<bennnnn>
a80 will have more binary blobs :( gmac(but same with a20), video, gpu, nand
<ganbold>
Turl: wow, nice
<Turl>
bennnnn: nand? :|
<Turl>
bennnnn: for previous SoC they released source on request I think
akaizen has quit [Read error: Connection reset by peer]
akaizen has joined #linux-sunxi
<bennnnn>
Turl: I don't like nand :)
<Turl>
bennnnn: did you make CT with eMMC / tSD?
<ganbold>
bennnnn: so a23 dev board can have such features like this Atmel Xplained board, of course if it is possible
<ganbold>
bennnnn: so when you think it will be available?
eagles0513875_ has quit [Ping timeout: 246 seconds]
<bennnnn>
Turl: CT can support tSD directly. I've tested it. And I have not test eMMC
<bennnnn>
ganbold: I am thinking how the board should be. If we have decided to make it, it will be very soon
<ganbold>
cool
apo has quit [Ping timeout: 264 seconds]
* ganbold
thinks maybe he can get CT to port gmac driver to FreeBSD
<Turl>
bennnnn: can you fix uart vampire leak on it? :p
<ganbold>
Turl: so does linux-sunxi has driver for this CAN controller?
* ganbold
thinks or maybe wait a23 or a80 boards
<Turl>
ganbold: I don't think so; probably neither does allwinner
<wens>
ganbold: a23 doesn't have ethernet
<ganbold>
ah, I see
<Turl>
it's like an A13
bennnnn has quit [Read error: No route to host]
<ganbold>
yeah, seems so
<ganbold>
so better wait a80 then
<ganbold>
or get CT
bennn has joined #linux-sunxi
<wingrime>
Turl: together with leak , CT have bad placement of wifi IC , that too close with VGA connector, VGA can have noice on some resolution when wifi working
<ganbold>
bennn: so count me the one who can try freebsd on a80 board when it is available :)
<bennn>
ganbold: got it :)
<Turl>
wingrime: yes, I saw note on wiki
<bennn>
wingrime: think for your report :)
<wingrime>
bennn: please consider it on next revisions)
<Turl>
wingrime: irssi ok?
<wingrime>
Turl: whait a sec
JohnDoe_71Rus has joined #linux-sunxi
apo has joined #linux-sunxi
<bennn>
ssvb: does xrandr work on linux-sunxi platform ?
<wingrime>
Turl: yes, irssi starts , but I have confiugre it
<wens>
bennn: maybe release some early design schematics for us to check out
akaizen has quit [Remote host closed the connection]
<bennn>
wens: good suggestion
joedj has quit [Ping timeout: 265 seconds]
TheSeven has quit [Disconnected by services]
[7] has joined #linux-sunxi
<Turl>
bennn: I don't think it is supported, but don't take my word for it
<mrnuke>
bennn: also, credit card and bank account information would be a nice thing to release
<Turl>
bennn: sunny723 is from cubietech? :)
<bennn>
how I can switch VGA and HDMI for X on the fly (linux-sunxi platfrom)?
<bennn>
Turl: yes
<Turl>
bennn: made same question today checking the logs for xrandr :)
<bennn>
Turl: we have college students :D
<bennn>
Turl: yes, I ask him to find out a solution
<Turl>
bennn: you can configure screen mode via script.bin or kernel cmdline
<Turl>
bennn: I think there is also tool to set fb resolution (fbset or something)
<wens>
libv: I will post my fex file, but you should do yours, too, then we can see if anything changed
joedj has joined #linux-sunxi
kuldeepdhaka has quit [Ping timeout: 240 seconds]
<bennn>
Turl: with script.bin, we cannot change the display mode on the fly
<bennn>
Turl: the same to change the resolution
<bennn>
Turl: you know, CT have HDMI and VGA port. but many people don't have HDMI
<bennn>
Turl: to build to image is not convenient
<bennn>
2 images
<Turl>
bennn: yeah :(
<bennn>
Turl: a simple way is to add a hdmi hotplug event trigger at driver code. udevd receive the event, make a switch by telling driver and killing X server
<bennn>
Turl: It works! but I think it's a bad solution
<bennn>
Turl: android can switch VGA and HDMI smoothly
<Turl>
bennn: you should talk with ssvb and libv :)
<Turl>
bennn: libv is writing KMS driver for sunxi to replace disp
<bennn>
Turl: great information!
<wens>
Turl: did you see Olof's mail on the rename series?
<ojn>
wens: linux-next isn't broken though, so I suspect we have a missing dependency
<ojn>
oh wait, -next is in bad shape too
<ojn>
hm
<wens>
ojn: oh, sorry didn't notice you were here
<ojn>
I need to pay closer attention
<ojn>
wens: no worries. people don't connect my nick to me always, and that's fine. :-)
<Turl>
ojn: I think we discussed this already
* Turl
hunts for convo
<ojn>
Note that I might be on this channel, but I rarely monitor discussions on it. :-)
<ojn>
anyway, i'm calling it a day here. please follow up over email
<Turl>
ojn: check if "arm-soc boot: 37 pass, 6 fail (v3.14-rc4-695-g7c79649)" is the same issue please
<ojn>
ah yeah, i even commented on that thread
<ojn>
ok, we need to merge in the corresponding clk changes then
<wingrime>
oliv3r: ping
<Turl>
ojn: I'll try to send the pull tomorrow night at the latest if you need it soon
<ojn>
Turl: yeah, things are closing down for the merge window, it's a bad idea to have all of sunxi broken in -next. it means we won't find any other, new, regressions that might be added from other trees and so on
<Turl>
ojn: yeah
<Turl>
ojn: I don't expect any other massive compatible rename on the horizon so we should be good to go
<ojn>
Turl: good. we usually don't like them much. :)
<Turl>
and we'll take more care with the dependencies in the future
<Turl>
ojn: :)
<Turl>
ok, time to get some sleep. Good night guys :)
<ssvb>
bennn: xrandr does not work at the moment, that's going to be provided by the kms driver later
<wens>
got a second hand cubieboard 1, and the power supply that came with outputs at 5.5V under load
<wens>
my boards make weird noises when I use that one
HeHoPMaJIeH has joined #linux-sunxi
Gerwin_J has joined #linux-sunxi
<ganbold>
maybe it is adapter that makes noise
<wens>
nah, it's from the board
bbrezillon has quit [Ping timeout: 264 seconds]
bbrezillon has joined #linux-sunxi
<JohnDoe_71Rus>
some litle noise cb2 then write to SD card
<swabbles>
wens: I've got the same for my Cubietruck with a certain adapter.
<swabbles>
But the thing is, that my Cubieboard 2 doesn't produce those noises as far as I know.
<wens>
swabbles: get my axp209 hwmon patch, and get a reading for ACIN
lag` has quit [Quit: Coyote finally caught me]
lag` has joined #linux-sunxi
lag` has quit [Changing host]
lag` has joined #linux-sunxi
lag` is now known as lag
Black_Horseman has quit [Ping timeout: 252 seconds]
FunkyPenguin has quit [Remote host closed the connection]
FunkyPenguin has joined #linux-sunxi
diego_r has joined #linux-sunxi
cubear has joined #linux-sunxi
_massi_ has joined #linux-sunxi
<mripard>
mrnuke: the issue is not spidev itself, but rather how you load it.
<mripard>
mrnuke: and yes, eventually we will support spi transfer sizes larger than the FIFO
<mripard>
most likely through DMA
newleaves_ has quit [Quit: Leaving]
nabblet has joined #linux-sunxi
FR^2 has joined #linux-sunxi
grafgustav has joined #linux-sunxi
setkeh has quit [Ping timeout: 256 seconds]
setkeh has joined #linux-sunxi
bonbons has joined #linux-sunxi
notmart has joined #linux-sunxi
maksimlin has quit [Quit: ChatZilla 0.9.90.1 [Firefox 27.0.1/20140218140359]]
Gerwin_J has quit [Quit: Gerwin_J]
<bonbons>
to select a given function for some of the pin-headers on cubietruck, do I have to change the fex or can I do that at runtime?
popolon has joined #linux-sunxi
popolon has joined #linux-sunxi
<grafgustav>
Hello! I use the cubieboard and just tried to load the crosscompiled module spidev into my kernel. I have copied the file to the correct directory and executed depmod -a. Then I called modprobe spidev and there still was no error. However lsmod does not display spidev as a loaded module, how is this possible? I have to admit that the versions are slightly different .79-sun71+ kernel and .79-sun7i module, but I hoped it worked anyways. Is this th
<gzamboni_>
grafgustav, do you have the spi dev in your /dev directory ?
<gzamboni_>
you have to modprobe the sunxi spi module
<gzamboni_>
if you're using spi0 on a A10/A20 the mux should be 2 as specified in the spidev tutorial
<grafgustav>
Now I finally get what those mux colums are for... yeah, the mux seems to be set right except for one. The cs1. We don't need that one so far though, because we just have one sensor attached right now
<gzamboni_>
no, you dont need if you dont have to use chip select
Gerwin_J has joined #linux-sunxi
<gzamboni_>
just be sure your devices have the chipselect enabled when you do start the comunication
<gzamboni_>
you can set always enabled once you only have one sensor
Dodger78 has joined #linux-sunxi
<gzamboni_>
next step is to check with an osciloscope if data is getting out from the mosi/miso pins
JohnDoe_71Rus has quit [Quit: Miranda IM! Smaller, Faster, Easier. http://miranda-im.org]
Black_Horseman has joined #linux-sunxi
Black_Horseman has joined #linux-sunxi
Black_Horseman has quit [Changing host]
leviathanch2 has quit [Ping timeout: 240 seconds]
focus has quit [Remote host closed the connection]
geecko has joined #linux-sunxi
leviathanch2 has joined #linux-sunxi
focus has joined #linux-sunxi
focus has quit [Remote host closed the connection]
Black_Horseman has quit [Remote host closed the connection]
kuldeepdhaka has quit [Ping timeout: 255 seconds]
pwhalen has quit [Quit: Leaving]
leviathanch2 has quit [Ping timeout: 264 seconds]
kuldeepdhaka has joined #linux-sunxi
deasy has joined #linux-sunxi
ganbold_ has joined #linux-sunxi
HeHoPMaJIeH has quit [Remote host closed the connection]
pwhalen has joined #linux-sunxi
<bonbons>
trying to build kernel (sunxi-3.4) with most sunxi drivers built-in I end up having symbol errors, is that expected? (e.g. bcmsdh_probe exported twice, ump_dd_secure_id_get undefined)
pwhalen has quit [Ping timeout: 252 seconds]
<libv>
yes, there is a symbol clash between two bcm drivers
<bonbons>
eek, bcmsdh_probe is exported 5 times through sunxi-3.4 tree!
<libv>
the joy of vendor drivers
<mrnuke>
i =5; i = 5; i=5; //Make sure I is 5 or the code below will crash
<bonbons>
:), any idea for the ump/mali area?
<libv>
bonbons: what do you mean with ump/mali "area"
<libv>
is that some special military base in new mexico?
<bonbons>
ump_dd_secure_id_get symbol missing and a second one
<libv>
bonbons: did you build a defconfig?
<libv>
with ARCH=arm sunXi_defconfig?
<bonbons>
I took the config from cubietruck ubuntu kernel, then switch some items from modular to built-in
<libv>
why?
<bonbons>
I like kernels with fewer modules so hw gets initialized more in paralell during early boot instead of serialized module loading
<libv>
...
cubear has quit [Quit: Leaving]
<bonbons>
but yes, that might be incompatible with distro aproeach and vendor drivers that only work as modules :/
<libv>
bonbons: please do measure bootup time.
* Wizzup
also boots a module-less kernel on sunxi, since it's annoying to copy over modules too ;-)
<Wizzup>
But not for the "startup speeds"
<libv>
i love modules, as it allows me to unload/load some and see the effects while developing
<bonbons>
yeah that's the other useful part of it, just a single file (kernel) to care about
<libv>
without rebooting _all_ the time
<Wizzup>
both have their merits
<bonbons>
I agree with both of you
pwhalen has joined #linux-sunxi
<bonbons>
will try again after lunch, first getting back to more modular configuration in the hope of getting something that builds
leviathanch2 has joined #linux-sunxi
<libv>
bonbons: store your current .config, then start with sun7i_defconfig, and then work your way back
<Turl>
libv: I'm pretty sure we already have that one
<libv>
gzamboni_: thanks, reading through it
<libv>
Turl: in u-boot and boards
rz2k has joined #linux-sunxi
<libv>
Turl: but not on the wiki.
Net147 has quit [Quit: HydraIRC -> http://www.hydrairc.com <- Po-ta-to, boil em, mash em, stick em in a stew.]
<Tartarus>
libv, Right, so the problem is that somewhere between v2013.01.01-sunxi and master support for this board broke
<Tartarus>
libv, I was trying to make sure I was building the right board, before seeing what files I need to grab to get going on top of Hans' and ijc's series for mainline
<Turl>
Tartarus: can you try v2013.10-sunxi?
<Tartarus>
Turl, yeah, working my way up in tags before starting off on a bisect
<Turl>
Tartarus: ok, I'll let you to it :)
<libv>
hrm.
<Tartarus>
Turl, good news, that one also works, even with a pull the power test for sanity
<Tartarus>
well maybe not, heh, this is USB powered, didn't realize that
<Tartarus>
yes, really sane
<Tartarus>
bisect time
Skaag has quit [Ping timeout: 252 seconds]
nove has joined #linux-sunxi
alexvf has joined #linux-sunxi
<grafgustav>
gzamboni_: Hello again, using SPI still does not work. I have checked the script.fex 4 times, the spidev0.0 is existent even though there is still no module spidev in lsmod (but I guess that just doesn't have to be there?). I tried using spidev_fdx.c instead of spidev_test.c but this program as well just gives me 00's. I've connected an oscilloscope to various ports (it took me some time to work out how it works) but the values I get only make p
<mrnuke>
modprobe spidev ?
<grafgustav>
Should I even have a device called spidev0.0 if there is no spidev module loaded?
<wens>
libv: wasn't aware of that, I'll look at it tomorrow
<Tartarus>
sigh, time to pause the bisect, systemd crashed and I ought to pass that bug report along
<gzamboni_>
grafgustav, only make ... ?
<Tartarus>
Turl, bisect says e5e2ed05813fd4bf9a21d84677cbee4db4a3b05e but I haven't been able to confirm that a simple revert fixes
<grafgustav>
What am I supposed to make?
<gzamboni_>
grafgustav, spidev shouldnt apear in lsmod, this is normal, you have to modprobe the sunxi-spi if its compiled as module
kuldeepdhaka has quit [Ping timeout: 264 seconds]
<gzamboni_>
put one point of the osciloscope in MOSI and the other in to the GND and execute the compiled spidev_fdx.c with the right parameters
<mrnuke>
what about SCLK ?
<gzamboni_>
i dont have anymore my spi device in my actuall workspace so i cant do it in my side, but i did manage to make it work and the code sample from the wiki worked with the configs in there in the tutorial
<gzamboni_>
mrnuke, grafgustav , yes, with SCLK it should be easier to see in the osciloscope
<grafgustav>
I've tried that. There are hardly any changes, definitely no logical changes. So I tried grabbing the Clock (which I assumed would change constantly) but there I get constant 4.7V
<gzamboni_>
4.7 ? :P
<mrnuke>
no SCLK transitions means no SPI transaction is under way
<gzamboni_>
theres something wrong
<grafgustav>
I am sorry if I appear very stupid, I usually have no idea about electrical engineering
<mrnuke>
grafgustav: do you have a cheap-ass FX2-based logic analyzer?
<grafgustav>
I don't think so...? :D
kuldeepdhaka has joined #linux-sunxi
<mrnuke>
that's pretty handy for debugging SPI. 8 probes @ 24MHz each
<grafgustav>
Well, we use the 5V Vcc from the board, since the sensor works with 5V. And 4.7 sounds like a pretty close guess...
leviathanch2 has quit [Ping timeout: 255 seconds]
<mrnuke>
There's one problem. Your SoC runs at 3.3V
<Turl>
Tartarus: it's highly unlikely that breaks your olinuxino
<Tartarus>
Turl, yeah, I agree
<Turl>
Tartarus: that's me on the commit btw :)
<grafgustav>
As well it's quite possible I just misused that spidev_fdx.c file... I have not found any proper documentation about the parameters...
<Tartarus>
The way the tree is constructed makes bisects hard, lots of skips to avoid mainline commits that just can't work
MarkusR has joined #linux-sunxi
<Turl>
Tartarus: btw, have you tried the latest head tip?
<Turl>
Tartarus: commited like a day ago?
<Tartarus>
Turl, thI started
<Turl>
Tartarus: it has a fix for a typo on the uart pin definitions for sun5i, that may be your bug
<gzamboni_>
grafgustav, the 5V is the vdd of your device, the spi comunication will occurs in 3.3V
<gzamboni_>
the CLK, mosi and miso from the A10 or A20 are in 3.3V
<gzamboni_>
if you mesure the CLK you have to do it with the GND and it should oscilates from 0V to 3.3V in the frequency set on your c app that uses the spidev
hansg has quit [Quit: Leaving]
kivutar has joined #linux-sunxi
leviathanch2 has quit [Remote host closed the connection]
leviathanch2 has joined #linux-sunxi
<grafgustav>
gzamboni_: Okay, thanks. I guess we connected it wrong then. We figured since the sensor works with up to 5.5V we can just connect it's Vcc to a 5V Vcc pin on the board
Skaag has joined #linux-sunxi
focus has joined #linux-sunxi
<gzamboni_>
grafgustav, for the vdd you can use the 5V from the board, it doesnt come from the allwinner chip
FreezingCold has joined #linux-sunxi
<grafgustav>
I see, I think I get it now. Partially. So right now I am not executing anything, I just have the sensor attached to the board and with a multimeter measure the voltage from CLK to GND. It still is 4.7V :D What am I even doing wrong, I even use the 3.3V Pin of the board now...
Gerwin_J has quit [Quit: Gerwin_J]
leviathanch2 has quit [Ping timeout: 240 seconds]
Gerwin_J has joined #linux-sunxi
<Turl>
w 2
<Tartarus>
Whee, this board seems moderately unstable :(
<Tartarus>
Leave it on for a while and now it's kinda working when it had been hanging before
<Tartarus>
of course, other hell so I'm not sure which it is..
<Tartarus>
brbish I hope
<oliv3r>
wens: a10-meminfo does not work on a23; you need a31 meminfo that i hacked on ages ago; checkout my github; it probably doesn't work :)
<oliv3r>
libv: we're still waitingon u-boot source to be fair
<oliv3r>
bennn: a23 is a very cool little chip. it does lack a lot of features i think; but it should pack quite some interesting performance. not only that, it is required as a dev platform to prepare for a60/a70
<Tartarus>
Turl, yup, I'm in hell, the board now consistently boots up
<Turl>
Tartarus: :(
<Turl>
Tartarus: maybe you have bad/missoldered ram
<Tartarus>
Perhaps
<Tartarus>
Easiest experiment for the day is to unplug, see what it does tomorrow morning
<Tartarus>
And do the build time part of porting this over to Hans' series today
leviathanch2 has joined #linux-sunxi
diego_r has quit [Remote host closed the connection]
diego_r has joined #linux-sunxi
<bonbons>
drivers/net/wireless/bcmdhd/wl_iw.c fails to build if WIRELESS_EXT is set but WEXT_PRIV is not...
<bonbons>
not sure yet what (other wireless driver?) from defconfig select WEXT_PRIV...
<wens>
oliv3r: would the fex file be enough?
<bonbons>
defconfig includes pretty much all available wireless drivers!
<bonbons>
adding "select WEXT_PRIV if WIRELESS_EXT" to config BCMDHD in drivers/net/wireless/bcmdhd/Kconfig fixes the build
<kivutar>
hi, i'm trying to port openelec to cubieboard2. Something goes wrong during the initramfs init. It's hard to debug because I don't get enough messages in the log. Strangely, if I remove console=ttyS0 the boot become verbose enough. I can see that my kernel is panicing during the initramfs. However, I suspect the panic to be a side effect of removing console=ttyS0 because I get the warning "unable to open an initial console"
<kivutar>
speakman, I see in the chan log that you had a similar issue and solved it
<Turl>
kivutar: that warning sounds like missing /dev/ nodes
<kivutar>
yes Turl
<kivutar>
maybe I am missing something in my kernel config to populate /dev with the serial nodes
<Turl>
kivutar: your ramdisk has an empty /dev/ ?
akaizen has joined #linux-sunxi
<kivutar>
I can't get a debug prompt... so I don't know
<Turl>
kivutar: add console and ttyS0 to your dev and try again
<Turl>
kivutar: cd dev; mknod -m 622 console c 5 1; mknod -m 622 ttyS0 c 4 64
<Turl>
bbl
<ccaione>
kivutar: have you enabled devtmpfs?
<kivutar>
ok, thanks, but it's not gonna be easy as the initramfs is in the kernel
<kivutar>
ccaione, let me check that
<kivutar>
CONFIG_DEVTMPFS=y
<kivutar>
CONFIG_DEVTMPFS_MOUNT=y
<mripard>
it's not mounted automatically in an initramfs
<mripard>
you have to mount it manually
grafgustav has quit [Quit: playing games with my heart]
<kivutar>
let me check my initramfs init script then..
<kivutar>
i'm doing a /bin/busybox mount -t devtmpfs devtmpfs /dev
jukivil1 has quit [Ping timeout: 265 seconds]
<ccaione>
kivutar: btw, the correct bootarg is console=ttyS0,115200