[Bug dynamic-link/24506] FAIL: elf/tst-pldd

["carlos at redhat dot com" <sourceware-bugzilla at sourceware dot org>]

https://sourceware.org/bugzilla/show_bug.cgi?id=24506

--- Comment #6 from Carlos O'Donell <carlos at redhat dot com> ---
(In reply to Adhemerval Zanella from comment #5)
> (In reply to Carlos O'Donell from comment #4)
> > Right, --enable-hardcoded-path-in-tests is not enough to make all tests
> > work, some tests require more setup e.g. env vars, containers (chroot),
> > special arguments.
> 
> Maybe we should disable such tests if --enable-hardcoded-path-in-tests is
> used?

That would be bad for test coverage, and --enable-hardcoded-path-in-tests is a
developer-only option, so you have to know what you're doing. The option allows
stand-alone tests to run just fine, all other tests need special options.

My current idea is as follows:
- Every test we execute, we record *how* it was executed.
- Provide a log at the end that can be parsed by some kind of simplistic test
runner, and you can use that to re-run tests by # or name.

The reason I want this is to be able to build a tests subpackage downstream and
use it for A/B patch testing and running the test outside of the testsuite. I
don't yet want to split the testsuite out of glibc because there is value in
having it integrated.

-- 
You are receiving this mail because:
You are on the CC list for the bug.