Path: csiph.com!fu-berlin.de!uni-berlin.de!individual.net!not-for-mail From: rbowman Newsgroups: alt.folklore.computers,comp.os.linux.misc Subject: Re: The joy of Linux Date: 25 Oct 2024 03:54:47 GMT Lines: 26 Message-ID: References: <8e3c519b-770e-e8e9-0d46-155863cf9e05@example.net> <885b7a57-c94b-f7a4-98ed-c3c9783fe172@example.net> Mime-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit X-Trace: individual.net m3e5cXcWdWFrO8+MHjwmfw8dcxrErFadkwy6xKPz0B/jm9skIe Cancel-Lock: sha1:dYpCSrLFQkRo/bKeROeCIykxRBE= sha256:0Zg/FVIJqqlkUldghcbXrC0JFiUpVTFkhkeQc6lFeao= User-Agent: Pan/0.149 (Bellevue; 4c157ba) Xref: csiph.com alt.folklore.computers:228149 comp.os.linux.misc:59980 On Fri, 25 Oct 2024 01:36:34 -0000 (UTC), Lawrence D'Oliveiro wrote: > On 25 Oct 2024 00:30:42 GMT, rbowman wrote: > >> Lawrence D'Oliveiro writes: >>> >>>On 24 Oct 2024 17:02:02 GMT, rbowman wrote: >>> >>>> RS6000 is big endian and the ONC-RPC XDR encoding is big endian >>>> across the network. The Linux boxes would swap the order when >>>> decoding the XDR. >>> >>>Linux runs on hardware of both kinds of endianness. This is why you >>>have functions like htons/htonl etc >> >> Yes, and they wouldn't be too helpful when dealing with XDR. That comes >> with its own set of data types and functions. > > This is the same XDR that was created by Sun, back in the day? They knew > their networking stuff. They were not stupid; they would have taken > endianness into account in their support functions; if you’re having > trouble with this, then I suspect you’re doing it wrong. At what point did I indicate I was having trouble with it?