On 01/24/2014 01:39 AM, Timothy Potter wrote:
Hi Kyosti, Thanks the pointers. I tried adding the changes for dbgp from the bug20-2.6.35-linaro to the 2.6.32 kernel I'm running on the phone. You can see the results here:
https://github.com/DengueTim/jordan-kernel/commits/AeroKernel
You still need the two patches from http://www.coreboot.org/EHCI_Gadget_Debug
Looks like you reverted them for this attempt.
On startup, in /proc/kmsg I see:
<7>[ 2.010284] dbgp gadget: setup: desc device <7>[ 2.010314] dbgp gadget: setup complete: 0, 8/8 <7>[ 2.010467] dbgp gadget: setup: desc device <7>[ 2.010467] dbgp gadget: setup complete: 0, 18/18 <7>[ 2.010589] dbgp gadget: setup: failure req 6 v 200 <7>[ 2.010681] dbgp gadget: setup: failure req 6 v 200 <7>[ 2.010772] dbgp gadget: setup: failure req 6 v 200 <7>[ 2.016204] dbgp gadget: setup: failure req 6 v 3ee
Ignore those, g_dbgp driver implements only a few USB requests required for this special debug gadget purpose.
I've attached the full log in case there's more clues as to why it's failing. Any more suggestions of what I should be looking at?
I requested "lsusb" and "sudo lsusb -v" and dmesg from the other end of the communication too. If you see a "Debug descriptor" listed there, you can connect and reboot the chromebook with usbdebug enabled and collect new dmesg from that android side.
Also run "cat /dev/ttyGS0" on that android before you restart your chromebook.
Kyösti