<benn>
wens: powerVR : the Free Software Foundation has put reverse-engineering of the PowerVR driver on a high-priority project list
<benn>
wens: I still don't understand why most of us don't like PVR ? do you know what is the key reason?
<wens>
benn: you will have to ask the others
<wens>
i am not familiar with the subject
<benn>
wens: ok :P
<wens>
benn: i suppose the reason is, no documentation whatsoever has ever been released for PVR
<benn>
hi everybody in the list, who can give any suggestions about the video enc/dec ? or what problem we have on a20 ?
<benn>
or what API we expected from their binary blob ?
<wens>
benn: from wiki: The Android glue code is implemented as a "media player" (parallel to Android's Stagefright multimedia framework) instead of as standard OpenMAX (OMX) components and API's.
<wens>
i think most people in this channel would say binary drivers are a bad thing
<wens>
but for maximum usability for users, binary drivers are needed for now
<wens>
benn: ask them to support OpenMAX
<benn>
wens: yes, i know. but we also need to realize that, allwinner will never release the video source code.
<tomboy64>
video source code = cedarX?
<benn>
tomboy64: yes. and they are willing to support openMAX, i think
<wens>
benn: not betting on them to release source code
<tomboy64>
last status about cedarx that i read about was that it's necessary to wrap it into libhybris.
<tomboy64>
which complicates things
<wens>
hopefully OpenMAX support will get rid of this requirement
<tomboy64>
but i haven't been able to actually get it running yet - too busy this week.
<tomboy64>
btw, did you try to get cubietruck working with the brcmfmac driver? their homepage states they support 4329-sdio
<benn>
tomboy64: not till now
<wens>
it still needs some glue for gpio and pinctrl, haven't had the time though
<benn>
wens, tomboy64: but at least we have had a working driver now :D
<tomboy64>
:) we do
<tomboy64>
it works quite fine for me, though i had to tweak wpa_supplicant
<tomboy64>
and i learned to hate network-manager - again.
<wens>
:p
<benn>
i found a terrible bug for cpufreq yestoday
<benn>
ondemand will make the system unstable
<wens>
may need to tweak some parameters for ondemand. i remember someone got it working
<benn>
it works, but unstable
<benn>
and I think it's hard to debug why this happen. what I can do is just tell everybody try to avoid using ondemand cpufreq :D
fredy has quit [Excess Flood]
fredy has joined #linux-sunxi
[7] has quit [Read error: Operation timed out]
prasannapete has joined #linux-sunxi
TheSeven has joined #linux-sunxi
JohnDoe_71Rus has joined #linux-sunxi
deasy has joined #linux-sunxi
<Turl>
benn: I believe their android sdk already has openmax in hardware/aw/
<JohnDoe_71Rus>
hi
<Turl>
benn: on what hw is cpufreq unstable?
hipboi has joined #linux-sunxi
<Turl>
hi JohnDoe_71Rus, hipboi
<hipboi>
Turl, hi
TheSeven has quit [Disconnected by services]
[7] has joined #linux-sunxi
<benn>
Turl: can we use that openmax on linux platform ?
<Turl>
benn: I don't know
<Turl>
benn: probably not, there is a blob .so there
<benn>
Turl: so we need a working version on linux, it's that right ?
<Turl>
maybe with libhybris
<benn>
Turl: libhybris and openmax, we need them all ?
<Turl>
benn: libhybris is a hack to use android libraries on normal glibc distro
<libv>
collabora again... why is it that when i see any company hacking in binary driver support, it is always them
<libv>
libhybris and wayland integration was also collabora
<Turl>
libv: it makes sense once you read between lines on their company name
<libv>
one of their noisiest employees is known to actively hinder proper open source projects, in favour of less free ones
<Turl>
hm, splitted-desktop pages seem to have been moved/killed
hipboi has quit [Ping timeout: 264 seconds]
Taduro has quit [Ping timeout: 246 seconds]
hipboi has joined #linux-sunxi
tomboy64 has quit [Ping timeout: 240 seconds]
tomboy64 has joined #linux-sunxi
gzamboni has quit [Read error: Connection reset by peer]
gzamboni has joined #linux-sunxi
panda84kde has joined #linux-sunxi
Taduro has joined #linux-sunxi
BluesBoy has joined #linux-sunxi
rellla has joined #linux-sunxi
Quarx has joined #linux-sunxi
eebrah has joined #linux-sunxi
FunkyPenguin has quit [Ping timeout: 246 seconds]
FunkyPenguin has joined #linux-sunxi
wolfy has joined #linux-sunxi
eebrah has quit [Ping timeout: 248 seconds]
n01 has joined #linux-sunxi
eebrah has joined #linux-sunxi
eebrah has quit [Read error: Connection reset by peer]
prasannapete has quit [Quit: prasannapete]
eebrah has joined #linux-sunxi
hansg has joined #linux-sunxi
enrico_ has joined #linux-sunxi
tzafrir has quit [Ping timeout: 246 seconds]
eebrah has quit [Read error: Connection reset by peer]
benn_ has joined #linux-sunxi
eebrah has joined #linux-sunxi
_massi_ has joined #linux-sunxi
benn____ has joined #linux-sunxi
benn_ has quit [Ping timeout: 265 seconds]
notmart has joined #linux-sunxi
<juanfont>
benn____ ping
<benn____>
juanfont: pong
<juanfont>
hi :) why do you think AW is willing to support OMX? Could we expect OMX for the A20 SoC or only for future models?
<arokux>
mnemoc: have you fixed it? (i haven't touched anything)
eebrah has quit [Ping timeout: 265 seconds]
adb_ has joined #linux-sunxi
<arokux>
hi benn____ I wish you nice time with allwinner :)
eebrah has joined #linux-sunxi
n01 has quit [Ping timeout: 272 seconds]
<mnemoc>
arokux: yes, yesterday issues with the nightlies are fixed. I wonder if anyone uses xen from there
<wens>
arokux: i think he's back already :p
<arokux>
wens: oh :)
n01 has joined #linux-sunxi
rellla3 has joined #linux-sunxi
<arokux>
then.. benn____: I hope you had nice time with allwinnner! did they mention sunxi.org?
rellla3 has quit [Client Quit]
eebrah has quit [Quit: Lost terminal]
benn____ has quit [Ping timeout: 264 seconds]
FR^2 has joined #linux-sunxi
prasannapete has joined #linux-sunxi
HeHoPMaJIeH has joined #linux-sunxi
HeHoPMaJIeH has joined #linux-sunxi
diego has joined #linux-sunxi
diego is now known as Guest20264
panda84kde has quit [Read error: No route to host]
lkcl has quit [Ping timeout: 252 seconds]
panda84kde has joined #linux-sunxi
Guest20264 has quit [Ping timeout: 264 seconds]
tzafrir has joined #linux-sunxi
<JohnDoe_71Rus>
arokux: Hi
wolfy has quit [Remote host closed the connection]
<[7]>
to what degree have people attempted to reverse engineer livesuit?
diego_ has joined #linux-sunxi
panda84kde has quit [Ping timeout: 248 seconds]
<libv>
[7]: i think hno did most of the work on getting fel mode figured out
<[7]>
has anyone attempted to mess with livesuit (or rather its components) themselves?
<libv>
[7]: i am pretty clueless about the whole topic, but it could be that livesuit is nothing more than an interface on top of fel
<[7]>
FEL by itself isn't much more than a code downloader and executor, which is fed with various little binaries that interface with each other and boot up some kind of firmware update environment on the device
<[7]>
all of that is controlled by livesuit, which is basically a lua engine
<libv>
[7]: so what is there to reverse engineer?
<[7]>
the lua parts of livesuit can very easily be decompiled, which will tell a lot about how the actual booting and flashing process works
<[7]>
sure, we could just rip it out and upload our own binaries (uboot spl etc.) instead and use those to reflash the device
<[7]>
but that would of course require some additional features in uboot (most importantly nand support)
<[7]>
those livesuit images are basically archive containers, similar to zip files
<libv>
[7]: right, first nand support, then someone just needs to go and make some dumbed down tool for android users
<[7]>
when you load such an image, livesuit executes the "UPFLYTLS" lua script inside that archive
<libv>
[7]: so is something open source capable of loading livesuit images what you are looking for?
<[7]>
what I'm looking for, in the long term, is an open source solution for nand access, android boot, and flashing via FEL
<[7]>
in the short term, I just need something that works :)
<libv>
surely you already have something that works?
<[7]>
i.e. that will flash my own android images and boot them (from nand)
<[7]>
so far I haven't managed to build a livesuit image that can actually be flashed
<libv>
[7]: so basicall, you want to reproduce the full livesuit
<[7]>
whatever source code olimex released for its android image for this board (olinuxino a20) produces completely different images compared to the image binary that they released
<[7]>
first of all, I want to understand how livesuit works and actually work with my images
<[7]>
in the long term, I wouldn't mind replacing it with something better
<libv>
[7]: i guess that you are on your own with that one, as everyone else is mostly happy to manually install their proper OS on their own devices :)
<[7]>
what is really making me angry is that the source code releases just can't be used to make working binaries
diego__ has joined #linux-sunxi
<libv>
[7]: working binaries which can only be loaded by the allwinner windows tool?
<[7]>
the binary seems to have been built with a much newer version of the allwinner packing tools, which I haven't found anywhere yet
<[7]>
libv: tell me a way how to properly reflash android to the nand, ideally via USB, and I'll be happy
<[7]>
I've tried messing with nand in uboot for that purpose, but so far not successfully
<[7]>
mostly due to the fact that the A20 nand driver is only available as a blob
<[7]>
unlike the A10 one, for whatever reason
<libv>
i really cannot help you there, i haven't ever needed that, as i was _very_ glad to finally get a mali based device with a proper linux :)
torbenh3 has quit [Ping timeout: 245 seconds]
<[7]>
well, if you give me a proper linux that works well with touchscreen-based input and support for most on-board hardware, I'd drop that android project right away :)
diego_ has quit [Ping timeout: 245 seconds]
juanfont has left #linux-sunxi ["Saliendo"]
juanfont has joined #linux-sunxi
juanfont has left #linux-sunxi ["Saliendo"]
juanfont has joined #linux-sunxi
torbenh3 has joined #linux-sunxi
Black_Horseman has quit [Ping timeout: 240 seconds]
<juanfont>
where can i find the AllWinner SDK? I'm looking for that rumoured libomx.so...
<hno>
[7], but never cared about livesuite protocol. It's very similar to FEL protocol but different, and we don't really need livesuit at all.
<hno>
just a little time to finish up NAND driver, which is considerably less and more productive than trying to reverse livesuit.
<[7]>
atsampson: that source doesn't match the blob used by all those android images
<hno>
[7], the android u-boot sources is in the A20 SDK, but full of GPL violations.
<hno>
and probably not matching the binary used either.
<[7]>
hno: the one in the android sources uses a nand driver blob
fredy has quit [Excess Flood]
<hno>
yes
<hno>
and in the lichee-dev-a20 tree we have the supposedly matching sources of that blob.
<hno>
from some point in time.
<[7]>
those had major differences
<[7]>
i.e. completely incompatible with the on-flash data structures
<[7]>
different log block table entry format, different partition table format, probably more that I haven't found yet
<[7]>
also, I'm not really trying to fully reverse engineer livesuit, rather trying to understand how the basics work, and fix it up to work properly with my images
fredy has joined #linux-sunxi
fredy has quit [Excess Flood]
panda84kde has joined #linux-sunxi
fredy has joined #linux-sunxi
Guest69324 has quit [Ping timeout: 272 seconds]
<[7]>
hno: so the plan is to go for MTD in the long run?
<[7]>
how will that work with android?
<torbenh3>
whats the problem for android to use proper mtd ?
eebrah has quit [Read error: Connection reset by peer]
<[7]>
IIUC you can't support proper MTD with this hardware, as it has severe constraints on how you can use the OOB bytes
<[7]>
which means that yaffs etc. won't work on top of it
<[7]>
IIUC the only file system that will work on top of this will be UBIFS, and I've never seen that being used with android so far
<torbenh3>
so ubifs works... the only thing i am interested in.
<[7]>
how well does it work with android?
<hno>
[7], don't have much trust in AW continuing to release usable NAND driver sources.
eebrah has joined #linux-sunxi
<torbenh3>
errm... its just an FS... i mean... android just works fine with an NFSroot. i dont see why there should be problems...
<[7]>
things like mount times, RAM usage, stability and durability for that use case, ...
<[7]>
flash file systems are kind of a beast
<hno>
[7], the AW NAND block emulation is no less of a beast.
prasannapete has quit [Quit: prasannapete]
prasannapete has joined #linux-sunxi
<[7]>
hno: sure, but in my experience FTLs are generally much better suited for large MLC flash media than flash file systems
<[7]>
it's kinda of a shame that there isn't a good open source FTL yet+
<[7]>
maybe I should go ahead and write one, if I ever find time to do so
<hno>
[7], there is reasons to that... US patents system.
<hno>
There is several quite potent FTL layers in the kernel even, but you are not allowed to use them on NAND.
eebrah has quit [Read error: Connection reset by peer]
<[7]>
I couldn't care less about patents... they're a major problem for device and chip manufacturers, but not really for hackers who want to replace the closed source crap on their devices with something better
<torbenh3>
ubi with fastmap is pretty fast in mounting...
<[7]>
how much RAM does it consume for a reasonably large flash device?
<[7]>
I once pondered to use it for another project, and came to the conclusion that it doesn't scale well enough
<[7]>
but I think that was on a device with just 32MB of RAM vs. 8GB of flash
eebrah has joined #linux-sunxi
<hno>
[7], AW libnand sources is ~17K lines with ~0 lines documentation.
<[7]>
(on a side note, I'm fairly sure that UBI would violate similar patents to FTLs)
<hno>
and no one to talk to about it..
<[7]>
hno: I might be biased because I have reverse engineered a whole FTL from a disassembly a while ago, so doing it from source seems comparatively easy.
eebrah has quit [Quit: Lost terminal]
<hno>
[7], it is.. all structures is there. But I am not sure we will see sources to next libnand major update, the sources we have is already way behind the binary blob of later SDKs.
<hno>
or that the actual sources that is being used actually do good things.. my first analysis of the version shipped for the A20 SDK raised many major questionmarks.
<[7]>
the blobs aren't that bad either, with debug symbols and printf's
<[7]>
yes, I spotted a few rather suspicious things as well
<[7]>
torbenh3: btw, another problem that would need to be solved is how to flash UBIFS to an android device
<[7]>
you'd need to implement support for unpacking a sparsed UBIFS image to a flash with bad blocks etc. into fastboot or something
diego_ has joined #linux-sunxi
<torbenh3>
[7]: you just need to skip bad blocks, when writing...
wolfy has quit [Remote host closed the connection]
<[7]>
that might be true for yaffs, but is it true for UBIFS as well?
fredy has quit [Excess Flood]
<[7]>
hm, might work if you don't use fastmap
panda84kde has quit [Read error: No route to host]
diego__ has joined #linux-sunxi
<torbenh3>
the fast map is generated, when that image is mounted first time.
panda84kde has joined #linux-sunxi
fredy has joined #linux-sunxi
diego_ has quit [Quit: Konversation terminated!]
fredy has quit [Excess Flood]
diego__ has quit [Ping timeout: 240 seconds]
fredy has joined #linux-sunxi
wolfy has joined #linux-sunxi
diego_ has joined #linux-sunxi
<[7]>
"local function downlaod_MBR(para)" - I love that kind of english :)
panda84kde has quit [Ping timeout: 272 seconds]
adb_ has quit [Ping timeout: 252 seconds]
diego_ has quit [Read error: Operation timed out]
diego_ has joined #linux-sunxi
fredy has quit [Excess Flood]
ZetaNeta has quit [Ping timeout: 246 seconds]
panda84kde has joined #linux-sunxi
fredy has joined #linux-sunxi
JohnDoe_71Rus has quit [Read error: Connection reset by peer]
diego_ has quit [Ping timeout: 272 seconds]
Taduro__ has quit [Ping timeout: 246 seconds]
adb_ has joined #linux-sunxi
Black_Horseman has joined #linux-sunxi
ganbold_ has joined #linux-sunxi
ZetaNeta has joined #linux-sunxi
penghb has quit [Read error: Connection reset by peer]
Black_Horseman has quit [Remote host closed the connection]
prasannapete has quit [Quit: prasannapete]
awafaa has quit [Remote host closed the connection]
jeremb_ has quit [Remote host closed the connection]
steev has quit [Remote host closed the connection]
ojn has quit [Remote host closed the connection]
awafaa has joined #linux-sunxi
geecko has joined #linux-sunxi
Taduro_ has joined #linux-sunxi
penghb has joined #linux-sunxi
steev has joined #linux-sunxi
hramrach has joined #linux-sunxi
jeremb_ has joined #linux-sunxi
fredy has quit [Excess Flood]
fredy has joined #linux-sunxi
adb_ has quit [Ping timeout: 252 seconds]
ojn has joined #linux-sunxi
fredy has quit [Excess Flood]
fredy has joined #linux-sunxi
panda84kde has quit [Remote host closed the connection]
ZetaNeta has quit [Ping timeout: 272 seconds]
panda84kde has joined #linux-sunxi
diego_ has joined #linux-sunxi
EGM has quit [Ping timeout: 246 seconds]
panda84kde has quit [Read error: Connection reset by peer]
HeHoPMaJIeH has quit [Remote host closed the connection]
wingrime has joined #linux-sunxi
EGM has joined #linux-sunxi
popolon has joined #linux-sunxi
diego_ has quit [Read error: Operation timed out]
diego_ has joined #linux-sunxi
naobsd has joined #linux-sunxi
bbrezillon has joined #linux-sunxi
nove has joined #linux-sunxi
rz2k has joined #linux-sunxi
bbrezillon has left #linux-sunxi ["Ex-Chat"]
<libv>
hrm, it's been a while since ssvb was around
<popolon>
Fix miscompilation of tar on armhf. LP: #1243656 <= on the last update of gcc-4.8 on *buntu 13.10
<popolon>
could resolve the probleme we have with the sunxi kernel and gcc 4.8 ?
<hglm>
popolon: gcc-4.8 worked for me sunxi a few months ago (after a kernel memset patch was appplied), but haven't tried 4.8 for the kernel recently (only 4.7).
<hglm>
I know for sure the gcc 4.8.2 on current Debian unstable is BROKEN because it fails with -O2 on some floating-point intensive C++ code...(works OK with -O).
<popolon>
ok
<popolon>
because I managed to compile a ~perfectly working kernel in begining of october with gcc-4.8
<popolon>
but seem than some updates (on linux-sunxi git and/or gcc) broken everything
<popolon>
every other kernel compilation/binary crash a lot for me
<hglm>
popolon: So current linux-sunxi kernel does not work when compiled with gcc 4.8.2?
<popolon>
don't know for last one
<popolon>
but one or two weeks ago, some said me that problem was identified between linux-sunxi kernel and gcc 4.8
<popolon>
gcc (Ubuntu/Linaro 4.8.1-10ubuntu9) 4.8.1 <= I've this one now
<hglm>
Well I fixed an issue with gcc 4.8.2 that I had with a C++ class that involved reinterpret_cast -- not sure if the code is actually wrong or it is a compiler bug. At least it didn't show up on earlier gcc versions like 4.7.
eebrah_ has quit [Read error: Connection reset by peer]
pirea has joined #linux-sunxi
<pirea>
hy
<pirea>
with libvdpau-sunxi subtitle doesn't work
<pirea>
why?
<hglm>
pirea: I think subtitle may be fairly complicated, because it involves something beyond the raw video buffer,
eebrah_ has joined #linux-sunxi
<pirea>
hglm tnx
<hglm>
pirea: I would assume libvdpau doesn't directly deal with subtitles.
FR^2 has quit [Quit: Connection reset by peer]
tomboy64 has quit [Ping timeout: 240 seconds]
<pirea>
hglm but with another vo like xv it's working
<hglm>
I think the decode layer before xv (for example gstreamer) blends the subtitles into the YUV data, then xv converts the YUV to RGB (hardware disp layer).
<hglm>
pirea: Which video player?
<pirea>
hglm mplayer
Guest6981 has quit [Ping timeout: 246 seconds]
tomboy64 has joined #linux-sunxi
panda84kde has joined #linux-sunxi
woprr has quit [Read error: Operation timed out]
<hglm>
pirea: I think it is up to mplayer to blend the subtitles -- maybe there is an option or work-around in mplayer to show subtitles.
woprr has joined #linux-sunxi
ZetaNeta has joined #linux-sunxi
ZetaNeta has quit [Changing host]
ZetaNeta has joined #linux-sunxi
<pirea>
i will try to compile xbmc with libvdapu
<pirea>
but ....:-?
eebrah_ has quit [Ping timeout: 252 seconds]
<pirea>
hglm i have the same problem with cubian(debian)
eebrah_ has quit [Read error: Connection reset by peer]
panda84kde has quit [Quit: Konversation terminated!]
pirea has quit [Quit: Leaving]
Taduro_ has quit [Read error: Connection reset by peer]
7CBAAIFWN has joined #linux-sunxi
FR^2 has joined #linux-sunxi
jinzo has joined #linux-sunxi
eebrah__ has quit [Read error: Connection reset by peer]
pfdm has joined #linux-sunxi
\\Mr_C\\ has joined #linux-sunxi
n01 has joined #linux-sunxi
eebrah_ has joined #linux-sunxi
n01_ has joined #linux-sunxi
n01 has quit [Ping timeout: 264 seconds]
n01_ is now known as n01
eebrah_ has quit [Ping timeout: 264 seconds]
eebrah_ has joined #linux-sunxi
von_fritz has joined #linux-sunxi
hglm has quit [Quit: leaving]
rellla3 has joined #linux-sunxi
n01 has quit [Ping timeout: 272 seconds]
eebrah_ has quit [Read error: Connection reset by peer]
jeremb_ has quit [Quit: Connection closed for inactivity]
eebrah_ has joined #linux-sunxi
Tumbledor has joined #linux-sunxi
Tumbledor has quit [Client Quit]
<mnemoc>
lovely... the 4.4 OTA (soft)bricked my nexus 7 :(
eebrah_ has quit [Read error: Connection reset by peer]
eebrah_ has joined #linux-sunxi
n01 has joined #linux-sunxi
tzafrir has quit [Ping timeout: 272 seconds]
von_fritz has quit [Quit: vonfritz leaves, don't panic]
shineworld has joined #linux-sunxi
shineworld has left #linux-sunxi ["Leaving"]
Fusing has joined #linux-sunxi
libv has quit [Ping timeout: 260 seconds]
rz2k has quit []
ZetaNeta has quit [Ping timeout: 252 seconds]
<ccube>
where to start when I want to control LED1 with integrated PWM from linux on olixulino a10s?
libv has joined #linux-sunxi
eebrah_ has quit [Read error: Connection reset by peer]
eebrah_ has joined #linux-sunxi
paulk-collins has joined #linux-sunxi
jinzo has quit [Ping timeout: 272 seconds]
rellla3 has quit [Remote host closed the connection]
eebrah__ has joined #linux-sunxi
eebrah_ has quit [Ping timeout: 260 seconds]
focus2 has joined #linux-sunxi
jinzo has joined #linux-sunxi
jinzo has joined #linux-sunxi
jinzo has quit [Changing host]
libv has quit [Ping timeout: 260 seconds]
libv has joined #linux-sunxi
Black_Horseman has joined #linux-sunxi
Black_Horseman has quit [Changing host]
Black_Horseman has joined #linux-sunxi
<focus2>
Ordered 3 x cubieboards yesterday - start of masterplan to make 3D stackable computing array :)
<focus2>
wondering what to call the computer
<focus2>
Squeeech (TM)
<focus2>
may be not :)
<focus2>
To be totally pretentious, one has to add a dyne to the end or it would be as wet as calling it a squeech (TM)
<focus2>
I know - i call it CubieDyne (TM)
AreaScout has quit []
<focus2>
hmmm... I like Squeeech better...
<focus2>
my squeeech has 6 cubieboards per case - each case is a cube
<focus2>
the cube itself is a puzzle made of 12 identical pieces that have to be assembled in correct order or it won't assemble into cube shape
<focus2>
inside the cube are brackets to hold one cubieboard and one 2.5 inch sata drive per cube face
<focus2>
plenty of space inside the cube to run cables between the 6 computers
<focus2>
the special design of the case also allows any number of cubes to stacked and cables like ethernet to be run into any cube (within reason)
<focus2>
design limit is 4 x 4 x 4 cubes assembled without cabling issues
<Turl>
focus2: cooling? ;)
<focus2>
60W per cube with 6 cubies + sata HDDs present - plenty of vent holes to take the air in and out as well as the cables - though too many cables could choke the vent holes which are in effect also the cabling ducts
<Turl>
you may want to add some fans to keep things cool
<focus2>
Nearly 4kW if constructing 4x4x4 array = 384 dual core CPUs + 384 TB disk space max - not bad I think!!!!!
<focus2>
linear dimensions of 4x4x4 array is estimated at 0.6M x 0.6M x 0.6M cubic
<Turl>
focus2: you'll need a couple kW more to cool the stuff :P
<jelly-home>
perhaps focus2 lives in Finland
<focus2>
possible to run heat pipes through the vent holes
<popolon>
hi focus2
<focus2>
each cube costs about USD1000 to make. The entire 384 CPU + 384TB disk system will cost 64 kilo dollars
<focus2>
hi popolon
<popolon>
did you see the patch on ubuntu 13.10 today
<popolon>
about the compilation bug of tar, introduced by another ubuntu patch
<focus2>
i saw the messages - broken gcc
<popolon>
perhaps , it will work again now for linux-sunxi kernel ?
<popolon>
and what about your project of 2*aw A20
<focus2>
popolon: is it worth recompiling the lot again - I'm trying to liven up cubietruck with ubuntu - no luck yet
<popolon>
I will retry, with last one update and with ~5 october that worked at this time
<focus2>
popolon: really? I tried 5 october - it crashed out with unable to access partition 2 on uSD card - it don't have it, and I don't know why it is looking there - it is supposed to boot from sata
<Turl>
focus2: 1kUS$ per cube? why so much?
<focus2>
6 x cubies + 6 x TB 2.5" drives
<focus2>
may about $600 if cut specs a little
<Turl>
hm I thought spinning rust was cheaper
<Turl>
yeah makes sense when you do the math :)
<focus2>
2.5" drives double price of 3.5"
<focus2>
nearly
<popolon>
with this version, I'm able to boot on SATA and uSD
<popolon>
I still use it
<popolon>
didn't have to much time to play with my cubie since few weeks
<focus2>
popolon: hmmm ... i must retrace the steps and try again - i got 2 cubietrucks to play with
tzafrir has joined #linux-sunxi
iamfrankenstein has quit [Ping timeout: 265 seconds]
<oliv3r>
mnemoc: btw, 4.4 will be pointless for sunxi (the memory savings exist allready in CM anyway)
eebrah__ has quit [Read error: Connection reset by peer]
<focus2>
first mission of the 3 cubieboard squeech (TM!) is to accept an uploaded html file, split the file into 3 server PCs and rewrite the URLs so that web content such as images are pulled from different servers
chkktri has quit [Ping timeout: 264 seconds]
<focus2>
the page turn speed of DDR is still 10MHz after all these decades - I should be able to get it up to 30MHz effective with this technique and serve content more quickly
paulk-collins has quit [Remote host closed the connection]
<focus2>
what i would really like to do is experiment
<focus2>
with 3D stacked computing arrays now that prices are coming crashing down
<focus2>
the highest prize is making sql query / transactions work on a parallel cpu system - worth a lot if it can be made to work
eebrah_ has joined #linux-sunxi
<mnemoc>
oliv3r: i'm from those who prefer AOSP over CM... but as I won't be doing it, I prefer CM over no sunxi.org-supported android
<mnemoc>
device/allwinner/common was abandoned long time ago :<
chkktri has joined #linux-sunxi
heffer has joined #linux-sunxi
heffer has joined #linux-sunxi
heffer has quit [Changing host]
naobsd has quit [Quit: Page closed]
eebrah_ has quit [Ping timeout: 246 seconds]
eebrah_ has joined #linux-sunxi
eebrah_ has quit [Read error: Connection reset by peer]
eebrah_ has joined #linux-sunxi
issueat has joined #linux-sunxi
<issueat>
Hi there! Does anyone have a working cubietruck kernel soure repository with included sunxi_gmac?
notmart has quit [Quit: notmart terminated!]
ganbold_ has quit [Remote host closed the connection]
FR^2 has quit [Quit: und weg...]
<mnemoc>
issueat: stage/sunxi-3.4 has gmac
enrico_ has quit [Quit: Bye]
eebrah_ has quit [Read error: Connection reset by peer]
issueat has quit [Remote host closed the connection]
eebrah_ has joined #linux-sunxi
<WarheadsSE>
mnemoc: says he didn't get that when he compiled
eebrah_ has quit [Read error: Connection reset by peer]
eebrah_ has joined #linux-sunxi
n01 has quit [Ping timeout: 265 seconds]
<hno>
did he enable it in kernel config?
arete74 has quit [Ping timeout: 245 seconds]
eebrah_ has quit [Read error: Connection reset by peer]
arete74 has joined #linux-sunxi
eebrah_ has joined #linux-sunxi
focus2 has quit [Quit: Leaving]
<WarheadsSE>
idk, he popped off our channel too
<[7]>
is anyone here aware of a update_mbr tool that is capable of generating 4KB-sized MBRs?
<[7]>
the ones that I have found (44229 bytes file size on linux) only generate 1KB ones
<[7]>
seems like we have at least 3 different MBR formats these days? 1KB, 4KB and 16KB?
arete74 has quit [Ping timeout: 245 seconds]
eebrah_ has quit [Ping timeout: 248 seconds]
eebrah_ has joined #linux-sunxi
arete74 has joined #linux-sunxi
eebrah_ has quit [Ping timeout: 252 seconds]
<[7]>
has anyone EVER been able to pack a working livesuit image for A20?
<[7]>
it seems like allwinner just didn't release all tools that are required