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

Skip the failing dynamic vs static tests on Win32 in blead #369

Merged
merged 1 commit into from
Sep 30, 2020

Conversation

toddr
Copy link
Contributor

@toddr toddr commented Sep 24, 2020

Work around for Perl/perl5#17601

See also https://rt.cpan.org/Ticket/Display.html?id=115321

This mirror's @tonycoz 's merge to blead found here Perl/perl5#18157

@karenetheridge
Copy link
Member

It would be good to get to the bottom of these test issues, rather than disabling them.

@toddr
Copy link
Contributor Author

toddr commented Sep 25, 2020

I agree. However, RT-115321 has been open for 4 years now.

Also to Quote @tonycoz:

%Config provides the information required, but the generated Makefile in this case is generated for LINKTYPE=static, but then run with LINKTYPE=dynamic for Linux that's not such a big deal, if perl is dynamic ccflags includes -fPIC as a matter of course, which just slows the resulting binary, but it is a problem on Win32 due to the way data imports work. I have trouble believing anyone is doing this in the wild ... it might be that the test just isn't practical on Win32, to make it work I think you'd have to use whatever conditional mechanisms the various Win32 makes support.

This has been causing noise on the smokers and I agree with him that turning the test off for a truly corner case like this makes sense until someone comes back and says: I care about Perl running this way on Windows. I am dubious that this will ever happen.

@Leont
Copy link
Member

Leont commented Sep 25, 2020

It would be good to get to the bottom of these test issues, rather than disabling them.

I think it's understood well enough. It's the sort of thing that is not very portable, and arguably people shouldn't be doing this anyway (but probably someone does).

That said, not passing through LINKTYPE may be a better solution.

@toddr
Copy link
Contributor Author

toddr commented Sep 25, 2020

That said, not passing on LINKTYPE may be a better solution.

@Leont To move this forward, would you be up for providing the alternative pull request?

OR: Can we merge this and you provide the alternative when it is available?

@Leont
Copy link
Member

Leont commented Sep 25, 2020

@Leont To move this forward, would you be up for providing the alternative pull request?

OR: Can we merge this and you provide the alternative when it is available?

Changing this is trivial, but I haven't figured out if anything would break. I can't think of anything, but this is MakeMaker we're talking about…

@toddr
Copy link
Contributor Author

toddr commented Sep 25, 2020

Changing this is trivial, but I haven't figured out if anything would break. I can't think of anything, but this is MakeMaker we're talking about…

And while it's critical, we also get feedback really fast when it breaks something. Given we've thought about it for 4 years. maybe the best option at this point is to try your suggestion and get feedback?

@xsawyerx
Copy link

Changing this is trivial, but I haven't figured out if anything would break. I can't think of anything, but this is MakeMaker we're talking about…

And while it's critical, we also get feedback really fast when it breaks something. Given we've thought about it for 4 years. maybe the best option at this point is to try your suggestion and get feedback?

I was going to respond with a similar question. Would it not be useful to put in the suggestion solution and try to receive feedback?

@bingos bingos merged commit ede92b7 into Perl-Toolchain-Gang:master Sep 30, 2020
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.

None yet

5 participants