11:02 <micken> ayaka: I have checked the DATA0 / 1 registers and they has the last hi /low values in them,

11:02 <micken> ayaka: after i2c write that is.

11:03 <micken> ayaka: so my only guess is that rk3288 needs som specific config for the phy i2c interface

11:03 <micken> ayaka: please advice

11:25 <ayaka> micken, I will check

11:29 <ayaka> micken, except those i2c register in hdmi phy, I didn't see the other register would effect it

11:30 <micken> ok

11:30 <micken> ayaka: I wonder what happends the error bit in the stat0 register is set after every write

11:31 <micken> the i2c write code is exactly how it is in u-boot

11:32 <micken> and in linux

11:32 <micken> the linux driver is derived from i.mx6 , as is mine

11:32 <ayaka> micken, it is not set by the driver but the device itself

11:33 <micken> ayaka: that was my other guess

11:34 <ayaka> Wait for the ii2cmasterdone interrupt in the ih_i2cm_stat0 register. before reading the result

11:39 <micken> isn't it phyi2cm_stat0?

11:40 <micken> i2cmphy

11:44 <patrik> I'm grabbing the rockchip kernel as instructed on http://rockchip.wikidot.com/linux-user-guide but the kernel build has been broken for quite some time

11:45 <patrik> is there another repo where kernel development is taking place or am I missing something?

11:49 <mmind00> patrik: depends on your platform ... for a lot of things the mainline kernel also is an option

11:51 <patrik> mmind00, I have the Sapphire + Excavator RK3399 board and is trying to build a kernel with working network and usb support. Couldn't get mainline to work so figured I'd test the rockchip kernel

11:52 <patrik> do you know if mainline is supposed to work on this board with a correct dtb?

11:52 <mmind00> patrik: I'm running mainline on my rk3399-firefly with usb support and still somewhat flaky network support

11:54 <mmind00> but that is likely a firefly-specific issue ... so yes, in general it should work

11:55 <mmind00> also mainline works just fine with my Kevin (the Chromebook Plus) as well

11:56 <patrik> ok nice, which dtb are you using for that?

11:57 <mmind00> patrik: Kevin is in the 4.12-rc kernels and the firefly is in my devicetree branch and thus part of linux-next

11:58 <patrik> Is that the rk3288 or rk3399 firefly?

11:59 <mmind00> patrik: the rk3399-firefly of course :-)

11:59 <patrik> just checking :)

11:59 <mmind00> https://git.kernel.org/pub/scm/linux/kernel/git/next/linux-next.git/tree/arch/arm64/boot/dts/rockchip/rk3399-firefly.dts

12:02 <patrik> thanks, now I should be able to put something together

12:03 <mmind00> patrik: with the default arm64 "defconfig" you should at least see some output [maybe needing an earlycon=...] commandline param

12:04 <patrik> serial or actual display?

12:05 <mmind00> patrik: serial ... actual display is on the todo list ... driver parts have landed, I just need to piece together devicetree stuff to hopefully make it work

12:07 <patrik> mmind00, ok, I can probably help out since that is what I'm trying to get working

12:07 <mmind00> patrik: also please check the serial baudrate (1500000 seems to be the default on most Rockchip boards)

12:08 <patrik> yep that is required on my board

12:13 <micken> ayaka: the error bit is set in both i2cmphy_stat0 and i2cm_stat0

12:13 <micken> ayaka: uboot check i2cmphy

12:25 <micken> i2cm_stat0 is for ddc, I wonder what might triggered that interrupt

13:00 <Gekko> How would one debug the initial kernel init sequence for a self-compiled kernel flashed onto a rk3288 device? I emailed the company that manufactures the device and got the kernel source for them, with two whole lines of instructions on how to build the kernel. I'm using rkflashkit to flash the "@boot" partition after building the kernel and using mkbootimg on arch/arm/boot/zImage with an

13:00 <Gekko> initrd.img they suppied

13:00 <Gekko> If I flash the original boot image back in the device, the device works fine. With the kernel I simply get nothing out of the device on my monitor, although the device's camera flashes a bit when the power is plugged in

13:00 <Gekko> with the kernel I built*

13:03 <Gekko> I extracted the original boot image from the device and went through some data in it. It would appear at least the initrd.img is identical to what I'm using with mkbootimg, with only the kernel inside it being different

13:03 <Gekko> according to "rkflashkit part" the device has @boot, @linuxroot and @backup partitions

13:07 <Gekko> I was considering using an USB-TTL chip to see if I can get some kind of serial connection, but I can't locate any pins for that on the board

13:20 <Gekko> this is essentially what I do to flash the bugger http://plantmonster.net/koodailut/arm/persee.txt