[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: reiserfs empirical study (very long)



On Wed, May 30, 2001 at 02:45:41PM +0200, Christoph Ewering wrote:
> 
> I´m from the mac-side of live, i do not like options, i like smart
> software that determinds the fs by its own.
> I like options to force a program to do what I want, maybe doing stupid things.
> So i prefer to let the kernel look at the partition-table (or another

the partition table is again the WRONG place for this.  partition
table data in no way reflects whats on that partition, i can make a
Apple_Free partition be swap for all the kernel cares.  

guess what partition type we usually use for ext2 on macs?
Apple_UNIX_SVR2, now guess what type we use for reiserfs? and for xfs?
ill give you a hint: Apple_UNIX_SVR2.  (though we should not use that
type at all, Apple_* is reserved by apple, but thats another issue). 

partition types have thier place, but identifying filesystems isn't it
IMO.  

the proper way for the kernel to deal with such things is by
inspecting what is really on the partition, its not that difficult to
figure out a filesystem type by magic numbers in the filesystem.  it
already does this for mounting / do you think that the partition type
has anything at all to do with what fs the kernel uses for mounting / ?  
it doesn't.  

> place on the disk) and let the kernel decide to use little or big-endian fs-drivers.
> So the kernel includes two fs-drivers for performance reasons, one for
> little- one for big-endian fs and which one to use is decided at boot-time.
> With this is it possible to have different endians on different partions
> on the same disk.
> (Does not makes a lot of sense to mix endianess, but it would be possible)
> For new filesystems the prefered endianess is determined by the
> architecture that runs mkfs. (Can be overwritten by options)

and this could all be done, but it involves adding all kinds of new
code to all filesystem drivers you volunteering to maintain it all? 

> Well, I do not want to change endianess on an existing filesystem
> (sounds very very dangerous to me :-))
> But it would be very nice to put a disk from a crashed system into any
> linux-box and repair the filesystem and let it run there until the
> original system is replaced by a new one.

which you can do now. 

> That´s what I mean, a lot of work at the big-endian-side to support
> software that was written with little-endian machines in mind.

thats how it goes, there are many more little endian machines out
there then big endian (because of x86) so were stuck with it.  

> But I think it will never happen, it is just a thought.

i don't either, unless your volunteering to do all the work and
maintenence now and forever more.  most of the kernel hackers use x86
so guess what endianess they give a damn about...

-- 
Ethan Benson
http://www.alaska.net/~erbenson/

Attachment: pgpJEtxfCAnHn.pgp
Description: PGP signature


Reply to: