"James C. McPherson" <[EMAIL PROTECTED]> wrote:
> Joerg Schilling wrote:
> > "James C. McPherson" <[EMAIL PROTECTED]> wrote:
> >>The design of ZFS is endian-neutral. Ufs is very much endian-dependent
> >>and as I understand it the work involved to remedy that is more than
> >>non-trivial.
> > Making UFS dual endian would not be really hard.
> > All instances in UFS that read/write data structures would need to call
> > a function instead of doing "minor work" in place.
> > At mount time the correctness of Bill Joy's birthday needs to be verified
> > and the UFS implementation needs to decide whether these functions will
> > just do the read or whether swapping is needed. The swapping could even be
> > done
> > in a way similar to what ufsrestore does......
> > I would gues this is 1-2 weeks of work + testing.
>
> Hi Jo"rg,
> well since you have cft I expect you'll have that for me by the 19th? :)
>
> Casper Dik and Frank Hofmann might have a word or two about endian-ness
> on ufs though.
I did already talk with both of them about this.
But I am sorry, as long as Sun is unable to integrate simple and long planned
projects like integrating/replacing of star & the rmt server/client code from
star
into Solaris, it does not make sense to start new projects.....
Jörg
--
EMail:[EMAIL PROTECTED] (home) Jörg Schilling D-13353 Berlin
[EMAIL PROTECTED] (uni)
[EMAIL PROTECTED] (work) Blog: http://schily.blogspot.com/
URL: http://cdrecord.berlios.de/old/private/ ftp://ftp.berlios.de/pub/schily
_______________________________________________
opensolaris-discuss mailing list
opensolaris-discuss@opensolaris.org
