Cyril Hrubis

It would probably make more sense to put it into glibc given that they started to add tests for mq_notify recently https://sourceware.org/git/?p=glibc.git;a=commit;h=4cc79c217744743077bf7a0ec5e0a4318f1e6641

This is the same as strcmp() tests, generally looks good, needs a few minor things to be sorted out.

That looks like the system is out-of-memory, but without further debugging we cannot know what happened and why there is not enough of free RAM.

The easiest way is probably to copy the controllers runtest file and delete everything you are not interested in from the copy. I guess that it would make sense to...

@richiejp Can you please have a look at this as well?

If we are going to remove these flags we also have to add a new tests that call the *chown() syscalls with additional the bits enabled, because otherwise we will...

Well LTP has quite a few testcases that test behavior like that, if they start to fail we talk to the kernel maintaners if the change was intentional or not...

Yes the fucntional test should be separated like that.

The CI just checks if the code could be compiled on different distributions, as writing portable code even for Linux is not easy if you need to cover popular distributions...

We are phasing out uclinux support, so yes we can drop it. Also these tests should be converted to use the new library as well.