frozenleaves/pytorch
1
0
Fork 0
mirror of https://github.com/pytorch/pytorch.git synced 2025-10-20 21:14:14 +08:00
Code Issues Packages Projects Releases Wiki Activity

Updated Codegen and Structured Kernels (markdown)

Manuel
2024-07-03 19:27:24 +02:00
parent 5e61168773
commit 4eda1f5069
1 changed files with 3 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files

3
Codegen-and-Structured-Kernels.md

@ -386,3 +386,6 @@ struct structured_add_out_functional final : public at::native::structured_add_o
You can see that it has its own definition of set_output() - in this case, it’s implementing the functional at::add kernel, so it needs to allocate a new tensor as the output (it does that using at::native::empty_cpu()).
That class corresponds to one of the leaves of the class hierarchy - a picture of the full class hierarchy can be found in the linked presentation (https://drive.google.com/file/d/16qPvpCF4Jbh7ss2lCQMk5hmcyzJvUyQj/view?usp=sharing)
## Next
Unit 4: Dispatcher, Structured Kernels, and Codegen - [Dispatcher and Structured Kernels Lab](https://github.com/pytorch/pytorch/wiki/Dispatcher---Structured-Kernels-Lab)