arokux2 has quit [Remote host closed the connection]
akaizen has quit [Remote host closed the connection]
alcides has quit [Remote host closed the connection]
hramrach has quit [Ping timeout: 240 seconds]
hramrach has joined #linux-rockchip
Astralix1 has joined #linux-rockchip
Astralix has quit [Ping timeout: 264 seconds]
akaizen has joined #linux-rockchip
Theueip has joined #linux-rockchip
massi has joined #linux-rockchip
rellla has joined #linux-rockchip
Tsvetan3 has joined #linux-rockchip
Tsvetan2 has quit [Ping timeout: 252 seconds]
Tsvetan3 has quit [Quit: Leaving]
atiti has joined #linux-rockchip
arokux2 has joined #linux-rockchip
arokux2 has quit [Remote host closed the connection]
deviker has joined #linux-rockchip
pfalcon has joined #linux-rockchip
Tsvetan has joined #linux-rockchip
<pfalcon>
Tsvetan: Hi, do you have a git repo with all available rk3xxx datasheets, like you have for a1x?
<arokux1>
do you guys know the specification of the battery which will fit radxa board?
<Tsvetan>
pfalcon there are links in the wiki to sites where this info is present
<Tsvetan>
rockchip asked me to sign NDA for the documents they gave me
<Tsvetan>
but basically these are same with the documents which are on the web already
<pfalcon>
Tsvetan: I see. Current links I randomly spread, and point to dropbox accounts which can be gone any time. Someone who didn't sign nda should make a more central kind of dump ;-).
<Tsvetan>
agree
ganbold_ has joined #linux-rockchip
<lunra>
What's the legal risk in doing so? I can chuck it on my crappy webserver just for redundancy.
<lunra>
(given that I did not sign an NDA, however it's still 'protected'?
<lunra>
)
tomCode_ has joined #linux-rockchip
<tomCode_>
Hi all, i am trying run u-boot on the rockchip rk3066! somebody able to do that before?
<tomCode_>
Does somebody know the mysterious loader.bin from rockchip? whats this?
pfalcon has quit [Ping timeout: 264 seconds]
pfalcon has joined #linux-rockchip
libv has quit [Ping timeout: 260 seconds]
pfalcon has quit [Ping timeout: 260 seconds]
eebrah|noPC has joined #linux-rockchip
libv has joined #linux-rockchip
vinifr has joined #linux-rockchip
pfalcon has joined #linux-rockchip
<pfalcon>
tomCode_: where did you get reference to *loader.bin* (exactly this name)? Otherwise, I started page to collect bootloading info: http://linux-rockchip.info/mw/index.php?title=User:Pfalcon#Bootloading_Questions
<tomCode_>
pfalcon: thanks, i will study this page
<tomCode_>
mmhh...i am wondering if the mask rom mode is something that is hardwired (lets say in hardware, programmed by the production of the chip) or it is something that is updated when a new image is flashed...
<tomCode_>
with loader.bin I was speaking about e.g RK30xxLoader(L)_V1.18.bin
vinifr has quit [Ping timeout: 245 seconds]
pfalcon has quit [Ping timeout: 240 seconds]
eebrah|noPC has quit [Read error: Connection reset by peer]
eebrah|noPC has joined #linux-rockchip
vinifr has joined #linux-rockchip
eebrah|noPC has quit [Read error: Connection reset by peer]
<tomCode_>
just want to be sure, that i do not brick my device
<tomCode_>
i do not understand the RK images...usually an android image starts with the header "ANDROID" but the RK3066 boot.img starts with KRNL which is different..
<tomCode_>
to make the confusion complete: the RK3066 recovery.img starts with ANDROID
<tomCode_>
the question is, what is inside boot.img
<naobsd>
some years ago, it can be bricked because sometimes proper RK..Loader...bin is not available, but for RK3x, it seems all devices uses compatible RK...bin, then it's impossible to brick it
<naobsd>
KRNL boot.img is ramdisk
<naobsd>
ANDROID! boot.img is kernel+ramdisk
<naobsd>
same for recovery.img
<naobsd>
ramdisk-only image need kernel.img in kernel partition
<tomCode_>
thanks
<naobsd>
you can convert image format
<tomCode_>
yes, theses tools i have (img_maker afptool ...)
<tomCode_>
is the linux kernel loaded directly from the RK3x loader?
<tomCode_>
yes, but propably still not 100% understood
<naobsd>
mask rom -> RK loader -> Linux kernel
<tomCode_>
where is the rk loader stored? nand-flash?
<naobsd>
if you have kernel+ramdisk image in boot partition, it will be used instead of kernel image in kernel partition
<naobsd>
it = kernel in kernel+ramdisk image
<naobsd>
nand
<tomCode_>
that means if i want to run u-boot i have to replace the RK3x loader, correct?
<naobsd>
I don't have right answer for now because no one tried yet. but is there any reason that RK3x loader can't load u-boot instead of kernel?
<tomCode_>
ok. i have seen that there is also the possibility to include a second stage bootloader in boot.img.
<deviker>
naobsd: Thanks you were right. It wasn't loading my kernel yesterday because I forgot to change boot.img and the previous boot.img already had a kernel inside so It didn't look at the kernel mtd.
<tomCode_>
maybe this is the easier way
pfalcon has joined #linux-rockchip
<naobsd>
Linux kernel is sometimes used as (some sort of) bootloader, there is kexec
<tomCode_>
this is my second option
<naobsd>
deviker: I see. kernel partition can be wiped if you have kernel in boot/recovery
<tomCode_>
some rockchip devices have a "recovery" button which starts an alternative system (partition /recovery i think). where is this switch evaluated? is it in the loader?
<naobsd>
NAND driver is binary only, it will be difficult to support NAND other than Linux(Android) kernel
<naobsd>
well, there is a line in parameter like "RECOVER_KEY: 1,1,0,20,0"
<naobsd>
I can't remember detail but probably it's related to ADC
<naobsd>
RK loader read parameter, then it will check ADC key state
<tomCode_>
i have got the # before this line in parameter
<tomCode_>
naobsd, this clever guy (sarg) obtained the bootrom (which is synonym for mask rom) from the RK3066, right?
<naobsd>
tomCode: I think so. he sometimes join here
<tomCode_>
i think in the reference manual there is a flow chart of the bootrom which says that an ID block is checked from NAND flash
<tomCode_>
this would mean that the rom code can read nand flash
<tomCode_>
(rom code = bootrom = mask rom)
<naobsd>
tomCode: I think so
<naobsd>
and from spi, emmc, uart, and usb
<tomCode_>
scrambled or not?
<naobsd>
well?
<tomCode_>
i think the loader which is booted from the bootrom must be scrambled, independent which memory/download is used
<naobsd>
probably, yes
<tomCode_>
would mean that a binary (lets call it u-boot.bin) cannot just be transmitted via UART and executed...it must be scrambled before...
<tomCode_>
one day i will try this...
<naobsd>
there is key in sarg's info
<tomCode_>
yes, i've seen it
<tomCode_>
next step is to bring my board into UART mode
<naobsd>
"erase IDB" erases RK loader on NAND, then you can enter mask rom mode
<tomCode_>
the other way is : shorten two pads (on the NEO X5 it is T23 and T24), which propaply disables the NAND-Flash and the cpu thinks there is no external memory attached...
<naobsd>
yes
<naobsd>
some device has a key to kill NAND
<naobsd>
hmm. reading rk30_clock_data_init() is hard task ;)
<naobsd>
oh, FreeBSD-CURRENT arm compiler is clang...
<naobsd>
(return to my own work...)
<tomCode_>
naobsd, cu, thanks for all answers
<sjoerd>
Heh, did the linux-rockchip-3.0 repository disappear ?
<naobsd>
it was just an archive of released/leaked kernel source. same as mirror of tarballs. no unique work was committed.
<sjoerd>
I see
<sjoerd>
Is there one main repository for these efforts ? iirc it was mentioned Gallands tree is the current suggested one
<naobsd>
there are some people, some groups, but (I think) no main "team"
<naobsd>
#linux-rockchip and linux-rockchip.info is very new
<Tsvetan>
naobsd is there any obstracle linux-rockchip repository on github to be made?
<Tsvetan>
linux-sunxi also at the beginning was spread through many repos hno/uboot amery/sunxi etc
<Tsvetan>
but later they decided to put everything on one place