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

Re: JFFS2 endianness.



ugh. The amount of folk with removable jffs2 media at present is trivial
compared to the amount that will have it.

Can we pick the right design and not break it cause a few people will
have to change?

I suggested this when the last jffs2 incompatible change happened. I
wish I had just taken the time to create a patch then.

Here's one vote for a small clean BE-only implementation.

I do not look forward to the "well which device did you format that on?
Perhaps it's a bug in the [LB]E swapping code?"

Jim Gettys wrote:
> 
> I think Nico has it exactly right:
> 
> The byte swapping gets lost on modern processors in the initial cache
> miss to load the data (if the code is good).
> 
> Worry about installed base (of removable media), and use that as your guide.
> 
> The other attitude is what we did in X, where the core protocol
> does "server makes it right", and byte swaps appropriately if needed.
> This isn't much of the overall code, if done well, and guarantees
> no overhead in the common "on machine" case.  It just isn't very hard
> at all... :-).

Mike Sundius wrote:
> that sounds like the best compromise. that way you can maintain
> compatibility w/ existing filesystems.....

-- 
Tim Riker - http://rikers.org/ - short SIGs! <g>
All I need to know I could have learned in Kindergarten
... if I'd just been paying attention.

To unsubscribe from this list: send the line "unsubscribe jffs-dev" in
the body of a message to majordomo@xxxxxxx.com