begin using attribute_hidden

[Peter S. Mazinger]

On Mon, 7 Nov 2005, Joakim Tjernlund wrote:

> > -----Original Message-----
> > From: uclibc-bounces at uclibc.org 
> > [mailto:uclibc-bounces at uclibc.org] On Behalf Of Peter S. Mazinger
> > Sent: 07 November 2005 09:59
> > To: uclibc at uclibc.org
> > Subject: RFC: begin using attribute_hidden
> > 
> > Hello!
> > 
> > What about start using attribute_hidden like:
> > 
> > attribute_hidden __x () {
> > }
> > 
> > strong_alias(__x,x)
> 
> hehe, I just replied to your previous mail, suggesting this :)

seen it, thx
> > 
> > Jocke, should libc/string/powerpc/string.c bzero not be 
> > changed to this 
> > too (the others are already done)? Why a weak_alias here?
> 
> hmm, that looks like a mistake. I think I just copied this from
> somewhere.
> I wounder though why it was a weak alias from wherever I copied it from.
> Does glibc define these functions as weak aliases or is this just some
> strange leftover
> from early days?

glibc's comment on __bzero (string.h)
We define this function always since 'bzero' is sometimes needed when the 
namespace rules does not allow this (bzero is ifdef'd by __USE_BSD, but 
__bzero not)
glibc has it everywhere as weak_alias.

Could this __bzero/bzero from powerpc not being used generally maybe as 
replacement of the one in libc/string?

Peter

-- 
Peter S. Mazinger <ps dot m at gmx dot net>           ID: 0xA5F059F2
Key fingerprint = 92A4 31E1 56BC 3D5A 2D08  BB6E C389 975E A5F0 59F2