[Bug nptl/18853] dynamically loading libpthread renders locked mutexes unusable

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

https://sourceware.org/bugzilla/show_bug.cgi?id=18853

--- Comment #3 from Carlos O'Donell <carlos at redhat dot com> ---
(In reply to Alexander Monakov from comment #2)
> Sorry, but I think calling the testcase outright invalid is unfair.
> 
> The "original application" is explicitly using pthreads only for the sake of
> *being a minimal testcase*.  A similar failure can be demonstrated with the
> original application using neither libdl nor libpthread, but instead using
> some libfoo.so, which in turn depends on libdl and loads libbar.so, which
> depends on libpthread.

If you use pthread_* functions you must link with -lpthread. If you have
another test case please provide that.

> Also, I don't see why you said "I agree that it would have been better if
> the linking of the application failed" â I never asked for linking to fail,
> and neither anybody on the libc-alpha thread, as far as I remember.

I was simply stating my opinion that using pthread_* functions should require
you link with libpthread.

> I hope I clarified the intent of the bugreport.  Please reopen if you see
> what I meant.

That the glibc project should make it impossible to get this wrong? I agree.
You should not be able to get into a situation where the stubs and real
implementations get mixed across a single mutex. Was that your intent?

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