Re: [opensoftware] [sr #110849] make check darwin.at test fails on Linux

[Christoph Prokop]

> Was this run on a git checkout or distribution tarball?

tarball:
https://ftp.gnu.org/gnu/libtool/libtool-2.4.7.tar.gz

> You bring up a good question. If it's a git checkout I think it makes sense 
> to run these since it may detect potential portability issues for Darwin 
> hosts that are observable from Linux systems. It may also make sense to run 
> some of these tests even in a distributed tarball if they are trying to 
> detect cross-compilation issues from a Linux build system for Darwin 
> hosts/targets.

I wanted to run make check to see if the build was successful and i was happy 
to see that there are a lot of tests.
I am not sure if i can expect to get an exit code 0 after that or if i just 
have to check the results if the failed tests really matter to me.
But i was curious since it was only the one test which i would have expected to 
be skipped in the first place.

> Overall it's a good question though and I'd like to understand more about why 
> they're running because it's certainly possible that they should be skipped 
> or that the test suite isn't detecting the build/host system properly.

Please let me know if you need any further information.
I am happy if i can help.