Messages in this thread | | | Date | Wed, 22 Oct 2008 11:36:15 +0200 | From | Ingo Molnar <> | Subject | Re: io resources and cached mappings (was: [git pull] drm patches for 2.6.27-rc1) |
| |
* Keith Packard <keithp@keithp.com> wrote:
> On Mon, 2008-10-20 at 13:58 +0200, Ingo Molnar wrote: > > > yes but note that by caching the whole mapping on 64-bit we get > > everything we want: trivially lockless, works from any CPU, can be > > preempted at will, and there are no ugly INVLPG flushes anywhere. > > I was assuming that on 64-bit, the map would be created at driver init > time and be left in place until the driver closed; if that's what you > mean by 'caching', then yes, we should cache the map.
correct.
> > 32-bit we should handle as well but not design for it. > > As long as we get kmap_atomic-like performance, and we get to simplify > our code, I'm up for it.
okay. So ... mind sending your io_mapping patch as a generic facility? It looks all good to me in its present form, except that it should live in include/linux/io.h, not in the drivers/gpu/drm/i915/io_reserve.h file :-)
also, please send at least two patches, so that we can look at (and possibly merge) the generic facility in isolation.
Ingo
| |