cheng has quit [Read error: Connection reset by peer]
<Tradie>
can anyone help with an a13 with screen issues
<Tradie>
i got told to find a guy who can help here
hipboi has quit [Ping timeout: 260 seconds]
<aexl>
@Tradie: and what is the problem
mSquare has quit [Read error: Connection reset by peer]
<aexl>
i'm no expert. let's keep this public. maybe others can jump in.
<Tradie>
no worries. well, my tablet died. i fixed it. theres two roms that take. one has no touchscreen. the other, the actual display is off to the left and up about 1 centimitre
cheng has joined #arm-netbook
<aexl>
i guess you don't have an original firmware from the vendor?
<Tradie>
they woulnt supply it
<aexl>
is touchscreen supposed to work in the first rom?
<Tradie>
but im getting the exact same tablet off a mate tomorrow, and ill grab the firmware from that. only problem is that i dont know how to make that into a flashable image
<Tradie>
and yes it is
<Tradie>
and another question, could i add the touchscreen modules from the second rom to the first one?
popolon has joined #arm-netbook
<aexl>
could be possible. can you boot from sd card?
<Tradie>
i have no idea... its just an android tablet if your wondering :P
hipboi has joined #arm-netbook
<aexl>
the only possible way that i know of to copy an existing system is to grab everything from nand via linux from sd card.
<Tradie>
i can do system dumps!
<aexl>
?how?
<Tradie>
theres is a tool that uses adb to make it all automated
<aexl>
what is the output format?
<Tradie>
each partition gets pulled to the following files
<aexl>
then it should be possible to build a new flashable image.
<Tradie>
but im not sure how
<aexl>
or do you want to use another rom?
<Tradie>
i want to add the ts modules from this dump, into a dump of the correct screen placement one, then repackage
hipboi has quit [Ping timeout: 252 seconds]
<oliv3r>
morning everybody :)
mSquare has joined #arm-netbook
<aexl>
hello.
<aexl>
@Tradie: hardware is defined in a file in the boot partition (script.bin or such) usually /dev/nanda or /dev/block/nanda. you could also try to edit its display section.
<Tradie>
@aexl ive tried that, but every time i go to open the file it is just a bunch of jibberish that notepad++ cant understand
<aexl>
copying modules from one to the other should also be possible. /lib/modules/{kernel version}/drivers/input/touchscreen
<oliv3r>
So i'm going over some code in drivers/video/sun4i/disp/de_bsp/de/disp_display.c and notice they very clearly define sizes (which is really good I suppose). The A10 manual mentions however very 'wide' sizes. for example a lot of registers are defined as 1KiB in size, when only 12Bytes are being used (See A10/PWM for example). Not now, but in the future, do we want to have proper definitions on the A10 registers page? Or leave them as per datasheet at
<Tradie>
ill be back to tell you how it goes later :)
<aexl>
good luck
hipboi has joined #arm-netbook
hipboi has quit [Ping timeout: 245 seconds]
Tradie has quit [Quit: irc2go]
Tradie has joined #arm-netbook
hipboi has joined #arm-netbook
<oliv3r>
someone able to share some insights? THere's a timer IRQ enable register. it has two options, 0 and 1; if I write 1 to it, it should enable the interrupt for the timer, but writting 0 has no effect. So once enabled, how would I disable said interupt?
<oliv3r>
if in doubt, check the source!
aexl has left #arm-netbook [#arm-netbook]
hipboi has quit [Ping timeout: 240 seconds]
<rm>
did you try "2" or "-1"? :P
<mnemoc>
oliv3r: try the linux-sunxi mailing list
<mnemoc>
-1 is a bit large...
<oliv3r>
i don' thave hardware :p
<mnemoc>
oliv3r: get one :)
<oliv3r>
i'm reading the datasheet, and it says 0 no effect, 1 interupt enable
<oliv3r>
mnemoc: I know!
<oliv3r>
mnemoc: a bit strapped on cash at the moment :)
<oliv3r>
and time :p
<oliv3r>
mnemoc: I will sign up to the linux-sunxi mailing list however :)
<oliv3r>
i'll study sunxi-wdt.c atm to find out how hno did it
<mnemoc>
:)
hipboi has joined #arm-netbook
hehopmajieh_ is now known as HeHoPMaJIeH
lerc has joined #arm-netbook
lerc_ has quit [Ping timeout: 256 seconds]
<mnemoc>
hipboi: hi, is it possible to get some feedback from matson hall about the last sata fix patch on the ML?
HeHoPMaJIeH has quit [Changing host]
HeHoPMaJIeH has joined #arm-netbook
hipboi has quit [Ping timeout: 245 seconds]
<mnemoc>
:(
hipboi has joined #arm-netbook
hipboi has quit [Ping timeout: 260 seconds]
<mnemoc>
hno: when you get a minute, can you comment on the latest "sata fix" patch from guillaume? It's totally out of scope for me so I don't know what to do
<hno>
oliv3r, the wdt is documented in my little register guide.
<oliv3r>
ah, should have looked there
<oliv3r>
well i haven't (yet) found any interupt enable/disable reference in the wdt, but i guess makes sense, if it's enabled/disabled globally on a higher lvl
<oliv3r>
IntCtl is undefined :(
hipboi has joined #arm-netbook
cheng has quit [Quit: Leaving]
hipboi has quit [Ping timeout: 264 seconds]
<oliv3r>
hno: nope can't find any reference in your code to 0x000; IntCtl (under Interrupt?)
cat2 has joined #arm-netbook
cat1 has quit [Ping timeout: 264 seconds]
hipboi has joined #arm-netbook
Tradie has quit [Quit: irc2go]
<lundman>
interupt control!
<lundman>
there, I helped
hipboi has quit [Ping timeout: 256 seconds]
sspiff has quit [Remote host closed the connection]
sspiff has joined #arm-netbook
sspiff has quit [Changing host]
sspiff has joined #arm-netbook
<oliv3r>
:D
hipboi has joined #arm-netbook
<oliv3r>
CNTL or CTRL; i think the linux kernel generally uses 'CTNL' as abbreviation?
hipboi has quit [Ping timeout: 244 seconds]
hipboi has joined #arm-netbook
hipboi has quit [Ping timeout: 252 seconds]
rsalveti has quit [Remote host closed the connection]
itamarjp has joined #arm-netbook
rsalveti has joined #arm-netbook
QingPei has joined #arm-netbook
Entropy512 has quit [Remote host closed the connection]
Entropy512_ has joined #arm-netbook
Entropy512_ is now known as Entropy512
cat_n9 has quit [Ping timeout: 246 seconds]
tuliom has joined #arm-netbook
mysteryname has quit [Ping timeout: 246 seconds]
cat_n9 has joined #arm-netbook
specing has joined #arm-netbook
<Turl>
CTRL sounds more natural to me, it's what's written on the 'Control' key
<mnemoc>
ctnl sounds more like "controller" to me
<mnemoc>
cntl*
Quarx has quit [Read error: Connection reset by peer]
Quarx has joined #arm-netbook
cheng has joined #arm-netbook
<oliv3r>
so CNTL for controller, and CTRL for control?
aexl has joined #arm-netbook
<aexl>
hello again
drachensun has joined #arm-netbook
aexl has quit [Ping timeout: 240 seconds]
lerc has quit [Ping timeout: 246 seconds]
lerc_ has joined #arm-netbook
mSquare has quit [Ping timeout: 246 seconds]
cheng has quit [Quit: Leaving]
tuliom has quit [Quit: Konversation terminated!]
sspiff has quit [Remote host closed the connection]
<cromo>
hi. I have set up xbmca10 and it seems to work except for a sound lagging behind and video stuttering a bit. It seems it only starts to stutter after the sound kicks in. I am not sure whether it's only me here or is just the current state of xbmca10? other than that, it's quite impressive I must say
<rm>
Turl, yes, and....
<rm>
yes.
<drachensun>
cromo: did you build on hard float libraries?
<Turl>
rm: nfs server right?
<cromo>
drachensun: AFIK there are none?
<rm>
Turl, my mk802 is the IPv6 VPN router in my LAN
<Turl>
nice
<cromo>
I mean, cedarx is only available as armel at the moment
<rm>
CONFIG_NFS_V3=y # CONFIG_NFSD is not set
<rm>
will enable nfsd next time
<Turl>
it'd be good to maintain a single "A10-server" config
<drachensun>
cromo: ok, so if the rest of my system is setup for hard float, will that cedar driver work with it?
<Turl>
I have my own too :P
<rm>
oh and it also handles a webcam and an LCD screen displaying a weather report
<cromo>
drachensun: I am same here, I run xbmc from within armel chroot
<rm>
and an USB temperature sensor
<rm>
so it's a "server" in a broad sense of the word, a home automation server of sorts :)
<drachensun>
ahh ok
merbzt has joined #arm-netbook
<Turl>
free on my mele shows 511820
<drachensun>
I think I will just setup a second card with this debian image and use that to try it
<cromo>
drachensun: what distro are you normally on?
revident has quit [Read error: Connection reset by peer]
revident has joined #arm-netbook
rmull has joined #arm-netbook
cat1 has joined #arm-netbook
<drachensun>
cromo: I'm using linaro alip 12.09. Which I think really means Im on ubuntu
<cromo>
use debootstrap and schroot
<drachensun>
I was thinking they probably had a non-hf package running around but I haven't found it. I'm starting to think they upgraded at some point and stopped having the armel version
<cromo>
non-hf package needs non-hd dependencies
<cromo>
*non-hf dependencies
<cromo>
add ubuntu sources to apt/sources.list
<cromo>
use debootstrap to create armel chroot
<cromo>
use schroot for ease things up with mounting dev, proc, sys
<drachensun>
Ok, I haven't used it before but I've got a page up now, it looks pretty straight forward
<cromo>
it is
<drachensun>
well thats a perfect solution then, i can keep it on the same image
<zub>
funny... stripping down defconfig I'm in a state that ifconfig and dhclient always die with some horrible kernel message. but defconfig itself works...
<zub>
CONFIG_BLK_DEV_INTEGRITY ... should this be enabled? it is in defconfig, and I had a config that dies on kernel startup when the option is off and works when it's on. but with a different config it seems to work when it's off. does the HW really have any support for this integrity stuff?
cat1 has quit [Ping timeout: 240 seconds]
<zub>
mnemoc: the same issue (using input_* from drivers/power/axp_power/axp*-sply.c) is present in allwinner-v3.0-android-v2 + patch applies cleanly
<zub>
-> I'd say yup, also for allwinner-v3.0-android-v2
<Turl>
mnemoc: do you want me to redo my patch in that other way? I don't have any preference for either format
<mnemoc>
Turl: I only made questions. I'm no dictator
<mnemoc>
zub: thank you! I'll apply it on both
<zub>
n/p, it's trivial anyway
<mnemoc>
trivial but missing
<Turl>
there's a lot of trivial stuff missing, and it's equally important :)
<mnemoc>
Turl: ML, but the first question remains unanswered, and the "likely" seems unneeded
<zub>
just the ifconfig process dies. the same happens for dhclient, so it seem anything accessing the network interfaces dies
<Turl>
mnemoc: the likely is a compiler optimization hint
<zub>
I can find the cause by going from defconfig (where it works) to mu stripped down config (where this happens), but I'm asking in case it was something obvious...
<mnemoc>
Turl: yes, to optimize branches, but in this case it doesn't give any gain imo
<mnemoc>
zub: weird
<Turl>
mnemoc: ==NULL happens only once on my testing
<Turl>
so it's unlikely
<mnemoc>
ok
<zub>
I'll try to figure the network issue tomorrow. 'night
<mnemoc>
i'm going to sleep too. good night
<mnemoc>
Turl: ML please ;-) there is an open thread to close :)
<mnemoc>
don't forget the reply-to argument
<Turl>
mnemoc: yep give me a minute
<Turl>
mnemoc: btw, "tty: sunxi-8250: fix something in the hackery with more hackery" doesn't seem to fix the issue with the console spamming those lines
<mnemoc>
:(
<mnemoc>
good to sleep now
<mnemoc>
good night
<Turl>
mnemoc: I'm sending it now
<Turl>
mnemoc: what should I type on message-id?
popolon has quit [Quit: Quitte]
<Turl>
rm do you have your server config somewhere online?
merbzt has quit [Ping timeout: 264 seconds]
<Marex>
Turl: that likely() is there mostly for intel
<Marex>
on intel, the in-CPU branch predictor can be programmed and can be hinted
<Marex>
so if you generate proper code, the CPU will execute the code a bit faster
<Marex>
if there's no hint, the CPU prefetches both results of the jump ... if there's hint, it prefetches only one (the expected one) and doesn't prefetch the other
<Marex>
in case the prefetch is incorrect, it's dropped and the correct one is fetched
<Marex>
but if it fetches the correct one n-times and 1 times the incorrect one, the sum gain is actually better