On Wed, 2014-10-08 at 09:17 +0800, Mark yao wrote: > Hi Steven > I'm glad to see you to discuss about rk29xx. > > On 2014年10月08日 06:26, Steven Newbury wrote: > > I've just been discussing how this relates to rk29xx on #etnaviv. > > I > > looked through the patch and it's good to see it's not limited to > > supporting Mali GPUs. I see no reason this wouldn't be compatible > > with etna_viv for rk29xx (or future Rockchip designs with > > alternative > > GPUs) as long as the rest of the infrastructure to support rk29 is > > in > > place iommu (ARM SMMU?) driver, for example. > Now the drm driver only test on rk3288, and future when we add > compatible for rk29xx, we > would consider to add a non-iommu path for it. That's great! I'm working on trying to update the rk29 driver support. Getting it supported where possible with the new drivers would make it much easier, hopefully such that simply creating a new dts file would produce a working kernel for any useful hardware out there. > > IMHO it's vital to keep a logical separation between the video > > display > > hardware and the graphics processor on SoCs. > Sure, the graphics processor maybe change in different solutions, > logical separation is that we > try to keep.
Attachment:
signature.asc
Description: This is a digitally signed message part
