svn commit: r40970 - head/en_US.ISO8859-1/books/porters-handbook

[Alex Dupre]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Pietro Cerutti ha scritto:
> On 2013-Feb-14, 15:41, Alex Dupre wrote:
>> Pietro Cerutti ha scritto:
>>> If a port insists on linking <literal>-lpthread</literal> +
>>> directly, patch it to use <literal>-pthread</literal>.</para>
>> 
>> Why bother?
> 
> What I have seen in some cases is that the flags are used during 
> compilation too. In this case, using -lpthread issues a warning, 
> while using -pthread does not.

If such patching is done only to silence a (not so common) warning I'd
say that it doesn't worth. If such patching could also require the
following patch:

> <para>If building the port errors out with +	<literal>unrecognized
> option '-pthread'</literal>, +	it may be desirable to use
> <command>cc</command> as linker by +	setting
> <makevar>CONFIGURE_ENV</makevar> to +	<literal>LD=${CC}</literal>.

I'd say we should remove the suggestion entirely.

It seems to me we are creating a problem ourself: a software is
designed to use 'ld' directly with -lpthread, we are suggesting to
modify it with -pthread and then modify the linking because we broke it.

- -- 
Alex Dupre
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2.0.19 (FreeBSD)
Comment: Using GnuPG with SeaMonkey - http://www.enigmail.net/

iEYEARECAAYFAlEc/EUACgkQgRXp2M5fVU3LmgCfXvPhyF46E5rIUcaP3vRf1koM
SqMAn22L9BrMNQi83n3P29m7ocGtDJUw
=EIUl
-----END PGP SIGNATURE-----