Skip to content

[JIT] X64 - Using cmovns for faster division - #81252

Closed
TIHan wants to merge 4 commits into
dotnet:mainfrom
TIHan:cmovns-opt
Closed

[JIT] X64 - Using cmovns for faster division#81252
TIHan wants to merge 4 commits into
dotnet:mainfrom
TIHan:cmovns-opt

Conversation

@TIHan

Copy link
Copy Markdown
Contributor

Will resolve#41549 once I figure a few things out

@ghostghost added the area-CodeGen-coreclr CLR JIT compiler in src/coreclr/src/jit and related components such as SuperPMI label Jan 27, 2023
@ghostghost assigned TIHanJan 27, 2023
@ghost

Copy link
Copy Markdown

Tagging subscribers to this area: @JulieLeeMSFT, @jakobbotsch, @kunalspathak
See info in area-owners.md if you want to be subscribed.

Issue Details

Will resolve #41549 once I figure a few things out

Author:TIHan
Assignees:-
Labels:

area-CodeGen-coreclr

Milestone:-

@TIHanTIHan changed the title [JIT] Using cmovns for faster division[JIT] X64 - Using cmovns for faster divisionJan 27, 2023
@jakobbotsch

Copy link
Copy Markdown
Member

@TIHan I'd suggest waiting for #81267 to make it in and then solve this in the way @EgorBo suggested in the issue:

My working (ugly) prototype for RyuJIT: EgorBo@9b1d149 (I believe it should be done the other way - I'd introduce a GT_SELECT/GT_CMOV operator so we can later use it for other cmov-based optimizations, e.g. remove branches)

It should be less specific/pattern matchy.

@TIHan

Copy link
Copy Markdown
ContributorAuthor

@jakobbotsch - Ok, I can wait.

@ghost

Copy link
Copy Markdown

Draft Pull Request was automatically closed for 30 days of inactivity. Please let us know if you'd like to reopen it.

@ghostghost locked as resolved and limited conversation to collaborators Mar 29, 2023
This pull request was closed.
Sign up for freeto subscribe to this conversation on GitHub. Already have an account? Sign in.

Labels

area-CodeGen-coreclrCLR JIT compiler in src/coreclr/src/jit and related components such as SuperPMI

Projects

None yet

Development

Successfully merging this pull request may close these issues.

RyuJIT: Optimize "X / POW2_CNS" via cmovns

2 participants

@TIHan@jakobbotsch