Skip site navigation (1)Skip section navigation (2)
Date:      Sat, 22 Dec 2018 08:41:38 +0200
From:      Konstantin Belousov <kostikbel@gmail.com>
To:        Yuri Pankov <yuripv@yuripv.net>
Cc:        freebsd-hackers <freebsd-hackers@freebsd.org>
Subject:   Re: is adding new private symbols to libc frowned upon?
Message-ID:  <20181222064138.GM60291@kib.kiev.ua>
In-Reply-To: <e120678a-359b-5131-3866-219d4c3cd796@yuripv.net>
References:  <e120678a-359b-5131-3866-219d4c3cd796@yuripv.net>

next in thread | previous in thread | raw e-mail | index | archive | help
On Sat, Dec 22, 2018 at 04:38:52AM +0300, Yuri Pankov wrote:
> Essentially, I need the __collate_equiv_value symbol to be visible to
> libregex for the changes in https://reviews.freebsd.org/D18531.  Is the
> following change OK (it works, at least), or should try to avoid that?
> 
> --- a/lib/libc/locale/Symbol.map
> +++ b/lib/libc/locale/Symbol.map
> @@ -212,6 +212,7 @@ FBSD_1.3 {
>  FBSDprivate_1.0 {
>         _PathLocale;
>         __detect_path_locale;
> +       __collate_equiv_value;
>         __collate_load_error;
>         __collate_range_cmp;
>  };
> 
Then libregex must always match the installed libc.

I looked at the the libregex/Makefile and my question is, what is the
difference between exports from libc/regex vs. libregex.  Can libregex
become ELF filter for libc ?





Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20181222064138.GM60291>