<rz2k>
new changes to sunxi-serial killed it for linaro (nothing after <6>sunxi-uart.0: ttyS0 at MMIO 0x1c28000 (irq = 1) is a U6_16550A) and even more, I have constant buffer scroll-up to last line, this means that serial actually sends some tty info
<rz2k>
cant clearly say what commit, because I had it yesterday before new stuff from Hans
<rz2k>
ArchLinux had same waaay long before.
drgreenthumb has quit [Read error: Connection reset by peer]
drgreenthumb has joined #arm-netbook
drgreenthumb has quit [Changing host]
drgreenthumb has joined #arm-netbook
pucko_ has quit [Quit: leaving]
<rz2k>
managed to get earlyprintk working with removing console=ttyS0
hg_5 has quit [Read error: Connection reset by peer]
pucko has joined #arm-netbook
<rz2k>
ssvb: techn: is '<6>disp clks: lcd 148500000 lcd_div 1 hdmi 148500000 pll 297000000 2x 0' ok? I see pastes with ' disp clks: lcd 74250000 lcd_div 1 hdmi 74250000 pll 297000000 2x 0'
<rz2k>
WarheadsSE: linaro now has same crap with serial output after some commits to sunxi-serial
<WarheadsSE>
mm, which gcc
<rz2k>
no idea what it is, I just removed console=ttys0 and enabled earlyprintk
<rz2k>
4.7.2
<WarheadsSE>
k
<Turl>
nemik: cd vendor/allwinner; cp -a zatab mk802ii; cd mk802ii; rename 's/zatab/mk802ii/' zatab*; sed -i 's/zatab/mk802ii/g' *mk
<Turl>
nemik: the ones from your device won't work, they're older (r2p4 iirc) and the kernel tree has r3p0 modules
xman has quit [Quit: Ex-Chat]
Sam4ever has joined #arm-netbook
<Sam4ever>
I have a question, when I connect the mini usb (press and hold the FeL key) to the windows XP 32 bit laptop, I always get one notice for installing USB driver but not 2, does this happen to anyone here as well? I can't flash the NAND using livesuit
<nemik>
Turl: ahhh ok so this build i made is useless then, i should do android clean?
mSquare has joined #arm-netbook
<Turl>
nemik: nah, just fix that and build again without cleaning, it should suffice
<Turl>
you can cd android and "make installclean" to be even more sure
<nemik>
Turl: cool, thanks. what about the kernel config in there though? in BoardConfig.mk. there's gyro stuff not applying to mk802
<nemik>
also, i had a problem with the TARGET_KERNEL_CONFIG := mk802ii_defconfig and had to change it to TARGET_KERNEL_CONFIG := sun4i_crane_defconfig
<nemik>
not sure if that was the right way to do it though
<Turl>
you should talk with servili007, he was working on that device
<Turl>
he probably hasn't commited all of his work yet or something
<servili007>
Hello
<nemik>
hello!
<nemik>
this is convenient
<nemik>
servili007: were you able to build an android livesuit image for the mk802ii? it's what i'm trying to do
* rz2k
posted topic @ linux-sunxi with his problems. feels like a total noob.
<rz2k>
(for not beign able to fix this on my own
<servili007>
nemik: What I did was make a CWM sd card tailored to the mk802ii instead. Livesuit images had a tendency to break it, and phoenixcard refused to flash on user made images
<libv>
rz2k: did Xorg work earlier
<rz2k>
yes
<servili007>
nemik: MY user made images, that is. There is a always a chance that I overlooked something
<nemik>
servili007: :( damn. i was hoping to replace the stock one with my own
<nemik>
i'm doing it all using the sunxi-bsp tools
<libv>
rz2k: so this is a regression
<servili007>
nemik: You're building android through sunxi-bsp?
<libv>
rz2k: this was not clear from your email
<rz2k>
somewhat, ssvb has everything fine
<rz2k>
sorry
<nemik>
servili007: i'm trying to, yes
<libv>
rz2k: kernel bisect?
<servili007>
nemik: Did you pull my configuration down from the allwinner-dev-team repo?
<nemik>
well yes earlier, but Turl told me to do "nemik: cd vendor/allwinner; cp -a zatab mk802ii; cd mk802ii; rename 's/zatab/mk802ii/' zatab*; sed -i 's/zatab/mk802ii/g' *mk"
<rz2k>
I believe that this affects rev.C only, I will try to revert Hans's commits to disp tomorrow and find when this did happen.
jeremb_ has quit [Quit: Connection closed for inactivity]
<servili007>
those are only the proprietaries
Sam4ever has quit [Quit: Page closed]
<nemik>
when i pulled yours down, it complained to me about not finding libMali.so so i copied it from my device
<nemik>
but turns out that probably won't work, so good thing i didn't try to flash the resulting image
<servili007>
well, it's pretty hard to perma-brick it, so no worries
<servili007>
on occasion the build will fail on libmali, clean out and retry
<servili007>
the stock files will work
<nemik>
hmm should i try to build libUMP and mali kernel stuff the sunxi-linux repo first? or add those somewhere?
<servili007>
I do however need to add a readme, there are a couple of edits you should make to the common device files to get things sorted
<nemik>
the bsp does a lot of magic so i'm not sure if it did all that
<servili007>
No, that isn't necessary, make android from sunxi-bsp will only work within the android folder, which handles everything
<servili007>
don't try to pull other items from sunxi-bsps other submodules
<nemik>
that's what i thought
<nemik>
ok, so when i did the cp rename thing with the zabat device, it first told me i had no mk802ii_full.makefile, so i copied that into it
<servili007>
you should be pulling down the whole thing into device/allwinner
<servili007>
and rename it to mk802ii
<servili007>
Turl: ping, got a minute?
tinti has joined #arm-netbook
<Turl>
servili007: pong
<servili007>
nemik: be aware, the overscan/underscan is locked to 95% of usable screen real estate, audio might not be working, and script.bin is your only way to set resolution
<servili007>
nemik: also, to install it, I can provide you with an SD card image for cwm and a script.bin for CM10
<nemik>
servili007: ouch...i was hoping i could come close to the stock image
<servili007>
nemik: keep an eye open for a readme I'm going to add to the mk802ii device repo, I'm going to detail some changes for partition layouts and possibly an audio fix
<rz2k>
ssvb: sorry for confusion!
<rz2k>
ssvb: I have rev.C
<rz2k>
you have older revision that has no problems, we had problems with SATA/Mali-400 clocks sharing on rev.C before.
<nemik>
servili007: awesome that sounds great. i've patched up the makefile in sunxi-bsp for it to create livesuit.img files so that's what i'm hoping to flash and try
<rz2k>
ssvb: EDID can be overriden with command line, I have same behaviour as VGA if I provide 720p as override.
<nemik>
ultimately, what i'd like to have is an mk802ii stick with kernel hci_usb drivers for bluetooth so i could enable bluetooth on it from USB fobs
<rz2k>
ssvb: and vga should be somewhat same.
<servili007>
nemik: it might not find what it's looking for. The building process is different for android
<nemik>
hmm ok i'll read up on it, although people were saying apparently cm10 did it? and the sunxi image is based on it so...maybe works?
<ssvb>
rz2k: ok, if it's the case of hardware revision dependent differences, I can only wish you good luck with debugging this
<ssvb>
rz2k: I'm generally trying to fix any problems that I manage to encounter myself
<rz2k>
yeah, I understand.
<TestModule>
Is the 3.4 stable ?
ZaEarl has joined #arm-netbook
<ssvb>
rz2k: but bisecting is really something that you can do easily
<TestModule>
So I am attempting this BSP, and I got this: "make: *** [u-boot] Error 2"
<TestModule>
on the CC
hipboi has joined #arm-netbook
hipboi_ has joined #arm-netbook
hipboi has quit [Ping timeout: 260 seconds]
rz2k has quit []
aholler has joined #arm-netbook
aholler_ has quit [Ping timeout: 248 seconds]
hno has quit [Ping timeout: 240 seconds]
Ershov has joined #arm-netbook
XenGi is now known as XenGi_
jardiamj has joined #arm-netbook
rz2k has joined #arm-netbook
Ershov has quit [Quit: Ershov]
Ershov has joined #arm-netbook
hno has joined #arm-netbook
jardiamj has quit [Ping timeout: 252 seconds]
jardiamj has joined #arm-netbook
Ershov has quit [Read error: Connection reset by peer]
rz2k has quit [Read error: Connection reset by peer]
Ershov has joined #arm-netbook
Ershov has left #arm-netbook [#arm-netbook]
rz2k has joined #arm-netbook
servili007 has quit [Read error: Connection reset by peer]
jlj has quit [Ping timeout: 248 seconds]
jlj has joined #arm-netbook
gimli has joined #arm-netbook
ZaEarl has quit [Ping timeout: 276 seconds]
muda has joined #arm-netbook
muda has quit [Client Quit]
gimli has quit [Ping timeout: 248 seconds]
Alex1269 has joined #arm-netbook
pcat has joined #arm-netbook
hansg has joined #arm-netbook
arokux has joined #arm-netbook
jardiamj has quit [Ping timeout: 256 seconds]
<Alex1269>
Hi, all. I found some info about sunxi-gpio driver in irc logs but can't find it on git... Is it usable ? I'm going to write normal one...
<mripard>
Alex1269: what do you need to write one for ?
<rz2k>
mripard: he wants to do it the right way with drivers/gpio for 3.0-3.4 kernels I believe.
<mripard>
because, I just wrote one that I expect to be merged for 3.9
<mripard>
so maybe, you can just backport my patches.
jardiamj has joined #arm-netbook
<mripard>
maybe only for 3.4 though, it uses pinctrl and I'm not sure it was there at the 3.0 time
<Alex1269>
I'm planning to use cubie as home automation computer to control ventilation, heating, env control, auto light, etc... So I need input and output...
<Alex1269>
I've allready made some kind a "driver" for onboard (and connectable) leds using generic leds-gpio as base (/dev/class/leds) to use with triggers - i put it on cubie forum...
<mnemoc>
it's from a recent source drop of allwinner code
hansg has quit [Ping timeout: 240 seconds]
gzamboni_ has joined #arm-netbook
lkcl_ has joined #arm-netbook
Jef92 has joined #arm-netbook
Jef91 has quit [*.net *.split]
hipboi_ has quit [*.net *.split]
lkcl has quit [*.net *.split]
TheLarch has quit [*.net *.split]
gzamboni has quit [*.net *.split]
pwhalen has quit [*.net *.split]
<hramrach>
libv: yes and no. GLX and EGL is different interface
pwhalen has joined #arm-netbook
TheLarch has joined #arm-netbook
TheLarch has quit [Changing host]
TheLarch has joined #arm-netbook
hipboi_ has joined #arm-netbook
<hramrach>
rz2k: you possibly need console=ttyS0,115200
<hramrach>
early prink uses whatever is set up by u-boot. after loading serial driver the port is reconfigured and when the paramaters are wrong the output stops
<hramrach>
and getty worked again because it again reconfigured the serial line
<rz2k>
interesting
<rz2k>
because it was working some time ago with default u-boot boot.scr
<rz2k>
that doesnt set the connection speed
<mnemoc>
can you test the latest -r1 ?
<mnemoc>
that's just before the hdmi/edid changes
<mnemoc>
either in 3.0 or 3.4
pcat has quit [Ping timeout: 240 seconds]
<TestModule>
i'm getting the same error
<mnemoc>
-r2 is after the changes but beofre the version jump
<hramrach>
maybe there is kernel config somewhere that sets the default speed
<hramrach>
and maybe the driver just grew the ability to set teh speed sometime in-btween
<mnemoc>
115200 is the default speed
<hramrach>
it is not
<mnemoc>
.oO
<hramrach>
you haveto set it on kernel command line
<hramrach>
otherwise the port won't work as console
<hramrach>
u-boot sets that speed but the kernel driver when loaded sets some other default unless the speed is explicitly set
<mnemoc>
so the kernel driver is screwing it up
<mnemoc>
good to know
<hramrach>
it is not. it does what you tell it and when you don't tell it what speed to set it sets something else
<mnemoc>
if you don't tell it, it should not touch it
<mnemoc>
and keep the hardware default of 115200
<hramrach>
it should configure the port. it's the driver. it's its job
<hramrach>
it amy be even that the serial console driver asks the serial driver to set the speed
jquip has joined #arm-netbook
<TestModule>
Are there any missing steps to the BSP related firststeps wiki entry?
pcat has joined #arm-netbook
rz2k has quit [Read error: Connection reset by peer]
<mnemoc>
TestModule: what do you miss?
<TestModule>
Oh, it was just as a question.
<TestModule>
Is it more of a summary or is it exact
<mnemoc>
by rule documentation is *always* some steps behind the code
<mnemoc>
and that's exactly why we have a wiki ;-)
<TestModule>
I'm on the wiki?
<mnemoc>
it's a cycle. you read the docu, try to do the stuff, report problems, we solve doubts and fix code, you update the wiki ;-)
tinti is now known as tinti_
<mnemoc>
TestModule: if not, register
<TestModule>
Ohhh. I don't know that I would be much help but I will try.
mSquare1 has joined #arm-netbook
tinti_ is now known as tinti__
tinti__ is now known as viniciustinti
mSquare has quit [Ping timeout: 240 seconds]
viniciustinti is now known as viniciustinti_
viniciustinti_ is now known as viniciustinti__
viniciustinti__ is now known as tinti
tinti has quit [Quit: Leaving]
tinti has joined #arm-netbook
<TestModule>
Ok so the git for the linux-sunxi won't clone but the rest will... ?? I recall it used to clone long ago but that was before i used BSP / who knows
<mnemoc>
also, if you have another close of linux you can cp -a it and then add the extra remote
<mnemoc>
it will be faster
<TestModule>
boom, works. I should've thought of that, but i'm just trying the listed ways
<mnemoc>
it's a wiki, improve it :)
<TestModule>
kk
Quarx has joined #arm-netbook
<mripard>
stefanro: ping?
hansg has joined #arm-netbook
<stefanro>
mripard: hi
<mripard>
stefanro: hi :)
<stefanro>
mripard: what's up?
<mripard>
fine, and you ?
<mripard>
(happy new year to all of you btw(
<stefanro>
yeah, thanks! same to you
<mripard>
I just wanted to know, what's the state of your wemac driver? did you had some time to work on it?
<stefanro>
ahh, no not at all
<stefanro>
do you want to take over?
<stefanro>
i won't have time to look at it in the next weeks
<mripard>
I'm pretty busy at the gpio and pinctrl right now, so maybe later, but not right now
<stefanro>
okay, just let me know if you need anything
<mripard>
it's just that it could be a great thing to have in 3.9, and we are quite close to the end of the time frame to expect a new driver to be merged in 3.9 :)
<stefanro>
yes, would be great - but i won't make it (buried deep on $$ work)
<mripard>
yes, I understand :)
jlj has quit [Ping timeout: 265 seconds]
<mripard>
do you have the patch somewhere?
<mnemoc>
damn $work$ always an obstacle to do important things in life
<mripard>
I'll try to take a look if I have some time
jlj has joined #arm-netbook
Reepicheep12 has joined #arm-netbook
<TestModule>
rofl, how many gigs is linux-sunxi?
<mnemoc>
my guess is "none"
<mnemoc>
err
<Reepicheep12>
Hey all, wondering if there is a uboot out there I can use to boot from nand (with the uboot also on nand), no mmc involved?
<TestModule>
Ah
<mnemoc>
read "girls"
<mnemoc>
Reepicheep12: work on (proper) nand support is still in progress
<Reepicheep12>
Ah alright so at the moment my only option is to have a uboot and uEnv.txt on the sd card and tell that to boot from the nand?
<Reepicheep12>
I'm on a Mele A1000G so the boot goes sd>nand>usb.
<mnemoc>
Reepicheep12: you can also use lichee-dev which is nand-only
<mnemoc>
but were is the fun when droping-in prebuilts...
<Reepicheep12>
Thanks I'll try that. So do I need to have any boot parameters in uEnv.txt or will look for uImage in the nand /linux/uImage by default?
<mnemoc>
lichee-dev doesn't support uEnv.txt/boot.scr
<Reepicheep12>
Yeah very true mnemoc but I've used a lot of technical elbow grease over the last two weeks so any respite is welcome.
<mnemoc>
it's just a free replacement with fixed ATAG_MEM
<Reepicheep12>
So is there any way to add boot parameters to the kernel? Can I recompile it with those parameters? I need to add mem=1024M.
<mnemoc>
uart -> interrupt u-boot -> adjust the env
<mnemoc>
Reepicheep12: mem= shouldn't be needed
<Reepicheep12>
Does that persist across reboots?
<mnemoc>
Reepicheep12: if you want to omit the mali/ve/fb reserves, disable them when building the kernel
<mnemoc>
if you saveenv, it will persist
jquip has quit [Quit: jquip]
<Reepicheep12>
I disabled them in script.bin and I was left with around 350. It seems to think I only have 512M for some reason.
<mnemoc>
disabling them in script.bin is useless
<mnemoc>
reserves happen before script.bin can be read
<mnemoc>
if you have 512M-reserves on a 1G device you are using the wrong bootloader
<Reepicheep12>
By that you mean uboot?
<Reepicheep12>
The only file I have changed on the nanda partition from the factory defaults is the uboot file.
<mnemoc>
if booting from uSD you need a u-boot-spl built for *your* device, with your proper dram params
jquip has joined #arm-netbook
<Reepicheep12>
I'm booting just from the nand (or attempting to) no sd involved.
<mnemoc>
if you are booting from nand, ram is initialized by the proprietray boot0/boot1... and you'll need to replace nanda's linux/u-boot.bin with lichee-dev's, because the stock one doesn't apps ATAG_MEM
<mnemoc>
and rebuild your kernel with mali/ve/g2d/fb disables to get your full 1G
<mnemoc>
disabled*
<Reepicheep12>
Alright, well I have no experience compiling uboot so this'll be a journey. Is the only thing I need to change the ATAG_MEM variable?
<Reepicheep12>
And I think I remember that there is a mele a1000 config option built in.
<Reepicheep12>
So I don't need to compile my own to get the full 1GB on the A1000G?
<TestModule>
but main is still on 3.0.39 or something, correct?
<mnemoc>
TestModule: but both sunxi-3.x are maintained in parellel atm
<TestModule>
ahhh
<mnemoc>
TestModule: no... v3.4.24 and v3.0.57
<Reepicheep12>
And are all the modules built into the kernel?
<mnemoc>
that's up to you
<mnemoc>
but the defconfig makes most of the platform drivers built-in
<Reepicheep12>
Ah okay, 'cause I need sw_ahci_platform, snd_usb_audio and the gpio driver to change the led colour (whatever that's called). Might have to compile them.
<mnemoc>
there is a real led driver for sunxi-3.0/3.4 posted on the cubieboard list. hopefully Alex1269 will submit it to linux-sunxi's ML soon
<Reepicheep12>
Very hopeful. I like working with obscure hardware like this but the fewer soldiers out there has its disadvantages. Finally have to sort things out for myself like back in the day.
<Reepicheep12>
I'm going to give this a go thanks greatly for your tips.
<mnemoc>
my pleasure
<hramrach>
I am running linux 3.4 on A10 and it resonably works
<hramrach>
self-compiles, runs gnome, HDMI and serial works
<hramrach>
did not try sata and nand
<mnemoc>
hramrach: using the latest hdmi/edid changes?
<hramrach>
up to the HACK v1
<hramrach>
sometimes locks up temporarily. not sure what/where is stuck
<hramrach>
but if it was the card surely at least mouse cursor would move
<TestModule>
mnemo, have you ever tried with LTO flags?
<mripard>
stefanro: thanks :)
hipboi_ has quit [Quit: Leaving]
<stefanro>
mripard: your welcome - again, please feel to contact me if you need anything - the driver still needs lots of cleanups
<mripard>
well, like I said, I'm not sure I'll be able to work on it before you do, but ok :)
<mnemoc>
TestModule: no
sspiff has joined #arm-netbook
<sspiff>
is there an AOSP or CM based jelly bean tree for allwinner A1x SoCs?
tinti has joined #arm-netbook
<mnemoc>
yes
<mnemoc>
CM
jquip has quit [Ping timeout: 252 seconds]
<sspiff>
mnemoc: where can I find it?
<oliv3r>
spiff: did you search on the wiki? it should be up there if i'm not mistaken
<mnemoc>
leowt: not aware of anyone doing a gst plugin for cedarx
<mnemoc>
Reepicheep12: neither. build your own
<TestModule>
Can I pass a flag for 3.4 in the BSP?
<mnemoc>
if you really want a prebuilt, stole it from a hdpack
<mnemoc>
TestModule: flag?
<Reepicheep12>
Yes I plan to compile one but I just want to see if I can get the uboot working first.
<Reepicheep12>
Thanks.
<TestModule>
-*
<sspiff>
oliv3r: all I seem to find is an ICS repo (the matson-hall one?)
<mnemoc>
Reepicheep12: that dir has timestamped subdirs with per-device hwpacks inside
<leowt>
so, mnemoc, in a X11 env, can i use a10's co-processor
<leowt>
?
<mnemoc>
Reepicheep12: you can take the kernel and modules from there.... but it's a boring defconfig
popolon has joined #arm-netbook
<mnemoc>
leowt: mali (3D), yes. 2D, not yet
<sspiff>
mnemoc: are the amery repos for android the ones to use?
<leowt>
mnemoc, what about hd video decoding?
<leowt>
in a X11 env
<TestModule>
Reepicheep12 , I just made the BSP directions easier ;)
<mnemoc>
sspiff: what do you mean by "amery repos"?
<leowt>
btw, dont they have a x11 driver?
<mnemoc>
video decoding has nothing to do with x11
<sspiff>
mnemoc: the android ones on the amery github account
<Alex1269_>
mnemoc, i was out (driving home)... Yes i'm regressor. I've read your sunxi-gpio code and yes - it is not for 3.0/3.4 so i'll write /dev/class/gpio basic driver.
<mnemoc>
sspiff: amery's github is a personal repo. use linux-sunxi's repos
<sspiff>
no android ones in there?
<sspiff>
but oliv3r said there was a wiki page describing the process?
<oliv3r>
lmgtfy
<mnemoc>
Alex1269_: great! but submit the led driver patch first ;-) also, sunxi-led, not sun4i-led :)
<oliv3r>
but it's old and not accurate anymore I think
<oliv3r>
nono, YOU fix it, i'm still messing with the templates. Well I stare at them, and chringe
<mnemoc>
afaik `make android` on sunxi-bsp mostly works... after you add the device tree for your board
<oliv3r>
kringe* too; I keep putting it off, because I really, REALLY don't wanna do it :(
<oliv3r>
I litterally have had it on my screen all day
<Alex1269_>
well, i'm goinng to use sys_config.c gpio routines from sun4i platform code and i'm not sure if sun3i/sun5i have this.
<leowt>
are you guys maintaining linux-sunxi wiki and git?
<oliv3r>
leowt: yeap
<leowt>
so this is the "official" channel?
<mnemoc>
Alex1269_: yes, they have
<oliv3r>
leowt: more or less
<Alex1269_>
I'll send led driver as a patch. What patch should I use ? git-like with all defconfig changes + Makefile + Kconfig, etc or just driver file ?
<oliv3r>
leowt: the linux-sunxi people kinda hijacked it
<leowt>
i see
<leowt>
cool
<mnemoc>
Alex1269_: the whole thing and a git formatted commit
<oliv3r>
and the linux-sunxi mailling list of course
<mnemoc>
leowt: but it's a taken-over channel anyway
<leowt>
u pirates
<leowt>
=P
<mnemoc>
we haven't been evicted yet :p
<sspiff>
hmmmmm, repo init won't accept my -b jb
<Reepicheep12>
I'm confused about where to put the lib .so files in ArchLinux (usr/lib or lib). Do I need them to boot or just the modules folder? They look graphics related and I don't need that. Oh and TestModule, thanks a bunch.
<mnemoc>
Reepicheep12: WarheadsSE is our resident arch expert
<WarheadsSE>
Reepicheep12: /lib is a symlink to /usr/lib
<Reepicheep12>
Yeah not in the old rootfs on the Archlinuxarm website which is still on the old glibc.
<Reepicheep12>
Unforunately.
<leowt>
what is the status of mali X11 driver?
<WarheadsSE>
I do need to make a new one
<WarheadsSE>
and a get a new kernel package put together
<Reepicheep12>
Going to try with just the modules. That would be a delight Warheads. Appreciate all your work and assistance so far.
<leowt>
do anyone uses X11?
<leowt>
in A10
ganbold_ has joined #arm-netbook
<TestModule>
Many do.
<TestModule>
It's included with all the non-headless premade images floating around.
<leowt>
TestModule, does it support 3d/2d?
<TestModule>
Acceleration? No idea. Is it usable? Yes.
<leowt>
TestModule, well, if it is usable its ok
<TestModule>
I've had Unity, E17, Gnome, KDE, and Openbox running fine.
<leowt>
cool
datagutt has joined #arm-netbook
<leowt>
Unity is sluggish?
<TestModule>
Yes
<Reepicheep12>
So I added the modules and I get the same thing, hangs at "[ 3.878] jump to".
<Reepicheep12>
Maybe I should try copying the kernel over again.
<leowt>
but, where you running from sd?
<TestModule>
LOL yes, but my SD pushes 31MB/s
<TestModule>
and sick 4k reads/writes
<TestModule>
Unity is just bloated crap.
<leowt>
TestModule, a sata disk can double that
<leowt>
i think, i didnt test it
<TestModule>
Yes, but that's faster than the NAND which for me came in at 24MB/s
<leowt>
btw, are you able to write in the nand?
<TestModule>
Yes, it's mountable.
<leowt>
cool
<leowt>
that means i can put uboot + kernel img there?
<TestModule>
No idea, you should prob look at a tutorial for what is required for NAND boot. miniand.com's forums have a few.
<leowt>
well, at least i want to get rid of SD
jquip has quit [Ping timeout: 265 seconds]
<leowt>
sure do
<leowt>
tnks
<leowt>
btw
<leowt>
vlc cedar plugin is usable?
<TestModule>
No idea, I code and surf the web on mine.
<oliv3r>
leowt booting from nand is tricky atm
<leowt>
oliv3r, i see
<leowt>
TestModule, websurfing experience? good?
<Reepicheep12>
So I was using the wrong u-boot.bin like a chump. using the right one now but I get a more exciting error message. http://sprunge.us/JGeb
<Reepicheep12>
Am I supposed to setenv something?
<mnemoc>
if anyone here is going to start a cedarx-compatiblity plugin for his favourite player, PLEASE make a libva plugin instead of something framework/player specific
<leowt>
wich browser
<TestModule>
Yes , works well in Midori
<leowt>
cool
<leowt>
No lag on scroll?
<oliv3r>
mnemoc: libva is the framework of choice yes? not openmax! :p
<TestModule>
No, but i have it set to cache to ram, no ads, smooth-scrolling, etc
<mnemoc>
oliv3r: you can see libva as omx done right
<TestModule>
report as ipad useragent
<oliv3r>
i know on the ML someone worked on a VLC library patch and i think ffmpeg too
<leowt>
cool
<mnemoc>
and it is supported by most players/framworks already
<oliv3r>
too true
<leowt>
anyways, all i want is to test as a embedded video player, remote controlled in a X11 env
<leowt>
and as a standalone voip server
<mnemoc>
cedarx also knows to decode/encode audio
<leowt>
Im RTFM just now
<TestModule>
leo, You can remote control the android env
<leowt>
btw, in your "FirstSteps", the u-boot im compiling is with mmc by default?
<leowt>
TestModule, i must not use android
<leowt>
=P
<TestModule>
;)
<mnemoc>
leowt: our u-boot is still mmc-only
<oliv3r>
mnemoc: any news on hipboi or A20 based devices? :)
<leowt>
TestModule, there's many things dependant of a debian based distro
<leowt>
mnemoc, got it
<leowt>
tnks
<mnemoc>
oliv3r: haven't heared a word about the topic
<oliv3r>
Can't be long though :( Hopefully we get a source dump soon too though
<andoma>
is there an uboot that can boot linux from nand?
<mnemoc>
leowt: we have a fixed branch of the original u-boot from allwinner (nand-only) but it doesn't get improvements
<oliv3r>
for A20 that is, couldn't care less atm about A31
<andoma>
on the a10, that is
<andoma>
mnemoc: that answered my question
<mnemoc>
andoma: the lichee-dev branch
<andoma>
mnemoc: ah ok thanks
<Reepicheep12>
This is the issue I was having originally. Trying to setenv but not sure what the boot parameters should be instead of "boot_normal=nand read 50000000 boot; boota 50000000". Tried "nand info" in the uboot command prompt but I don't get any partitions back. Is this expected or is it unable to read the nand?
<leowt>
mnemoc, is it capable of getting kernel image and rootfs from sda1?
<mnemoc>
leowt: rootfs mounting has nothing to do with u-boot
<mnemoc>
Reepicheep12: nand uboot loads it's env and kernel from raw partitions
<Reepicheep12>
Ah golly so I have to set them up. No idea how to go about that but I'll look it up. Should I be wiping nanda entirely to make a new partition for the kernel?
<mnemoc>
Reepicheep12: the proprietary bootloader on your nand (boot0/boot1) needs that nanda. you change it, it dies
<mnemoc>
you can replace u-boot.bin, edit script.bin and add a uImage. but not wipe out nanda entirely
<mnemoc>
once our u-boot learns to do sunxi-nand properly we will be able to replace boot0/boot1, but not there yet
<oliv3r>
mnemoc: who's mostly working on that besides hno?
<Reepicheep12>
Exciting, would like to be a part of something like this if I had the intimate knowledge. One day soon.
<Reepicheep12>
So this part "fatload nand 0 0x48000000 uImage" is looking for the uImage at location 0x48000000? How do I know where mine is located? 'Cause it finds the script.bin but not the uImage.
<leowt>
where/how do i put u-boot in sdcard without using adb?
<mnemoc>
oliv3r: slapin
<leowt>
forget the adb part
<leowt>
XD
<mnemoc>
Reepicheep12: that's the *destination* address
<mnemoc>
bbl
<Reepicheep12>
Yes I just found that out via internet sorry should have searched more extensively.
<Reepicheep12>
Oh and the default boot commands in this u-boot image is this "bootargs=console=ttyS0,115200 root=/dev/nandbinit=/init loglevel=".
<WarheadsSE>
help fatload should have said it too
<Reepicheep12>
Yes my apologies Warheads. Not sure if only I get that typo or you guys already knew about it but I thought I would mention it.
<oliv3r>
mnemoc: slapin too? Thought he was working on mdt for linux bit, not on uboot stuff AND that they gave up :(
<Reepicheep12>
Had to change it to "bootargs=console=ttyS0,115200 root=/dev/nandb init=/init loglevel=".
<slapin>
hi, all!
<oliv3r>
slapin!
<slapin>
have anybody recovered from celebraion?
<oliv3r>
no :(
<TestModule>
So BSP should be bootable OOTB?
<oliv3r>
mnemoc: can you delete a wikipage for me?
<TestModule>
once finished
<oliv3r>
mnemoc: Template:Device_List_Medium_Detail_Header (i recreated it with a different name)
* slapin
enjoys first working day after NY holidays, and have LOTS of LOLs and ROTFLs around. Can't work though :(
<oliv3r>
you start now? Been working for a full week allready!
<slapin>
anybody seen hno lately?
<oliv3r>
uhm .. yes! yes! He is hard at work on the mtd driver and was hoping you could work on it too! O:-)
<TestModule>
Ok so, do I still have to debootstrap if a full deb filesystem is already made and i use the linaro toolchain?
<TestModule>
Because BSP worked great, but the image won't load in the device.
<slapin>
oliv3r: are you sure?
<slapin>
oliv3r: I don't see much
<oliv3r>
slapin: yeah :( I was hoping i could convince you!
<oliv3r>
do you have a wiki account yet? Still need to send you boobies
<buZz>
boobies \o/
<slapin>
oliv3r: I was preparing a bunch of patches for some time now, sick of them, actually, will need to have private review as I don't see obvious errors and stupid things.
<oliv3r>
I know the feeling, stupid wikipedia templates; making me very angry :S
<Reepicheep12>
Oh wow guys it boots. Can't hardly believe it. Thanks for all your guidance.
Reepicheep12 has left #arm-netbook ["Leaving"]
Reepicheep12 has joined #arm-netbook
<Reepicheep12>
What a trek.
<oliv3r>
good job :)
<TestModule>
BSP loads the HW pack?
<Reepicheep12>
Yeah, the precompiled one anyway. Thanks.
<hansg>
Hmm, talking about nand booting, do we know where the nand uboot stores its environment ? Then we could maybe use mkenvimage and dd to automatically overwrite it with a u-boot environment what does what we want (ie tells uboot to honr uEvnt.txt, etc.)... This assumes that the already present u-boot.bin on the pre-installed android knows how to do fatload ...
<Reepicheep12>
So those fatload command to write the kernel and script.bin, is there are way to make them permanent. I gather I can use "nand write", but I want to confirm 'cause that sounds scary and I don't know exactly how to go about it.
<WarheadsSE>
be wary of nand write
<WarheadsSE>
You might be looking for saveenv
<Reepicheep12>
I tried that. It saves the bootcmds but not the fatload writings.
<TestModule>
The BSP doesn't work correctly for me (as in, booting afterwards).
<hansg>
Reepicheep12, then you need to put the fatload commands into bootcmd, let me copy and paste an example uEnv
<WarheadsSE>
Reepicheep12: nand write is for writing images to nand. Not saving environments.
<hansg>
Reepicheep12, this is a slightly modified standard setup (nore for mmc adjust accordingly !!)
<Reepicheep12>
I see thank you I think I can work it out from that.
<hansg>
Reepicheep12, also note how the commands in bootcmd expand to other variables. If you want you can make things more kiss and simply put everything directly in bootcmd
<hansg>
Reepicheep12, once you've the necessary fatload, etc. commands in bootcmd, saveenv should make it boot from nand with your config by default :)
<TestModule>
Do I have to debootstrap in addition to BSP when mounting a premade armhf debian filesystem?
<hansg>
Reepicheep12, when you've that working please do a little readme documenting what you did, others are interested in this too :)
* hansg
wants to create an "install to nand" button in the Fedora-18 sdcard images he is working on
<Reepicheep12>
Yes I surely will. I have only the bare minimum of knowledge now though to get this working so perhaps it'll read half-baked.
TestModule has quit [Quit: Page closed]
<Reepicheep12>
And now it boots without intervention holy moly. Thanks hansg. Will sort out that readme in the near future to the best of my abilities.
<mnemoc>
oliv3r: url to the page you want me to delete?
pcat has quit [Read error: Connection reset by peer]
<Reepicheep12>
Now that I'm setting this all up is there anything I should be worrying about re: nand usage? I
<Reepicheep12>
I've read that sd might be preferred over nand due to nand wear. If I'm mounting with noatime/relatime is there anything else I should be doing to prevent wear to the one nand I've got?
<Reepicheep12>
I'd use SD boot if it worked for me although I'm glad I finally got this working.
<Reepicheep12>
Thanks again mnemoc.
<mnemoc>
and any decent SD card will be faster than the nand in your a10-based device
<leowt>
what is the stable kernel branch?
<mnemoc>
sunxi-3.0
<leowt>
tnks
<Reepicheep12>
Yeah I would prefer to use it but I kept getting input/output errors after using the sd card for a few hours, at which point half the files from the sd card would disappear and I'd have to write them again. Maybe some stage in the future I'll try it again but for now I think I'll stick to the nand provided it keeps working.
<Reepicheep12>
Tried a bunch of cards too.
<mnemoc>
it's not uncommon to have dust on the uSD slot
<Reepicheep12>
I could try cleaning that had not thought of that. Although the failure was pretty consistent. Booted a fresh install, installed mpd-git in ArchLinux, it installed the dependencies, started compiling mpd-git, and about a minute in suddenly couldn't read anything on the drive.
<Reepicheep12>
But I'll check it out inside at a later date thanks for the suggestion.
<leowt>
anyone uses uart with mele a1000
<leowt>
?
<Reepicheep12>
I certainly do.
<leowt>
Reepicheep12, works ootb?
<Reepicheep12>
For me it did.
<WarheadsSE>
if you wire it right, yeah
<leowt>
ok, going to give it a try
<WarheadsSE>
Reepicheep12: why mpd-git vs git as it was?
<Reepicheep12>
You mean mpd-git versus mpd? It was a while ago when I made the decision (while before the mele came along) but I think the core mpd didn't support staying paused when the server started.
ganbold_ has quit [Remote host closed the connection]
<WarheadsSE>
ah
tinti is now known as maluta
mSquare1 has left #arm-netbook [#arm-netbook]
maluta is now known as tinti
<nemik>
Reepicheep12: i had the same problem, uSD's get corrupted very quickly
<WarheadsSE>
i havent hit that problem
* WarheadsSE
shrugs
tinti is now known as maluta
<nemik>
it took a couple weeks with Ubuntu, but it all crashed hard when it did
<Reepicheep12>
Did you find a solution? Or is it maybe a hardware thing?
maluta has quit [Quit: Leaving]
<nemik>
so i think i broke my mk802ii. built custom livesuit image and tried to flash it with phoenixcard on SD thing. did not work :(
<leowt>
is there any doc talking about the boot sequence of the A10 chip
<leowt>
?
tinti has joined #arm-netbook
<nemik>
Reepicheep12: no solution, i'd try using the NAND, it was level-wearing functionality and should work much better
<WarheadsSE>
I compiled the kernel & the mali drivers ~ 20 times on mine
<WarheadsSE>
no problems
<nemik>
WarheadsSE: hmm maybe you have a better SD card. what brand/size?
<Reepicheep12>
Well I'm using the newer Mele A1000G revision so perhaps that's it. What was yours nemik? Nice to hear I'm not the only one.
<nemik>
Reepicheep12: i was using mk802 with 1gb of RAM and now MK802II also
<WarheadsSE>
Reepicheep12: I have an A100
<WarheadsSE>
hipboi sent it to me, i think.
<WarheadsSE>
nemik: sandisk ultra class 10
<WarheadsSE>
I think it is 8G
<Reepicheep12>
Ah okay. Oh and you probably already know this Warheads but your latest kernel panics with nand boot (but not sd) like this: http://sprunge.us/CSUj
<Reepicheep12>
Yeah I was using the same card (although they all behaved the same).
<WarheadsSE>
NAND drivers weren't exactly grand at that time :)
<mnemoc>
they still the same
<WarheadsSE>
mm, then idk, that 3.0.39? kernel was a while ago
<Reepicheep12>
:)
jlj has quit [Ping timeout: 256 seconds]
<nemik>
has anyone succeeded in making an mk802-compatible livesuit image using the sunxi-bsp? i built one but it didn't flash
jlj has joined #arm-netbook
leowt has quit [Remote host closed the connection]
<Reepicheep12>
Cockroach just crawled right onto my mele's pcb and sat there having a right old time. Thoroughly irked.
<WarheadsSE>
nice and warm
<WarheadsSE>
also: eww
* slapin
hates cockraches
<traeak>
i've only seen cockroaches at an indian restaurant here
<traeak>
must be the arid conditions
<slapin>
they come when you drink too much then make large break, which is very bad sign
<traeak>
they dont' seem to like -20C conditions very much
<WarheadsSE>
:P true
<slapin>
also, cockraches seems to be immune to everything you do to them, they leave when they want to leave, that's all
<traeak>
although have to admit we've had some problems with mice
<traeak>
there's a certain kind of dust made from crystanthemums (or whatever) that should kill them
<traeak>
i took out a yellow jacket nest with that dust this past summer
<traeak>
anyways
<Reepicheep12>
Pity it's arid and +40C around here (well during the day it was). Watched an episode of the X Files the other night where a cockroach walks across the screen, not part of the action or anything, breaking the fourth wall like, just to freak you out. Nearly jumped out me chair.
<Reepicheep12>
Then a real one head for me as the credits were rolling. I hate them too.
<slapin>
the mice problems are all simple, just get yourself a cat or a girlfriend, regular wives won't do without proper training...
<slapin>
but cockraches are true evil
* slapin
lived onec in old house which was full of cockraches everywhere, they were eatine everything, from your food to your closthes, wallpaper, walls under wallpaper (your wallpaper is alive and clicks, clicks everywhare!!!)
<slapin>
brrrrr
<slapin>
and evry appartment was the same, only demolisment of building stopped that misery
<slapin>
s/evry/every/
<ibot>
slapin meant: and every appartment was the same, only demolisment of building stopped that misery
hansg has quit [Remote host closed the connection]
<mnemoc>
so they moved to the next?
<Reepicheep12>
Reminds me of that kid's horror story show I saw a while back, fella goes to a cheap motel, enjoys a nice water bed mattress, chops it open in the morning to see why it's so unusually ripply, million cockroaches for stuffing.
<slapin>
mnemoc: I think so
<slapin>
mnemoc: but I moved not to the same building, so never seen them again
<mnemoc>
do you miss them? :p
<mnemoc>
i've been luck. i've only had rats and spiders, never cockraches
<mnemoc>
lucky*
ZaEarl has joined #arm-netbook
<Reepicheep12>
Guys sorry to bring the topic off insects, but I'm finding it impossible to copy a new kernel to the /dev/nanda partition. There used to be a reasonable amount of room in the 16mb partition, but now it says there's only 3.2M left, and the kernel's 4M. Tried rebooting any ideas? Not sure how that partition works but this is very unusual.
<mnemoc>
fsck?
rellla has joined #arm-netbook
<Reepicheep12>
"fsck: error 2 while executing fsck.vfat for /dev/nanda"
<Reepicheep12>
Would format it if I could.
<slapin>
Reepicheep12: check with du -sk * what do you have there
<slapin>
Reepicheep12: what df shows?
<Reepicheep12>
When it is mounted df shows "/dev/nanda 16342 13016 3326 80% /mnt/1"
<Reepicheep12>
I haven't really done anything unusual to the partition besides copy a few kernels onto it (three on there at a time or so). I could try "nand-part /dev/nand 'LROOTFS 4069376'", like I did when I was first installing to nand, although that would wipe my nandb partition.
<mnemoc>
"copy a few kernels" is like to be the cause ;-)
<mnemoc>
you barely have room for 1 in there
<mnemoc>
so make some room
<mnemoc>
also, you can copy everything out, reformat, and copy back
<Reepicheep12>
Well I deleted them all there's only u-boot.bin there now.
<Reepicheep12>
So I can formate nanda as long as I copy everything back on? Just with a regular mkfs.vfat?
tinti_ has joined #arm-netbook
<Reepicheep12>
It seemed like every kernel I deleted added a tenth of its size to free space. Going to make a dd image just in case the formatting goes awry.
tinti_ has quit [Read error: Connection reset by peer]
tinti_ has joined #arm-netbook
tinti has quit [Remote host closed the connection]
tinti_ has quit [Read error: Connection reset by peer]
tinti has joined #arm-netbook
<Reepicheep12>
Still not working after a mkfs.vfat reformat. Going to try one more time.
<Reepicheep12>
Nope.
<Reepicheep12>
My next plan is to boot from that official mele android image and recreate the nand. Is there anything that might work besides this? seems silly to start all over again just because of one file.
Jef91 has joined #arm-netbook
gimli has joined #arm-netbook
jardiamj has joined #arm-netbook
<Reepicheep12>
Alright just tried a few more times using different methods, works now. Took a dd image for the future.
<drachensun>
I recently had problem where mkfs.vfat was making the disk larger than the partition it was on, so now when I used mkfs.vfat I am setting the block size manually
<hramrach>
but u-boot defaults to 115200 or somesuch baud on most devices
<hramrach>
whatever that means
popolon has quit [Quit: Quitte]
<leo__>
should i get some image without rootfs?
<hramrach>
what for?
TheLarch has quit [Quit: <puff of smoke>]
<leo__>
compiled uboot and kernel
L84Supper has joined #arm-netbook
<leo__>
and followed instructions to put on sd
<hramrach>
you want to put them on the nand or ?
<leo__>
but i didnt prepare rootfs
<leo__>
SD for now
<leo__>
and boot rootfs from sda1 later
<hramrach>
you can just boot he kernel but it will panic immediately without an init
<hramrach>
you create Debian or Ubuntu rootfs with debootstrap
<leo__>
so it is booting the kernel
<hramrach>
not sure others
<traeak>
holy crap a pristine unactivated oem windows install is 30GB
<traeak>
anyways
<hramrach>
would fit on a SHHC ;-)
<mnemoc>
and lacks of most basic drivers...
<hramrach>
*SDHC
<leo__>
but should it output serial/fb without rootfs?
<mnemoc>
if you made fbcon built in and pass the right console=, yes
<hramrach>
serial yes, fb no
<mnemoc>
fbcon should show the panic about no rootfs
<hramrach>
the rootfs is where you store the kernel
<hramrach>
you can set up serial console
<mnemoc>
I use the boot partition for that ;-)
<leo__>
shouldnt it be working by thefault
<leo__>
the serial output
<WarheadsSE>
traeak: duh
<leo__>
?
<mnemoc>
yes, you'll have greetings and such
<leo__>
im getting nothing out of it
<mnemoc>
same for the kernel.... if your u-boot found something to load
<leo__>
with sd card led stays blue and red
<leo__>
and no output
<traeak>
WarheadsSE: just looking at giving this unactivated license of windows to someone else
<WarheadsSE>
depends on if it was an OEM
<WarheadsSE>
has to be reail to transfer
<mnemoc>
leo__: test it without SD first. boot0/boot1 also dump logs to uart0
<traeak>
even if its never been activated?
<WarheadsSE>
yes
<WarheadsSE>
legalese*
<leo__>
mnemoc, with the original rom?
<mnemoc>
leo__: yes
<WarheadsSE>
^
<leo__>
nothing
<WarheadsSE>
When i received my mele A100, that was how I brought up Arch Linux ARM on it, was via serial
<WarheadsSE>
Do you have the pinout right leo__ ?
<leo__>
yep
<WarheadsSE>
You're dead sure of that
<leo__>
going to try with another serialttl iface
<mnemoc>
also test if your uart dongle works. and NEVER connect the VCC pin
<WarheadsSE>
^
<WarheadsSE>
zap happens
<mnemoc>
leo__: easy test is to jump the TX/RX of the dongle.... should get you a nice echo if the dongle works
<nemik>
woot, finally got images flashing with LiveSuit, but the one i built does nothing :(
<slapin>
nemik: how? in Windows or Linux?
<nemik>
slapin: Windows in VMWare
<nemik>
had to add this to .vmx though: usb.quirks.device0 = "0x1f3a:0xefe8 skip-reset"
<nemik>
i flashed the stock firmware and it's booting and nice. but the one i built myself using sunxi-bsp isn't working, just blank screen. and unit i have doesn't have easy tx/rx pic access so i can't see what's going on really
<slapin>
nemik: tried with Linux livesuit?
<nemik>
not yet, i have it installed on one of my machines though
<nemik>
oh weird, so i flashed with "en-us-NEW2-ak003.img" and it is all Chinese this time...
pcat has joined #arm-netbook
<nemik>
anyone know what the latest stock image for mk802ii is? and where i could download it?
rellla has quit [Quit: rellla]
leo__ has quit [Remote host closed the connection]
leo__ has joined #arm-netbook
<rm>
not unusual for images to be chinese by default
<rm>
you can try finding your way to the android settings menu
<rm>
to change the language
<leo__>
its working
<leo__>
something wrong with one of my serial ttl interface
<libv>
not that the chip isn't pure evil by using a pvr :)
<mnemoc>
wingrime: no, sunxi-tools
<mnemoc>
libv: :)
<wingrime>
so, we can drop allwiner proprietary boot loader (pre-uboot) except BROM ?
freakazoid0223 has joined #arm-netbook
<mnemoc>
wingrime: not yet
<mnemoc>
wingrime: but that is not related to DTS support. once our u-boot learn to do nand properly we will be able to replace boot0/boot1 on 3.0/3.4 based systems too
<wingrime>
it possible now?
<wingrime>
only nand?
<mnemoc>
not yet. once our u-boot learn to do nand properly.
ibrah has quit [Read error: Connection reset by peer]
leo__ has quit [Ping timeout: 240 seconds]
Alex1269_ has quit [Ping timeout: 245 seconds]
Alex1269 has quit [Ping timeout: 245 seconds]
Yaku has quit []
<slapin>
have anybody tried to run livesuit under VirtualBox?
tinti has quit [Quit: Leaving]
<mnemoc>
people claims success with that thing from oracle
sgtpepper has joined #arm-netbook
<sgtpepper>
hello
<sgtpepper>
new hackberry owner here :d
<sgtpepper>
:D
<sgtpepper>
quick question, how do you turn this headless images into regular images... was the Kernel compiled without ttys?
<hramrach>
more likely withuot framebeffer support
<hramrach>
to save the memory used by graphics
<sgtpepper>
hmm... I'm checking linaro now... but probably I'll end up starting from scratch
<ZaEarl>
slapin, yes, I run LiveSuit in XP inside of VirtualBox just fine.
<slapin>
ZaEarl: cool, thanks. Do you use proprietary USB plugin?
<ZaEarl>
not sure, I just have virtualbox installed from the Ubuntu 12.10 repos
<slapin>
ZaEarl: from virtualbox.org or from standard repos?
popolon has joined #arm-netbook
<ZaEarl>
standard repo
ibrah has joined #arm-netbook
dyoung is now known as dyoung-away
ibrah is now known as eebrah
eFfeM has quit [Quit: Leaving.]
<slapin>
ZaEarl: thanks a lot
xman has joined #arm-netbook
<wingrime>
slapin: usb in VBox now common feature in opensource version