[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: fix for 30094 breaks HURD compilation again
From: |
Yavor Doganov |
Subject: |
Re: fix for 30094 breaks HURD compilation again |
Date: |
Fri, 11 Jun 2010 15:02:54 +0300 |
User-agent: |
Wanderlust/2.15.9 (Almost Unreal) SEMI/1.14.6 (Maruoka) FLIM/1.14.9 (Gojō) APEL/10.7 Emacs/23.1 (i486-pc-linux-gnu) MULE/6.0 (HANACHIRUSATO) |
Yavor Doganov wrote:
> Riccardo Mottola wrote:
> > now using __GLIBC__ causes troubles on Hurd with GCC (and possibly
> > other platforms which I don't remember),
>
> The Debian archive is full of code like this, so it's strange that it
> causes trouble. What is the specific error?
Without having access to GNU/Hurd, I think I realize why the build
fails with my initial patch. On GNU/Hurd, __GLIBC__ is defined in
<features.h>, as it is on GNU/Linux. On GNU/kFreeBSD, it is a
preprocessor macro.
By contrast, __linux__ is always guaranteed to be defined on GNU/Linux
(no matter what headers you include); likewise for __GNU__ on
GNU/Hurd.
So what I said in the bug log is wrong, and I stand corrected:
__GLIBC__ is always guaranteed to be defined on GNU/kFreeBSD, and on
all GNU platforms iff <features.h> is included.
Defining _GNU_SOURCE should address this issue, I hope.