or_array_layers: array_layers, |
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.
I'm glad they clarified this - I remember asking in the wgpu users room if depth applied to the layer count 😆
@@ -56,29 +56,29 @@ pub use wgpu_upstream::{ | |||
util::{self, BufferInitDescriptor}, | |||
vertex_attr_array, Adapter, AdapterInfo, AddressMode, Backend, BackendBit, BindGroup, | |||
BindGroupDescriptor, BindGroupEntry, BindGroupLayout, BindGroupLayoutDescriptor, | |||
BindGroupLayoutEntry, BindingResource, BindingType, BlendFactor, BlendOperation, BlendState, | |||
Buffer, BufferAddress, BufferAsyncError, BufferBindingType, BufferCopyView, BufferCopyViewBase, | |||
BindGroupLayoutEntry, BindingResource, BindingType, BlendComponent, BlendFactor, |
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.
This file is just due to cargo fmt
I've just been testing the latest commits and all seems good, though I'm having some issues running the |
I must have missted that, as I've been running examples. I'll debug it and see what's going on. In the meantime we also need to publish the conrod changes so this doesn't depend on my branch but a proper version. I've just discovered There's a bunch that are failing and i'll track fixes here:
|
|
Fixed the conrod issue, will raise a PR soon |
I think that conrod issue should be fixed as of publishing version |
Yup, i just realied i was pointing to the wrong branch. The only remaining thing is the instancing example. |
Done :) It seems that the instancing demo requires locations to be mentioned by name and the matrix to be reconstructed manually. I've pushed the fix |
OK just pulled down the latest commit and it looks like the
Going to sleep now, but will check this tomorrow |
I've uploaded the vertex shader. It seems I just uploaded the code, but not the compiled version. This explains why my tests passed locally. I double checked that no shader binaries have been missed and tests still pass. |
This is impressive! I noticed one other validation error with the example
|
That's weird, I didn't notice that one in the output, I'll take a look in the following days and fix it. |
@danwilhelm @AlexEne I opened a PR with a patch for that a week or two ago here AlexEne#1 - sorry I should have linked it here when I opened it! I thiiiink that might be the last patch before we can merge this, but I'll try running the rest of the examples now to double check. Edit: Can confirm, everything else appears to run smoothly - once that PR above is merged I'll merge this. |
Use correct TextureViewDimension for single array layer
@mitchmindtree thanks for that. Obviously life got in the way of me working on this last weekend :D. But it should be good to go. I merged it |
Thanks again @AlexEne, I think this is good to land! |
Opening a PR for 0.8. There are a few kinks left as tracked in the issue here: #734
I'm opening this as it will be easier to track changes as part of a PR than an issue.