<TuxboxGuru>
ssvb: thanks for the tip... the Cubie2 dies when I run the memory tester
blsd has joined #linux-sunxi
<ssvb>
TuxboxGuru: you can resolve this by either reducing the dram clock frequency in u-boot or by increasing the dcdc3 voltage in fex
<TuxboxGuru>
ssvb: Is there a 'Really dummies guide for dumb people' .. like me? :-P
<ssvb>
TuxboxGuru: the guide is yet to be written :) for dcdc3 voltage just edit your fex and change "dcdc3_vol = 1250" to "dcdc3_vol = 1275" or "dcdc3_vol = 1300"
<ssvb>
TuxboxGuru: until the memory test starts running happily without failures
<TuxboxGuru>
ssvb: thanks... will take a look now... but yes, the board runs fine for a few days and then just hangs :-P sometimes the heartbeat continues - sometimes not
<ssvb>
TuxboxGuru: and currently I'm hacking the dram controller code in u-boot to check if it can be configured better
<TuxboxGuru>
ssvb: so the FEX would be suggested method of fixing?
<TuxboxGuru>
(still in the dark what the FEX is :-P .. hopefully google will light the path)
<studentol_>
Sorry I reread the wiki.And make myself. like wiki said err() dpg() can't work. I note it. but when I try Livesuit.sh : I already install libpng12 and it in /usr/lib32/libpng12.so.0 libpng12.so.0: cannot open shared object file: No such file or directory
<ssvb>
TuxboxGuru: the fex file (in text format) is getting converted to the binary format in script.bin file, the script.bin file is used by the kernel to get various information about your board
<TuxboxGuru>
ssvb: ah yes... the one is also responsible for SATA enabling and such things...
<TuxboxGuru>
I did read about it... didnt realise that was the fex
wickwire has joined #linux-sunxi
<ssvb>
the dcdc3 voltage is also configured there
avsm has joined #linux-sunxi
hansg has joined #linux-sunxi
hansg has quit [Client Quit]
vector80 has joined #linux-sunxi
<vector80>
Hi everyone, yesterday finally I got my hi253 camera working under debian with 3.4.79 kernel. I would like to capture video, encode to h264, and stream out via RTSP. Is there a tutorial for that ?
kuldeepdhaka has joined #linux-sunxi
paulk-collins has joined #linux-sunxi
avsm has quit [Read error: Connection reset by peer]
chazy- has joined #linux-sunxi
codyps1 has joined #linux-sunxi
codyps has quit [*.net *.split]
chazy has quit [*.net *.split]
setkeh has quit [*.net *.split]
chazy- is now known as chazy
avsm has joined #linux-sunxi
setkeh has joined #linux-sunxi
setkeh has quit [Max SendQ exceeded]
setkeh has joined #linux-sunxi
kuldeepdhaka has quit [Ping timeout: 252 seconds]
<rellla>
ssvb: has un-forking linux-sunxi been reverted?
slapin has joined #linux-sunxi
studentol_ has quit [Quit: Leaving]
paulk-collins has quit [Remote host closed the connection]
payne_ has joined #linux-sunxi
* TuxboxGuru
starts the memtester using 1.35V settings
payne has quit [Ping timeout: 264 seconds]
kuldeepdhaka has joined #linux-sunxi
leviathanch2 has joined #linux-sunxi
deasy_ has quit [Ping timeout: 258 seconds]
deasy_ has joined #linux-sunxi
leviathanch2 has quit [Ping timeout: 240 seconds]
kuldeepdhaka has quit [Remote host closed the connection]
bgal has joined #linux-sunxi
maksimlin has quit [Quit: ChatZilla 0.9.90.1 [Firefox 29.0/20140428193838]]
cubear has joined #linux-sunxi
leviathanch2 has joined #linux-sunxi
avsm1 has joined #linux-sunxi
avsm1 has quit [Client Quit]
avsm has quit [Ping timeout: 276 seconds]
<ssvb>
rellla: yes, it looks like this
<ssvb>
rellla: after the change, I noticed that the repositories which used to be 'linux-sunxi' forks have changed to be 'torvalds/linux' forks
<ssvb>
rellla: I reported this to the github people, but never heard anything back
<ssvb>
rellla: apparently they got everything reverted
<ssvb>
rellla: which is pretty weird, because my small repository https://github.com/ssvb/cpuminer got detached from upstream and retained its forks just fine
cubear has quit [Ping timeout: 276 seconds]
<ssvb>
rellla: I suspect that this is just a manual process at github and human errors are possible, that's why the results are not consistent
_whitelogger has joined #linux-sunxi
leviathanch2 has quit [Ping timeout: 240 seconds]
bgal has quit [Ping timeout: 252 seconds]
leviathanch2 has joined #linux-sunxi
leviathanch2 has quit [Ping timeout: 240 seconds]
tgaz has quit [Ping timeout: 245 seconds]
tgaz has joined #linux-sunxi
<libv>
amitk: ah, the reason why your gmac is not working is most likely because you are using a cubieboard2 "image"
<libv>
amitk: when you work through the new device howto fully, you will, at the end, be pointed to the manual build howto, which might give you a more completely working device
payne_ has quit [Remote host closed the connection]
<amitk>
libv: I've replaced the kernel and uboot with my own (started a new defconfig for the qt840a and a new boards.cfg entry in u-boot)
<amitk>
libv: and a new fex file too
<amitk>
libv: based on what I grabbed from the device
<libv>
amitk: the fex file is going to be the key
<amitk>
libv: will send out a patch to add the new fex file later today
<libv>
amitk: cool
<libv>
amitk: can you also make a picture of the bottom of the device/case?
cosm has quit [Ping timeout: 276 seconds]
Fusing3 has joined #linux-sunxi
<amitk>
libv: I've read that working GMAC needs uboot to activate gmac too. Do you know if these two are the only config options I need in uboot: SUNXI_GMAC, RGMII ?
<amitk>
libv: nothing on the bottom, not even a sticker :)
Black_Horseman has quit [Quit: Zwi se logou mou!!!]
<libv>
amitk: still, even if it is perfectly flat, it might be helpful to distinguish this device from others
<amitk>
libv: ack
<libv>
amitk: since you have a camera out anyway, you might as well go all the way :)
_massi_ has joined #linux-sunxi
<libv>
amitk: it somehow seems that the side with the android logo on it, is the bottom
<amitk>
libv: the printed characters are upside down
<libv>
amitk: but all the connectors are too
<amitk>
and the damn case is a dust magnet
<libv>
amitk: yeah :)
Fusing3 has quit [Ping timeout: 258 seconds]
diego_r has joined #linux-sunxi
leviathanch2 has joined #linux-sunxi
bard3307 has joined #linux-sunxi
JohnDoe_71Rus has quit [Quit: Miranda IM! Smaller, Faster, Easier. http://miranda-im.org]
jebba1 has quit [Ping timeout: 258 seconds]
jebba1 has joined #linux-sunxi
jebba2 has joined #linux-sunxi
cosm has joined #linux-sunxi
leviathanch2 has quit [Ping timeout: 240 seconds]
ganbold_ has joined #linux-sunxi
leviathanch2 has joined #linux-sunxi
jebba1 has quit [*.net *.split]
<Montjoie>
oh at least 3DES is far better with SecuritySystem
<Wizzup>
now we can hash our passwords if we go back 20 years! ;-)
<Wizzup>
On a more serious note, if you have some code I can look at, hack at, let me know :-)
<vector80>
I really really need to know, how to capture the camera and stream out.....
<wens>
amitk: you will only need RGMII if the PHY you have is of the RGMII type
<wens>
lookup the part number
shineworld has joined #linux-sunxi
<wens>
also the phy might need to be powered up by toggling some gpio, the fex file should show
<wens>
as for wifi, the AP6330 seems to have a bcm40183 chip inside
<wens>
using the bcmdhd driver from broadcom should work, as that driver supports auto-detection of their chips, unlike the brcmfmac driver in mainline
<TuxboxGuru>
ssvb: three iterations of Memtester and still going strong.... CPU is nice and toasty
<ssvb>
TuxboxGuru: sounds good, which of the fixes have you applied?
<TuxboxGuru>
the fex voltage
<TuxboxGuru>
1.25 to 1.35 change
<ssvb>
oh, maybe you should also try something lower and avoid going straight to 1.35V
<libv>
amitk: when you're not touching the page, give me a heads up, there are a few minor changes i want to make
<TuxboxGuru>
ssvb: I can do.. I'll try 1.30
<ssvb>
TuxboxGuru: it's best to find the lowest voltage that passes tests and increase it by 0.025 to have some safety margin
<TuxboxGuru>
ssvb: Cool :D very fine resolution
<TuxboxGuru>
ssvb: whats the typical ranges you have seen? (I read quite a few different values in the IRC logs
<Montjoie>
Wizzup, I am writing the header of the mail:) I will sent it today I think
<Wizzup>
guess I should join the list then ;-)
<ssvb>
TuxboxGuru: I guess 1.3V would be a good choice for dcdc3, unless you want to do extreme dram overclocking well beyond 480MHz :)
<TuxboxGuru>
ssvb: I don't need any overclocking, I added a porcupine heatsink for fun ('cos I CAN!) ... I can run it a little slower if needed even
<TuxboxGuru>
ssvb: how quickly did you get the Memtest to fail? (just for reference)
leviathanch2 has quit [Ping timeout: 240 seconds]
leviathanch2 has joined #linux-sunxi
<libv>
does anyone know what arisc is?
<libv>
there are some binary blobs labelled as that in the a23 sdk
<libv>
then there is libisp and a libnand for each new sunxi version
<wens>
libv: arisc is the OpenRISC core (or AR100)
<wens>
at least that's what I make of it
<atsampson>
there's some source to talk to it under drivers/arisc...
shineworld has quit []
<libv>
wens: perhaps throw a few words about it in the wiki?
<libv>
wens: to what extent have you been playing with the a23 sdk kernel source?
<Montjoie>
does it is safe to replace writel/readl by memcpy ?
<Montjoie>
it seems yes, but I fear to not handle all concequences
<TuxboxGuru>
Montjoie: keep in Mind writel/readl are not cached... memcpy might be
<amitk>
libv: I'm not editing currently
<libv>
amitk: cool
<amitk>
wens: ok, is the phy info typically printed in kernel logs? /me doesn't recall seeing such a thing. I guess I need to study the chips on the board
<Montjoie>
thanks TuxboxGuru
<Montjoie>
because by using memcpy, I have PERFORMANCE with SS
<Montjoie>
I need to launch my crypt tester
kuldeepdhaka has joined #linux-sunxi
<TuxboxGuru>
Montjoie: maybe marking the variables as Volatile will negate the cache (it should)
eagles0513875 has quit [Changing host]
eagles0513875 has joined #linux-sunxi
eagles0513875 has quit [Changing host]
eagles0513875 has joined #linux-sunxi
<amitk>
wens: so I need uboot to initialise the gmac HW for it to work, correct? IOW, the kernel GMAC driver isn't poking all the right registers yet?
<wens>
libv: mostly grepping through it
<libv>
amitk: is it really gmac hw?
<libv>
amitk: have you located the phy already?
<libv>
wens: ok
jinzo has joined #linux-sunxi
popolon has quit [Ping timeout: 264 seconds]
vector80 has quit [Quit: Konversation terminated!]
<libv>
pfff, this tree was not a vanilla 3.4.39
<libv>
i hope the android 3.4.39 comes a lot closer
popolon has joined #linux-sunxi
popolon has quit [Changing host]
popolon has joined #linux-sunxi
paulk-collins has joined #linux-sunxi
leviathanch2 has quit [Ping timeout: 240 seconds]
xavia has joined #linux-sunxi
leviathanch2 has joined #linux-sunxi
codekipper has joined #linux-sunxi
deasy_ has quit [Quit: Nom d'un quark, c'est Edmonton !]
<codekipper>
Hi all
<codekipper>
I'm thinking about looking into the DMA on A10/A20
<codekipper>
does anybody have any experience with this?
<codekipper>
I started trying to harmonize the 3.4 kernel but it was becoming a head fuck
<ccaione>
codekipper: mripard already submitted some patches about DMA IIRC
<codekipper>
so I was I started looking at the mainline effort and whether I could piggy back on the bits that maxime has done on the A31
<codekipper>
Yeah...I see that, but he's got the cutting edge dev boards whereas I'm limit to my trusty A10/A20 devices
<Montjoie>
codekipper, Turl will work on A20/DMA thought a GsoC
<codekipper>
The hardware blocks look exactly the same comparing the A10/A20/
<codekipper>
Turl: anything I can do/help out with?
xinj has joined #linux-sunxi
xavia has quit [Ping timeout: 252 seconds]
<codekipper>
Montjoie, do you know if there is a link to the project?