RayFlower has quit [Read error: Connection reset by peer]
RayFlower has joined #linux-rockchip
RayFlower has quit [Read error: Connection reset by peer]
RayFlower has joined #linux-rockchip
lioka has joined #linux-rockchip
RayFlower has quit [Read error: Connection reset by peer]
RayFlower has joined #linux-rockchip
lioka has quit [Ping timeout: 258 seconds]
lioka has joined #linux-rockchip
naobsd has joined #linux-rockchip
levd has joined #linux-rockchip
lioka has quit [Ping timeout: 258 seconds]
levd1 has joined #linux-rockchip
levd has quit [Ping timeout: 244 seconds]
lioka has joined #linux-rockchip
lioka has quit [Ping timeout: 258 seconds]
lioka has joined #linux-rockchip
FreezingCold has quit [Ping timeout: 258 seconds]
RayFlower has quit [Read error: Connection reset by peer]
RayFlower has joined #linux-rockchip
levd has joined #linux-rockchip
levd1 has quit [Ping timeout: 255 seconds]
levd1 has joined #linux-rockchip
levd has quit [Ping timeout: 255 seconds]
cnxsoft has joined #linux-rockchip
RayFlower has quit [Ping timeout: 255 seconds]
Tony_ has joined #linux-rockchip
<Tony_>
Isaac, /proc/emmc_info is here on my device (rk3188 MID).
<Tony_>
it based the driver of eMMC. I guess. maybe it is in your other file on your board.
levd1 has quit [Remote host closed the connection]
levd has joined #linux-rockchip
neyder has joined #linux-rockchip
neyder has left #linux-rockchip [#linux-rockchip]
neyder has joined #linux-rockchip
<neyder>
hi there i'm all new in developing for this
<neyder>
so nice to meet you+
<Tony_>
neyder, nice to meet you too. :P
<neyder>
to compile images does i need 12.04 ? i hace 14.04
<neyder>
*have
<Tony_>
I think both they can works fine.
<neyder>
thanks
<neyder>
I got a Q7 device, and sd images from linuxium doesnot recognize wifi or ethernet, so i think need to compile by myself
Tony_ has quit [Remote host closed the connection]
RayFlower has joined #linux-rockchip
levd1 has joined #linux-rockchip
levd has quit [Ping timeout: 264 seconds]
Tony_ has joined #linux-rockchip
<Tony_>
Yep, maybe.
<Tony_>
naobsd, there is a question about if it's possible to enable USB debug via serial command ?
<naobsd>
Tony_: well, do you want to enable (check) adb on android firmware?
<Tony_>
Not really, I just want to type some command to enable USB DEBUG, like Settings -> enable USB debug.
<naobsd>
it sounds you want to enable adb
<naobsd>
please make your question more clear
<Tony_>
Yes, enable adb. but with command.
<naobsd>
I cannot try android now. please try "start adbd"
RayFlower has quit [Read error: Connection reset by peer]
RayFlower has joined #linux-rockchip
levd has joined #linux-rockchip
levd1 has quit [Ping timeout: 245 seconds]
levd1 has joined #linux-rockchip
levd has quit [Ping timeout: 256 seconds]
Astralix has joined #linux-rockchip
Astralix1 has quit [Ping timeout: 250 seconds]
RayFlower has quit [Read error: Connection reset by peer]
RayFlower has joined #linux-rockchip
RayFlower has quit [Read error: Connection reset by peer]
RayFlower has joined #linux-rockchip
levd has joined #linux-rockchip
levd1 has quit [Ping timeout: 264 seconds]
RayFlower has quit [Read error: Connection reset by peer]
RayFlower has joined #linux-rockchip
levd1 has joined #linux-rockchip
levd has quit [Ping timeout: 264 seconds]
GriefNorth has joined #linux-rockchip
levd has joined #linux-rockchip
levd1 has quit [Ping timeout: 255 seconds]
ganbold_ has quit [Remote host closed the connection]
<Tony_>
naobsd, thank you. but it don't work. I enable it by GUI.
RayFlower has quit [Read error: Connection reset by peer]
RayFlower has joined #linux-rockchip
levd1 has joined #linux-rockchip
levd has quit [Ping timeout: 255 seconds]
antoinemaillard has joined #linux-rockchip
AstralixNB has joined #linux-rockchip
levd has joined #linux-rockchip
levd1 has quit [Ping timeout: 264 seconds]
RayFlower has quit [Read error: Connection reset by peer]
RayFlower has joined #linux-rockchip
RayFlower has quit [Read error: Connection reset by peer]
RayFlower has joined #linux-rockchip
levd1 has joined #linux-rockchip
levd has quit [Remote host closed the connection]
RayFlower has quit [Read error: Connection reset by peer]
RayFlower has joined #linux-rockchip
RayFlower has quit [Read error: Connection reset by peer]
RayFlower has joined #linux-rockchip
dlan has quit [Ping timeout: 272 seconds]
dlan has joined #linux-rockchip
RayFlower has quit [Read error: Connection reset by peer]
RayFlower has joined #linux-rockchip
<AstralixNB>
good morning
<AstralixNB>
I did some SD Loader investigation and found a thing that makes me thinking...
<rperier>
hi
<AstralixNB>
hi!
<AstralixNB>
I prepared an SD card with a linux image and booted it on radxa
<AstralixNB>
works fine
<rperier>
Someone could try linux-next (20141201) on the radxa rock ? it's very unstable on RR, I get random crashes when I boot from usb, unable to boot from nfs. The marsboard rk3066 works perfectly fine (with usb or nfs
<rperier>
)
<AstralixNB>
I booted it on MK908IV and it boots fine
<rperier>
just to confirm that it's not just me
<AstralixNB>
rperier, later no problem
<AstralixNB>
give me a second to complete my finding
<rperier>
ok thanks
<AstralixNB>
but give me the link to your linux-next mirror and I start clone
<AstralixNB>
Ok, so... I made a linux SD image and started it on radxa rock and it works fine.
<AstralixNB>
I put it into MK908IV and it boots up fine
<AstralixNB>
I put it into Odys Iron tablet and the loader just throws "SHA Error" and then stays in USB Loader mode.
<AstralixNB>
I put the image into MK809III and again it boots fine.
<AstralixNB>
I got a remark that the image might be at wrong address, but that is not possible as the very same card boots in 3 out of 4 devices and the device that doesn't accept the image can read SD cards perfectly (tested by booting from NAND and doing some tests with a different SD card)
<naobsd>
AstralixNB: what you investigated?
<AstralixNB>
I just did the 3...4 steps above and wonder why the tablet throws SHA Error, while all other boards and sticks I have boot fine
akaizen has quit [Read error: Network is unreachable]
akaizen has joined #linux-rockchip
<AstralixNB>
So what is the SHA Error and why does it only appear on the Tablet? If it is enforced security, where is this set? If it is eFuse, is there a documentation how to check / set / reset this enforced security?
<naobsd>
I guess if secure boot is enabled and locked loader is flashed on NAND, loader on SD can be booted, but I'm not sure
<AstralixNB>
I did not check the version messages of the Loader... Good point! I might check that cause the Loader used in the SD image is an older version compared to the one written to NAND
<naobsd>
I think that error said sha hash in boot.img is wrong
<naobsd>
you have kernel+ramdisk boot.img in boot on SD, right?
<AstralixNB>
yes
<naobsd>
and you used different boot.img for each device (because kernel cannot be shared), right?
<AstralixNB>
This SD (exactly this SD with exactly the same content) works on three other devices
<naobsd>
you used same boot.img for all 4?
<AstralixNB>
Yes, cause at least the ARM cores should start fine
<naobsd>
hm
<AstralixNB>
It will crash later on wrong DC/DC controllers and other things but it must boot half way through the kernel
<AstralixNB>
This is one of the reasons I did not publish my build script... It doesn't work in all scenarios....
<naobsd>
E:Invaid tag(0x4D524150)! after loading boot.img is also strange...
antoinemaillard_ has joined #linux-rockchip
<naobsd>
can I see "success" log?
<AstralixNB>
No, did not made one
antoinemaillard has quit [Ping timeout: 264 seconds]
antoinemaillard_ is now known as antoinemaillard
<AstralixNB>
I can log that this evening
<AstralixNB>
What I found out too is some bad news for the linux guys...
<AstralixNB>
If you use the usual image tools to create an upgrade.img for Android and then use the windows tool to build an SD card image from it, it works many times. But if you exchange the Loader to a newer one, the image doesn't boot anymore.
<AstralixNB>
If you try to just boot linux with (boot) and (rootfs) many image tools fail as they miss the other Android images.
<AstralixNB>
And the linuxium script is hard coded to Android too.
<AstralixNB>
There are still a lot of unknown magic behind the loader process...
<rperier>
AstralixNB: okay, could you try to boot an usb stick ,
<rperier>
?
<AstralixNB>
What?
<rperier>
you booted from a sdcard right ?
<AstralixNB>
Yes
<rperier>
could you try to boot your rootfs from an usb key ?
<rperier>
(on the RR)
<rperier>
on my RR I boot either from usb or from the network (nfs)
<AstralixNB>
Never heard that this would work...
<AstralixNB>
If you use uboot you can boot from this media
<AstralixNB>
but with the original loader this is not working?
<rperier>
if you change kernel cmd line you can boot from this media too
<AstralixNB>
rperier, I have to boot Loader and kernel from SD or NAND, the rootfs can be wherever I like
<naobsd>
you said relation between loader and rknand.ko?
<AstralixNB>
naobsd: I respected the relationship between Loader and rknand.ko
<rperier>
AstralixNB: I load my kernel from the network, I load it on the fly with kexec and then I boot from whever I like too
<AstralixNB>
And in the stage where the SHA error is thrown, the loader has no access to the rknand.ko
<AstralixNB>
Ah, ok, that is a second way
<naobsd>
AstralixNB: no, what I asked about "exchange the Loader"
<AstralixNB>
rperier, which gcc to use for kernel?
<naobsd>
loader can be changed
<AstralixNB>
naobsd, I can try 2.08 (original RR Loader in Toms images), 2.10 and 2.16 if you like
<rperier>
usb driver is the same for both marsboard and RR, it works perfectly on marsboard, and it works on the RR if you boot from another media (for example If I boot from the network and then I plug my usb key, it is detected and I can mount it)
<AstralixNB>
In the NAND there is currently a 2.16, on the SD there is the original RR 2.08
<rperier>
however it works randomly when you want to boot from it directly (only on RR)
<naobsd>
AstralixNB: I just want explanation about you said
<rperier>
AstralixNB: gcc 4.9
<naobsd>
AstralixNB: how "if you exchange the Loader to a newer one, the image doesn't boot anymore." can happen?
<AstralixNB>
rperier, I do have linaro 4.9 hf here, is that ok? gcc-linaro-arm-linux-gnueabihf-4.9-2014.09_linux
ganbold_ has joined #linux-rockchip
<naobsd>
if you just said compatibility between loader and rknand.ko, it happen regardless of SD card tool for Windows
<rperier>
I use yocto, the toolchain works fine
<rperier>
this is the same toolchain I use for all my boards, other boards work just fine
<naobsd>
I can exchange loader on SD and no problem
<AstralixNB>
naobsd, I just tried to do that and failed. I used an Android AOSP 4.4.2 image that boots on RR
<AstralixNB>
I unpacked the recover and boot and put in the new 2.16 rknand.ko
<AstralixNB>
I exchnaged the Loader.bin and repacked the upgrade.img
wildea01 has joined #linux-rockchip
<AstralixNB>
Then I used the SD Windows tool to convert from upgrade.img to SD upgrade.img
<AstralixNB>
Booted: failed
<naobsd>
how failed?
<AstralixNB>
But as it was just a series of tests I tried to find a stable way of doing that, I did not save all the logs
<naobsd>
tool may be wrong, or your procedure may be wrong, I have no idea
<naobsd>
but "But if you exchange the Loader to a newer one, the image doesn't boot anymore." sounds very generic, it sounds like it must happen on everyone
<naobsd>
generic? natural?
<naobsd>
mm
<naobsd>
it's time to return to home
<AstralixNB>
Hey, I just followed wikis, image tool explanations and explanations read here.
<AstralixNB>
I tried to do step by step and it doesn't work
<naobsd>
which step?
<AstralixNB>
So either I am an idot, or my computer is broken or the explanations on the net are incomplete
antoinemaillard has quit [Quit: antoinemaillard]
<AstralixNB>
+i
<naobsd>
I guess "explanations on the net are incomplete"
<AstralixNB>
Thank you...
<AstralixNB>
Have a nice trip home :)
antoinemaillard has joined #linux-rockchip
<AstralixNB>
rperier, aren't there any rk3xxx defconfigs anymore?
naobsd has quit [Quit: Page closed]
<rperier>
AstralixNB: it's rk30_rk31_defconfig now (in devel/workbench, I mean mine is based on this one)
<AstralixNB>
Couldn't you make a compressed version of it and push it to pastebin? Doesn't make sense if I test with my version
<AstralixNB>
And there hasn't been push any version of it to mainline yet...
<rperier>
mmind00: you might be interested by my branch "drm-noimmu" for your work about older lcdc :) (I need to patch gem but the interface is there)
<AstralixNB>
rperier, where do you have this rk30_rk31_defconfig
<AstralixNB>
I thought havinf initramfs build into the kernel is enough to use it as boot.img
<naobsd>
boot is for kernel+ramdisk mkbootimg or ramdisk rkcrc image
<naobsd>
kernel may have initramfs in it, it's kernel with or without initramfs
<naobsd>
and kernel rkcrc image is not for boot even if it has initramfs
<naobsd>
kernel is no more than kernel even if it has initramfs
<AstralixNB>
So I change (boot) to (kernel) and it should work?
<AstralixNB>
At least in NAND that should work. I have tried with SD but SD requires kernels in boot and recovery
<naobsd>
AstralixNB: probably no, loader will want kernel and ramdisk partition
<naobsd>
it's not matter of NAND or SD
<naobsd>
loader issue
<naobsd>
you can put kernel in kernel partition on SD
<naobsd>
rule is quite simple
<naobsd>
loader needs kernel and ramdisk
<naobsd>
1: kernel in kernel and ramdisk in boot(or recovery for recovery boot)
<naobsd>
2: kernel+ramdisk in boot(or kernel+ramdisk in recovery for recovery boot)
<naobsd>
that's all
<AstralixNB>
Yes, I see
<naobsd>
you may change loader behavior
<AstralixNB>
The error messages of the loader are constantly out of focus. Thy do not represent anything usefull to the user reading them
ganbold_ has joined #linux-rockchip
<naobsd>
yes, but we already know the rule :)
<AstralixNB>
by renaming (boot) to (kernel) in parameter, the Error just changes to twice the old error messag
<AstralixNB>
+e
<AstralixNB>
start_linux=====393561
<AstralixNB>
E:Invaid tag(0x4D524150)!
<AstralixNB>
Load failed!
<AstralixNB>
E:Invaid tag(0x4D524150)!
<AstralixNB>
Load failed!
<AstralixNB>
The funny thing is, that this message doesn't have to do anything with the real error, cause it is claming about PARM, but ther first time the error was that KRNL was in boot, but no initramfs was in boot, now the double message of PARM represents a valid KRNL in kernel but no (boot) partiton defined
<naobsd>
you need kernel and boot partition unless you use kernel+ramdisk mkbootimg
<AstralixNB>
I asked why there are no more desciptive error codes and the answer was: It is not intendend for user, only for mass production.
<naobsd>
hehe
<AstralixNB>
head->desk
<naobsd>
no product has serial console ;)
<naobsd>
no one see that error
<AstralixNB>
Yes, they only see nothing and wonder why their product is dead
<AstralixNB>
however, I need a initramfs and mkbootimg
<AstralixNB>
let me check, what I have on my drives
<naobsd>
?
<naobsd>
ah, you cannot add one more partition?
<AstralixNB>
To test the mainline kernel I need to program it to the NAND and boot it from there
<naobsd>
I thought you already made it
<AstralixNB>
But without any boot.img I cannot as the Loader avoids that
<naobsd>
I explained
<naobsd>
prepare kernel and boot partition
<naobsd>
flash kernel.img to kernel
<AstralixNB>
yes
<naobsd>
flash kernel.img which has initramfs to kernel
<naobsd>
flash dummy but valid boot.img to boot
<AstralixNB>
hmmm...
<naobsd>
if you configured kernel which uses included initramfs, no need to flash ramdisk to boot
<naobsd>
only valid signature is needed, i.e. rkcrc -k dummy boot.img
<naobsd>
or you may flash kernel.img to boot, it's valid dummy
<AstralixNB>
cool idea
<naobsd>
valid signature (rkcrc -k or mkbootimg) is needed for the parittions where loader loads i.e. kernel/boot/recovery
<naobsd>
but loader doesn't understand it's payload
<naobsd>
you can put any code e.g. *BSD kernel as kernel body
<naobsd>
if kernel want ramdisk, you may put ramdisk into boot/recovery, then loader loads it to ram
<naobsd>
or if kernel has initramdisk in it, loader loads boot/recovery but kernel just ignores it
<AstralixNB>
I will need to split initramfs from kernel for /l)ubuntu, but I wanted only a short test of the mainline kernel. And for that I prepared a working minimal initramfs from debian
<AstralixNB>
So I just dd if=/dev/zero of=boot bs=1 count=1024
<AstralixNB>
then rkcrc -k boot boot.img
<AstralixNB>
Now thw loader does something, but he doesn't tell anymore
<naobsd>
I cannot remember but
<naobsd>
there is few more rule for dummy
<naobsd>
e.g. rkcrc -k 0byte_file boot.img doesn't work
<AstralixNB>
correct
<naobsd>
I'm not sure 1K is enough
<AstralixNB>
flashed kernel.img as kernel.img and as boot.img and the kernel started
<naobsd>
good :)
<AstralixNB>
we really should carry that all together and put it in the wiki
<AstralixNB>
or complete uboot and ignore this mass production only loader
<naobsd>
only advanced developer tries such a strange image ;)
<naobsd>
but yes, docs is needed
<naobsd>
ah yes, complete u-boot too :)
<AstralixNB>
lol
<AstralixNB>
rperier: kernel lives... for about 0.25 seconds
<AstralixNB>
but I thought I should test this as it works in 3066 but not in 3188 and he wants to exclude that it is a local failure of his personal board.
<AstralixNB>
So I can confirm that it is not a failure of his personal board, as I can reproduce a dump here. But I am not sure, wht he wants to test exactly, so I wait till he is back online
<AstralixNB>
But from this moment on, I love the SD boot, as I flash test kernels to NAND and test, and after lunch break, I put back in my working SD card and I am back to work in less than a minute
<naobsd>
he never said about led dump
<naobsd>
what he said is usb/emac random crash
GriefNorth has joined #linux-rockchip
<AstralixNB>
You're right...
Tony_ has quit [Quit: Leaving]
RayFlower has quit [Read error: Connection reset by peer]
RayFlower has joined #linux-rockchip
<hramrach_>
naobsd: how does the loader tell if it should load kernel from kernel?
<hramrach_>
if you flash kernel+initramfs to both kernel and boot I would expect it to run the kernel from boot .. if it can tell it's a kernel
<AstralixNB>
That I know: If system should start in recovery, it checks for kernel in recovery. If it fails it falls back to kernel
<AstralixNB>
Same for boot
<AstralixNB>
If it should start from recovery or boot is set in misc.img
<naobsd>
first few bytes in boot/recovery are enough for checking kernel+ramdisk or just ramdisk
<naobsd>
i.e. ANDROID! or KRNL
antoinemaillard has quit [Quit: antoinemaillard]
<AstralixNB>
Did you check that?
<naobsd>
actually loader will check completeness not only header
antoinemaillard has joined #linux-rockchip
<naobsd>
I can read u-boot source but
<AstralixNB>
Cause I signed a NULL file of 1k, 100k, 1M and they where not accepted
<naobsd>
do you really think checking ANDROID! or KRNL is difficult or pointless?
<naobsd>
then size is not important
<naobsd>
something is needed
<AstralixNB>
So I can test echo "ANDROID!" > dummy and then rkcrc -k dummy boot.img
<naobsd>
it will be KRNLANDROID!
<naobsd>
== it's KRNL
<AstralixNB>
It must be something with the size too
<AstralixNB>
cause that KRNLANDROID! is not enough
<AstralixNB>
Loader hangs without any firther information
<AstralixNB>
i=u
<naobsd>
it's just 8 bytes data
<naobsd>
I think 12345678 will be same
<naobsd>
you are just checking rule for dummy, not kernel or kernel+ramdisk
<naobsd>
AstralixNB: please think well, you can write code to check "is it kernel(KRNL) or kernel+ramdisk(ANDROID!)?", right?
<naobsd>
it should be easy for you
<AstralixNB>
yes shure
<AstralixNB>
sure
<naobsd>
and I think RK can write it too
<AstralixNB>
But you explained, that they check for the KRNL or ANDROID!
<naobsd>
ah, sorry, "ramdisk or kernel+ramdisk" ;)
<AstralixNB>
And they check for something like size too
<naobsd>
same thing
<AstralixNB>
ramdisk is ANDROID! and kernel is KRNL
<naobsd>
rkcrc-ed ramdisk(KRNL) or mkbootimg-ed kernel+ramdisk(ANDROID!)
<naobsd>
no no
<AstralixNB>
and kernel+ramdisk is KRNL too, but somewhere inside it needs an additonal ANDROID!
<naobsd>
rkcrc -k is KRNL, mkbootimg is ANDROID!
<naobsd>
no no no
<naobsd>
just run rkcrc -k and mkbootimg and see first some bytes
<AstralixNB>
I know.
<naobsd>
rkcrc -k is KRNL, for kernel only image and ramdisk only image
<naobsd>
mkbootimg is ANDROID!, for kernel+ramdisk image
<AstralixNB>
But the Loader boots the kernel.img from (kernel) even there is a kernel.img in (boot)
<naobsd>
boot partition should have either ramdisk only image(KRNL) or kernel+ramdisk image(ANDROID!)
<AstralixNB>
And the kernel.img in (boot) is invalid as it doesn't have a bootimage
<naobsd>
loader check boot(or recovery) first,
<naobsd>
if it's ANDROID!, it includes kernel. loader loads kernel and ramdisk from boot
<AstralixNB>
Ok, again:
<AstralixNB>
I just flashed a kernel-only kernel.img to boot and to kernel
<naobsd>
if it's KRNL, it's ramdisk(even you flashed dummy, loader doesn't understand it's payload), then loader loads kernel from kernel partition
<naobsd>
"kernel-only kernel.img to boot" === dummy, loader loads it, but no one use it
<AstralixNB>
so the Loader considers KRNL as kernel in kernel and as ramdisk in boot even there is a kernel
<naobsd>
yes
<AstralixNB>
Blody bastard Loader
<AstralixNB>
sorry
<naobsd>
very clear logic, flashing kernel to boot is just workaround from our laziness :)
<naobsd>
we just do non-standard way
<AstralixNB>
meeting... cu later
<naobsd>
so it looks like strange
<naobsd>
later
GriefNorth has quit [Quit: WeeChat 1.0.1]
ganbold_ has quit [Ping timeout: 255 seconds]
ganbold_ has joined #linux-rockchip
antoinemaillard_ has joined #linux-rockchip
antoinemaillard has quit [Ping timeout: 250 seconds]
antoinemaillard_ is now known as antoinemaillard
mrcan_ has joined #linux-rockchip
mrcan has quit [Ping timeout: 272 seconds]
antoinemaillard has quit [Quit: antoinemaillard]
antoinemaillard has joined #linux-rockchip
linuxium has joined #linux-rockchip
<neyder>
hey hey, good morning , please need your help with my -config for my kernel
<linuxium>
naobsd: I successfully test sd card boot on rk3288
<naobsd>
neyder: what I said is _for your device_
<linuxium>
naobsd: I got my bricked (and useless) R28 to boot Android from SD card
<naobsd>
linuxium: oh, good!
<naobsd>
what was the problem?
<neyder>
linuxium, you are so smart!
<AstralixNB>
Hey linuxium, good work, but we need to extend functionality a bit :)
<neyder>
naobsd, so where can I got _for my device_ ? any bit of light?
<naobsd>
neyder: ask vendor, or write it from scratch
<neyder>
i think i'm writing from scratch (so 'make menuconfig' for it) I think for learn and for a project
<linuxium>
AstralixNB: sure, I put it out as a starter to help anyone who was having problems ... from the forums I saw many people struggling and suddenly I needed to access Android on a stick which had Linux installed to NAND and I didnt want to overwrite Linux
<AstralixNB>
neyder: No original vendor image to rip apart (check backlog and google for rkunpack tools )
<AstralixNB>
Yes, sure
ganbold_ has quit [Ping timeout: 255 seconds]
<AstralixNB>
I am currently trying the imgrepackrk tool from RedCorpio... It can unpack images down to file-system level of boot and recovery
<AstralixNB>
But it fails on images that are for linux (only boot and linuxroot) or that you try to modify with means of kernel images
<AstralixNB>
Would be great ot combine these tools and be able to create Android and Linux images for NAND, eMMC and SD just by a script call..
<linuxium>
naobsd: stupid typo from me which I didn't see and not having a serial UART connection on the box meant it took a long time to pick up
<naobsd>
linuxium: so, my image just work?
<naobsd>
as like as on firefly
ganbold_ has joined #linux-rockchip
<linuxium>
naobsd: I used idb_sector_0.enc idb_sector_1 FlashData.bin FlashBoot.bin and then create an sd card with full android on it
<naobsd>
linuxium: I see
<naobsd>
if you need you can build FlashBoot.bin from source
<naobsd>
but current one should be enough for normal usage
<linuxium>
yes, big drawback is onboard emmc taking priority over sd card when using third party software (i.e. with precompiled kernel)
<naobsd>
ah, yes
<naobsd>
btw kernel detect your broken emmc? can you read/write it?
<naobsd>
oops
<linuxium>
I didn't load rk30xxnand so cannot see emmc
<naobsd>
surely rknand.ko can be used for emmc driver, but I guess 3.10 kernel uses in kernel mmc driver for emmc...I'm not sure default configuration for R28
<naobsd>
anyway, it's not so important
<naobsd>
I just curious what happened on your emmc
<naobsd>
oops
<naobsd>
my PX2 board seems unstable :(
<naobsd>
random crashes :(
<AstralixNB>
if the dwmmc driver is enabled and the right entries are in dtb, the eMMC should be detected by kernel
<linuxium>
I'm going to look into it further ... tomorrow ... as it is bedtime now!
<AstralixNB>
If you want to avoid interference by a possibly dead eMMC, you should disable the dwmmc or at least ste slot that is connected to eMMC first
<AstralixNB>
the slot
<naobsd>
linuxium: good night
<naobsd>
mmm, what happened on my PX2 board... it was fine :(
<naobsd>
maybe I shorted pads wrongly to go mask rom mode :(
<naobsd>
there is no way to go to loader/mask rom mode by default :(
<naobsd>
I had to short pads after I flashed non-functinal image...
<AstralixNB>
Check with glasses if there is a shortage of the pins made by solder
<AstralixNB>
I had that on a device, where some solder tin was pressed between two pins shorting them caused by not so luck pin shorting trial
<AstralixNB>
Normally I use a detist's tool to short pins or repair SMD. But that day I couldn't find it and used a normal board pin. And the pin scratched off some solder pressing it between the pins.
<AstralixNB>
dentist's tool
linuxium has quit []
cnxsoft has quit [Read error: Connection reset by peer]
cnxsoft has joined #linux-rockchip
<field^Mop>
naobsd: AstralixNB: I will put together a wiki page containing all those bits and pieces about loader, boot, krnl, ramdisk, initrfs etc.
cnxsoft has quit [Client Quit]
<AstralixNB>
naobsd? Which wiki should he use? rockchip.info?
RayFlower has quit [Read error: Connection reset by peer]
RayFlower has joined #linux-rockchip
antoinemaillard has joined #linux-rockchip
mrcan__ has joined #linux-rockchip
mrcan__ has quit [Read error: Connection reset by peer]
mrcan__ has joined #linux-rockchip
mrcan_ has quit [Ping timeout: 264 seconds]
mrcan__ is now known as mrcan
AstralixNB has quit [Ping timeout: 250 seconds]
RayFlower has quit [Read error: Connection reset by peer]
RayFlower has joined #linux-rockchip
RayFlower has quit [Read error: Connection reset by peer]
RayFlower has joined #linux-rockchip
irsol has quit [Ping timeout: 244 seconds]
<field^Mop>
ok, i'd start with the Boot_Sequences page and add the loader stuff to the wiki
irsol has joined #linux-rockchip
<field^Mop>
Astralix: what do you mean by "full loaders" from radxa site? the other site has an extensive loader list, too
<field^Mop>
afk
RayFlower has quit [Read error: Connection reset by peer]
RayFlower has joined #linux-rockchip
RayFlower has quit [Read error: Connection reset by peer]
RayFlower has joined #linux-rockchip
wildea01 has quit [Quit: leaving]
antoinemaillard has quit [Ping timeout: 272 seconds]
field^Mop has quit [Ping timeout: 264 seconds]
AstralixNB has joined #linux-rockchip
AstralixNB1 has joined #linux-rockchip
AstralixNB has quit [Ping timeout: 250 seconds]
bengal has joined #linux-rockchip
RayFlower has quit [Read error: Connection reset by peer]
RayFlower has joined #linux-rockchip
RayFlower has quit [Read error: Connection reset by peer]
RayFlower has joined #linux-rockchip
FreezingCold has quit [Remote host closed the connection]
RayFlower has quit [Read error: Connection reset by peer]
RayFlower has joined #linux-rockchip
RayFlower has quit [Read error: Connection reset by peer]
RayFlower has joined #linux-rockchip
RayFlower has quit [Read error: Connection reset by peer]
RayFlower has joined #linux-rockchip
gb_master has joined #linux-rockchip
RayFlower has quit [Read error: Connection reset by peer]
RayFlower has joined #linux-rockchip
markm_ has quit [Ping timeout: 272 seconds]
RayFlower has quit [Read error: Connection reset by peer]
RayFlower has joined #linux-rockchip
RayFlower has quit [Read error: Connection reset by peer]
<Astralix>
What could this mean...
<Astralix>
start_linux=====66051 E:CRC failed! Load failed! Begin recover... GetRemapTbl flag = 1 E:Invaid tag(0x00000000) in backup!
<Astralix>
oha, CR/LF...
RayFlower has joined #linux-rockchip
<Astralix>
start_linux=====66051
<Astralix>
E:CRC failed!
<Astralix>
Load failed!
<Astralix>
Begin recover...
<Astralix>
GetRemapTbl flag = 1
<Astralix>
E:Invaid tag(0x00000000) in backup!
<Astralix>
And the loader is back in USB mode
<naobsd>
redo rkcrc?
<naobsd>
I thought my PX2 board was broken, but it seems my u-boot somewhat makes kernel unstable
<naobsd>
I flashed my u-boot to another PX2 board, same thing happened
<naobsd>
and original u-boot compiled from chipsparks repo doesn't boot at all :(
<naobsd>
here is 2 bricks now :)
field^Mop has joined #linux-rockchip
<field^Mop>
naobsd: brickmaster? ;)
mrcan_ has joined #linux-rockchip
mrcan has quit [Ping timeout: 258 seconds]
<naobsd>
sounds honorable :)
<naobsd>
I have to go office
<naobsd>
later
naobsd has quit [Quit: Page closed]
<field^Mop>
l8
<field^Mop>
wow, this is time skew.. I'm about to leave for home, you're in for office at the same time! 8-|
<Astralix>
I'm about to head for bed...
<Astralix>
But I am actually a bit frustrated now with this mess of images
RayFlower has quit [Read error: Connection reset by peer]
RayFlower has joined #linux-rockchip
RayFlower has quit [Read error: Connection reset by peer]
RayFlower has joined #linux-rockchip
<field^Mop>
Astralix: tomorrow is another day
<field^Mop>
Astralix: with better images :)
<Astralix>
Yes, looks like
gb_master has quit [Quit: Leaving]
<Astralix>
Ah... I am not sure what the problem is
<Astralix>
The tablet insists on posting SHA Error to each kind of image I flash
<Astralix>
Even I am kurrently flashing images that already ran fine on this tablet
<field^Mop>
give it a rest, the tablet is tired, too..
<field^Mop>
;)
<Astralix>
Yes... Just one last test
<Astralix>
instead of using the upgrade_tool. I try windows tool...
<field^Mop>
Astralix: so? did it work this time?
<Astralix>
Hehe.. No, I need to copy it from one PC to another one first, but the memory stick is not willing...
<Astralix>
I moved to plan b) and switched to TV and beer
phh_ has joined #linux-rockchip
phh has quit [Ping timeout: 252 seconds]
<hramrach_>
heh, bricksmaster
RayFlower has quit [Read error: Connection reset by peer]
<hramrach_>
freemasonry reinvented
RayFlower has joined #linux-rockchip
dlan has quit [Ping timeout: 252 seconds]
<Astralix>
Do you mean me?
dlan has joined #linux-rockchip
<hramrach_>
who is not bricksmaster here?
<hramrach_>
bricking devicesd all the time ^.^
<Astralix>
Ah, I don't believe it is bricked
<hramrach_>
that's why you are a bricksmaster. You can unbrick it too \o/
<Astralix>
Ah, it is a game... roll the dice, read the book and see if you can proceed to the gate...
nighty-_ has quit [Read error: Connection reset by peer]
<field^Mop>
Astralix: good switch, more relaxing than make images fail to boot.
<Astralix>
However... The memory stick doesn't want to memorize anymore and I am over with my TV series and the beer is empty... good point to head over to the bed. A good book is waiting there while the rest of the family is already sleeping well.
<Astralix>
See you guys tomorrow
AstralixNB1 has quit [Ping timeout: 250 seconds]
bengal has quit [Quit: Leaving]
RayFlower has quit [Read error: Connection reset by peer]