On Sat, Apr 13, 2013 at 5:45 PM, Thierry Reding thierry.reding@avionic-design.de wrote:
On Sat, Apr 13, 2013 at 08:54:22AM -0400, Rob Clark wrote:
On Mon, Mar 4, 2013 at 1:46 PM, Tony Lindgren tony@atomide.com wrote:
drivers/gpu/drm/tilcdc/tilcdc_slave.o:(.data+0x54): multiple definition of `__mod_of_device_table' drivers/gpu/drm/tilcdc/tilcdc_tfp410.o:(.data+0x54): first defined here drivers/gpu/drm/tilcdc/tilcdc_panel.o:(.data+0x54): multiple definition of `__mod_of_device_table' drivers/gpu/drm/tilcdc/tilcdc_tfp410.o:(.data+0x54): first defined here drivers/gpu/drm/tilcdc/tilcdc_drv.o:(.data+0x184): multiple definition of `__mod_of_device_table' drivers/gpu/drm/tilcdc/tilcdc_tfp410.o:(.data+0x54): first defined here
Rob, I assume you'll do a patch for this one?
oh, I apologize for the late reply, I didn't see this email...
There is a patch that we can merge to make tilcdc bool instead of tristate[1], which I suppose is ok for a temporary fix. Although I'm all-ears if someone has a better idea about how to fix this. The problem is that we have multiple sub-devices for different possible panel drivers, so that depending on DT tables, the correct ones get loaded for the hw. But they are all built into a single module. Splitting them into multiple modules will be problematic, as panel drivers which are present really need to get probed before the toplevel drm device..
You could look at the Tegra driver. I had to solve a similar problem there. What I did is basically parse the DT in the host1x driver and add all device nodes which are required by DRM to a list. Later when the individual devices are probed they are removed from that list, so that when the list becomes empty we are sure that all required devices are there and only then call the drm_platform_init() function.
thx, ok, I'll have a look at this
This fits very well with how Tegra hardware is designed because host1x is the parent for all DRM subdevices (DC, RGB/LVDS, HDMI, ...). So it is probed before any of its children and it can easily parse the DT upfront and initialize the list of required devices.
I suppose in theory it is possible to make drm cope better with dynamically loaded outputs, but I'm not sure that there is any way to do this without breaking userspace which expects that all of the connectors/encoders are present once the drm device is loaded.
I had been thinking about this on and off for a while, but I haven't come up with anything concrete. Ideally we could just have some kind of event that userspace would listen for, so that new outputs can be dynamically added and userspace informed about them. Last time I checked most of the helpers assumed that the complete output configuration is known when the DRM device is registered, so some major rework will be required to efficiently make use of such dynamicity.
I'm less worried about the kernel re-work.. more worried about the fact that we have no way to know whether userspace knows to listen for this new event. So anything down this path could, I think, be considered as breaking userspace.
I think in the end, we need some way to have sort of "dummy" connectors for output drivers which might or might not be probed, so that from userspace perspective, non-present panels appear as displays that are not plugged in.
BR, -R
Thierry