-
Notifications
You must be signed in to change notification settings - Fork 1.2k
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
domeLight.glslfx doesn't support Apple's Metal #1656
Comments
Filed as internal issue #USD-6970 |
pixar-oss
pushed a commit
that referenced
this issue
Nov 16, 2021
HgiShaderFunctionDesc, allowing client to specify the desired number of threads in the local compute work group for a computer shader. HgiShaderGenerator now handles the writing of this local size to the shader, if necessary. The HgiComputeCmds::Dispatch() function for HgiGL and HgiVulkan will now use the local size provided by the HgiShaderFunctionDesc to determine number of work groups to dispatch. Calls to Dispatch() should specify the total size of the compute work (and not the total work size / local work size), and the backend will handle determining the number of work groups. Also adds some limit checks on work group size and number of work groups based on device limits. See #1656 (Internal change: 2199330)
pixar-oss
pushed a commit
that referenced
this issue
Nov 25, 2021
…akes some progress towards getting the dome light computations to run on Metal. See #1656 (Internal change: 2200704) (Internal change: 2201198)
lkerley
pushed a commit
to imageworks/USD
that referenced
this issue
Jan 7, 2022
HgiShaderFunctionDesc, allowing client to specify the desired number of threads in the local compute work group for a computer shader. HgiShaderGenerator now handles the writing of this local size to the shader, if necessary. The HgiComputeCmds::Dispatch() function for HgiGL and HgiVulkan will now use the local size provided by the HgiShaderFunctionDesc to determine number of work groups to dispatch. Calls to Dispatch() should specify the total size of the compute work (and not the total work size / local work size), and the backend will handle determining the number of work groups. Also adds some limit checks on work group size and number of work groups based on device limits. See PixarAnimationStudios#1656 (Internal change: 2199330)
lkerley
pushed a commit
to imageworks/USD
that referenced
this issue
Jan 7, 2022
…akes some progress towards getting the dome light computations to run on Metal. See PixarAnimationStudios#1656 (Internal change: 2200704) (Internal change: 2201198)
pixar-oss
pushed a commit
that referenced
this issue
Feb 25, 2022
fullscreen shader pipelines. Contribution: Jon Creighton See #1656 (Internal change: 2217440)
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
When using latest USD on macOS, the domeLight.glslfx compute shader fails to compile on Metal.
This is due to a few GL hard-coding in the shader which makes the shader generator fail to recognize/replace tokens, such as hd_GlobalInvocationID.
Based on other uses of HdStGLSLProgram::GetComputeProgram, seems like there's something to pass to setup the kernel parameters, including output texture
Log:
The text was updated successfully, but these errors were encountered: