hno changed the topic of #linux-sunxi to: /Allwinner/sunxi development discussion - Don't ask to ask. Just ask and wait! - See | | Logs at
rz2k has quit []
<arokux2> Turl,
Seppoz has quit [*.net *.split]
forcev has quit [*.net *.split]
mouchon has quit [*.net *.split]
Black_Horseman has quit [*.net *.split]
Dreadlish has quit [*.net *.split]
bfree has quit [*.net *.split]
egbert has quit [*.net *.split]
Tsvetan has quit [*.net *.split]
\\Mr_C\\ has quit [*.net *.split]
[7] has quit [*.net *.split]
woprr has quit [*.net *.split]
bamvor has quit [*.net *.split]
CoSM0 has quit [*.net *.split]
arete74 has quit [*.net *.split]
tgaz has quit [*.net *.split]
hramrach has quit [*.net *.split]
akaizen has quit [*.net *.split]
WarheadsSE has quit [*.net *.split]
ynezz has quit [*.net *.split]
jlj_ has quit [*.net *.split]
arokux2 has quit [*.net *.split]
brain__ has quit [*.net *.split]
tm512 has quit [*.net *.split]
arokux1 has quit [*.net *.split]
tzafrir has quit [*.net *.split]
drachensun has quit [*.net *.split]
Super-noob has quit [*.net *.split]
jeremb_ has quit [*.net *.split]
uwe_ has quit [*.net *.split]
atsampson has quit [*.net *.split]
Nikolas has quit [*.net *.split]
jelly-home has quit [*.net *.split]
wigyori has quit [*.net *.split]
Montjoie has quit [*.net *.split]
pirea has quit [*.net *.split]
leviathanch has quit [*.net *.split]
gzamboni has quit [*.net *.split]
steev has quit [*.net *.split]
FergusL has quit [*.net *.split]
servili007 has quit [*.net *.split]
focus_it has quit [*.net *.split]
wens has quit [*.net *.split]
moofree_ has quit [*.net *.split]
book` has quit [*.net *.split]
mdp has quit [*.net *.split]
on1x has quit [*.net *.split]
setkeh has quit [*.net *.split]
pitillo has quit [*.net *.split]
dwilkins has quit [*.net *.split]
ssvb has quit [*.net *.split]
techn_ has quit [*.net *.split]
HeHoPMaJIeH has quit [*.net *.split]
buZz has quit [*.net *.split]
torindel has quit [*.net *.split]
aep has quit [*.net *.split]
zumbi has quit [*.net *.split]
tkoskine has quit [*.net *.split]
diego71 has quit [*.net *.split]
Tartarus_ has quit [*.net *.split]
MadSpark has quit [*.net *.split]
deasy has quit [*.net *.split]
dlan has quit [*.net *.split]
Turl has quit [*.net *.split]
ganbold has quit [*.net *.split]
mnemoc has quit [*.net *.split]
ojn has quit [*.net *.split]
mripard has quit [*.net *.split]
naobsd has quit [*.net *.split]
cajg has quit [*.net *.split]
Brod_ has quit [*.net *.split]
CaCtus492 has quit [*.net *.split]
Olaff has quit [*.net *.split]
binaryferret has quit [*.net *.split]
specing has quit [*.net *.split]
andoma has quit [*.net *.split]
xenoxaos has quit [*.net *.split]
ChanServ has quit [*.net *.split]
drachenphone has quit [*.net *.split]
eebrah|FTS has quit [*.net *.split]
awafaa has quit [*.net *.split]
mturquette has quit [*.net *.split]
libv has quit [*.net *.split]
hno has quit [*.net *.split]
jukivili has quit [*.net *.split]
fredy has quit [*.net *.split]
speakman has quit [*.net *.split]
fldc has quit [*.net *.split]
eagles0513875 has quit [*.net *.split]
oliv3r has quit [*.net *.split]
stekern has quit [Ping timeout: 351 seconds]
Amitabha has quit [Ping timeout: 246 seconds]
lkcl_ has quit [Ping timeout: 246 seconds]
_whitelogger has joined #linux-sunxi
<brain__> Got most of the major bugs worked outta my CM9 A10 tablet build. All hardware works it seems now from testing. not sure if hardware accel works tho, videos play fine no out of sync issues using MX player on HW decoder mode, but things like youtube have no video. netflix audio only...getting there.
<arokux2> brain__, wifi works?
<brain__> wifi worked on the first build yes.
<arokux2> brain__, which module? aw's?
<brain__> yes
<brain__> pulled from stock rom
<arokux2> brain__, ah, I've thought you're using our sunxi-3.4
<brain__> i wish :( but i'd lose touchscreen due to a custom module i have no source for. i'm still looking into that as well.
<brain__> well no working source that is
<arokux2> brain__, i see :( try to ask for code
<brain__> I plan to send an email but i dont have my hopes up on them releasing it to me, the company who made the tablet is small from what i've read. not sure if it was them or focaltech who customized it so much.
<arokux2> brain__, :(
<brain__> actually now that i think about it, it was the tablet company. other modules having similar signature strings i am seeing in the touch driver but the open source versions work fine for those modules, mainly just want that darn touch code. i'll see what happens.
<arokux2> Turl, ping
<Turl> arokux2: pong
<arokux2> Turl, how did you know emac-power is 3.3v?
<Turl> GPIOs are 3v3
<Turl> arokux2: ask maxime, he wrote the other regulator :p
<arokux2> Turl, so mine should be too?
<arokux2> Turl, but it doesn't really have some influence, right...?
<arokux2> Turl, since it is on/off
<arokux2> Turl, can you help me understand something? I've found some differences...
<arokux2> Turl, what is this? ---> <pullup/down> is 0 = disabled;
<arokux2> Turl, emac_power has it set to 1.
<arokux2> Turl, whereas usb_gpio has it set to 0
<arokux2> Turl, my question is how it would be resembled in code?
<Turl> arokux2: it's configured on your pinctrl declaration
<Turl> allwinner,pull = <...>
<arokux2> Turl, ah, but emac has it set to 0
<arokux2> Turl, I mean in pinctrl
<arokux2> Turl, but in fex it is 1...
naobsd has quit [Quit: Page closed]
egbert has quit [Disconnected by services]
egbert has joined #linux-sunxi
<arokux2> Turl, do you know if things like "port:PH06<1><0><default><0>" are parsed somewhere other then in a driver?
<Turl> arokux2: the fex parser thing handles those I think
<arokux2> good night Turl and thanks
<Turl> night, yw :)
rm has joined #linux-sunxi
ijc has joined #linux-sunxi
rah has joined #linux-sunxi
heffer has joined #linux-sunxi
arokux2 has quit [Remote host closed the connection]
nedko has joined #linux-sunxi
* nedko wants to store non-android linux on olinuxino-a20 nand
jukivili has quit [Ping timeout: 260 seconds]
jukivili has joined #linux-sunxi
drachenphone has quit [Remote host closed the connection]
[7] has quit [Read error: Operation timed out]
TheSeven has joined #linux-sunxi
leviathanch has quit [Ping timeout: 264 seconds]
leviathanch has joined #linux-sunxi
ykchavan has joined #linux-sunxi
leviathanch_ has joined #linux-sunxi
leviathanch has quit [Read error: Connection reset by peer]
leviathanch_ has quit [Quit: No Ping reply in 180 seconds.]
leviathanch has joined #linux-sunxi
JohnDoe_71Rus has joined #linux-sunxi
servili007 has quit [Quit: Leaving]
hipboi has joined #linux-sunxi
eebrah|FTS has quit [Ping timeout: 260 seconds]
eebrah|FTS has joined #linux-sunxi
atiti has joined #linux-sunxi
<oliv3r> morin'
n01 has joined #linux-sunxi
eebrah|FTS has quit [Quit: .]
gnufan has joined #linux-sunxi
eebrah has joined #linux-sunxi
rellla has joined #linux-sunxi
<mnemoc> arokux1: ssh for ? give me your ssh pub key
panda84kde has joined #linux-sunxi
\\Mr_C\\ has joined #linux-sunxi
massi_ has joined #linux-sunxi
\\Mr_C\\ has quit [Client Quit]
\\Mr_C\\ has joined #linux-sunxi
<oliv3r> mnemoc: how do I create a 'default config'
<oliv3r> mnemoc: using the bsp ideally :)
<mnemoc> on a dir contaning .config: make savedefconfig; cp defconfig arch/arm/configs/foo_defconig
<mnemoc> in the case of bsp, go into linux-sunxi
<mnemoc> and make ARCH=arm CROSS_COMPILE=arm-linux-gnueabi- O=../build/linux..... savedefconf
<mnemoc> the O= tells where the output/bins are stored
<n01> uhm there is something pre-cooked in linux kernel for doing something like (((x) & (mask << shift)) >> shift)?
<n01> mnemoc oliv3r any complain on this ?
<steev> what is that used for? something seems wrong there
<mnemoc> extract a part a register
<n01> yep
<n01> I know it seems ugly. But it is better than defining 10 different masks
<oliv3r> mnemoc: ok so i need to manually do the make with savedconf appended; so make linux-config; make savedefconfig; rgr
<oliv3r> n01: i think so
<oliv3r> n01: maybe not, i'm confused
<mnemoc> oliv3r: after make linux-config to tweak build/linux-whatever/.config you need to `cd linux-sunxi` and MANULLY make ARCH=arm CROSS_COMPILE=arm-linux-gnueabi- O=../build/linux-whatever savedefconf
<n01> steev: mnemoc the original code is here
<oliv3r> yeah manually
<mnemoc> then cp ../build/linux-whatever/defconfig arch/arm/confgs/whatever_defconfig
<oliv3r> got it
<mnemoc> :)
<oliv3r> cherry picked my driver allready
<mnemoc> \o/
<mnemoc> please make a sunxi_defconfig in sunxi-3.10
<oliv3r> so 3.10, only add allwinner drivers/arch or do we wanna build a big multiplatform kernel by default (slow compiles and not really needed)
<oliv3r> mnemoc: that's what i'm doing :)
<mnemoc> :D
wingrime has joined #linux-sunxi
<mnemoc> soon weca have real-life users of the DTS-based kernel
<mnemoc> and hopefully `fexc` extended to dump .dts fles
<steev> oliv3r: multiplatform, might as well since upstream will want it
<oliv3r> steev: ?
<oliv3r> steev: the kernel IS multiplatform allready
<oliv3r> steev: but do we want to build a kernel that can run on rockchip, amx, tegra; By Default?
<steev> i can build all armv6 armv7?
<oliv3r> steev: i don't think that's needed per say, as it only slows down the build :p
<steev> well it's 3.10 which is lts
<oliv3r> steev: the kernel IS multiplatform
<oliv3r> steev: but do we want every sunxi-user compiled kernel be able to run on all platforms?
<oliv3r> steev: this. is. sunxi.
<steev> i know the kernel IS, in theory, what i meant is something like, fedora likes to build one kernel to rule them all
<steev> their boot.scr is quite a sight to behold
<steev> oliv3r: what i meant is, it would be nice to be able to "side port" the sunxi stuff to distribution kernels
<steev> for your own work, nah
<oliv3r> steev: fedora won't use our defconfig anyway
<steev> get it working before worrying about interop
<oliv3r> steev: also distributions won't ever use our sunxi-3.10
docice has joined #linux-sunxi
<oliv3r> steev: 3.10 is only for us, a stepping stone for users and developers
<oliv3r> i say, our defconfig, in our kernel, should only be for our users
<steev> one distro might, but only because i do the sunxi stuff for them :P
<oliv3r> so drop all other platforms and their drivers :)
<oliv3r> steev: :)
<mnemoc> steev: 3.10 is supposed to become an "stableish" .dts incldun the crapy drivers too, so usable by real peple
<oliv3r> and that
<oliv3r> the crappy drivers
<steev> incldun?
<mnemoc> including*
<oliv3r> i wonder if we can load fex AND dts simultaniously
<steev> oh
<mnemoc> i can't type dotay
<mnemoc> f*
<oliv3r> heheh
<steev> :)
<mnemoc> today
<oliv3r> german keyboard ftw
<mnemoc> :D
<steev> i prefer japanese keyboard layout myself
<n01> dvorak ftw
<n01> ;)
<oliv3r> i really wanna learn dvorak one of these days
<oliv3r> i tried it once, but without key caps its hard
<n01> It is better without keycaps
<steev> das keyboard
<n01> yep
<n01> das keyboard blank
<n01> cherry mx blue ftw
<mnemoc> my das keyboard in my warehouse in spain :(
ykchavan has quit [Ping timeout: 260 seconds]
<mnemoc> hopefully shippingall that stuff here next wek
<n01> steev: now I want to take a wasd v2 with cherry mx clear
<steev> that's pretty hot
<steev> but i use laptops, so, almost impossible to get one with blank keys
<n01> steev: that's mine
<steev> and considering how each manufacturer likes to jerk keys around, kinda hard to have all blank
<mnemoc> steev: buy black stickers :p
<steev> "is this one page up? nope"
<n01> with some coloured keycaps from wasd
<n01> oliv3r: anyway learn dvarak. It's worthy it
<oliv3r> n01: oh i know, i tried it for a few days
<n01> it takes 3 months to me to get to the same speed as on qwerty
<n01> *took
tzafrir has quit [Ping timeout: 245 seconds]
<oliv3r> i just need a keyboard to learn it on ;)
<oliv3r> and especialyl at the start; I find the keycaps help enormously
<oliv3r> maybe I should borrow a few keyboards
<n01> oliv3r: it is better to learn without keycaps IMO. That's help when you are back writing on a qwerty keyboard
<n01> oliv3r: try a mechanical keyboard
<oliv3r> n01: on qwerty i type 'blindly'
<oliv3r> but i have an idea; i'll get a keyboard here and reorder the keys :)
<oliv3r> n01: that site is pretty cool
<n01> yup
<steev> meh, i highly doubt i can hit 140+ with dvorak
FR^2 has joined #linux-sunxi
<n01> 140wpm??
deasy has quit [Quit: Nom d'un quark, c'est Edmonton !]
fredy has quit [Excess Flood]
<steev> yeah, i used to do data entry
fredy has joined #linux-sunxi
<n01> that's fast
popolon has joined #linux-sunxi
notmart has joined #linux-sunxi
notmart has joined #linux-sunxi
<oliv3r> there, swapped all keycaps on a spare keyboard :)
wingrime has quit [Read error: Operation timed out]
<oliv3r> steev: once you get proficient with it; you might get faster
* mnemoc nows imagines steev as a sexy secretary
<mnemoc> uhm.... better not
<steev> s/sexy/scruffy/
naobsd has joined #linux-sunxi
DigiDaz_ has joined #linux-sunxi
<mnemoc> scruffy can be sexy too ;-)
Black_Horseman has quit [Remote host closed the connection]
<oliv3r> mnemoc: i'll send my first sunxi_defconfig to the mailing list; the driver itself i'll push as a dicator
<arokux1> Tsvetan, ping
<arokux1> is Olimex doing any board donations? I'd like to have A13 based board...
<steev> mnemoc: hah, well, my ex claims i'm sexy still, and we've not been together for ~3 years
<oliv3r> and i did it wrong so i have to redo it
<mnemoc> oliv3r: +1
<steev> i think it's because she wants to get in my pants still but regardless, that's a bit off topic for here i think
<mnemoc> steev: I was thinking in a scruffy female secretary ;-)
<mnemoc> hehehe
<oliv3r> shemale? *shiver* mnemoc you dawg
<mnemoc> -_-
<n01> well some shemales are really hot, just saying
<oliv3r> O.o
<oliv3r> I guess whatever rocks your boat :D
hipboi has quit [Quit: Leaving]
<n01> ahaha I'm plain straight but in some cases you cannot really see the difference
<buZz> wut
<buZz> #linux-sunxi-sextalk ?\
<n01> hahha it's just monday. back to work and IT ;)
<buZz> your work is not IT?
<n01> yep
<n01> IT = in topic
<buZz> yes its not?
<buZz> or yes it is
<buZz> IN topic?
<buZz> wtf do you want to put in it
<buZz> ?
<n01> yes it is. and I meant that it is better to get back on topic
<buZz> ah, ON topic
<buZz> yes that makes sense
<arokux1> jukivili, no sunxi hacking anymore?
<oliv3r> Error while writing of the configuration. Your configuration changes were NOT saved.
<oliv3r> crapyness
tzafrir has joined #linux-sunxi
leviathanch has quit [Quit: No Ping reply in 180 seconds.]
leviathanch has joined #linux-sunxi
<oliv3r> mnemoc: make[3]: *** No rule to make target `savedefconf'. Stop.
<mnemoc> savedefconfig
<mnemoc> in the kernel tree, with O= pointing to the build dir
<mnemoc> it might be good to send a diff between sunxi_defconfig and multi_v7_defconfig too
<oliv3r> yeah all that i did, but you sad conf, not confIG :p
<oliv3r> yeah
<mnemoc> ok, my fault :(
<n01> make help ;)
fredy has quit [Excess Flood]
soldoKyn has joined #linux-sunxi
fredy has joined #linux-sunxi
_whitelogger has joined #linux-sunxi
<n01> oliv3r: lol, mnemoc try to suggest make savedefconfig && rm -fr /
<mnemoc> eh?
<mnemoc> I rm -rf the build dirs, not the source dirs or / :p
<mnemoc> <3 make O=
<oliv3r> :p
<oliv3r> i only did a compile test; that'll have to do for now; don't have hardware
Soru has joined #linux-sunxi
arete74 has quit [Read error: Connection reset by peer]
arete74 has joined #linux-sunxi
<mnemoc> oliv3r: it's experimental, you can push it when it builds :p
<mnemoc> oliv3r: and reduce work duplication while allowsing others to test
<oliv3r> well it built and tested; before commiting the sunxi_defconfig it'll go through ML
granite_crusher has joined #linux-sunxi
<mnemoc> =)
<granite_crusher> is it possible to corrupt SD card with UART connection?
<granite_crusher> I had few times so that board wouldnt restart until I wouldnt deatach debug UART cables
<arokux1> granite_crusher, disconnect power cable of your uart
<arokux1> granite_crusher, yep, I had the same
<granite_crusher> arokux1: power? mine have just 3pins: gnd, rx, tx
<jukivili> arokux1: sorry, I'm busy with other things.. most notably, trying to get a job
<oliv3r> mnemoc: i probably have to manually attach the diff :S
<arokux1> jukivili, I see. good luck.
<arokux1> granite_crusher, ah, ok. well there were some problems like this, they disappeared because I do not mess up with sd cards anymore, I boot over network, so no need to overwrite the sd card
<mnemoc> oliv3r: in a manual reply?
<granite_crusher> too bad there is no way to load kernel image from usb in u-boot...
<mnemoc> oliv3r: please also take some minutes to explain people the point of doing that branch
<mnemoc> =)
<arokux1> granite_crusher, there is!
<oliv3r> mnemoc: if/when they start complaining
<oliv3r> the core devs. are in here; and know the reasoning :)
<mnemoc> ok
<mnemoc> fair point
<oliv3r> mnemoc: anyhow, i would have, but it's too late now :)
<arokux1> granite_crusher, I've just submitted patches to our ml
<granite_crusher> I am talking about u-boot
<arokux1> granite_crusher, me too. I've added usb to u-boot too.
<oliv3r> mnemoc: so I did one thing of my list; and prioritized your preference
<oliv3r> mnemoc: now I need your help
<granite_crusher> when?
<mnemoc> oliv3r: I can not check the patches from here...
<arokux1> granite_crusher, take a look at our ML plz
<mnemoc> doing IRC is already against the rules :<
<arokux1> oliv3r, how can config be tested if there are no *.dts(i) yet?
<oliv3r> mnemoc: no, i only need your brain
<oliv3r> arokux1: 3.10 has dts
<oliv3r> arokux1: 3.9 was the first patches for a10, 3.10 added a20 i think
<Turl> oliv3r: what are you doing? :P
* Turl semizombie still
<Turl> oh, 3.10 defconfig
<Turl> oliv3r: there's a minimalistic sunxi_defconfig on my tree :)
<oliv3r> Turl: too late!
<oliv3r> you can diff it against the one on the ML if you want more things/less things in it
<oliv3r> Turl: and good morning
<oliv3r> Turl: since mnemoc isn't lending me his brain; i need yours
<Turl> oliv3r:
<oliv3r> Turl: that your brain?
<Turl> oliv3r: no, that's the config :p
<oliv3r> that's a small config :p
<oliv3r> /home/emilio/buildroot? :p
<Turl> for big we have multi_v7
<oliv3r> Turl: i did a multi_v7_defconfig -> sunxi_defconfig
<oliv3r> strip useless drivers; add sunxi drivers
<Turl> :)
<oliv3r> Turl: how easy do you think it will be to backport your clk patches (when more or less done)?
<oliv3r> Turl: so i'm guessing it'll be pretty much the same
<oliv3r> CONFIG_EEPROM_AT24?
<Turl> oliv3r: Olinuxino A10S
naobsd has quit [Ping timeout: 250 seconds]
<granite_crusher> arokux1: sorry for silly question, I can't find it, Is it not v2013.10-rc2? from git linux-sunxi/u-boot?
<Turl> oliv3r: we can backport the full clk subsystem I guess
<Turl> I can give it a try
<oliv3r> Turl: the drivers you can just push to the branch
<arokux1> granite_crusher, ML = mailing list, I've posted patches yesterday, it is not in repos yet.
<oliv3r> the defconfig made sense to have reviewed i suppose
<oliv3r> and i forgot initramfs support :)
<Turl> oliv3r: what's the branch name again? :)
<oliv3r> experimental/sunxi-3.10
<oliv3r> iirc
<mnemoc> yes
<oliv3r> bah no sun4i_emac yet
<mnemoc> currently == reference-3.10
<oliv3r> gotta be backported too
<mnemoc> reference-3.10 = android-3.10 + v3.10.12
eebrah|noPC has joined #linux-sunxi
<oliv3r> mnemoc: got the first complain mail allready :p
<mnemoc> :p
<oliv3r> mnemoc: so brain; can I borrow yours for a bit?
<mnemoc> remember, stable (LTS, no rebases) experimental branch, and crappy drivers DTSized and ported to common frameworks in a common central repo
<mnemoc> oliv3r: depends
<mnemoc> so real-life testing by brave users is possible
<mnemoc> not only devs
<Turl> oliv3r: bfree I think did the backporting already for debian
eebrah|noPC has quit [Ping timeout: 260 seconds]
<oliv3r> or starting kernel probably, i forgot
<oliv3r> Turl: or merge a lot back;
<oliv3r> bfree: ^
<oliv3r> mnemoc: when I use my own boot.img with my own kernel (uImage) I get 'loading kernel ...' and then nothing :S
<oliv3r> bfree: ping
<oliv3r> Turl: even better; so we can just merge that in?
<oliv3r> mnemoc: I really want to replace the GPL violating kernel from my tablet. I have currently a u-boot installed that prints to the sdcard's serial port. U-boot + kernel log; all's good
<oliv3r> though yesterday evening, I had an idea, i can always place the image on the fat partition and just use u-boot commands to load the kernel and initramfs intself to see if they boot; and then just see if bootimg generation is what's failing
<andoma> oliv3r: you compile with LLDEBUG and early printk?
<Turl> oliv3r: using lichee dev uboot?
<Turl> oliv3r: 3.4 doesnt support the automagic ram detection from AW bootloaders
tinti has joined #linux-sunxi
<mnemoc> oliv3r: the kernel needs to be told what uart to use for output
<mnemoc> and be properly configured in uboot and script.bin
naobsd has joined #linux-sunxi
_whitelogger has joined #linux-sunxi
<oliv3r> andoma: to help with debugging problems; yes :)
<oliv3r> Turl: yep, lichee-dev sun4i_sd
<mnemoc> oliv3r: android has always broken ipv6=m, it has to be y
<oliv3r> ah, ok so do we want that?
<oliv3r> multi_v7 is default ipv6=m :)
<mnemoc> sure, but we are androidized
<mnemoc> and ipv6 is importnat
<oliv3r> okay
<mnemoc> so y is better than n
<mnemoc> we can assume no one wants a networkless sunxi device
<oliv3r> ipv4=y :p
ganbold_ has joined #linux-sunxi
<oliv3r> but sure, if that's 'the fix' that's fine with me
<oliv3r> i just got some compiler errors, and didn't wanna dive into them
<mnemoc> if you have time, try with y
<mnemoc> at least worked on 2.6.36, 3.0, 3.3 and 3.4
<mnemoc> where m was broken
<mnemoc> reading MLs in gmail for android is PAIN
<mnemoc> and thought gmail for web was bad enough already...
_whitelogger_ has joined #linux-sunxi
_whitelogger has quit [Remote host closed the connection]
Black_Horseman has joined #linux-sunxi
Black_Horseman has joined #linux-sunxi
<n01> oliv3r: if you have time and you are willing to do it, could you please add a wiki page with description of what each branch is supposed to be for. I'm lost with all the branches on linux-sunxi :((
<oliv3r> n01: we have that allready ;)
<Turl> n01: the wiki already has one
<Turl> :p
<n01> gh, sorry for being retarded
<oliv3r> mnemoc: i patched sunxi-bsp allready; i just need to clean it and submit it :)
<n01> link or page name?
<oliv3r> Turl: ^
<mnemoc> oliv3r: we need to make the kernel-version branching thing anyway
<n01> tnx Turl
<oliv3r> i get a warning for ipv6 but working sorta
<oliv3r> mnemoc: yeah sounds far more usable
<oliv3r> Turl: heh, see I remember seeing it on the github wiki page ;)
<oliv3r> there, added some initial 3.10 stuffs
<oliv3r> Warning: The database has been locked for maintenance, so you will not be able to save your edits right now.
<oliv3r> mnemoc: ^
<mnemoc> uhm
<mnemoc> the backup script should have removed the lock....
<mnemoc> at 3am...
_whitelogger_ has joined #linux-sunxi
<oliv3r> mnemoc: yep
<oliv3r> Turl: ok so automagic ram stuffs, would I get debug output from th at?
<Turl> oliv3r: nope
<oliv3r> i should have tested the kernel from SD card obviously :)
<Turl> or maybe earlyprintk stuff and a big explosion
<oliv3r> Turl: so step one, is adding that
<oliv3r> how do I get that to work?
vinifr has joined #linux-sunxi
<oliv3r> Turl: andoma: But i think one of the problems with getting debug to work, is that I can't choose the sunxi serial port in 3.4 for earlyprintk
<Turl> oliv3r: enable debug_ll and earlyprintk and pass earlyprintk and that's it I think
<oliv3r> andoma: turl: the only options are 'no uart', 'EmbeddedICE' and semihosting
<oliv3r> Turl: so we don't need the: kernel low level debug output via: option?
<Turl> oliv3r: I think it's provided by the mach selection on 3.4
<oliv3r> let me double check that
<oliv3r> i would love to liberate my tablet :)
_whitelogger_ has joined #linux-sunxi
<Turl> oliv3r: it has a rotating camera :P
<andoma> Turl: URL?
<oliv3r> Turl: ahh i see your arch thing
<oliv3r> it's under system settings; allwinner sunxi uart debug (CONFIG_SW_DEBUG_UART)
JohnDoe_71Rus has quit [Quit: Miranda IM! Smaller, Faster, Easier.]
<oliv3r> lets hope that overrides the 'none' bit :)
<granite_crusher> arokux1 or arokux2: nice u-boot patch :) I am trying it no: first impression: U-boot see that there is two usb where one is hub of 3, and I can see content of usb soon will try to boot mainline :)
<granite_crusher> *now
<arokux1> "is hub of 3"?
<arokux1> granite_crusher, post output of "usb tree", plz
<oliv3r> strange, mUSB (otg) should be completly ignored
<granite_crusher> when I do usb tree?
<arokux1> granite_crusher, to load the kernel from usb stick do: fatload usb 0:...
<arokux1> granite_crusher, in U-Boot
<oliv3r> unless of course you actually have a hub :)
<granite_crusher> usb otg I think is device 1
<granite_crusher> well not fatload but ext4load
<oliv3r> usb otg has no driver, so shouldn't be any device ;)
<arokux1> granite_crusher, could be.
<arokux1> granite_crusher, first do usb start, then usb tree and show the output to me
<granite_crusher> ok
<granite_crusher> moment
<arokux1> oliv3r, plz do not confuse ppl there are some different problems
<oliv3r> arokux1: i clearly stated OTG, remember, for some people USB is USB and A10 has 3 usb ports :)
<arokux1> oliv3r, he has A13. let me handle USB stuff myself.
\\Mr_C\\ has quit []
<oliv3r> yes god-of-usb, i shall not be worthy.
<arokux1> oliv3r, thanks :p
<granite_crusher> arokux1:
<arokux1> granite_crusher, (use
<arokux1> granite_crusher, can you also please post "usb info"
<oliv3r> sprunge isn't easy to paste stuff with; and sunxi doesn't have a preferred pasting site policy
<Turl> oliv3r: they have a web form with a button like all others :p
<Turl> but yeah, as long as it doesn't require me to sign up to download pastes I'm ok with it
<Turl> (ubuntu paste I'm looking at you)
<mnemoc> oliv3r: has a link to open a post form
<mnemoc> oliv3r: so it's not command-line only
<oliv3r> yeah but you still have to click to get to it and you can't setup anything for it; but i suppose it IS duable
<oliv3r> Turl: ubuntu paste requires a login? gawd that's horrible
<oliv3r> :)
<oliv3r> Turl: try copying that url btw though ;)
<arokux1> oliv3r, I just do not like to see pastebins not in raw format.
<tgaz> lovely way of opening a form
<oliv3r> arokux1: pasteie isn't the nicest one; debian, fedora all have nice paste services, when you 'repaste'you can plain text i think
<mnemoc> oliv3r: why would we need to make our own paster?
<oliv3r> mnemoc: because we can!
<mnemoc> is just awesome
<mnemoc> oliv3r: :p
<mnemoc> first make 3.10
<mnemoc> then mailman
<oliv3r> mnemoc: that's done allready?
<oliv3r> allready pushed the first commit too
<mnemoc> oliv3r: everything backported?
<oliv3r> defconfig is 'us under review':p
<oliv3r> mnemoc: pfft; i pinged bfree about that :) he did the work appearantly
<oliv3r> allready*
<mnemoc> lazy :p
<steev> woo, works-ish
pirea has quit [Quit: Leaving]
<oliv3r> steev: what worksish?
<steev> oliv3r: the mini-x
<oliv3r> steev: awesome
<granite_crusher> arokux1:
<oliv3r> granite_crusher: it is, you can actually do cat something | sprunge (if oyu've set it up right)
<oliv3r> steev: you got a 512mb device?
<steev> 300ish mb of ram because so much is saved for video i guess
<granite_crusher> this sprunge is awsome :)
<steev> oliv3r: yeah, it's old, it's one of the originals, without any vents so it gets nice and toasty
<granite_crusher> well I use puTTY (on debian) for UART comunication (it works out of box, I tried other I couldn't ), so anyway I copy text from putty..
<arokux1> granite_crusher, ok, thanks for the pastes. I'll try to adapt code to A13, can you test new patches later sometime?
<steev> granite_crusher: screen /dev/ttyUSBX 115200
<arokux1> granite_crusher, try to boot the kernel from USB :)
<oliv3r> steev: ahh well 512 mb is tight, my tablet has 512 mb too; really shamefull, as it's really not enough; im even thinking of dropping cedarX's memory reservation to get more ram, maybe even disabeling opengl, to get more ram
<oliv3r> granite_crusher: i use screen for UART, screen /dev/tty* 115200
<granite_crusher> arokux1: ok, just one thing in your path (I think line 37,38: COBJ := $(COBJ-y)..... needs tab nstead of spaces (as in original), else patch fail to apply
<granite_crusher> *patch
<steev> oliv3r: yeah, i know 512 can be tight, i don't run X on it (myself), but it does run fine with fbdev, not even sure what the point of cedarX is
<arokux1> granite_crusher, here? +COBJS-$(CONFIG_USB_EHCI_SUNXI) += ehci-sunxi.o
<arokux1> granite_crusher, where do you see spaces there?
<arokux1> granite_crusher, around +=?
<granite_crusher> no not that
<granite_crusher> no added lines
<granite_crusher> but two lines after
<arokux1> granite_crusher, ok, i'll take a look. thanks
<granite_crusher> maybe my git version is very pedantic :)
<arokux1> granite_crusher, btw, why are you unhappy about booting from an sd card?
<granite_crusher> arokux1: while I have no network, except usb dongle for eth, taking out sd card puting it into adapter, then to laptop, and all this uart ataching/deataching ruitine, and if something goes bad, again to put uboot... and writing speed to SD is terrible
<granite_crusher> with this setup I can have board on with UART atached all the time just usb take out put new kernel put back
<granite_crusher> no hassles with uart
<granite_crusher> less chance to corrput sd
<arokux1> granite_crusher, do you use external usb hub?
<granite_crusher> no
<granite_crusher> this is Olimex-A13
<granite_crusher> it have one build in
<granite_crusher> I just booted kernel :)
<arokux1> granite_crusher, with sd card you do not need to poweroff the board. i get to u-boot prompt then take out sd card and rewrite it. then stick it in and do: "reset" in u-boot.
<granite_crusher> well, my bad expierence (two cards tottaly damaged) just lays on my desk and reminds me to put board off
<steev> arokux1: you can just do mmc rescan rather than reset, unless you mean you're writing a whole new u-boot
<arokux1> steev, oh, good tip, thanks
<arokux1> granite_crusher, the safest way not to wear out anything is probably to connect an external usb hub
<arokux1> granite_crusher, but even more safest is to boot over NFS (if you only had ethernet.....)
balage has joined #linux-sunxi
<oliv3r> there, a nother thing done on my todo list
<oliv3r> update my fosdem abstract :)
balage has quit [Quit: leaving]
balage has joined #linux-sunxi
<arokux1> granite_crusher, i'm thinking. which kind of usb dongle do you have? usb2ethernet? if so, you could boot over NFS, i think.
<granite_crusher> AX8877
<arokux1> granite_crusher, ah, shit. usb2ethernet needs to be supported by u-boot :(
<granite_crusher> what about UART?
<arokux1> I know nothing about booting over UART, maybe somebody else knows...?
<oliv3r> arokux1: you'dneed to add 'all'usb ethernet drivers to u-boot; i don't know if there's any there atm.
<oliv3r> granite_crusher: wingrime boots via uart; so it's quite possible
<arokux1> I'll ask him to write a tutorial.
<arokux1> granite_crusher, anyway, what are your goals with this board and
<arokux1> oliv3r, A13 doesn't have an emac built into the SoC?
<balage> Hello, is this the right channel to ask for help about an issue I have on my hackberry?
<oliv3r> arokux1: nope nor sata, nor hdmi
<oliv3r> arokux1: a13 has very very few pins
<arokux1> balage, I guess so. Just ask.
<oliv3r> balage: #hackberry probbaly; but shoot and maybe we can help
<granite_crusher> now... I dont know anymore... I wanted to built functional field equipment (call it something like geopad) for messurements in geology
<arokux1> granite_crusher, why you do not want to build it anymore?
<balage> Thanks. I've compiled sunxi-3.4 for my hackberry yesterday. source of kernel tree, u-boot, etc are from official github. Mainly it's okay, but when i loaded rtl8192cu i got this: ERR: script_parser_fetch usb_controller_type failed
<granite_crusher> well it takes more time to build this thing whan would be to crawl on my feets in the field and map it simplly using hammer pencil and rubber
<granite_crusher> of course there is some crazy allternatives (crazy=overpriced) as Trimble
<arokux1> balage, but wifi works?
<granite_crusher> but no linux
<granite_crusher> linux without android
<balage> I found out that it tries to look for usb_controller_type in fex file but hackberry doesn't have that. My question is this, is it intentional or a bug? wlan0 cannot be seen btw.
<arokux1> granite_crusher, I see. but you need gps module etc..?
<granite_crusher> yeah
<granite_crusher> I got it all working on 3.4
<arokux1> granite_crusher, cool. so what is the next problem?
<granite_crusher> when SD card failed...
vinifr has quit [Quit: Saindo]
<granite_crusher> main problem is AXP209
<arokux1> granite_crusher, why is axp a problem?
<balage> 5~
<granite_crusher> this simply do not work as supposed, it underdischarges batteries
<balage> no wifi doesn't work (it worked with 3.0) I am using emac connection right now
<arokux1> balage, yesterday I've tried rtlwifi driver, the one from mainline, not from Relaltek, it worked like a charm.
<granite_crusher> some kind of leak somthere
<arokux1> balage, but you should make sure you enable a proper usbc in fex.
<arokux1> granite_crusher, I see. does it work with shipped android?
<balage> arokux1, I used the one from sunxi-boards and have only changed MAC, should I change anything else?
<arokux1> balage, "the one" -- what is this?
<arokux1> balage, let me check
<balage> arokux1, yes this one sys_config/a10/hackberry.fex
<arokux1> balage, ah, fex
<granite_crusher> arokux1: I bought it without NAND, because I am more familiar with Debian, and I saw there was quite many problems with GPS-android-gps deamon-gis application-postgis database
<arokux1> balage, fex is fine.
<arokux1> granite_crusher, I see. you can try to get support from Olimex, after all they've sold you the product...
<arokux1> granite_crusher, or you can try to take a look at AXP driver.......... but it is complicated.
<arokux1> balage, so just try to use kernel's rtlwifi maybe?
<arokux1> balage, btw, is it 8192cu from Realtek? (I was assuming it is..)
<balage> arokux1, so the problem is with the driver rtl8192cu (i was a bit confused as I've seen rtl8291cu_sw too) isn't and not with the fex. cool
<arokux1> balage, I haven't seen rtl8291cu_sw
<tgaz> the wiki has the pages I2C IIC and I²C ;)
<arokux1> balage, the last one is from Realtek
ykchavan has joined #linux-sunxi
<arokux1> balage, only 8192cu and rtl8192cu
<tgaz> (they all redirect to the same place, but i think we have all corners covered)
<balage> arokux1, i've got this in my config : CONFIG_RTL8192CU=m and CONFIG_RTL8192CU_SW=m though I don't know much about the latter one, is this the mainline version or is it from realtek?
<arokux1> balage, and its module name is 8192cu, make sure you load a proper one.
<granite_crusher> arokux1: Olimex sells development boards, not the end product; AXP... once I got such a troll from AXP - It was unable to shutdown, just keeping rebooting... or pretending it shuted down, but getting bizare consumtion of power (even twice more than in on mode with everything atached)
<oliv3r> balage: backreading, but the wifi is connected to the hackberry via USB, it just didn't route out the pins to a port; internally its connected to USB :)
<arokux1> granite_crusher, I see, still you are supposed to get a working driver for an AXP or a doc you can use to develop your own, don't you?
<oliv3r> balage: also note, that this driver is 'modified' by allwinner to add the fex stuff to it
<oliv3r> arokux1: i do'n't think olimex will support hackberry, as olimex doesn't sell hackberry :p miniand is of hackberry fame :)
<arokux1> oliv3r, we are talking about an A13 board from olimex (bought by granite_crusher )
<oliv3r> balage: so the _SW version is probably the SoftWinner-ized (Allwinner) version
<oliv3r> arokux1: balage is talking about hackberry, by miniand
<balage> arokux1, ok, I found out 8192cu loaded automatically, used rmmod and modprobe rtl8192cu did not get error any messages, though I haven't got any wlan interface either :)
<arokux1> oliv3r, but i was't suggesting him to contact olimex, was I?
<oliv3r> balage: chances are great you need the _CW version
<oliv3r> arokux1: you are correct, you suggested granite_crusher ;)
<tgaz> jukivili: i saw your name in the wiki. are you still working on musb? i'm interested in getting serial gadget console working, but don't seem to be even getting any ep0 RX interrupts...
<arokux1> tgaz, check out last message of him today in this channel.
<tgaz> arokux1: ah. :)
<arokux1> balage, everything worked for me. post your kernel log, plz (use if it is ok for you...)
<arokux1> oliv3r, I want to add EMAC to the table on the main page, I'm not sure how this row should be named
<oliv3r> arokux1: and now i know why he should contact olimex :) underdis-charging batteries
<arokux1> oliv3r, EMAC?
<oliv3r> arokux1: EMAC and GMAC we have, but why not LAN?
<arokux1> oliv3r, hm.. I mean this table: Comparison table for "A"-Series Allwinner SoC's
<arokux1> oliv3r, there is no EMAC
<oliv3r> Connectiviity? Networking i suppose
<arokux1> oliv3r, maybe Ethernet? Networking is too general?
<oliv3r> granite_crusher: anyway, the discharging isn't the AXP's fault nor Olimex, the axp gets configured by FEX from the linux driver (i don't know the details) but the driver should configure the axp appropiately
<oliv3r> granite_crusher: the problem however is, you need to know many details about your batteries to enter into the fex. Normally OEM's do that for the device (tablet) they sell (well they guess usually) but since you are using a custom battery, you have to define those settings
<arokux1> oliv3r, apparently the driver is buggy. so whos fault is this?
<oliv3r> granite_crusher: the AXP is very versatile
<oliv3r> arokux1: ours :D
<oliv3r> arokux1: well that or allwinner :)
<arokux1> oliv3r, no way. first it is the problem of olimex, then they forward it to allwinner.
<oliv3r> arokux1: if you use a 3.4 kernel, it's kinda ours; if it's the lichee kernel; it's allwinners fault; but good luck in getting help from them
<oliv3r> arokux1: no, olimex makes hardware; the software is all allwinner/us
<oliv3r> it's like calling microsoft that your motherboard says the CPU temperature is 200 C
<oliv3r> or calling asus that your program keeps segfaulting :)
<arokux1> oliv3r, if asus hardware is broken - yes.
<oliv3r> olimex doesn't write or supply drivers nor supports them
<oliv3r> but the board isn't broken, nor is the AXP; the driver/u-boot is broken
<granite_crusher> I thought it supports developers with hardware?
<oliv3r> anyway, that's not even the case probably
<balage> arokux1, here it is:
<oliv3r> granite_crusher: yep it gives away dev boards to improve support :)
<oliv3r> granite_crusher: which we work meticiously on ;)
<arokux1> balage, this is not kernel boot log. post the full kernel boot log.
<oliv3r> granite_crusher: anyway, you can't just randomly connect a battery and expect it to work. you need the fex parameters for your battery
<granite_crusher> arokux1: I am doing now more testing on U-boot USB
<oliv3r> balage: arokux1 means: dmesg | | curl -F 'sprunge=<-'
<oliv3r> with | removed
<arokux1> oliv3r, +1
hipboi has joined #linux-sunxi
<arokux1> granite_crusher, why? something is broken?
<granite_crusher> well because it is hub, what will happen if there will be more usbs connected...
<oliv3r> good testcase
<arokux1> granite_crusher, everything will work no matter how many hubs you connect.
<granite_crusher> arokux1: by the way hub is 1:4 (1 reserved for WiFi)
<arokux1> oliv3r, can you please fill in your knowledge here in Ethernet row?
<balage> arokux1, okay I managed to get it:
<oliv3r> granite_crusher: they used a USB hub to extend the 1 port of the a13? neat; i wonder if the a10 would have been cheaper
<oliv3r> arokux1: sure; i may have dropped from the most commits so extra commits are good :)
<oliv3r> arokux1: no matter how many hubs? there's a limit of 128 devices isn't there :)
<arokux1> oliv3r, :$
<balage> oliv3r, thanx, much easier than the web form :)
<oliv3r> arokux1: done
<arokux1> balage, hm.. where are your SUNXI_EHCI things? you need to enable them
<mnemoc> sprunge's web form is just a js hack
<arokux1> balage, CONFIG_USB_SUNXI_EHCI
<oliv3r> arokux1: he probably didn't think it was connected via USB :)
<balage> arokux1, it's enabled in my kernel CONFIG_USB_SUNXI_EHCI=y
<arokux1> balage, post lsusb
<arokux1> granite_crusher, why did you give a reference to that pdf? and why do you think there is something reserved?
<balage> oliv3r, i know it is connected to the USB, actually soldered as far as I know.
<arokux1> balage, +1
<balage> arokux1, ahaa, that's it: Bus 001 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub
<arokux1> balage, too bad. this one is OTG
<balage> arokux1, nothing else
<arokux1> balage, is this on? CONFIG_USB_EHCI_HCD
<balage> arokux1, I'll post my kernel .config soon
<balage> arokux1, full config is here:
<arokux1> balage, yes, should be.
<oliv3r> balage: also make sure you enable USB power via the FEX file
<granite_crusher> arokux1: oh its because my ignorance it works superb; I was trying to get to 3 pluged in usb sticks in u-boot with devices 0:1 0:2 0:3 instead of 0:1 1:1 2:1
<oliv3r> balage: the power supply for USB is connected to a swithch, which is turned on/off via a GPIO, which is defined and setup in the FEX file. The driver reads this setting and enables power. That is the reason you need the right fex-settings, and the correct driver (to actually perform this action)
docice_ has joined #linux-sunxi
<oliv3r> arokux1: is that's the one he is actually using?
docice has quit [Ping timeout: 264 seconds]
docice_ is now known as docice
docice has quit [Client Quit]
<arokux1> oliv3r, hope so. he was saying it.
<oliv3r> :D
<balage> oliv3r, that explains it, but isn't this something should have been done by the hackberry.fex?
<arokux1> balage, ... there was a fix recently. have you updated your fex?
<balage> oliver, I meant by default
<arokux1> balage, in any case something like sunxi ehci/ohci should show up.
<oliv3r> balage: make sure you use ^ linked fex file for your tests
<balage> arokux1, as far I know yes, i git pulled yesterday, but can double check it @ home
<arokux1> balage, you could insert several printk's into the kernel source and recompile it to see what happens
<oliv3r> balage: fex2bin and then copy it to nanda
<oliv3r> arokux1: or SD1 :p
<oliv3r> arokux1: i dunno how he boots :)
<oliv3r> mmc0p1 actually
<arokux1> oliv3r, why nanda? :)
<arokux1> oliv3r, why do you rush advising then...
<oliv3r> so many people, so many questions :(
<oliv3r> arokux1: nanda because of the nand question on ML :)
<oliv3r> typo is all
<balage> oliv3r, I am booting from sd card, so mmcblk0p1 actually. so as far as I understood the problem is not about kernel config but something around the fex file? I'll check it..
<balage> olive3r, arokux1, thanks a lot guys, I'll check it, if I'll manage to do it I'll let you know, if not, then the same :)
<arokux1> balage, yes, let us know.
<arokux1> balage, still there?
<arokux1> can you post the output of: cat /proc/ccmu ?
<arokux1> oliv3r, back to our olimex discussion. I think they need to provide a decent doc how to configure axp in fex
<balage> arokux1, yes for a couple of minutes.. here it is
<arokux1> balage, ok, usbclocks are not enabled.. so something is just not enabled.
<balage> arokux1, you meant in the fex file or in the kernel config?
<arokux1> balage, looks like in fex. btw, after pulling stuff, did you do fex2bin and copy it to the sd card?
<balage> arokux1, as far as I can remember yes, though last time i've done this was with 3.0 some months ago, so I'll make sure doing that too
<arokux1> balage, ok
popolon has quit [Remote host closed the connection]
popolon has joined #linux-sunxi
popolon has quit [Client Quit]
<balage> arokux1, gotta go now..
<arokux1> balage, cu
<balage> arokux1, bye
balage has quit [Quit: leaving]
FDCX has joined #linux-sunxi
<granite_crusher> arokux1: is there plans for AXP drivers in mainline?
<arokux1> granite_crusher, yes, I've started to work on it. but i'm a noob, and i do not have any battery equipment (unless someone donates) so do not expect to see something big and do not expect to see something working in a week. it can take month(s)
<arokux1> granite_crusher, usb patches I've submitted do not work 100% also :( usb storage works, but on-board wifi module does not (it is connected through usb)
ganbold_ has joined #linux-sunxi
ganbold_ has quit [Remote host closed the connection]
<granite_crusher> arokux1: well I have other idea for battery management (especially LiFePO4 alternative is tempting which is rather impossible on AXP); however AXP do ther not less important things as: CPU scaling - power saving, proper shutdown and restart. If these would work AXP would be at least useful, and not a bother
<granite_crusher> arokux1: I do not have Wifi module
<arokux1> granite_crusher, yeah, but I have and want it to work in mainline :(
<granite_crusher> arokux1: where is your usb kernel pathes in devel?
<arokux1> granite_crusher, they are here: but they probably won't work for you.
<arokux1> granite_crusher, first try to boot sunxi-next, if you want to play with mainline
<granite_crusher> arokux1: I said I already did
<arokux1> granite_crusher, ah! I thought it was sunxi-3.4, sorry.
<arokux1> granite_crusher, so you used cpio for initramfs and it worked?
<granite_crusher> yeah
<granite_crusher> thanks to this u-boot usb now life gets easer
<granite_crusher> easier*
<arokux1> granite_crusher, nice. my usb patches won't work for A13 at all, there is no binding in the device tree.
<arokux1> granite_crusher, do you want me to add it? so that you can test?
<arokux1> granite_crusher, honestly I do not understand what is the point for you to waste your time on mainline. grab feature full sunxi-3.4 and move along with your project.
<granite_crusher> winter is comming
<granite_crusher> I think it is too late to go to fields this year
mcbrick has joined #linux-sunxi
<arokux1> granite_crusher, so you are in the hacking mood? :)
<granite_crusher> arokux1: well, I think so :) but I need to learn a lot (I know just how to program in awk; seriouslly)
<arokux1> granite_crusher, do you have something else except A13-based board, btw?
<arokux1> granite_crusher, something else alwinner-based.
<granite_crusher> fortunatelly/unfortunattly no
<granite_crusher> :)
<arokux1> granite_crusher, alright. so do you want me to add support for A13 usb in mainline? (say yes, only if you'll test it till it works :p)
<arokux1> granite_crusher, (because I do not have hardware)
<arokux1> Tsvetan, there?
<granite_crusher> arokux1: yes of course
<arokux1> granite_crusher, ok. tonight I can do it, I think. if not, tomorrow in the evening.
<granite_crusher> arokux1: take your time
\\Mr_C\\ has joined #linux-sunxi
<arokux1> granite_crusher, do you know electronics? i know nothing about it, so I may need help with axp driver..
<granite_crusher> arokux1: I have very minimal knowledge
<arokux1> granite_crusher, ok
<granite_crusher> arokux1: AXP209 is used with A13 and in theyr user manual it have some memory addreses given for PMU (it looks different kind of information than in axp209) but it looks it is about the same AXP209, because twi address is the same; did you saw that manual? would it be useful?
<arokux1> granite_crusher, I think AXP152 is used with A13
soldoKyn has quit [Quit: Konversation terminated!]
<granite_crusher> no
<arokux1> granite_crusher, if you have something else that is here: then please show it :)
<arokux1> granite_crusher, ah, you r right.
<granite_crusher> arokux1: no, atleast with olinuxino-A13
<granite_crusher> I can read it on chip - it is AXP209
<arokux1> granite_crusher, yes, I've mixed it. 10s uses 152.
<arokux1> granite_crusher, ok, thanks, good to know. btw, you can find all the info on AW chips here:
<arokux1> granite_crusher, here is info about A13:
wingrime has joined #linux-sunxi
<Tsvetan> arokux1 pong
<arokux1> Tsvetan, could you maybe donate an A13 board to me?
<Tsvetan> arokux1 what you will do with it?
<arokux1> Tsvetan, test my mainline work.
<arokux1> Tsvetan, so far I've added usb support (usb storage works, wifi module not yet), also I've added usb to u-boot.
<arokux1> Tsvetan, A13 is a bit different from the others, so i should test on it.
<Tsvetan> arokux1 ok, send me your address
<arokux1> Tsvetan, I'm getting Cubietruck donated from Cubietech, and I own Mele A1000, so I'll have: A10, A20 and A13 (if you donate) for testing.
<Tsvetan> btw how you guys going to mainline linux-sunxi with FEX inside?
<arokux1> Tsvetan, we're writing everything from scratch
<Tsvetan> each time they release something new will be merged to mainline
<arokux1> if you want, you can ask me to work on specific driver first, no guarantees though.
<Tsvetan> as they refuse to use DT
<Tsvetan> arokux1 this is not necessary
<arokux1> "no guarantees" I can manage it in some time i mean.
ZetaNeta has joined #linux-sunxi
<Tsvetan> A13 have everything we need in 3.4
<Tsvetan> for the hardware support
<arokux1> Tsvetan, so why are you donating? :)
<Tsvetan> arokux1 so why you work for Allwinner for free ;)
<Tsvetan> hehehe
<arokux1> Tsvetan, fair enough.
<Tsvetan> open source is state of mind, not necessary should give you immediately benefit :)
<granite_crusher> if I could interact, well it was not everything: for example latest usb u-boot support! this is really cool; even with 3.4 (super blazing boot through usb); but this came as side effect of mainling
<granite_crusher> *mainlining
hipboi has quit [Quit: Leaving]
ZetaNeta has quit [Quit: ZetaNeta]
<arokux1> thanks granite_crusher . it is very pleasant to know my patches are useful for somebody :)
<granite_crusher> arokux1: just compare: SD 10class about 15-20MB read 1-5MB write vs good USB2.0 pendrive 40+MB/s read 15MB/s write
<arokux1> granite_crusher, yes it is a bit faster, but you are not loading so much stuff by u-boot, only 3-7MB (kernel image)
Soru has quit [Read error: No route to host]
<arokux1> granite_crusher, on the other hand presence of USB support in the kernel lets you use USB external drives as storage and rootfs -- this is indeed very cool. taking into account there is no other storage yet.
rellla has quit [Quit: Nettalk6 -]
<granite_crusher> arokux1:what about UART0 which gets muxed with some SD pins; after u-boot loaded from SD, confilicting pins probbably could be switched to UART0 for debuging? freeing UART1 from that duty for peripherals as GPS
<arokux1> granite_crusher, sorry, I'm still learning, this question is too advanced for me, maybe oliv3r can help
<granite_crusher> oliv3r: are you familiar with Olimex-A13 UART0/SD muxing?
n01 has quit [Quit: Lost terminal]
<mripard> arokux1: it's not deprecated, it's superseeded on ARM
<arokux1> mripard, superseded means? I should better use device tree, right?
<arokux1> mripard, also I have a question about GPIOs.. if you have several spare minutes.
<mripard> arokux1: superseeded means it's not the preferred way of doing it
<mripard> and yeah, shoot.
<arokux1> mripard, ok. here emac_power has <default> for a pull, and USB GPIO has 1, what parameter should be used for ",pull" in pinctrl in both examples?
<mripard> I'm not sure it makes much sense to set up a pullup for the power pins.
<mripard> and what is USB_GPIO?
popolon has joined #linux-sunxi
<arokux1> mripard, USB GPIO something that needs to be poked so there is power on VBUS
<arokux1> mripard, about pull up.. I do not know anything about it, the problem is I've discovered usb patches are buggy a bit. usb storage works, but communication with wifi module works only partially.
<arokux1> mripard, this is kernel log: as you see the wifi module gets correctly identified, but as soon as I do ip dev link wlan0 up I get errors:
<arokux1> sunxi-ehci 1c1c000.ehci1: detected XactErr len 0/0 retry 23
<arokux1> note, the hardware works with sunxi-3.4
<bfree> oliv3r: pong ;) too much backlog there right now to figure out why you were pinging, just skimmed it, but all I did for debian was push the cherry-picked emac patches from 3.11 to a wishlist bug report so they were included in their 3.10
<oliv3r> granite_crusher: unfortunatly no, but we have a wiki page for A13/PIO i'm guessing :)
<mripard> arokux1: yeah, well, the rtl8192 is quite a pita to work with
<mripard> arokux1: but I don't really know what could be going on, sorry.
<arokux1> mripard, well, everything works with 3.4 (with mainline driver, not realtek one)
<mripard> so, let's rephrase
<granite_crusher> olive3r: I mean I know there TX and RX lines of UART0 is data3 and sck lines for SD; what I would want to know if it is possible/safe to switch to UART0 with SD card pluged in (but not used), if comunication with UART0 wouldn't write some garbage to SD?
<arokux1> mripard, no problem, i'm just looking for possible issues. and found this pullup thingy.
<mripard> with a completely different USB driver, hardware support code, wifi support, everything works ?
<bfree> well ok, I prepared some deb packages (for debian unstable) of linux-sunxi kernel and u-boot (and sunxi-tools kinda) also ... they are a little out of date now, had been waiting for a20 support to land in sunxi stable and am now away until start of october, at which point I'll try and get them working and updated for a10 and a20 (cubie1 and cubie2 is all I can test though)
<mripard> arokux1: I'm not quite sure we can conclude anything from that :)
<arokux1> mripard, yes it is true. at least it confirms hardware is working.
<arokux1> thanks mripard I'll continue to look for solution.
<oliv3r> granite_crusher: ok so lets see, you wanna boot from SD and use UARt0 from those exact same pins?
<arokux1> mripard, btw, how is submitting of the patches to mainline, storage works... can I mention there are problems with wifi and submit it?
_BJFreeman has joined #linux-sunxi
<granite_crusher> oliv3r: with latest ability to switch to USB in u-boot, yeah
<oliv3r> bfree: can you link me the wishlist? so I can cherry-pick the same patches :)
_BJFreeman is now known as BJfreeman
<oliv3r> granite_crusher: those pins work either/or; I don't think it's wise to have an SD card in there, when you configure it as uart. there's absolutly no guarantee your card will remain undamged etc. It 'might' but I wouldn't count on it at all
<bfree> oliv3r: they are probably a little stale now (i.e. there have no doubt been more improvements on top of it) but
<oliv3r> granite_crusher: but if youw ant to boot from SD, to chainload SD, and then use uart from those pins; that's no problem
<granite_crusher> oliv3r: chainload? you mean take out SD?
<oliv3r> granite_crusher: BROM configures those pins per default to be an SD reader. then uboot comes along, reads its files and hands over to the kernel. at that point you can safely remove the SD card
<oliv3r> granite_crusher: the kernel then loads, and reconfigures the pins (if you told it to do so, via dts for example) and the kernel will do so
ganbold_ has quit [Remote host closed the connection]
<oliv3r> granite_crusher: BROM 'chainloads' u-boot. U-boot 'chainloads' the kernel
<oliv3r> :)
<mripard> arokux1: I'd still try other USB devices
<mripard> 50% of working devices is not that great.
<oliv3r> mripard: welcome back!
<granite_crusher> oliv3r: thats simplest thing, but not elegant
<arokux1> mripard, ok, I'll see what I can try, I have almost no usb devices :(
<oliv3r> granite_crusher: what i'm saying is, you can reconfigure the pins so they are different in the kernel and in u-boot and in BROM, but you can't leave the SD card in and hope all is well :)
<oliv3r> mripard: jetlagged much?
<oliv3r> bfree: don't think too many tbh
<mripard> oliv3r: mostly sick :)
<mripard> jetlag is ok
panda84kde has quit [Quit: Konversation terminated!]
<oliv3r> mripard: sick? from all the alcohol? :p
<mripard> I wish
<oliv3r> flue?
<mripard> from all the air conditionning
<oliv3r> ahh, yeah
<mripard> a bad cold
<mripard> I still don't get why you would set the air conditioning to 15-17 when it's warm outside.
Cubear has joined #linux-sunxi
<Turl> mripard: 22-24 ftw
<mripard> exactly
<mripard> I mean, some air conditioning when it's 35 degrees outside make sense
<mripard> but come on
<oliv3r> mripard: luxury i suppose; or just crazyness
<oliv3r> mripard: i like it at 15-17 ... at night when i go to sleep
<oliv3r> 22-24 is warmish, 21 is nice
<oliv3r> then again, when you live i awarm enviroment; yeah 22-24 is fine
<mripard> US and their usual waste of resources
massi_ has quit [Quit: Sto andando via]
<mripard> pretty much like the "hey guys, let's build a city in the middle of thte desert just to play poker and get wasted!"
<oliv3r> hahah yeah
<oliv3r> i have mixed shared feelings
<oliv3r> there's things I admire and like
<oliv3r> there's things I dispise and hate
<buZz> mripard: lets call this city Taxes!
<buZz> 'rightyo Texas it is!'
<oliv3r> :p
FR^2 has quit [Quit: Connection reset by peer]
<Turl> mripard: yay acks :)
atiti has quit [Ping timeout: 245 seconds]
<arokux1> Turl, so you'll be submitting PLL6 soon? :p
<Turl> arokux1: yeah, the PLLs are next in queue
<arokux1> Turl, +1
rellla has joined #linux-sunxi
Cubear has quit [Quit: Leaving]
<oliv3r> Turl: pll6 is in your sunxi-clk allready, right?
<Turl> oliv3r: yes, for all the things except A31
<oliv3r> crap :(
<Turl> oliv3r: you're working on an A31?
<oliv3r> not atm
<oliv3r> it depends on what A40 will look like
<oliv3r> a51 will be powerVR crapness too
<oliv3r> so ignoring that for now
<Turl> they announced new cpus? :o
<oliv3r> IF a40 has the same dualchannel memory controller; i'll resurect a31 u-boot-ness
<Turl> last I heard was A23
<oliv3r> oh a51 (or whatever it may be called) is coming after a23
<oliv3r> powerVR
<Turl> I think it'll be some time until we see an aw chip
<arokux1> oliv3r, why are you so limited to GPU? there is a need for headless systems too.
<arokux1> for which 4-core is a cool thing, isn't it?
<oliv3r> Turl: a23 is just around the corner, a23
<Turl> oliv3r: after a23 that is
<oliv3r> won't be long :)
Tsvetan2 has joined #linux-sunxi
<oliv3r> inside info :)
<arokux1> oliv3r, you should ask for docs :p
<oliv3r> arokux1: LOL funny man
<oliv3r> 'use a31 docs, they are the same
<oliv3r> they write docs when needed; not ahead of time
<Turl> oliv3r: A31 docs are not GA still
<Turl> nor are A10S :(
<arokux1> oliv3r, yes, I suspected they do not have any docs actually, only sources.
<arokux1> oliv3r, then ask for git history :p
<arokux1> oliv3r, ask for something.........
Tsvetan has quit [Ping timeout: 251 seconds]
<oliv3r> they don't give us anything nor care generally
<arokux1> oliv3r, still they give you some info
<oliv3r> sheer luck :)
<arokux1> keep looking for luck oliv3r :)
fldc has left #linux-sunxi [#linux-sunxi]
atiti has joined #linux-sunxi
ykchavan has quit [Ping timeout: 252 seconds]
n01 has joined #linux-sunxi
<oliv3r> 3.4 debug time
ykchavan has joined #linux-sunxi
hno has joined #linux-sunxi
hno has quit [Excess Flood]
BJfreeman has quit [Ping timeout: 240 seconds]
ykchavan has quit [Ping timeout: 240 seconds]
lkcl_ has quit [Ping timeout: 252 seconds]
notmart has quit [Quit: notmart terminated!]
FDCX has quit [Ping timeout: 276 seconds]
lkcl_ has joined #linux-sunxi
Black_Horseman has joined #linux-sunxi
tzafrir has quit [Ping timeout: 245 seconds]
eebrah|noPC has joined #linux-sunxi
<granite_crusher> Tsvetan2: ping
eebrah|noPC has quit [Ping timeout: 260 seconds]
deasy has joined #linux-sunxi
FR^2 has joined #linux-sunxi
arokux2 has joined #linux-sunxi
* atsampson peers oddly at the GPL header in the 3.10 device trees (and, it appears, in various other bits of arch/arm code)
<atsampson> shouldn't the "Version 2 or later" be in the first sentence, not the second one?
FR^2 has quit [Quit: und weg...]
n01 has quit [Quit: Lost terminal]
rellla has quit [Ping timeout: 260 seconds]
rellla has joined #linux-sunxi
akaizen has quit [Read error: Connection reset by peer]
akaizen has joined #linux-sunxi
<arokux2> any body knows some USB kernel experts? :)
rellla has quit [Remote host closed the connection]
<arokux2> leviathanch, hacking time? :)
<oliv3r> arokux1: i've booted stock android; and noticed this in my logs:
<oliv3r> [ 1113.450000] WRN:L841(drivers/usb/host/ehci_sun4i.c):ERR: not support sw_usb_disable_ehci
<oliv3r> [ 1113.460000] WRN:L727(drivers/usb/host/ohci_sun4i.c):ERR: not support sw_usb_disable_ohci
<oliv3r> it's from the wifi module btw; not important
<oliv3r> rtl8192c, point is, therew is/was ehci AND ohci
tinti has quit [Remote host closed the connection]
<oliv3r> arokux1: could be, that wifi is giving you troubles, because it falls back to 1.1 mode for some operatiosn
<leviathanch> arokux2: well
<leviathanch> more like sleeping time
<leviathanch> had 12hours lectures today
<leviathanch> >_>
<leviathanch> arokux2: and I still haven't managed to get the memory access working
<oliv3r> sleep time indeed
<arokux2> oliv3r, sunxi-3.4 works without ohci.
<leviathanch> hmm
<oliv3r> arokux2: i think hansg is/was USB expert :)
<oliv3r> but he is on ML only
<oliv3r> arokux2: he did a lot of mainline usb work/webcam work
<Turl> leviathanch: pss :p
<oliv3r> arokux2: i'll test it this week
<oliv3r> arokux2: ah i do see there was a unification patch; that unifies ohci and ehci!
<oliv3r> arokux2: done in okt. 2012; so about a year ago
<oliv3r> arokux2: that explains a LOT :)
<arokux2> oliv3r, I have a several simple facts. #1 hardware is the same #2 sunxi-3.4 (EHCI only)+rtl8192cu works, #3 mainline + my driver *doesn't* work.
<oliv3r> sunxi-3.4 is NOT ehci-only :)
<oliv3r> sunxi-3.4 is ohci + ehci unified
<arokux2> oliv3r, I've commented out ohci.
<oliv3r> hmm
<oliv3r> you break things :p
<oliv3r> anyway, it explains why there's only ehci driver in 3.4 now
<oliv3r> here's lsusb from stock android: shell@android:/ $ lsusb
<oliv3r> Bus 001 Device 001: ID 1d6b:0001
<oliv3r> Bus 002 Device 001: ID 1d6b:0002
<oliv3r> Bus 002 Device 002: ID 0bda:8176
<oliv3r> Bus 003 Device 001: ID 1d6b:0001
<arokux2> oliv3r, there is no "only" ehci, there are two.
<arokux2> sunxi-ehci.c and sunxi-ohci.c
<oliv3r> 1d6b:0001 is a Linux Foundation 1.1 root hub
<oliv3r> arokux2: those where unified in okt.
<oliv3r> see the patch; dunno if it got merged btw, true that
<arokux2> oliv3r, no. check the tree!
<oliv3r> arokux2: damn :(
<oliv3r> 1d6b:0002 is a Linux Foundation 2.0 root hub
<oliv3r> obda is the rtl wifi
<arokux2> oliv3r, how can ohci and ehci get unified?!
<oliv3r> but! stock android has both usb1.1 and usb2.0 hubs
<oliv3r> that'st he unification patch
<arokux2> oliv3r, don't even want to look at it :p
<oliv3r> it's from a fellow sunxi member ;)
<oliv3r> ah no i read it wrong
granite_crusher has quit [Quit: Page closed]
<oliv3r> it's sun4i + sun5i = sunxi unification
<oliv3r> so nvm
<oliv3r> well atleast i tried fixin' the wifi issue :)
<oliv3r> and that at this late hour :)
<oliv3r> bed time for now :)
<oliv3r> nn all
alcides has joined #linux-sunxi
<leviathanch> Turl: the card answers
<leviathanch> it initializes
<Turl> leviathanch: :)
<leviathanch> but the MemoryAccessDone bit never is set within the status register
<leviathanch> which leads to the timeout error
<leviathanch> Turl: is there a documentation describing the mmc registers?
<leviathanch> Turl: since the user manual only tells me to read the mmc standard...
<leviathanch> -.-
<arokux2> leviathanch, chinese source code :p
<leviathanch> arokux2: lol
<leviathanch> arokux2: that's what I'm working with already :-P
<leviathanch> anyway
<leviathanch> I'll sleep
<arokux2> leviathanch, or better: chinese sauce :D
<leviathanch> >_>
<Turl> leviathanch: if user manuals don't say a thing, then no
<leviathanch> glas noodle soup code
<leviathanch> >_>
<leviathanch> hmm
<leviathanch> ok
<leviathanch> Turl: well then
<leviathanch> in this case
<leviathanch> I'll take a nap
<leviathanch> and will try to get it working tomorrow afternoon or so
<leviathanch> n8
<leviathanch> :-)
alcides has quit [Remote host closed the connection]
<arokux2> gn
<arokux2> hm.... where is fex2bin?!
<arokux2> ok, ignore it, it needs to be built....
<mnemoc> fex2bin is a symlink to fexc
<mnemoc> good night
<Turl> night mnemoc
deasy_ has joined #linux-sunxi
deasy has quit [Ping timeout: 246 seconds]
popolon has quit [Quit: Quitte]
tinti has joined #linux-sunxi
wingrime has quit [Ping timeout: 245 seconds]
rah has quit [Ping timeout: 246 seconds]
rah has joined #linux-sunxi
naobsd has quit [Quit: Page closed]