Clean up extern inline

Marcin Dalecki martin@dalecki.de
Sat Feb 3 00:07:00 GMT 2007


Wiadomość napisana w dniu 2007-02-03, o godz00:42, przez Andreas  
Schwab:

> Marcin Dalecki <martin@dalecki.de> writes:
>
>> Not doing so is fine with me. But *pretending* too is wrong.
>> (See: glibc/CONFORMANCE document over there).
>
> You don't know what you are talking about.  That file has nothing  
> to do
> with the implementation.

Ah I see one of the higher priests from the echelon has spoken. We  
the uninitiated
should bow in respect?

Again: What's the problem again just to define which version of GCC  
is suited to use
with which version of GLIBC and fixing your ifdef/__attribute__ hell  
in header files,
instead of pretending "compatibility and portability over the whole  
range" and
constantly winning about the improvement of C standard conformance  
support in GCC?

As a matter of fact dependance on particular tool versions is already
indirectly the STATUS QUO in glibc, due to dependencies between  
compiler version
and kernel version. (Yes I know glibc pretends to be kernel version  
independent
too... but that doesn't match reality too.)

∎ Marcin Dalecki ∎




More information about the Gcc-patches mailing list