git: drm - Update to Linux 4.20.17

[Sergey Zigachev]

> Sorry for my late response. With Driver "intel" and i915_load=YES in
> rc.conf, I get a segfault when starting Xorg (I use Xorg from
> ravenports):
> 
> 	     70.937] (EE) Segmentation fault at address 0x0
> 	[    70.937] (EE)
> 	Fatal server error:
> 	[    70.938] (EE) Caught signal 11 (Segmentation fault). Server aborting
> 	[    70.938] (EE)
> 	[    70.938] (EE)
> 	Please consult the The X.Org Foundation support
> 		 at http://wiki.x.org
> 	 for help.
> 	[    70.938] (EE) Please also check the log file at "/var/log/Xorg.0.log" for additional information.
> 	[    70.938] (EE)
> 	[    70.942] (EE) Server terminated with error (1). Closing log file.
> 
> Driver "intel" works when I leave out i915_load.
> Driver "modesetting" works regardless of i915_load.

This is a known issue, I think there's a race between loading the kernel 
module and Xorg trying to use it.

> I am using Xorg-server from Ravenports. I think the ports there lack
> these patches.
> 
> How can I test if hardware acceleration is turned on? And will the
> "intel" driver be more performant than "modesetting"? How can I test
> this?

I'm not familiar with how Ravenports work. However, intel xorg driver 
should work with SNA hardware acceleration enabled, it just runs slower 
without the patch. Modesetting xorg driver doesn't enable acceleration 
without the patch, because previous patch force-disabled acceleration. 
I.e. it ignores xorg.conf settings.

To check if acceleration enabled, you can use `glxinfo | grep renderer`.
If hardware acceleration enabled you should see something with mesa but 
without mentioning llvmpipe.

That said, intel xorg driver is barely maintained upstream these days, 
so I think you should see better results with modesetting driver.
I didn't compare them though.

-Sergey