[Linux-ia64] Re: Reiserfs (was: Re: __ia64_syscall ?)

[Linux-ia64] Re: Reiserfs (was: Re: __ia64_syscall ?)

[Bill Nottingham]

Martin Wilck (Martin.Wilck@fujitsu-siemens.com) said: 
> This is the source of the endian-patch that makes ReiserFS work
> on IA64 (although originally developed for PPC).
> 
> http://penguinppc.org/~jeffm/

Why does an *endian* patch make reiserfs work on ia64, when
ia64 is little endian like x86?

Bill, genuinely curious...

Re: [Linux-ia64] Re: Reiserfs (was: Re: __ia64_syscall ?)

[Andreas Schwab]

Bill Nottingham <notting@redhat.com> writes:

|> Martin Wilck (Martin.Wilck@fujitsu-siemens.com) said: 
|> > This is the source of the endian-patch that makes ReiserFS work
|> > on IA64 (although originally developed for PPC).
|> > 
|> > http://penguinppc.org/~jeffm/
|> 
|> Why does an *endian* patch make reiserfs work on ia64, when
|> ia64 is little endian like x86?

The patch not only contains endian fixes, but the ia64 fixes were folded
into it.  It's just that the patch is now misnamed....

Andreas.

-- 
Andreas Schwab                                  "And now for something
SuSE Labs                                        completely different."
Andreas.Schwab@suse.de
SuSE GmbH, Schanzäckerstr. 10, D-90443 Nürnberg
Key fingerprint = 58CA 54C7 6D53 942B 1756  01D3 44D5 214B 8276 4ED5

Re: [Linux-ia64] Re: Reiserfs (was: Re: __ia64_syscall ?)

[Martin Wilck]

> Why does an *endian* patch make reiserfs work on ia64, when
> ia64 is little endian like x86?
>
> Bill, genuinely curious...

I don't know - it works for me, at least I have seen no data corruption
yet.

What the patch does is introduce generic disk<->cpu conversion macros
everywhere. My guess is that they not only provide correct endianness,
but also fix other machine-dependent type issues.

Try it, and call me whatever you want if it fails.

Martin

-- 
Martin Wilck     <Martin.Wilck@fujitsu-siemens.com>
FSC EP PS DS1, Paderborn      Tel. +49 5251 8 15113