Hi Marcel,
On 16 January 2016 at 17:50, Marcel Ziswiler marcel@ziswiler.com wrote:
Hi Simon
On Sat, 2016-01-16 at 10:14 -0700, Simon Glass wrote:
<snip>
I wanted to give that patch set a shot but could not figure out against what to get it cleanly applied. Could you reveal its baseline or better yet a git branch ready for testing?
Yes, this plus the next series is at u-boot-dm/rkf-working. Sorry I didn't point that out.
OK, that looks much better now. However I noticed a few things:
Apalis T30 has Ethernet broken but that's the same on master. Good news your u-boot-dm/eth-working branch actually makes it work againTM (;-p). So I assume your just about to post some fixes in that regard.
Yes I sent those.
Colibri T20 is completely broken. Bisecting it showed a simple configuration issue which I will report ASAP on resp. patch.
OK will fix, thanks.
Unfortunately with that Colibri T20 still does not show anything on the screen anymore but rather reports the following:
stdio_add_devices: Video device failed (ret=-22)
Are you familiar with what could be the issue there?
-EINVAL. Most likely this is a failure to read from the device tree.
There is a possibility that dc@54240000 needs to be disabled in the device tree (as seaboard). But that doesn't explain why nothing shows up on the screen. Can you check where the error is generated? Perhaps add #define DEBUG at the top of drivers/video/tegra.c?
But do check that your stdout env variable is correct. It should have vidconsole now, not lcd.
Also try 'dm tree' to see what devices probed successfully.
Colibri T20/T30 have Ethernet broken as well but that's the same on master. Unfortunately I don't think u-boot-dm/eth-working does anything on that front I guess. I will report this ASAP in a separate email.
Thanks Simon.
Thanks for helping test this.
Regards, Simon