On Wed, Mar 26, 2014 at 1:24 PM, Ben Skeggs <[email protected]> wrote:
> On Mon, Mar 24, 2014 at 6:42 PM, Alexandre Courbot <[email protected]> wrote:
>> Add a GR device for GK20A based on NVE4, with the correct classes
>> definitions (GK20A's 3D class is 0xa297).
>>
>> Most of the NVE4 code can be used on GK20A, so make relevant bits of
>> NVE4 available to other chips as well.
> This will need a bit of a rebase on top of the tree I mentioned
> earlier (also queued for drm-next now), where I've further split out
> and named the various chunks of state.
Will do that.
>
> Does GK104 match entirely correctly, or just happen to work? I could
> probably hunt down the GK20A netlist images and check that actually :)
Do you mean, the init sequence? I haven't checked in detail (we are
certainly doing things differently in the non-DRM driver), but the
registers seem to match and the GPU is able to render after that. I
admit I have not looked much further for now.
The only register that does not exist on GK20A is 0x260, but when
accessing it Nouveau will be able to continue unharmed after a memory
fault.
On Thu, Apr 3, 2014 at 12:03 AM, Alexandre Courbot <[email protected]> wrote:
> On Wed, Mar 26, 2014 at 1:24 PM, Ben Skeggs <[email protected]> wrote:
>> On Mon, Mar 24, 2014 at 6:42 PM, Alexandre Courbot <[email protected]> wrote:
>>> Add a GR device for GK20A based on NVE4, with the correct classes
>>> definitions (GK20A's 3D class is 0xa297).
>>>
>>> Most of the NVE4 code can be used on GK20A, so make relevant bits of
>>> NVE4 available to other chips as well.
>> This will need a bit of a rebase on top of the tree I mentioned
>> earlier (also queued for drm-next now), where I've further split out
>> and named the various chunks of state.
>
> Will do that.
>
>>
>> Does GK104 match entirely correctly, or just happen to work? I could
>> probably hunt down the GK20A netlist images and check that actually :)
>
> Do you mean, the init sequence? I haven't checked in detail (we are
> certainly doing things differently in the non-DRM driver), but the
> registers seem to match and the GPU is able to render after that. I
> admit I have not looked much further for now.
I meant the arrays of register data, there's generally been some
differences for most major chipset bumps. Where can I find the
netlist firmware packages that go with the android driver? I can pull
all the required info out of there to check :)
>
> The only register that does not exist on GK20A is 0x260, but when
> accessing it Nouveau will be able to continue unharmed after a memory
> fault.
I have it in my queue to fix that too, the register doesn't exist on a
couple of the other chips we support too.