Re: [patch 2/2] backport of sba sg list management to ccio-dma

On Thu, Oct 25, 2007 at 09:13:18AM +0200, Joel Soete wrote:
...
> I attached 2 test files (GetOrder-o.c using the original kernel get_iovp_order
>  routine and GetOrder-f.c using supposed faster release) and here are some
> results:
...
> I didn't check in more details the difference of results between not-optimized
> and optimized (-O2) compile but it's clear it's faster 

Thanks for running those...I didn't expect it to be 40x faster (in the
first pair of runs).

> (why not push this in upstream? [I mean ./include/asm-generic/page.h])

Erm, push what?
The code I was referring to is parisc assembly.
That doesn't belong in asm-generic.
Or did you somehow overlook that?

...
> > I'm not sure we have to handle that case...or if we do, make sure it's
> handled correctly in get_iovp_order().
> >
> I just hope so that linux check to never call get_order(0) 

"hope?" Bad idea. :)
Please add an ASSERT or verify ccio can never invokes get_order(0).

thanks,
grant