Skip to content

Conversation

@ggerganov
Copy link
Member

depends on #7856

Unary op implementations on the GPU assume that src0 is contiguous, so we now GGML_ASSERT this in the backend code

The CPU implementation actually supports ggml_is_contiguous_1(src0) (i.e. elements within the rows are contiguous and dims 2, 3 are also contiguous, but there could be gaps between the rows). Maybe in the future the GPU backends can be extended to also support that, but for now their support_op are updated to return false in such cases

@github-actions github-actions bot added testing Everything test related Nvidia GPU Issues specific to Nvidia GPUs Vulkan Issues specific to the Vulkan backend ggml changes relating to the ggml tensor library for machine learning SYCL https://en.wikipedia.org/wiki/SYCL - GPU programming language Kompute https://github.com/KomputeProject/kompute/ labels Jun 10, 2024
@mofosyne mofosyne added the Review Complexity : Medium Generally require more time to grok but manageable by beginner to medium expertise level label Jun 12, 2024
@ggerganov ggerganov force-pushed the gg/unary-non-cont branch from b64daed to 8412561 Compare June 12, 2024 12:25
@ggerganov ggerganov merged commit a9cae48 into master Jun 12, 2024
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

ggml changes relating to the ggml tensor library for machine learning Kompute https://github.com/KomputeProject/kompute/ Nvidia GPU Issues specific to Nvidia GPUs Review Complexity : Medium Generally require more time to grok but manageable by beginner to medium expertise level SYCL https://en.wikipedia.org/wiki/SYCL - GPU programming language testing Everything test related Vulkan Issues specific to the Vulkan backend

Projects

None yet

Development

Successfully merging this pull request may close these issues.

4 participants