Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

[X86] add missing TP and latency information for rcl_N/rcr_N #110415

Open
Ruihan-Yin opened this issue Dec 4, 2024 · 2 comments
Open

[X86] add missing TP and latency information for rcl_N/rcr_N #110415

Ruihan-Yin opened this issue Dec 4, 2024 · 2 comments
Assignees
Labels
area-CodeGen-coreclr CLR JIT compiler in src/coreclr/src/jit and related components such as SuperPMI
Milestone

Comments

@Ruihan-Yin
Copy link
Contributor

https://github.com/dotnet/runtime/blob/main/src/coreclr/jit/emitxarch.cpp#L18813

As of now, JIT does not have the throughput and latency information rcl_N and rcr_N, we will need to add these information to make sure the instructions are properly emitted and measured when they are used.

As of now, I don't run into any issue with this in testing, so it means rcl_N/rcr_N may not be used at all now, and might indicate some optimization for rotation might be missing. This part may need further investigation.

@dotnet-issue-labeler dotnet-issue-labeler bot added the area-CodeGen-coreclr CLR JIT compiler in src/coreclr/src/jit and related components such as SuperPMI label Dec 4, 2024
@dotnet-policy-service dotnet-policy-service bot added the untriaged New issue has not been triaged by the area owner label Dec 4, 2024
Copy link
Contributor

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

@JulieLeeMSFT JulieLeeMSFT added this to the 10.0.0 milestone Dec 6, 2024
@JulieLeeMSFT JulieLeeMSFT removed the untriaged New issue has not been triaged by the area owner label Dec 6, 2024
@Ruihan-Yin
Copy link
Contributor Author

https://uops.info/html-instr/RCR_R32_I8.html#SKL
https://uops.info/html-instr/RCL_R32_I8.html#SKL

Looking at how INS_rcr/INS_rcl (the CL variation) are defined in JIT now, I would suppose it should be fine to use the same tp/latency data for INS_rcr_N/INS_rcl_N if we want a rough estimation. While there are some variation from the experiment, I wonder how we usually handle this.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
area-CodeGen-coreclr CLR JIT compiler in src/coreclr/src/jit and related components such as SuperPMI
Projects
None yet
Development

No branches or pull requests

3 participants