Re: 9.99.38: pinebook11 drm Cannot find any crtc or sizes

[Tom Ivar Helbekkmo <tih%hamartun.priv.no@localhost>]

I wrote:

> A full build from December 15th works fine for me, but updating the
> kernel (only) to January 15th causes X to fail (showing a black screen
> with a white rectangle (cursor?) in the top left corner).

Had another attempt at this today.  Still running 9.99.32, from December
15th, and it's working fine.  Attempting to boot a fresh kernel from
today (9.99.42), I get the above, and the only seemingly relevant
difference in dmesg output during boot is this:

-sunxifb0: framebuffer at 0xffffffc043200000, size 1920x1080, depth 32, stride 7680
 +sunxifb0: framebuffer at 0xffffffc043400000, size 1920x1080, depth 32, stride 7680
 
The Xorg.0.log shows pretty much the same stuff, too, except that the
bit where it logs the physical details of the actual display hardware is
missing when running the new kernel.  That's the bit that begins:

-[   163.702] (II) modeset(0): Output eDP-1 has no monitor section
-[   163.702] (II) modeset(0): Output HDMI-1 has no monitor section
-[   163.807] (II) modeset(0): EDID for output eDP-1
-[   163.809] (II) modeset(0): Manufacturer: NCP  Model: 0  Serial#: 0
-[   163.809] (II) modeset(0): Year: 2016  Week: 0
-[   163.809] (II) modeset(0): EDID Version: 1.4

With the new kernel, the log file just stops at the point where the old
one has the above (plus a lot more detail).

While X is running, the new kernel keeps spitting out these,
approximately once per second:

[  1414.004149] sunxitwi0: autoconfiguration error: gttwsi_wait(): send-start timeout, control=0xe0, error=35
[  1414.014149] sunxitwi0: autoconfiguration error: send STOP failed, status=0xf9

Stopping X takes a long time, and the above error logging then ceases.

-tih
-- 
Most people who graduate with CS degrees don't understand the significance
of Lisp.  Lisp is the most important idea in computer science.  --Alan Kay