> On Tue, 2002-04-30 at 04:46, Dirk Eddelbuettel wrote: > > > > A little while back, I applied a patch to afio by our own Norbert Veber, and > > blessed upstream, to enable afio to work on archives >2gb. This was #110853. > > It added some sort of 64bit arithmetic flag > > > > We now have #144986 -- it appears that powerpc is confused by this. Could > > some kind soul with powerpc knowledge shed some light on this. Does the > > 32/64 > > bit nature of powerpc play into this? > > Let's start with the usual suspects, in particular endianness. Are all > the types correct? If 64 bit variables are treated as 32 bit, that could > work on little endian systems but not big endian...
Excellent point. Any volunteers willing to do some tests or study afio's source? Dieter: Do you want to try? I'm sure Michael would gladly assist with a few emails. Dirk (tied up this week) -- According to the latest figures, 43% of all signatures are totally worthless. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
