[prev in list] [next in list] [prev in thread] [next in thread] 

List:       glibc-alpha
Subject:    Re: Make linknamespace tests check only relevant libraries
From:       Joseph Myers <joseph () codesourcery ! com>
Date:       2014-11-29 9:29:14
Message-ID: alpine.DEB.2.10.1411290927150.5143 () digraph ! polyomino ! org ! uk
[Download RAW message or body]

On Fri, 28 Nov 2014, Carlos O'Donell wrote:

> Some of them are the `profil_counter` failure you noted, but
> others remain the __pointer_chk_guard_local inclusion and
> subsequent libpthread analysis, which I think is not correct,
> but haven't reviewed in detail.
> 
> FAIL: conform/POSIX/assert.h/linknamespace
> ~~~
> [initial] __assert_fail -> [libc.a(assert.o)] __fxprintf -> [libc.a(fxprintf.o)] \
> __vfwprintf -> [libc.a(vfwprintf.o)] _IO_fwide -> [libc.a(iofwide.o)] \
> __pointer_chk_guard_local -> [libc.a(libc-start.o)] __pthread_initialize_minimal -> \
> [libpthread.a(nptl-init.o)] __nptl_setxid_error -> [libpthread.a(pthread_create.o)] \
> __resp -> [libc.a(res_libc.o)] __res_iclose -> [libc.a(res_init.o)] fgets_unlocked

You'll note the pthread.h tests (for example) are XFAILed for all 
standards with pthreads at present.  So naturally when pthreads code is 
brought in, even without any explicit use of pthreads in the program, 
you'll get lots of failures that are logically the same as those already 
XFAILed.

-- 
Joseph S. Myers
joseph@codesourcery.com


[prev in list] [next in list] [prev in thread] [next in thread]