[RFC PATCH] drm/ttm: force cached mappings for system RAM on ARM

Benjamin Herrenschmidt benh at kernel.crashing.org
Thu Jan 17 05:59:53 UTC 2019


On Wed, 2019-01-16 at 07:35 +0000, Koenig, Christian wrote:
> No, but you answer the wrong question.
> 
> See we don't want to have different mappings of cached and non-cached on 
> the CPU, but rather want to know if a snooped DMA from the PCIe counts 
> as cached access as well.
> 
> As far as I know on x86 it doesn't, so when you have an un-cached page 
> you can still access it with a snooping DMA read/write operation and 
> don't cause trouble.

Hrm... well, if you map it uncached on the CPU on powerpc, a snoop DMA
will work fine too, it won't hit any cache. The only problem I'm aware
of is a core (or CAPI device) emiting non-cached load/stores colliding
with a cache snooper.

> > The old hack of using non-cached mapping to avoid snoop cost in AGP and
> > others is just that ... an ugly and horrible hacks that should have
> > never eventuated, when the search for performance pushes HW people into
> > utter insanity :)
> 
> Well I agree that un-cached system memory makes things much more 
> complicated for a questionable gain.
> 
> But fact is we now have to deal with the mess, so no point in 
> complaining about it to much :)

I wish we could just sent the HW designers home and tell them we won't
support that crap... oh well.

Ben.

> Cheers,
> Christian.
> 
> > Cheers,
> > Ben.
> > 
> > 



More information about the dri-devel mailing list