RayFlower has quit [Read error: Connection reset by peer]
RayFlower has joined #linux-rockchip
<naobsd>
"Shipping: US $47.53toJapan via DHL" only...
<naobsd>
from US??? ;)
hipboi has joined #linux-rockchip
<hipboi>
mmind00, rperier i already got one firefly
<hipboi>
it's well made
<naobsd>
I asked firefly team about special offer for developers some times ago, but no response about it ;)
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
<naobsd>
ah, not from US, just "US $"
RayFlower has quit [Read error: Connection reset by peer]
RayFlower has joined #linux-rockchip
<rperier>
naobsd: +1 for special offer
RayFlower has quit [Read error: Connection reset by peer]
RayFlower has joined #linux-rockchip
AstralixNB has joined #linux-rockchip
Avasz has joined #linux-rockchip
<Avasz>
hello anyone around? I got problem booting picuntu in mk908II (rk3188).
<AstralixNB>
What problem?
<Avasz>
AstralixNB: it doesn't boot at all. I bought mk908II 5-6 months back, picuntu runs fine in that. and a couple of months ago I bought another 5 of the same.
<Avasz>
none of these 5 boots picuntu at all
<AstralixNB>
AFAIK there was a hardware change...
<Avasz>
I know of RK3188 & RK3188-T. anything else after that too?
<Avasz>
all six of them are rk3188-T. and the same android runx fine.
<AstralixNB>
Hard to tell, probably you need to open at least one of the sticks and connect a serial port to it
<Avasz>
that would be difficult :(
<Avasz>
well I can open the first one. but not other five right now.
<Avasz>
is there any tutorial/readme on it?
<AstralixNB>
Hmmm... Not sure. I never needed a tutorial to check the testpointzs for the serial port...
<AstralixNB>
Did you flash the NAND with picunto or do you use a SD card?
<Avasz>
but not sure if it is only for radxa rock or works on all rk3188
<AstralixNB>
I am testing these on radxa and SD currently. Not sure if they all work well
<Avasz>
ok. Iguess I am out of luck for now then.
<Avasz>
does loader make any differene?
<AstralixNB>
AH, don't give up that fast.
<Avasz>
I tried with 1.2, 2.03 and 2.13.
<Avasz>
2.13 gave busybox shell even in the old working one.
<Avasz>
isn't there any other method to check about hardware difference without connecting serial pins? I can boot android all fine. maybe some tools in android?
<AstralixNB>
The ugly part of these MK809 devices is, that there are many out there that doesn't have to do anything with the original thing
<AstralixNB>
there are MK809III but IMHO tronsmart never produced them
<Avasz>
mine is 908II not 809.
<AstralixNB>
Yes, that is not a tronsmart but a Rikomagic
<AstralixNB>
if you get a shell, check the kernel log for what is probably not working
<Avasz>
no, I can't get to shell. Nothing shows up at all in these new sticks.
<Avasz>
"No Signal Found" message.
<AstralixNB>
There is absolutely no reaction on the screen? Not even 1 oto 4 Tux (Penguins) appearing?
<Avasz>
no
<Avasz>
nothing at all
<Avasz>
in all of them. except the one I bought separately 5-6 months back.
<Avasz>
I tried with various loaders.
<AstralixNB>
Ah...
<AstralixNB>
You tried loaders?
<AstralixNB>
Why?
<Avasz>
picuntu comes with rk3188loaderv1.2 by default
naobsd has quit [Ping timeout: 246 seconds]
<AstralixNB>
If you tried an 2.xx and then went back to 1.xx you probably killed the FTL and now you need to recover the FTL by a lower format.
<Avasz>
first I tried with the default loader. 1.2, it didn't work
<Avasz>
so triend 2.03 and 2.13 respectively
<Avasz>
both of which did not work too
<AstralixNB>
If you linux doesn't show up, it is probably everything but the loader
<Avasz>
ah ok.
<AstralixNB>
All RK3xxx can be bootet with 2.08 or 2.16. Both I tested on several tablets and sticks.
<Avasz>
i tried with 2 kernels, one default and another by rdavenport which fixes flickering issue in rk3188-T models
<AstralixNB>
2.xx has the benefit that it can boot from an SD card, so you can leave Android on the NAND and use Linux from a card
<Avasz>
haven't tried with different boot.img and system.img as can't find them.
<AstralixNB>
Do you use linux or windows for flashing?
<Avasz>
tried with both
<Avasz>
but windows first, as rkandroid tool is easier to use than rkflashkit.
<AstralixNB>
I use the current linix upgrade_tool and it works fine. rkflashkit is open source and does not support proper bootloader and FTL recovery handling
<AstralixNB>
It cannot handle FTL as this part is closed source by RK. But upgrade_tool is the official RK linux flash tool.
<Avasz>
yeah used upgrade_tool to flash a complete .img file. didn't know it can flash separate ones.
<AstralixNB>
First of all you should try to get the devices back into a working state
<AstralixNB>
Ok, Loader is working
<Avasz>
yes. I got it back by flashing android 4.4
<AstralixNB>
cool.
<Avasz>
so after that, now flashing -p,-b, -k & -s
<AstralixNB>
Yes
<Avasz>
does it need to be in specific order?
<AstralixNB>
no
<Avasz>
btw, aloksinha suggests to erase nand before flashing picuntu. I tried without erasing to test what happens, seems it fails during checking.
<Avasz>
right now also I haven't erased nand.
<AstralixNB>
that is if there is a bad block that was missed by FTL handling. So you need to repeat the erase and flash procedure until you have no further error
<AstralixNB>
Unfortunately there is no "format flash" function that erases, checks and updates the FTL in a single run.
<Avasz>
so after completion, just RD is enough to reboot?
<AstralixNB>
yes
<Avasz>
*sighs* didn't work.
<AstralixNB>
Ah, we just upgraded the tools you need to do further tests in a reliable way.
<AstralixNB>
But I fear that without a serial console there is not much chance to find a quick way.
<AstralixNB>
You only could try to search around in the usual forums.
<AstralixNB>
Most people here use more developer kit like units like radxa rock or even original RK developer kits
<karlp>
most, but not all ;)
<karlp>
(though most of the people who can help out the most do)
<AstralixNB>
hi karlp, yes
<Avasz>
difficult for endusers :)
<AstralixNB>
However, even I do have a tronsmart like 809 and RK 908 stick I am pretty sure that they do not match the hardware Avasz has.
<Avasz>
I have two 2 sets(1+5) of mk908II bought form same vendor, and seems they don't match too. :D *sad*
<AstralixNB>
What do you expect? It is chinese hardware.
<Avasz>
yeah.
<AstralixNB>
Name !0 brand != type != case != hardware
<Avasz>
I suggested to buy these in my office so that we can deploy as low powered computers in schools located ate remote places.
<Avasz>
I guess, need to find a workaround with android now.
<AstralixNB>
You really should consider buing radxa rocks for such things, cause you get schematics, updates, information and serial port is available in a manor that everyone can easyly contact.
<AstralixNB>
And radxa rock isn't more expensive
<AstralixNB>
ask hipboy for a price if you buy a bundle
<AstralixNB>
The radxa lite starts at 59$
<karlp>
re: pcb colour of the firefly, that's the same colour as the minix neo x5 mini pcb I have, andyeah, I thgoutht the samething :)
<Avasz>
first there was raspberry pi, which was discarded because it was too slow.
<Avasz>
this was taken as good solution because it comes with nice nand storage, wifi, bluetooth, casing.
<karlp>
no nand on radxa lite...
<AstralixNB>
karlp, I would nopt recommend Firefly for a business project as there is not enough support now. And radxa 4-core performs way better than a Raspberry PI
<Avasz>
yeah. read just now
<karlp>
(I'm not looking at firefly for a business project, I was just commenting on someone else's comment on the pcb colour)
<AstralixNB>
If you use a Sandisk Extreme 45/45MB/s SD card for the linux image, it performs great... feeling at least twice as fast as if it works from NAND
<AstralixNB>
sure. If Ian couldn't get your sticks to work, probably no-one can help :)
<Avasz>
I am searching for instructions on installing it directly to Nand instead of from sd
<AstralixNB>
Just test it with SD, if you have a working image, come back here and I help you to put it from SD to NAND
<Avasz>
ah ok. will do that.
<AstralixNB>
But I have to do some daytime job now. Just ping any Astralix in here, it is all me.
<karlp>
AstralixNB: you can do that with mainline stuff? or are you just talking about taking an android tree and getting the right ramdisk to get the rk_nand modules in it?
<AstralixNB>
There is probably no desktop linux based on mainline...
<Avasz>
I will do it tomorrow. Will leave for it to download tonight. don't have good bandwidth during daytime.
<AstralixNB>
But as current Loaders support SD card loading, I will probably focus on getting a ubuntu desktop version going and then upgrade the kernel bit by bit.
<karlp>
so you reckon, with one of the RKLoaderxxx 2.x, you can boot from SD on a rk3066 as well?
cristian_c has joined #linux-rockchip
<AstralixNB>
I haven't tested 3066 but can do this tonight. I have a dead tablet with 3066 and I do have 2.16 loaders for 3188 and 3066
<AstralixNB>
If it works I can report
<AstralixNB>
(btw the tablet isn't dead, but there is no source code for it available and it uses RK PWM DC/DC and I do not have source and voltage tables for it)
<cristian_c>
Hi
<AstralixNB>
Hi
<cristian_c>
I can't mount /system in cwm recovery
<cristian_c>
I get invalid argument
<cristian_c>
either No such file or directory
<cristian_c>
Any ideas?
<AstralixNB>
Actually no
<AstralixNB>
Is your system.img correct?
<AstralixNB>
cristian_c can you copy the command line you used here?
<cristian_c>
AstralixNB, mount -t ext4 /dev/block/mtd/by-name/system /system
<AstralixNB>
did you try to use the simple /dev/mtdX instead. cause mtd/by-name/ only works if the by-name script has worked
<cristian_c>
AstralixNB, there aren't mtd devices in /dev
<AstralixNB>
so there is no rknand_ko.ko loaded and you cannot access the NAND
<cristian_c>
AstralixNB, there are four files
<cristian_c>
oh, no, two .ko.ko files
<cristian_c>
rk30xxnand_ko.ko.3.0.36+
<cristian_c>
rk30xxnand_ko.ko.3.0.8+
<AstralixNB>
you need the module that matches your SOC and your Loader and you kernel. So if you have a 3.0.8 kernel you need the 3.0.8 driver.
<AstralixNB>
Same for the 3.0.36+
<AstralixNB>
Check if it is loaded or if the /var/log/kern.log or dmesg contains some errors about loading this driver
<cristian_c>
ok
<cristian_c>
AstralixNB, only a question: /var/log in my pc system?
<AstralixNB>
nope, in cwm
<cristian_c>
there isn't a /var directory
<AstralixNB>
just call dmesg and see
<cristian_c>
ok
<AstralixNB>
I assume you do not have a serial console tapped to the device?
<cristian_c>
AstralixNB, I'm using adb
<AstralixNB>
that works fine
bengal has joined #linux-rockchip
bengal_ has joined #linux-rockchip
bengal has quit [Remote host closed the connection]
bengal_ has quit [Remote host closed the connection]
bengal has joined #linux-rockchip
else- has joined #linux-rockchip
<cristian_c>
AstralixNB, always same error messages, repeated cyclically
<cristian_c>
<4>[ 4.194238] [<c043f390>] (unwind_backtrace+0x0/0xf8) from [<bf00f784>] (FlashReadErrorDump+0xb0/0xe4 [rk30xxnand_ko])
<cristian_c>
<4>[ 4.194305] [<bf00f784>] (FlashReadErrorDump+0xb0/0xe4 [rk30xxnand_ko]) from [<bf010f70>] (FlashReadEnhanced+0x238/0x270 [rk30xxnand_ko])
RayFlower has quit [Read error: Connection reset by peer]
<AstralixNB>
Looks like you should erase and flash all new
RayFlower has joined #linux-rockchip
<cristian_c>
lol
Avasz has quit [Read error: Connection reset by peer]
<AstralixNB>
Ups
<AstralixNB>
FlashReadRetry error...
<AstralixNB>
Ah... did you change the Loader?
<AstralixNB>
The Loader and the rknand_ko.ko must be et least of the same generation, so 1.xx or 2.xx
<cristian_c>
AstralixNB, no, i've installed only cwm recovery
<cristian_c>
I didn't change the loader
<AstralixNB>
The cwm has its own rootfs and needs to have the matching rknand_ko.ko inside.
<AstralixNB>
You need to fetch the working version from your system.img and put it in the rootfs of your cwm
<cristian_c>
ok
bengal has quit [Ping timeout: 245 seconds]
naobsd has joined #linux-rockchip
RayFlower has quit [Read error: Connection reset by peer]
RayFlower has joined #linux-rockchip
<naobsd>
well
<naobsd>
loader 1.x can boot from SD too
<naobsd>
but there is no reason to use 1.x loader
<naobsd>
^about RK3188
<naobsd>
ganbold_: did you try boot from USB (via rkflashtool)?
<ganbold_>
naobsd: on RR? no
<AstralixNB>
You're right. But 2.xx feels like beeing somewhat faster
<naobsd>
ganbold_: I heard it's handy to run NetBSD kernel :)
<ganbold_>
naobsd: you mean like something like fastboot?
<naobsd>
ganbold_: there is fastboot function in u-boot for RK, probably it may work. but what I tried is different protorol (=implemented in rkflashtool)
<AstralixNB>
I am currently trying to get a desktop on my RRs. But these RRs are early-birds. So I tried almost all images provided by RR and none really works in a usable way
<naobsd>
AstralixNB: I didn't try recent images from Radxa (Aug/Sep)
<naobsd>
AstralixNB: old image for full/lite works fine on my ES ver.
<AstralixNB>
I thought so, as they work... somehow... but not really good.
<naobsd>
AstralixNB: at least LCDC configuration must be same
<AstralixNB>
The non-pro desktop version throw an error about lcdc1 and do not start a screen
<AstralixNB>
the server version works, and if you apt-get install xubuntu-desktop you get e desktop
<naobsd>
AstralixNB: but recent Linux image uses stable-3.0 branch, it should have issues around framebuffer
<AstralixNB>
but networking, bluetoth and even simple user-management fails to work without some hand-crafting
<naobsd>
stable-3.0 is came from Rockchip tablet SDK
<naobsd>
and Radxa patched(import RBOX code) a lot
<naobsd>
they're fixing issues in stable-3.0
<AstralixNB>
Ok, nice hint.
<AstralixNB>
I am pretty sure that there are much mor things to fix that do not link to the kernel
<ganbold_>
naobsd: via u-boot? cool
<naobsd>
I cannot count how many issues there ;)
<naobsd>
ganbold_: RK's u-boot is very limited, don't expect too much ;)
<AstralixNB>
As I said, putting xubuntu on top of the server gives a login and a desktop, but simple system-settings applet crashes
<naobsd>
I didn't try stable-3.0 based recent images
<AstralixNB>
I'll check that. I am just collecting some data.
<naobsd>
I'm using -kk based kernel (probably, I forgot it) for long time as lubuntu desktop
<naobsd>
I'm writing driver on it ;)
<naobsd>
(not for it!)
<naobsd>
^on my job
<naobsd>
if I can use RR as a target board on my job, it should be nice ;)
<naobsd>
(unless customer wants GPU/VPU thing ;)
<AstralixNB>
I want to use it at home as a replacement for an old notebook.
<naobsd>
4core A9 is faster than old single core Pentium4 (HT) ;)
cristian_c has quit [Quit: Sto andando via]
<AstralixNB>
Sure, if only GPU and VPU would work a bit.
RayFlower has quit [Read error: Connection reset by peer]
RayFlower has joined #linux-rockchip
nighty^ has joined #linux-rockchip
<AstralixNB>
naobsd, you say that I use the KK kernel mentioned above for booting lubuntu
<AstralixNB>
ok, I check that tonight
<naobsd>
radxa-dev-kk branch is came from RBOX SDK, it doesn't have issues in recent stable-3.0 based image
<naobsd>
(I never said it's bug-free;)
<AstralixNB>
Hmm... I can live with some bugs. But the non-pro radxa images I tried even didn't show any Tux while booting. So there must be either something really different between preview version and radxa lite or pro or there must be something pretty damaged.
<AstralixNB>
Probably I'll check schematics too
AstralixNB has quit [Remote host closed the connection]
<userx__>
ahh .. it seems that SPI driver is not working properly on RK3188. Is there any know issue about that ?
<naobsd>
userx__: I have no answer, but I think you should explain which kernel are you using (at least)
<userx__>
naobsd: it's kernel from android kitkat package. Version 3.0.36+, but I did not find any changes in SPI code (rk29_spim.c) with other kernels from repository
GriefNorth has quit [Ping timeout: 272 seconds]
<userx__>
naobsd: my problem is that i'm not able to send content less then 32 bytes. Looking with oscilloscope i can't see any clock and data.
<naobsd>
userx__: then non-dma transfer is broken?
naobsd has quit [Quit: Page closed]
RayFlower has quit [Read error: Connection reset by peer]
RayFlower has quit [Read error: Connection reset by peer]
RayFlower has joined #linux-rockchip
<userx__>
naobsd: I dont know, but it is somehow related, because can send 33 bytes and not 32.
<userx__>
mmind00: interesting, but I even don't have such file spi-rockchip.c in my kernel tree
<karlp>
he wasn't suggesting to look at it in your tree
<karlp>
he was suggesting comparing what the mainline spi driver, submitted by rockchips staff, is like vs what's in your (presumably) old vendor hacked android 3.0.x or 3.10.x kernel
c0d3z3r0 has quit [Ping timeout: 272 seconds]
<mmind00>
yep exactly :-)
c0d3z3r0 has joined #linux-rockchip
RayFlower has quit [Read error: Connection reset by peer]
RayFlower has joined #linux-rockchip
<userx__>
those files spi-rockchip.c and rk29_spim.c are totally different.
<userx__>
Anyway what is the latest kernel on Radxa?
<karlp>
whatever you install?
naobsd has joined #linux-rockchip
RayFlower has quit [Read error: Connection reset by peer]
RayFlower has joined #linux-rockchip
cyrozap has quit [Ping timeout: 244 seconds]
RayFlower has quit [Read error: Connection reset by peer]
cyrozap has joined #linux-rockchip
RayFlower has joined #linux-rockchip
<userx__>
yeah, but I need something relatively stable, that can run Android 4.4.2. It seems that mainstream kernel have issues to be solved (regarding radxa wiki).
mrcan_ has quit [Ping timeout: 240 seconds]
RayFlower has quit [Read error: Connection reset by peer]
RayFlower has joined #linux-rockchip
<rperier>
userx__: radxa wiki is not up-to-date about this part. I can run a yocto-based distribution with systemd throught nfs on my RR
<rperier>
the mainline is still in development but several stuffs work just fine on radxa
<karlp>
yeah, but he wants andr0iD 4.4+++xxx!!
<rperier>
oh he wants absolutely android ^^
<karlp>
can android jsut be run as a usermode process?
<userx__>
in fact i found some repo .. /lsk-android-14.04-radxa-rock
<naobsd>
it's not maintained
<mmind00>
naobsd: especially the documents/*/*User manual*.pdf files in that repository
<naobsd>
mmind00: yes
<karlp>
heh, jon smirl's already filing bugs on missing sections ;)
<naobsd>
2nd impact from allwinner ;)
<karlp>
naobsd: what doo you mean?
userx___ has joined #linux-rockchip
<naobsd>
karlp: impressive!
RayFlower has quit [Read error: Connection reset by peer]
RayFlower has joined #linux-rockchip
userx____ has joined #linux-rockchip
userx__ has quit [Ping timeout: 246 seconds]
userx___ has quit [Ping timeout: 246 seconds]
RayFlower has quit [Read error: Connection reset by peer]
RayFlower has joined #linux-rockchip
naobsd has quit [Quit: Page closed]
naobsd has joined #linux-rockchip
RayFlower has quit [Read error: Connection reset by peer]
RayFlower has joined #linux-rockchip
<rperier>
mmind00: would it make sense for you to submit a RFC on devicetree ML for "system-power-controller" ? (I plan to explain the issue, propose a generic property and a patch) I mean a RFC just for a property...
<rperier>
s/would/does/
<rperier>
(we agree that something must be proposed for that, I am just talking about the RFC itself)
bengal has joined #linux-rockchip
bengal has quit [Client Quit]
bengal has joined #linux-rockchip
hramrach_ has quit [Remote host closed the connection]
hramrach_ has joined #linux-rockchip
RayFlower has quit [Read error: Connection reset by peer]
RayFlower has joined #linux-rockchip
GriefNorth has joined #linux-rockchip
FreezingCold has joined #linux-rockchip
<mmind00>
rperier: in general people want to see code, so simply implement a variant of what you're proposing [maybe even include a patch enabling it on the act8846] and send this as rfc
<mmind00>
this way you'll need a lot less words to "explain" what you want to do and more likely get poeple to comment. Of course not only include the devicetree list itself but also relevant maintainers [the devicetree guys and also Mark among them]
rz2k has joined #linux-rockchip
AstralixNB has quit [Ping timeout: 250 seconds]
RayFlower has quit [Read error: Connection reset by peer]
RayFlower has joined #linux-rockchip
FreezingCold has quit [Ping timeout: 245 seconds]
<rperier>
mmind00: ok, will do. thanks for your feedbacks ;)
wildea01 has quit [Quit: leaving]
RayFlower has quit [Read error: Connection reset by peer]
RayFlower has joined #linux-rockchip
akaizen has quit [Remote host closed the connection]
akaizen has joined #linux-rockchip
akaizen has quit [Ping timeout: 272 seconds]
RayFlower has quit [Read error: Connection reset by peer]
akaizen has joined #linux-rockchip
RayFlower has joined #linux-rockchip
GriefNorth has quit [Ping timeout: 272 seconds]
cyrozap has quit [Quit: Client quit]
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
iPadre has joined #linux-rockchip
RayFlower has quit [Read error: Connection reset by peer]
RayFlower has joined #linux-rockchip
rz2k has quit []
akaizen_ has joined #linux-rockchip
RayFlower has quit [Read error: Connection reset by peer]
RayFlower has joined #linux-rockchip
akaizen has quit [*.net *.split]
<karlp>
is there a way to use i2c tools to talk to the power regulators that the kernel is using?
<karlp>
Error: Could not set address to 0x2d: Device or resource busy seems to imply the kernel doesn't want to share
<karlp>
nvm, i2cget -f -y works.
FreezingCold has joined #linux-rockchip
RayFlower has quit [Quit: minab476vw3 tedgfjkhjl,kyjtrheg6s]
RayFlower has joined #linux-rockchip
akaizen has joined #linux-rockchip
RayFlower has quit [Read error: Connection reset by peer]
RayFlower has joined #linux-rockchip
akaizen_ has quit [Ping timeout: 272 seconds]
akaizen_ has joined #linux-rockchip
akaizen has quit [Ping timeout: 260 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
nighty^ has quit [Quit: Disappears in a puff of smoke]
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
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
bengal has quit [Ping timeout: 258 seconds]
RayFlower has quit [Read error: Connection reset by peer]
RayFlower has joined #linux-rockchip
<dlezcano>
mmind00: what is the upstream stable branch I can use to boot my rk3288 ?
<mmind00>
dlezcano: what is "stable" for you ... I guess still devel/workbench in https://github.com/mmind/linux-rockchip ... + maybe devel/for_broken_bootloaders merged in
<mmind00>
dlezcano: if I remember correctly, the evb you had stopped booting at some point and we never found the source of this
<dlezcano>
mmind00: right, we deduced it was a hardware issue
<dlezcano>
mmind00: so I got another one I am impatient to boot
<dlezcano>
mmind00: I tried on the v3.17-next/rk3288
<dlezcano>
mmind00: and it boots to "calibrating delay loop..." and then get stuck
<dlezcano>
mmind00: dunno if this is the limit of the support
<mmind00>
which is caused by the broken bootloader not enabling the timer block, that supplies the architected timer
<mmind00>
dlezcano: i.e. naobsd added code for this to the publc uboot and I also have a uboot binary for testing, that does this
<dlezcano>
mmind00: ok. So I can alternatively update my u-boot or take your patch, right ?
<mmind00>
correct ... uboot being the preferred long-term solution, the patch being the short term fix :-)
<dlezcano>
mmind00: so what branch can I use in git.kernel.org ?
<mmind00>
dlezcano: I guess linux-next would be your best option for the most support [there is a ton of stuff that is queued for 3.18 and split over a lot of trees] ... for basic support just use master of Linus' tree directly
<dlezcano>
mmind00: you switched from git.kernel.org to to github ?
<mmind00>
dlezcano: nope ... I simply split where I push stuff ... git.kernel.org for stuff for mainline, github for the dirty code ;-)
<mmind00>
(especially, as the git.kernel.org branches are build-tested these days)
<dlezcano>
mmind00: ah ok
<dlezcano>
mmind00: ... but still a bit lost ... :)
<dlezcano>
mmind00: assuming I want to send a patch, against which tree/branch shall I send it to have it merged upstream ?
<dlezcano>
mmind00: I see different 3.18-next/... but I don't know which one to use
<mmind00>
dlezcano: currently I only carry devicetree changes ... so for regular dt-changes it would be v3.18-next/dts
<dlezcano>
mmind00: hmm, ok. Thanks for the info.
<dlezcano>
mmind00: FYI, the board successfully booted (finally)
<dlezcano>
mmind00: it fails to "VFS: Cannot open root device "mmcblk0p1"
<dlezcano>
mmind00: but that's another story
<mmind00>
dlezcano: yeah :-D
<mmind00>
dlezcano: should've been yay :-)
<dlezcano>
:)
<mmind00>
dlezcano: just to finish explaining the branches, arm-soc (which takes my branches) generally splits changes into topic-branches like "soc", "dt", "drivers", etc so I keep a similar split ... and for 3.18 there were only devicetree changes going this way
<dlezcano>
mmind00: aah, ok. Hence, the usage of linux-next.
<naobsd>
I should try devel/workbench again to check cpufreq/regulator issue...
<naobsd>
mmind00: what's "upstream uboot"?
RayFlower has quit [Read error: Connection reset by peer]
RayFlower has joined #linux-rockchip
<mmind00>
naobsd: not sure I understand :-) ... there is the public uboot source you added the timer support for ... and I have a binary containing this and some other preliminary changes, but no sources .. but hopefully this will trickle down to the public sources once finished
<mmind00>
anyway, going to sleep now :-)
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
<naobsd>
mmind00: I see. good night :)
RayFlower has quit [Read error: Connection reset by peer]
RayFlower has joined #linux-rockchip
<naobsd>
there is u-boot binary ver. 2.15.01 and 2.15.02 in SDK, but source is not updated
<naobsd>
I'm not sure I tried arch timer with 2.15.0[12]
<naobsd>
hmm, 3.18 merge window? when it'll be closed?