#include <hallo.h>
* Bill Allombert [Tue, Nov 28 2006, 06:40:53PM]:
> On Sat, Nov 18, 2006 at 01:01:18AM +0100, Eduard Bloch wrote:
> > I could not find it here, and even if I could I doubt they would be
> > easily applicable. Instead, I took the current version in sid and
> > modified it again. The attached version is almost okay any still have
> > some little glitches:
> >
> > - it was not tested on real data, only sparse file to /dev/null
> > transfer
> > - detection of LFS support is not added, I hardcoded good values into
> > config.h.in
> > - I have used off_t as the main type everywhere. On some places even
> > for "unsigned int" where it looked like it would not hurt at the
> > first glance
>
> Hello Eduard,
> It seems the ranges are still 32bit only with your patch.
> This causes some gcc warnings.
>
> I plan to upload a NMU without you patch applied because I don't trust
> myself to deal with the LFS stuff correctly, and the time for Etch
> is short. Sorry about that.
Hm. Ranges stuff itself should be not a big problem, it is fixed like
other stuff, see attachment.
One thing I noticed when compiling on amd64 this time is that it shows
warnings about off_t not beeing a long long int expected by printf.
Which OTOH is understandable since off_t seems to be a long which is
already 64bit wide on amd64. Not sure whether this will break some stuff,
but I guess it won't.
Eduard.
--
<HE> <caro> Ganneff: passt auf, ich bin blond, habe keine ahnung von computern,
aber einen client kann ich einrichten, sogar alleine. *stolz guck*
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
