ChanServ changed the topic of #linux-rockchip to: Rockchip development discussion | Wiki at http://linux-rockchip.info | Logs at http://irclog.whitequark.org/linux-rockchip | ML at http://groups.google.com/group/linux-rockchip
<edolnx> which is surprising to me. I need to solder my connectors to the board to get a good stable link, let me do that and post a log.
kapouer has quit [Quit: kapouer]
<edolnx> Here is the serial port output during poweron http://edolnx-public.objects.dreamhost.com/CX-R8/minicom-log0001.txt
<edolnx> hehe "Press any key to abort autoboot" seems to be a lie...
<karlp> any key might be a specific key but wasn't changed :)
<edolnx> indeed :D
<edolnx> Now the question is, which one?
<edolnx> Since the line is "Hit any key to stop autoboot: 0" that could mean it's disabled too
<edolnx> Oh, hello there. When you boot into Android recovery, and the "reboot into bootloader" the bootloader says "reboot to rockusb." which _may_ mean it's trying to boot to a USB device. Now I just need to build an image to test that
naobsd has joined #linux-rockchip
<naobsd> edolnx: no need to investigate/assume about bootloader so much
<naobsd> the reason count down start at 0 is because it's defined and compiled so. you can change it by recompiling
<naobsd> and you'll get not so much useful u-boot prompt
<naobsd> you can use rockusb protocol with rkflashtool
<naobsd> you can enter rockusb mode by "reboot bootloader" command from (serial console|adb) shell on (Android|recovery)
<naobsd> or if your box has a button which is used for firmware update, then you can press&hold it while power on. it's rockusb mode.
<naobsd> basically rockusb mode is for flashing images, not for booting
<naobsd> if you have any experience about (more useful) bootloader with non-Rockchip SoCs, I recommend to forgot about it.
<edolnx> naobsd, all good information to know :D
<edolnx> naobsd, So, the only issue is I'm not sure where/if there is an OTG port. I'm thinking the one wired directly to the SoC may be, but the only USB A-A cable I have didn't seem to work...
<naobsd> I tought USB A-A works... is there any factory/update firmware image and tool which requires OTG for your box?
<naobsd> well, one point, the key you pressed and got recovery mode is for rockusb mode. when entering rockusb mode, if there is no OTG connection, bootloader try to boot recovery (see "Usb Timeout, Return for boot recovery!" in your console output)
<naobsd> are you sure OTG is connected to PC while pressing that key and powering on?
<naobsd> edolnx: OTG is nearest SD slot, right?
<naobsd> mm, it might be wrong, but anyway,
<naobsd> usb hub is connected to ehci, and no usb hub for otg
<naobsd> OTG port should be run as HOST while running Android. you have to change setting to connect PC with Android
<naobsd> oh, is eMMC TOSHIBA? I'm curious about performance ;)
<edolnx> Yes, the one I'm guessing is OTG is the one next to the SD slot
<edolnx> Let me plug in my cable and try again
<edolnx> oh hey, that is working!
<naobsd> nice :)
<naobsd> it might be nice if I2C/SPI holes can be configured as another UART
<edolnx> OK, now I just need an image and the tool to send it
<naobsd> is wifi chip AP6330?
<edolnx> How would you like me to test the flash performance?
<edolnx> Yes, WiFi and BT is from the AP6330
<edolnx> The antenna is only 2.4GHz is looks like, which is disappointing but I'm using wired ethernet so no problem
<naobsd> then your box should be Sunchip CX-R8
<naobsd> s/box/board/
<edolnx> probably, I'm sure they are all the same board just different logos on the cases :D
cnxsoft has joined #linux-rockchip
<naobsd> there is some possibility that some parts e.g. wifi/emmc/etc can be different
<naobsd> for customized products
<edolnx> The Android dmesg shows at least 3 serial ports, I'm not sure where they all go
<naobsd> 3 holes near wifi/bt antenna are probably 2 gpio and 1 gnd...it looks like for dual color LED
<edolnx> alright, so I need rkflashtool and some form of image. Let me checkout and build those
naobsd has quit [Quit: naobsd]
<edolnx> Ran Androbench and Geekbench on the Android install while waiting for the firmware to download. Results are posted here: http://edolnx-public.objects.dreamhost.com/CX-R8/index.html
<edolnx> (couldn't figure out how to get the raw results files, sorry)
rk___ has joined #linux-rockchip
<rk___> ls
rk___ has left #linux-rockchip [#linux-rockchip]
cnxsoft has quit [Remote host closed the connection]
cnxsoft has joined #linux-rockchip
kapouer has joined #linux-rockchip
nighty^ has joined #linux-rockchip
markm has quit [Ping timeout: 272 seconds]
markm has joined #linux-rockchip
cnxsoft has quit [Ping timeout: 246 seconds]
cnxsoft has joined #linux-rockchip
<mmind00> pulser: btw. the missing battery data for me, was just caused by it being module and me being dumb and not rebuilding them
<mmind00> pulser: when simply switching them to =y, I again get battery-infos on my jerry (upower -d)
<pulser> OK interesting mmind00 - I will quickly try again. I don't think it was modular when I checked
<pulser> (I use a shell script to clear the old modules, make a new folder, install them to it, then I punt them to the SD)
AstralixNB has joined #linux-rockchip
<pulser> mmind00: OK, so I tested it again on veyron-minnie - built from your tree, but made the sbs config line "y" rather than "m" to get it inside the kernel. When I do "upower -d", the only device I get is gpio-charger ahd DisplayDevice - daemon says it is not on battery
<mmind00> pulser: ah ... you're on a Minnie, sorry I forgot that detail
<pulser> oh no worries - do I need to tweak something in my config then? I guess different battery driver
<mmind00> pulser: minnie has a bq27500 battery-thingy connected to the i2c provided by cros_ec
<pulser> right - that makes sense then why it caused an issue
<pulser> let me force enable that :)
<mmind00> it should be ... at least I made sure in kconfig that it was enabled
<pulser> OK - it was set to "m" in your latest kernel config
<pulser> I will rebuild with it as "y", as I do not see a module made for it in my modules directory
<mmind00> pulser: yeah, I fixed that 10 minutes ago :-)
<pulser> (I assume the name would include "bq" (grep -i bq failed)
<mmind00> just need to push
<pulser> ah cool, thanks - will force enable and try that
<pulser> I'm sure I did that already, but oh well
<pulser> mmind00: battery confirmed working on 178053b (HEAD)
<pulser> thanks :)
<pulser> I shall continue to look into the mali stuff, although I am not likely to find anything (I suspect), as it's a slight blind-hunt for me :)
<mmind00> nice :-) ... I'll look too hopefully later today
<pulser> yeah - I posted the source of the errors on the dmesg, but it's going backward to the root cause that's harder
<pulser> I might find it if it's something directly linked to the error; if it's more subtle I'll not find it
cristian_c has joined #linux-rockchip
cnxsoft has quit [Quit: cnxsoft]
bamvor has quit [Ping timeout: 264 seconds]
bamvor has joined #linux-rockchip
naobsd has joined #linux-rockchip
AstralixNB has quit [Remote host closed the connection]
kapouer has quit [Quit: kapouer]
markm has quit [Ping timeout: 260 seconds]
markm has joined #linux-rockchip
ssvb has joined #linux-rockchip
nighty^ has quit [Ping timeout: 244 seconds]
jas-hacks has joined #linux-rockchip
markm_ has joined #linux-rockchip
JohnDoe_71Rus has joined #linux-rockchip
nighty^ has joined #linux-rockchip
markm has quit [Ping timeout: 252 seconds]
kapouer has joined #linux-rockchip
chris2 has quit [Ping timeout: 252 seconds]
chris2 has joined #linux-rockchip
chris2 has quit [Ping timeout: 264 seconds]
chris2 has joined #linux-rockchip
JohnDoe_71Rus has quit [Quit: KVIrc 4.2.0 Equilibrium http://www.kvirc.net/]
cosm has joined #linux-rockchip
<edolnx> OK, I think I have all the parts, but I'm having a heck of a time figuring out how to build an image for rkflashtool. Any pointers?
chris2 has quit [Ping timeout: 264 seconds]
<cristian_c> edolnx: what?
<edolnx> I'm trying to build a firmware image for an RK3368 board to run debian. I have the loader, and the kernel, and the initrd. Just not understanding how to put all that together.
<cristian_c> edolnx: I suppose micro-sd card
<cristian_c> edolnx: maybe you have got too 'parameter' file
<cristian_c> *parameter file too
<edolnx> I have those
<cristian_c> you should transfer them into micro-sd, I suppose
<cristian_c> (in linux environment, by dd, for example)
<cristian_c> edolnx: you also could search in radxa rock documentation
chris2 has joined #linux-rockchip
libv has quit [Ping timeout: 260 seconds]
<cristian_c> ah, sorry, true, I forgot that repack with crc is needed in some cases
<cristian_c> rkcrc, f0r example
<cristian_c> -k
<cristian_c> either mkbootimg
libv has joined #linux-rockchip
mrueg has quit [Remote host closed the connection]
mrueg has joined #linux-rockchip
Xeta has joined #linux-rockchip
jas-hacks has left #linux-rockchip [#linux-rockchip]
<Xeta> Hi! Anyone where have experience installing linux on a MK 902 II box? Is it possible to write an OS image to the boxes internal harddrive from my PC?
<edolnx> Right, I'm just not understanding how the images are put together. It looks like the firmware image as another image in it with the kernel and ramdisk. Ideally I'd like to put a second stage bootloader in there so I can change out kernels more easily (read from USB,etc) but right now just putting in a kernel and initrd would be a great start
Inode has quit [Quit: ]
<edolnx> It looks like rockchip-mkbootimage/img_maker makes the "firmware.img" and then rockchip-mkbootimage/mkbootimage makes the boot payload that would be inside the "update.img" inside the "firmware.img"?
<edolnx> There also appear to be specific name requirements, but I'm not sure
cosm has quit [Quit: Leaving]
nighty^ has quit [Ping timeout: 250 seconds]
gb_master has joined #linux-rockchip
Xeta has quit [Ping timeout: 246 seconds]
gb_master has quit [Quit: Leaving]
naobsd has quit [Quit: naobsd]
cristian_c has quit [Quit: Bye]