<Offshore>
what's the difference between a13_nuclear_defconfig and a13_defconfig?
<Offshore>
cant find any description about them
hipboi has joined #linux-sunxi
\\Mr_C\\ has joined #linux-sunxi
\\Mr_C\\ has quit [Client Quit]
\\Mr_C\\ has joined #linux-sunxi
vicenteH has quit [Ping timeout: 248 seconds]
\\Mr_C\\ has quit [Client Quit]
\\Mr_C\\ has joined #linux-sunxi
Offshore has quit [Ping timeout: 256 seconds]
\\Mr_C\\ has quit []
egbert has joined #linux-sunxi
egbert_ has quit [Ping timeout: 276 seconds]
\\Mr_C\\ has joined #linux-sunxi
\\Mr_C\\ has quit []
_BJFreeman has joined #linux-sunxi
bsdfox_ has quit [Ping timeout: 264 seconds]
_BJFreeman is now known as BJfreeman
\\Mr_C\\ has joined #linux-sunxi
_BJFreeman has joined #linux-sunxi
BJfreeman is now known as Guest73243
_BJFreeman is now known as BJfreeman
Guest73243 has quit [Ping timeout: 252 seconds]
wingrime has joined #linux-sunxi
[7] has quit [Disconnected by services]
TheSeven has joined #linux-sunxi
jlj has quit [Ping timeout: 264 seconds]
jlj has joined #linux-sunxi
vrga has joined #linux-sunxi
vrga has left #linux-sunxi [#linux-sunxi]
utente has joined #linux-sunxi
BJfreeman has quit [Quit: had a good time]
andoma has joined #linux-sunxi
jlj has quit [Ping timeout: 256 seconds]
jlj has joined #linux-sunxi
Quarx has joined #linux-sunxi
Tsvetan has joined #linux-sunxi
rellla has joined #linux-sunxi
atiti has quit [Ping timeout: 256 seconds]
Tsvetan has quit [Quit: Leaving]
<ykchavan>
What is difference between livesuit and phoenixsuit?
<ykchavan>
??
hipboi has quit [Remote host closed the connection]
shineworld has joined #linux-sunxi
<shineworld>
ykchavan, livesuit is a program to write an image in nand when the board is in FEL mode (powered with FEL button pressed). Phoenixsuit should be a SD maker which put image in SD and at boot write image in nand
<shineworld>
never used 2nd
<ykchavan>
shineworld, that is phoenixcard. phoenixsuit does same as livesuit.
<shineworld>
thank you for fix ! :)
<ykchavan>
Just now Tom Cubie explained me offline. Phoenixsuit is Livesuit2.0
<ykchavan>
Surprisingly, I could not find this information anywhere. :-))
<shineworld>
could you add that in linux-sunxi wiki for all ?
<hramrach__>
if it's old image the driver might be just broken. if it's recent there is still possibility your screen does not work with rounded timings
<hramrach__>
which image?
<shineworld>
I can see the note about Phoneixsuit is the livesuit 2.0
<rachie>
The latest Debian wheezy as of a couple days ago. I also tried openSUSE 12.3.
<rachie>
What exactly is livesuit?
<shineworld>
LiveSuit is a program to flash pre-made images files (IMG) on nand using a PC and cubieboard booted in FEL mode (pressing FEL before to attach OTG cable to PC)
<rachie>
I've tried several different displays, and tried VGA as well. The default android shows up fine, but regular Linux never does.
<rachie>
Curiously, it seems to boot up when I watch it over the serial port.
<rachie>
What window manager does it run? I was thinking of something lightweight like LXDE.
bsdfox_ has joined #linux-sunxi
<hramrach__>
no idea
<rachie>
That's why I wanted the Debian, but now I can't find a download link.
<oliv3r>
imo the 'what is the differnece between livesuit and phoenix suit is a generic allwinner faq item, has absolutly nothing specifically with cubieboard to do.
ykchavan has quit [Read error: No route to host]
ykchavan has joined #linux-sunxi
<rachie>
Okay, a different Debian image is being written to the SD card now.
shineworld has quit [Quit: Leaving]
Quarx has quit [Read error: Connection reset by peer]
<rachie>
The new Debian gives a bunch of errors in the serial log, but manages to boot up to a login prompt with HDMI.
<rachie>
And of course I now discover I don't have a spare USB keyboard.
atiti has joined #linux-sunxi
<mripard_>
Turl: pong pong pong :)
megal0maniac_afk is now known as Guest51791
Guest51791 has quit [Killed (rajaniemi.freenode.net (Nickname regained by services))]
megal0maniac_afk has joined #linux-sunxi
<rachie>
I like pong...
<Turl>
mripard_ a little late now, about to sleep :P
<rachie>
I should go rest too. I need to finish my patent tomorrow.
<Turl>
mripard I'm having trouble mapping the sram, the system somehow decided it's not mapped 1:1
<lunatik>
Hello. I have an A13-based tablet and I'm trying to boot it from SD card. The manuals on the wiki did not work (dunno why), so I want to clone the internal NAND to the microSD. Can anybody give me advice on doing so?
<lunatik>
basically, I just need to dd it to sd card, but I can not find the internal nand device
<lunatik>
there are only nand[a-j] available
Tsvetan has joined #linux-sunxi
<oliv3r>
that is what you want/need
<oliv3r>
you need to create a bootable SD card via the usual way
<oliv3r>
extract the kernel from nandc
<oliv3r>
(and run something like split_bootimg.pl) if you want to put the kernel as file on the sd card (probably recommended)
<oliv3r>
nandb holds your u-boot enviroment
<oliv3r>
nandd is your android /system partition, so you can dd that over (or copy the contents, which should be recommended)
<oliv3r>
and then optionally, /nande is /data and nandg is your CWM recovery partition
<oliv3r>
you cannot dd if=/dev/nand of=/dev/sdcard and think it will work
<oliv3r>
hno: u-boot doesn't even compile from them, for some reason some directories aren't created (usb/nand ones mostly). I am building via the bsp :p
<oliv3r>
e.g. the output dir is different from u-boot
<oliv3r>
manually created the output directories works though
bsdfox_ has joined #linux-sunxi
bsdfox\ has quit [Ping timeout: 260 seconds]
_BJFreeman has joined #linux-sunxi
_BJFreeman is now known as BJfreeman
BJfreeman has quit [Ping timeout: 252 seconds]
liu has quit [Ping timeout: 250 seconds]
ykchavan has quit [Quit: Leaving]
Black_Horseman has joined #linux-sunxi
notmart has joined #linux-sunxi
notmart has joined #linux-sunxi
<Offshore>
[04:30] <Offshore> what's the difference between a13_nuclear_defconfig and a13_defconfig?
<Offshore>
cant find any description about them
_BJFreeman has joined #linux-sunxi
FunkyPenguin has quit [Remote host closed the connection]
_BJFreeman is now known as BJfreeman
buZz has quit [Read error: Connection reset by peer]
FunkyPenguin has joined #linux-sunxi
bsdfox\ has joined #linux-sunxi
bsdfox_ has quit [Ping timeout: 246 seconds]
BJfreeman has quit [Quit: had a good time]
notmart has quit [Quit: notmart terminated!]
megal0maniac_afk is now known as Guest74321
Guest74321 has quit [Killed (adams.freenode.net (Nickname regained by services))]
megal0maniac_afk has joined #linux-sunxi
lujnan has joined #linux-sunxi
lujnan has quit [Client Quit]
tinti has joined #linux-sunxi
lujnan has joined #linux-sunxi
Black_Horseman has quit [Quit: Zwi se logou mou!!!]
<jemk>
wingrime: mpeg12 only and only common parameters
<jemk>
wingrime: dumping memory is done by replay for example
<wingrime>
Trul: jemk: do you find any new contol bits?
<wingrime>
Trul: some times on top wiki shows warning according cache
Soru has quit [Client Quit]
Soru has joined #linux-sunxi
<wingrime>
Trul: also two days ago I tryed add 'lang' header no a31 page
<Turl>
wingrime: yeah but that has nothing to do with spam
<Turl>
I dunno what that is
<wingrime>
but edit was rejected with spam message
<wingrime>
I have not even tryed add link
<wingrime>
my edit not constain any lunk
<oliv3r>
a20 u-boot is dram-less
<oliv3r>
a20 needs boot0, but we have that on hno's github
<oliv3r>
hno: forgot to fix what?
<Turl>
wingrime: should not happen again
<oliv3r>
hno: ohh the directories; i suppose so :)
<hno>
u-boot nand & usb output directories.
<oliv3r>
i can give you an overview
<hno>
I know them...
<hno>
AW makefiles is not perfect..
Soru has quit [Client Quit]
<Turl>
wingrime: btw, I suggest you use tab autocomplete, you wrote my nickname wrong a lot of times :P
Soru has joined #linux-sunxi
<hno>
Offshore, that sounds like some I/O task getting blocked.
<wingrime>
hno: about AXP, I see bug with batt on any a10 a13 tablet it shows 0% always with batt connected
<nove>
wingrime, still i don't get, the reserved buffer is what?
<wingrime>
hno: I checked sysfs files and it looked normal
<wingrime>
hno: It seems to be kpower not understand kernel
<hno>
wingrime, not familiar with battery stuff.
<wingrime>
nove: we have registers that constain addresses
<drachensun>
wingrime: I've noticed none of the monitors work either, if you find a spec for how the module should present the data for kpower or other battery monitors please let me know
<hno>
and AW drivers is tailored for Android only. Very likely they do not expose something normal Linux userspace expects..
heffer_ is now known as heffer
<techn__>
drachensun: you may want to check if hansg has implemented those features into his driver
<techn__>
wingrime also ^ :)
<drachensun>
I guess he does tend to do things the right way
<wingrime>
drachensun: I looked at kpower source code but it have strange strategy for choose witch sys fs fille parse and how
<Offshore>
hno, but i do nothing with the device -- it just gets blocked
<wingrime>
techn__: hansg writed only regulator driver (not batt)
<wingrime>
drachensun: ^
<wingrime>
drachensun: I too not understand witch sysfs file for battery nessesory
Soru has quit [Quit: No Ping reply in 180 seconds.]
<wingrime>
techn__: also do you tesed cm on 3.4 ?
Soru has joined #linux-sunxi
<techn__>
wingrime: I think I have tested it.. but it was ~5-6 months ago
<wingrime>
techn__: I still have hope Quarx with you port current cm to 3.4
<techn__>
why quarx?
<wingrime>
techn__: he worked on cm on a10 (now more with mot defy)
<techn__>
hmm.. looks like quarx is still merging stuff back
<jemk>
wingrime: I pushed ->alpha<- mpeg poc. Test it, it will have many bugs.
<wingrime>
jemk: wikiwiki
<nove>
wingrime, what if instead, the replay tool generates a .c file with the trace, and then that file can be edited to change/dump any register?
<wingrime>
jemk: I mean, add new known bits
<jemk>
wingrime: There are no new yet, all i know i documented
<jemk>
rest is magic constants...
<wingrime>
nove: replay must constain all stuff that go to cedar reserved memory for real work
gzamboni has quit [Ping timeout: 245 seconds]
Soru has quit [Client Quit]
Soru has joined #linux-sunxi
<wingrime>
nove: reserved spaces = lineral memory space reserved at boot for cedar pruposes, any data for cedar go to it (source undecoded data)
<wingrime>
nove: blob uses that memory for SRC data and DST butters
<wingrime>
nove: cedar hw have own DMA engine that can read and write result without cpu, you must only say where SRC and where DST buffers
Soru has quit [Client Quit]
Soru has joined #linux-sunxi
<wingrime>
nove: 'say' I mean write right register for example MACC_MPEG_REC_LUMA 0x148 to it you must write address where cedar will store reconstruct buffer luma
<wingrime>
jemk: may be you can explain better .... ^
<nove>
wingrime, i understand, but not what exactly you what the tracer to dump
<jemk>
nove: i wouldn't let the tracer dump more, thats a job for replay tools
<jemk>
wingrime: explain better? I simply don't know why i write some things to some regs
<wingrime>
nove: more precisly, when blob (mpeg case) write specific bits MACC_MPEG_CTRL2 (that mean do start decoding process) or when blob return from VE_WAIT ioct (frame finished) you must dump physical memory at offset in some regs for exmple MACC_MPEG_REC_LUMA (many buffers)
<wingrime>
jemk: now you code like magic , jpeg mostly avoid it
<nove>
jemk, i also agree, but i want to know more information wingrime says is need
<jemk>
wingrime: I simply didn't find any mpeg that changes those bits, so i can't figure out what they could mean
<wingrime>
jemk: also witch tool you used for dumping mpeg headrs for comparea with regs
<jemk>
wingrime: if you finde some let me know
<wingrime>
jemk: I can try find more defenetly but I must know where search ( witch function mostly important )
<wingrime>
jemk: witch regs are interested for you ?
<wingrime>
jemk: for example I can firstly say hardcoded that or not
<jemk>
wingrime: 0x114 mostly
<wingrime>
jemk: I investigate it before
<wingrime>
jemk: I can say that blob use 0x7c for clean pengind irq (0) enable irq (1)
<wingrime>
jemk: other bits
<wingrime>
jemk: used with mpeg4
<jemk>
wingrime: I simply write 0x800001b8 there for now, but i think there is data from mpeg involved
<wingrime>
jemk: It used for mpeg4 widely
BJfreeman has quit [Ping timeout: 252 seconds]
<wingrime>
jemk: witch tool are you used for mpeg headers dumping ?
<jemk>
wingrime: and some bits from reg 0x100 are guessed too, but i think they are correct becaus their order matches mpeg spec
<wingrime>
0x100 not used for mpeg4
<jemk>
wingrime: no tool, i first wrote my own parser (mpeg.c) to understand mpeg format
<wingrime>
jemK: i thinking about patch mpleyer for it or simular
<wingrime>
0x100 mostly known (mpeg header)
<jemk>
wingrime: i wrote this table you mean with mostly known, so i know, but some are guessed
<wingrime>
nove: not think that do strange stuff here
<wingrime>
jemk: ^
<wingrime>
jemk: 0x100 look realistic, more stange trigger reg can ctrl
<wingrime>
jemk: about 0x114
<wingrime>
I noticed that 800851fc for P+B 800841fc for I frame in mpeg4
<wingrime>
8008 look like color format
<wingrime>
but not sure
<wingrime>
nove: do you understnad what we want?
<wingrime>
jemk: about jpg poc please replace //?? with new info
<nove>
wingrime, this is going too fast than i can keep with, if you want me to do something, it must be clear explained.
zeRez has quit [Read error: Connection reset by peer]
zeRez has joined #linux-sunxi
zeRez_ has joined #linux-sunxi
<nove>
wingrime, by the way, when you write "constian" you meant "contains"
<jemk>
wingrime: the decoding is documented as good as i know
<wingrime>
jemk: may be define header with registers offsets
<jemk>
wingrime: the code around it are only fast written hacks that shouldn't really be used
<jemk>
wingrime: header with regs would be good idea, but as long as everything changes all the time it would mean huge maintaining effort
<wingrime>
jemk: we now have mostly stable regnames
<wingrime>
jemk: but I think ctrl2 must be trigger
zeRez has quit [Ping timeout: 256 seconds]
<wingrime>
jemk: I think you can make header with register names for mpeg engine
<wingrime>
jemk: only regs not bits
<wingrime>
jemk: how many files work now?
<jemk>
wingrime, oliv3r: btw, why are regs named MACC_ when engine/wiki name is VE
<wingrime>
jemk: with mpeg1/2
<wingrime>
jemk: oliv3r name it for some reason
<wingrime>
jemk: feel free fix it and stabilize
<wingrime>
jemk: a10 name cedarx VPU
<wingrime>
nove: 1)replace offsets with reg names, secondly 2) bit-fields for known regs 3) mark frame decodig process begin/end 4) ability show buffers at frame begin/end
n01 has quit [Ping timeout: 276 seconds]
<wingrime>
jemk: how much files work with mpeg example?
<jemk>
wingrime: all i tried, but i know some will not work because i simply ignore some things
<wingrime>
for example:
<jemk>
for example custom quantization tables are ignored due too lazyness
n01 has joined #linux-sunxi
<wingrime>
jemk: it make invalind colors on output?
<nove>
wingrime, ok, for 3) 4) ideally should be the task of the player(player -> ammt -> trace)
<jemk>
wingrime: i will fix many of these when i have more time again (august)
<jemk>
wingrime: i don't know what happens, haven't had anything to test
<wingrime>
nove: don't forget define iocrt
<wingrime>
*ioctl
n01 has quit [Ping timeout: 240 seconds]
n01 has joined #linux-sunxi
dwilkins has joined #linux-sunxi
dwilkins has quit [Read error: Connection reset by peer]
Soru_ has joined #linux-sunxi
Soru has quit [Read error: Connection reset by peer]
ssvb has quit [Read error: Connection timed out]
ssvb has joined #linux-sunxi
dwilkins has joined #linux-sunxi
wingrime has quit [Ping timeout: 240 seconds]
CountryGeek has joined #linux-sunxi
dwilkins has quit [Ping timeout: 240 seconds]
CountryGeek is now known as dwilkins
FunkyPenguin has quit [Ping timeout: 240 seconds]
FunkyPenguin has joined #linux-sunxi
<utente>
hello.
<utente>
i on cubieboard. i fount a istallation "cubian", a debian for cubieboard. it works, but kernel didt support all hardware i have. so i recompiled 3.4.43 (patched for cubie). i used the .confg of 3.4.29 using "make oldconfig" options, then i did make menuconfig to activatre what i need. at the end i moved uImage and /lib/modules/3.4.43 on cb. all works fine the new hardware support, but i lost the hdmi audio (that of course prev
<utente>
iously works fine).
<utente>
who to do to reactivate hdmi audio? afaik, the 2 config are identical related to audio menu :-(
Black_Horseman has joined #linux-sunxi
jemk has quit [Remote host closed the connection]
tzafrir has joined #linux-sunxi
<tzafrir>
my mk802 (Debian, kernel "3.0.62+" feels aufully slugish.
<tzafrir>
top takes some 10% CPU, which is something I had in my long-lost 486
tinti has quit [Read error: Connection reset by peer]
dwilkins has quit [Remote host closed the connection]
<oliv3r>
hansg: i'll try to review that motherload tomorrow or so
<oliv3r>
Turl: what's interesting about the cubieboard2 src drop is that atleast allwinner granted them permission to redistribute everything; sot hat's good
nove has quit [Quit: nove]
hark_ has quit [Ping timeout: 245 seconds]
<oliv3r>
wingrime: jemk: nove: does cedarX have some deinterlacing thing in it? or is that the task of some other component in the SoC?
hark has joined #linux-sunxi
<Turl>
utente: check your alsa settings, cards might have shuffled places
<utente>
Turl, i ONLY changed uImage and kernel modules, not else. so - i suppose - the problem is into kernel settings.
<Turl>
utente: yes, but new kernel can have different alsa card #s for HDMI and analog
<Turl>
so check your alsa settings
<utente>
Turl, whare are alsa settings?
<Turl>
utente: usually 'sound' settings on gnome/xfce
<Turl>
I don't know what you're using
<utente>
lxde
<utente>
at least, tell me what is the program to install to do the job.
dwilkins has joined #linux-sunxi
<Turl>
utente: check /proc/asound/cards and make sure you've got the hdmi card there
bsdfox\ has quit [Ping timeout: 248 seconds]
<oliv3r>
alsamixer -c 0
<oliv3r>
alsamixer -c 1
<oliv3r>
alsamixer -c 2
<oliv3r>
that should cover them all
<utente>
i will see
<hramrach__>
also alsamixer has a nice card selection menu
<utente>
oliv3r, -c 1 has not setting, -c 2 invalit card nujmber
<utente>
Turl, i did asound.conf
<utente>
test works!!! YESSSSSSSSS!!!!!!! yes ytes yes!!!!!!
<utente>
Turl, a beer 4 u :D
<utente>
c|_|
<oliv3r>
-c 1 should open the 2nd card, e.g. hdmi
<utente>
mmm speaket-test say "right feft" in hdmi audio output, but if i try youtube, it is mute, no audio :( (yes, gnash audio cursor is to the maximum)