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

[R] Enable prefetch on windows configure script #11141

Open
wants to merge 3 commits into
base: master
Choose a base branch
from

Conversation

david-cortes
Copy link
Contributor

ref #9810
ref #11124

This PR adds prefetching to the compilation flags on windows when supported.

Since CRAN only uses MinGW for windows builds and the msvc build here was dropped, it enables the GC builtin unconditionally, since no other compiler will be used on windows.

@david-cortes
Copy link
Contributor Author

@trivialfis I'm not sure what to do about the CMake scripts that are failing in the CI here. This PR creates a file "Makevars.win.in" from which "Makevars.win" is created, but I guess the configure script that generates "Makevars.win" will not get executed by CMake.

@trivialfis
Copy link
Member

Looks like apple clang internal error for the Python test. On the other hand, do we need to worry about R stringi? https://github.com/dmlc/xgboost/actions/runs/12609778121/job/35143648760?pr=11141

@david-cortes
Copy link
Contributor Author

Looks like apple clang internal error for the Python test. On the other hand, do we need to worry about R stringi? https://github.com/dmlc/xgboost/actions/runs/12609778121/job/35143648760?pr=11141

No idea about the stringi error. But besides the apple issue, there's a problem with the cmake build trying to use a fixed Makevars.win, which I am not sure how to solve. Perhaps it could create that file through an additional target or similar, but don't know what'd be the best way.

@trivialfis
Copy link
Member

That's weird, I will look into it.

@trivialfis
Copy link
Member

I'm not sure what to do about the CMake scripts that are failing in the CI here.

Could you please point me to the CI failure? I couldn't find it.

@david-cortes
Copy link
Contributor Author

Looks like the error got solved. This was the log:
https://github.com/dmlc/xgboost/actions/runs/12609648878/job/35143402945

However I don't see any windows cmake R job in the CI, and I'm not sure if it'd work under such a combination (windows + cmake + R + non-msvc compiler).

@trivialfis
Copy link
Member

I have done some tests today with Rtools43. Need some updates to get cmake+mingw+Windows working. Not sure if it worths the effort to maintain.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

2 participants