Re: adjust prefetch in free_one_pgd()

[David Woodhouse <dwmw2@infradead.org>]

torvalds@transmeta.com said:
>  My personal opinion is that if a prefetch has semantic meanings
> outside the "speed up subsequent accesses", it should not be exposed
> to the rest of the kernel (it might still be useful inside
> architecture-specific routines like optimized memcpy etc). 

Prefetch generally means 'bring it into the cache'. On architectures with
non-cache-coherent DMA, doing a prefetch on wild address which happens to be
a DMA buffer for which we've just said 'drop it from the cache' is generally
a bad thing.

Not that I'm necessarily disagreeing with you -- but can you confirm that 
you are including all architectures with non-cache-coherent DMA in the 
'broken hardware' category below, or point out what I'm missing?

> I'd rather speed up non-broken machines and let the broken hardware
> hopefully slowly die away.


--
dwmw2