Skip to content

Navigation Menu

Sign in
Appearance settings

Search code, repositories, users, issues, pull requests...

Provide feedback

We read every piece of feedback, and take your input very seriously.

Saved searches

Use saved searches to filter your results more quickly
Appearance settings

speed up f16 on CUDA with hmul2, hadd2, etc. #2752

New issue
Copy link
New issue
Copy link
Open
Open
speed up f16 on CUDA with hmul2, hadd2, etc.#2752
Copy link
Labels
CUDAimprovementperf
@ghost

Description

@ghost
ghost
opened on Feb 9, 2020

f16 CUDA can be faster with specialized half ops that operate on vector inputs. Correct me if I'm wrong, but I don't see them being used in the source code.

Metadata

Metadata

Assignees

No one assigned

    Labels

    CUDAimprovementperf

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions
      Morty Proxy This is a proxified and sanitized view of the page, visit original site.