0001-mingw-w64-time-functions.patch

[MSP-Greg]

I'm fairly certain that __MINGW64_VERSION_MAJOR is defined for the current release of MinGW / MSYS2.

See https://gist.github.com/jonforums/968522.

Hence, I suspect the patch 0001-mingw-w64-time-functions.patch is no longer needed.

I haven't been using it with my local builds for quite a while, and I'm around ten total failures and errors using --with-out-ext=pty,syslog.

Again, thanks for all your work.

[MSP-Greg]

@larskanis Thank you
See 46e1803

[larskanis]

@MSP-Greg Thanks for the hint! This issue has been the trigger to re-check the patches, but I didn't remember it.

[MSP-Greg]

@larskanis Not being a c type, I thought having you check it would be appropriate.

I've been building parallel for while, but at some point I added include_ruby_defines.h.patch. It's very similar to 0002-use-gnu-printf.patch, but changes include/ruby/defines.h.

I haven't tried a recent build with it removed, but it doesn't seem to cause any issues. If you get a chance to look at it, let me know if it's required...

FYI, there are a few recent trunk commits that affected encoding and temp files. My test-all has gone from 5 failures & 0 errors to 6 failures & 1 errors. test-rubyspec ran parallel previously, now it will not... ruby-loco...

Lastly, I recently build eventmachine with SSL support using my MinGW OpenSSL 1.1.0f package. I should probably update it, as OpenSSL stable has had quite a few commits. Should work with the standard MinGW package also.

[larskanis]

@MSP-Greg Did you file the gnu-printf issue on the ruby bug tracker, already?

[MSP-Greg]

@larskanis

I mentioned them right at the top of 13542 as the only two patches used for building. All my other patches are for testing.

Considering that the configure.in patch has been around for quite a while, I assumed (possibly incorrectly) that the issue had already been addressed.

@luislavena is listed as a maintainer for MinGW and he's listed as a committer/developer.

If he has time to look at it, the patches are: configure.in and defines.h...

[luislavena]

@luislavena is listed as a maintainer for MinGW and he's listed as a committer/developer.

I'm no longer actively involved with Ruby to be able to weight in or commit any changes. It will be better reach out to nobu instead.

[MSP-Greg]

@luislavena thanks for the reply. Nobu has helped with several commits affecting parallel builds and more.