-
Notifications
You must be signed in to change notification settings - Fork 280
Commit
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
Add launch latency benchmarks for triton.CompiledKernel and inductor
Summary: There are a number of more detailed views into launch latency that we can get, in addition to the path we get from `triton.JitFunction`: - `triton.compiler.CompiledKernel`, which is the lowest-level interface used by triton - Inductor's `CachingAutotuner.run`, which is the lowest-level lauch interface used by inductor - launching a mostly-nop inductor kernel (can't be truly nop because inductor won't generate a kernel with nothing in it) Reviewed By: xuzhao9, chenyang78 Differential Revision: D56073036 fbshipit-source-id: c72b80eb016a5c2ea27717664e8a1ff0f35c705a
- Loading branch information
1 parent
5c0d0ce
commit 9ff1725
Showing
1 changed file
with
87 additions
and
0 deletions.
There are no files selected for viewing
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters