<derethor>
oliv3r, the otg has another pin to detect if you are the host or the device
<derethor>
it is on the connector
<derethor>
probably, you are not connecting the pin
naobsd has quit [Quit: Page closed]
gzamboni has quit [Ping timeout: 264 seconds]
fldc has quit [Quit: quit]
fldc has joined #linux-sunxi
\\Mr_C\\ has joined #linux-sunxi
dwilkins has quit [Read error: Connection reset by peer]
derethor_ has joined #linux-sunxi
dwilkins has joined #linux-sunxi
derethor_ has left #linux-sunxi [#linux-sunxi]
derethor has quit [Quit: Leaving]
servili007 has quit [Quit: Leaving]
<vincenzo>
goodix_touch_3f.ko anyone found it somehere?
derethor has joined #linux-sunxi
<derethor>
it is possible to compile uboot and the kernel using the official gcc-4.7-arm-linux-gnueabihf from ubuntu?
<derethor>
i guess, it will be slower, but it is possible at least?
<derethor>
i got this error : error: required section '.rel.plt' not found in the linker script
<derethor>
eps
<derethor>
my fault, sorry
<derethor>
i didnt put the a13olinuxinom parameter when I call make
CaCtus492 has quit [Ping timeout: 256 seconds]
CaCtus491 has joined #linux-sunxi
fldc has quit [Quit: quit]
fldc has joined #linux-sunxi
unioah has joined #linux-sunxi
unioah has quit [Remote host closed the connection]
TheSeven has quit [Disconnected by services]
[7] has joined #linux-sunxi
unioah has joined #linux-sunxi
unioah has quit [Ping timeout: 260 seconds]
unioah has joined #linux-sunxi
_BJFreeman has joined #linux-sunxi
_BJFreeman is now known as BJfreeman
ganbold has quit [Remote host closed the connection]
ganbold has joined #linux-sunxi
soul has quit [Read error: Operation timed out]
Soru has joined #linux-sunxi
piyushverma has quit [Quit: Konversation terminated!]
piyushverma has joined #linux-sunxi
BJfreeman has quit [Quit: had a good time]
rz2k has joined #linux-sunxi
tzafrir has joined #linux-sunxi
dipra has joined #linux-sunxi
dipra has left #linux-sunxi [#linux-sunxi]
<stekern>
hno: I thought it might be appropriate to move our discussion from #openrisc into here, since people here might be more interested (or just be more able to help me out with my sunxi questions ;))
<stekern>
others: quick summary - I'm investigating the ar100 openrisc core on the a31 soc, I've put up my findings so far here: https://github.com/skristiansson/ar100-info
<stekern>
oh, I also see that you've created a wiki page for ar100, I presume I can update that as I make progress?
<stekern>
anyways, you were right about the clocking being configurable, possible clock sources are: LOSC (31 kHZ), HOSC (24 MHz) and PLL6
<rz2k>
stekern: update wiki without questions, any help is appreciated :p
<oliv3r>
so just to confirm, as i expected of course, i have to solder pin 4 to GND when connecting a slave device (such as a keyboard)
cnf has joined #linux-sunxi
unioah has quit [Ping timeout: 268 seconds]
rellla2 has joined #linux-sunxi
wingrime has quit [Ping timeout: 240 seconds]
wingrime has joined #linux-sunxi
rellla2 has quit [Remote host closed the connection]
eebrah|away is now known as eebrah
focus has joined #linux-sunxi
tzafrir has joined #linux-sunxi
wingrime has quit [Ping timeout: 264 seconds]
_BJFreeman has joined #linux-sunxi
_BJFreeman is now known as BJfreeman
notmart has joined #linux-sunxi
notmart has joined #linux-sunxi
vicenteH has joined #linux-sunxi
hramrach_ has quit [Ping timeout: 240 seconds]
hramrach_ has joined #linux-sunxi
rellla2 has joined #linux-sunxi
gzamboni has joined #linux-sunxi
<hno>
hi stekern
<hno>
stekern, yes please update the AR100 wiki page. I tried to collect what I could understand from the #openrisc discussion.
<hno>
wonder how high it can be clocked.. pll6 is fairly fast normally. Does it also have a clock divisor?
<stekern>
yes, the ar100 clock configuration register have a 5-bit divider part
<stekern>
I'm not sure if information from this can be applied to a31: http://linux-sunxi.org/A10/CCM, but where does that 240 MHz - 2 GHz limitation that is mentioned on PLL6 come from?
<oliv3r>
stekern: the usermanual :)
<stekern>
I'm also curious how fast it can be clocked, I did a quick test with PLL6 in the morning before running off to work.
<stekern>
I thought the values on N, M and K where 0, 0 and 0 and 1, 1 and 1
<stekern>
but I had a bug in my test software, so N was actually 24
<stekern>
and N = 24, M = 0 and K = 0 didn't crash and N = 24, M = 1 and K = 1 did crash
<stekern>
that gives 300 MHz vs 600 MHz
<stekern>
so >= 600 MHz is at least _not_ possible =P
<oliv3r>
I think if you connect PLL6 to something that can actually run at 2GHz
<oliv3r>
and nothing else
<oliv3r>
it shoudl work
<stekern>
yeah, I was speaking of the limits of the ar100 core here
<oliv3r>
lol yeah that won't work at 2 GHz idon't think :D
<stekern>
most probably not ;)
naobsd has joined #linux-sunxi
<Turl>
stekern: going by the A1X definition of PLL6 that's not 300 vs 600Mhz is it?
<Turl>
stekern: both of them look like 96MHz to me
<stekern>
(24 * (24 + 1) * (1 + 1))/2 = 600, no?
<Turl>
stekern: N doesn't take +1
<Turl>
and yeah, if you measure for 'other module' (I was calculating the sata output) it'd be 288 vs 576
<stekern>
mmm
<stekern>
from the allwinner sun6i linux code it looks like N should be +1 though
<stekern>
but that's pretty uncertain
<stekern>
I need to improve my clock frequency measuring routine to get some more definite information
<Turl>
I haven't looked at the sun6i code, but it's possible, yes
<stekern>
this is the part where I got the +1 from: http://git.rhombus-tech.net/?p=linux.git;a=blob;f=arch/arm/mach-sun6i/clock/sys_clk.c;h=e38ffec401b99d1703d075f2714d233c408c7df1;hb=4d6537b9e473af95baef30c2cdf232b5253e0640#l734
BJfreeman has quit [Quit: had a good time]
Soru has joined #linux-sunxi
hansg has joined #linux-sunxi
<derethor>
the A31 has a openrisc inside? that totally blows my mind
<rm>
heheh
<rm>
I wonder can you just run the OS on the OpenRISC
<rm>
not even touching any of the 4 cores
<oliv3r>
i just noticed i forgot to include the sunxi-ML on my patch :( added it now though
<derethor>
that confirms my idea that the only thing that stops allwinner to the complete world domination, is the lack of documentation
<oliv3r>
derethor: yeah, i think it'll be used for the baseband, akak phone function
<oliv3r>
derethor: lol tell lkcl :)
shineworld has joined #linux-sunxi
<derethor>
lkcl works in allwinner?
<derethor>
i dunno
<derethor>
btw, do you have a mailing list guys?
<stekern>
rm: unfortunately, the openrisc core is configured without an MMU, otherwise it'd be completely possible to boot the existing openrisc Linux port on it
<derethor>
but it has interruptions? like realtime interrupts?
<derethor>
because that will be totally awesome
<derethor>
i see a /dev/openrisc or someting with realtime features
<derethor>
shit, now i want an A31
<stekern>
I'm not sure what's connected to the irq lines of it, but it does have an interrupt controller implemented
<derethor>
i see, your code boots the openrisc, but it is possible the communication between the arm and the openrisc?
<derethor>
i mean, they share an addr space or soemthing?
<derethor>
or, you can only run one architecture at the same time?
<stekern>
well, you can access the DRAM from the openrisc core (and the internal SRAM at 0x0 + virtually all I/O addresses)
<stekern>
so, yes, you can communicate between them
<derethor>
omg
<derethor>
that is geek porn
<derethor>
so, you only need a locking mecanism, and you can queue commands, and run them in realtime, from linux
<stekern>
I meant internal SRAM at 0x44000
<derethor>
in your code, is see #ifdef CONFIG_SUN6I_FPGA
paulk-desktop has joined #linux-sunxi
<derethor>
now tell me that the a31 has also a fpga
<derethor>
:)
<derethor>
i guess it is their fpga for testing
edeloget has joined #linux-sunxi
rz2k has joined #linux-sunxi
rellla2 has joined #linux-sunxi
\\Mr_C\\ has quit []
rellla2 has quit [Remote host closed the connection]
<hno>
derethor, thats remains from when they tested the sun6i design on FPGA.
hramrach_ has quit [Ping timeout: 240 seconds]
<derethor>
hno: which compiler should i use to build the kernel? i am in linux mint
hramrach_ has joined #linux-sunxi
<hno>
derethor, I use Linaro 4.7 toolchain, but there is many other ARM toolchains.
<hno>
just avoid 4.8 for now.
hansg has quit [Remote host closed the connection]
hansg has joined #linux-sunxi
<edeloget>
hno: is there any issue with 4.8 ?
<oliv3r>
edeloget: it produces garbage that doesn't run properly, we don't know why yet
<rz2k>
edeloget: mainline arm exynos kernel doesnt launch on odroid
eebrah is now known as eebrah|away
<rz2k>
with 4.8
<rz2k>
now imagine our sunxi old kernel with ton of hacks
<oliv3r>
stekern: soft-mmu? :D Wasn' there uCLinux that did just that
<oliv3r>
stekern: if i'm not mistaken, those patches are even mainlined
<oliv3r>
derethor: we suspect the ar100 will be used to control a modem of some sort e.g. baseband radio which needs a small fast realtime kernel
<hno>
oliv3r, ?
<hno>
The current use in the kernel is pretty clear and far from that. Have you found some other?
<derethor>
and, do you know a good .config ? i have an a13 olinuxino micro
<oliv3r>
it was suspected that with phablet devices, where there's actual phone functionality it will be used for that
<oliv3r>
right now, yes it only does standby doesn't it
<hno>
yes
<hno>
there is no radio in the sunxi chips that I know of.
<edeloget>
rz2k: bad. do you have any link ?
tinti has joined #linux-sunxi
<oliv3r>
well the radio would be connected to the AR100 so would be outside
<derethor>
nice, they can send them directly to my pcb assembly suppliers... i will contact them
<derethor>
i was thinking to ask directly to allwinner, I didnt think in alibaba
<derethor>
but sure, there are comapnies with bulk chips for sale, sure
<derethor>
thank you
<hno>
derethor, you don't get chips directly from Allwinner unless you either are buying volume or already is an established customer.
<derethor>
i imagine that, i already asked some pcb assemblers, and of course, they said they can get them, but i am not very sure
<derethor>
it would be easier if i can buy them, and send them to the pcb assembler
<derethor>
i am thinking to assemble a simple board to test if i can boot it :) before anything strange
\\Mr_C\\ has joined #linux-sunxi
<hno>
derethor, please do an open hardware design if you can. And also speak to Olimex on the matter.
n01 is now known as n01|away
Soru has quit [Ping timeout: 252 seconds]
rz2k has quit []
soul has joined #linux-sunxi
fluo75_ has joined #linux-sunxi
edeloget has quit [Ping timeout: 250 seconds]
notmart has quit [Quit: notmart terminated!]
tzafrir has joined #linux-sunxi
n01 has joined #linux-sunxi
nove has joined #linux-sunxi
sanka has joined #linux-sunxi
sanka has quit [Remote host closed the connection]
Yaku321 is now known as Yaku-noob
notmart has joined #linux-sunxi
notmart has quit [Changing host]
notmart has joined #linux-sunxi
<stekern>
I can confirm that the a31 PLL6 formula uses N+1 (i.e. (24*(N+1)*(K+1))/2)
eebrah|away is now known as eebrah
<hno>
steev, no M?
<hno>
sorry, meant stekern ^
<stekern>
there is an M factor, but it's not used for the ar100 clocking at least
<stekern>
it's the same as in the "Others" formula for a10 PLL6 (except the +1 on N)
<hno>
I would not pe supriced if it's really +1 on the A10 as well. A multiplier of 0 makes no sense, and documentation aroung PLL6 is a bit confusing...
<hno>
soul, you may need to start adb as root once.. pkill adb, then start adb as root.
<hno>
it starts a daemon, and after that you can run adb as normal user.
<soul>
I did it as root
<soul>
Don't worry hno, I want to delete Android so... xD
<hno>
odd. maybe adb is not enabled in your android system settings. But never mind, it's not much you need to do so easier to just do it than figuring out why you have no adb..
<soul>
Correct xD
<soul>
So... hno, can I do it this way or what I have to doy? Sorry if I disturb :P
fldc has quit [Ping timeout: 245 seconds]
tinti_ has joined #linux-sunxi
<hno>
soul, you can use any path you like, just find somewhere writeable that can take executables.
<hno>
mount -o remount,rw /
<soul>
ok thanks hno
<hno>
might do the trick for / if it's not writeable.
fldc has joined #linux-sunxi
<hno>
/data/cache is just an example of a path that often is writeable and can take executables.
<Turl>
hno: on android / is a ramdisk, remounting rw does the trick, but the android mount binary usually sucks so you may need to pass an extra garbage argument
<Turl>
like mount -o remount,rw nevermind /
<hno>
Aha.. they got free braindamage from nih, nothing extra charged. Or busybox mount -o remount,rw / to keep some sanity.
<Turl>
hno: I think it's more like "not apache licensed" rather than nih
<soul>
Ok
<soul>
I did the first
<soul>
I have the meminfo dump!! :D (I don't know for what is it but I have it!! xD )
<soul>
Now I have to pick the script.bin
<hno>
good.
<hno>
and then transfer them both so you can mail them.
<soul>
But I don't understand how to do the steps for script.bin.
<hno>
what part?
<soul>
There is no script.bin file like a10-meminfo-static
<hno>
The script.bin file is on the device. You need to extract it.
<soul>
What's nanda?
<hno>
one partition of the nand flash inside your device.
<hno>
See step 8 in the script.bin instructions.
<soul>
oh
<soul>
Let me try it, thanks and sorry
fldc has quit [Read error: Connection reset by peer]
fldc has joined #linux-sunxi
<soul>
Ok done!
<soul>
Oh, I forgot that somebody told me that I have to create a wiki page for my tablet
<soul>
But... If I can't install a distro, It will be empty :S