This is the mail archive of the
libc-alpha@sourceware.org
mailing list for the glibc project.
Re: [PATCH]: hurd: Do not hide rtld symbols which need to be preempted [Was: hurd: Hidden symbols in rtld]
- From: "H.J. Lu" <hjl dot tools at gmail dot com>
- To: Samuel Thibault <samuel dot thibault at ens-lyon dot org>
- Cc: GNU C Library <libc-alpha at sourceware dot org>
- Date: Sun, 20 Mar 2016 10:14:40 -0700
- Subject: Re: [PATCH]: hurd: Do not hide rtld symbols which need to be preempted [Was: hurd: Hidden symbols in rtld]
- Authentication-results: sourceware.org; auth=none
- References: <20160320164214 dot GA21096 at var dot home> <20160320170528 dot GO3635 at var dot home>
On Sun, Mar 20, 2016 at 10:05 AM, Samuel Thibault
<samuel.thibault@ens-lyon.org> wrote:
> Samuel Thibault, on Sun 20 Mar 2016 17:42:14 +0100, wrote:
>> AIUI, this is because there are weak definitions in
>> sysdeps/mach/hurd/dl-sysdep.c which need to be preempted by the libc
>> one, right? Is __access perhaps missing in the list? There is a weak
>> definition for __access in dl-sysdep.c.
>>
>> Also, 6d56699d7e808419ccf244150ecba122156932ba ('Mark internal fcntl
>> functions hidden') made __open hidden, but dl-sysdep.c has a weak
>> definition, which AIUI needs to be preempted too, so that this commit
>> should be split to a dl-fcntl.h that sysdepas/mach/hurd/ can provide,
>> right?
>>
>> (Just asking the question, I'll handle commiting etc.)
It is not about weak nor not. By default, there is no difference
between weak defined symbol and non-weak defined symbols
at run-time.
The question should be asked is if __access in dl-sysdep.c can
be used in ld.so after bootstrap on Hurd.
> Just to explicit things, attached is the fix I'm considering.
>
> Samuel
--
H.J.