On Thu, Nov 27, 2014 at 11:51:08AM -0500, Rob Clark wrote: > On Thu, Nov 27, 2014 at 11:39 AM, Thierry Reding > <thierry.red...@gmail.com> wrote: > > Tegra K1 and later use a GPU that can be driven by the Nouveau driver. > > But the GPU is a pure render node and has no display engine, hence the > > scanout needs to happen on the Tegra display hardware. The GPU and the > > display engine each have a separate DRM device node exposed by the > > kernel. > > > > To make the setup appear as a single device, this driver instantiates > > a Nouveau screen with each instance of a Tegra screen and forwards GPU > > requests to the Nouveau screen. For purposes of scanout it will import > > buffers created on the GPU into the display driver. Handles that > > userspace requests are those of the display driver so that they can be > > used to create framebuffers. > > > > This has been tested with some GBM test programs, as well as kmscube and > > weston. All of those run without modifications, but I'm sure there is a > > lot that can be improved. > > > > TODO: > > - use Nouveau headers to get at the prototype for creating a screen > > - implement enough support to seamlessly integrate with X > > - refactor some of the code to be reusable by other drivers > > I haven't looked too carefully at the implementation yet, but couldn't > you just put in src/gallium/drivers/shim ? I guess you'd just want a > small if/else ladder where you create the *actual* screen, to create a > nouveau screen for tegra, an etnaviv screen for imx, armada, etc..?
That's not how Mesa's loader works. Typically only a file descriptor is passed in and helpers determine a name for the driver to load, then the DRI code will load <name>_dri.so. I don't see how this could be made to work with a single Gallium driver. There's also the fact that these drivers need to become very HW-specific at some point, so sharing a single driver will likely just cause an explosion of conditionals to special-case for each and everyone of the drivers. Lastly, Tegra is also somewhat special in this regard because earlier generations did have a GPU that's controlled via the same DRM device as the display part. If ever a Mesa driver shows up for that older GPU we are going to have a conflict between the shim and the Tegra driver that isn't going to be easy to untangle. With a separate driver we can use SoC-specific knowledge to determine whether the driver is running on an old SoC generation or Tegra K1 and later. Thierry
pgprALhlTTEI0.pgp
Description: PGP signature
_______________________________________________ Nouveau mailing list Nouveau@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/nouveau
