<mirko>
fortunately i don't know any server-/desktop linux distribution which is having the ash as default shell
<wpwrak>
is registering project pervertix, a new linux server and desktop distro
<bartbes>
wpwrak: *another* one?
<mirko>
with ash as default shell? ;)
<bartbes>
and without package system
<wpwrak>
mirko: ash as shell, dropbear for ssh, busybox for all the major utilities, gcc-1.4, ka9q for internet access, chimera for the web, haven't decided what to use for the windowing system
<bartbes>
and without make, gcc, etc installed
<wpwrak>
bartbes: ubuntu already cover the "no dev whatsoever by default" market :)
<bartbes>
yeah, but they have a package manager
<bartbes>
if you strip that as well
<bartbes>
and then stop using ELF
<wpwrak>
that's a good plan
<bartbes>
you get an amazingly useful distro!
<wpwrak>
yup. a.out is more than enough.
<wpwrak>
in fact, i could write all major scripts in, say, tcsh
<alcy>
me warns for noob alert
<alcy>
warns for noob alert. wow
<alcy>
ok so how do i get to the terminal from the graphical env ?
<tuxbrain2>
alcy: alt+f2
<alcy>
not working ?
<bartbes>
alt+f5
<bartbes>
oh
<bartbes>
wait
<bartbes>
from
<bartbes>
ctrl+alt+f1-f4
<mirko>
wpwrak: and openwrt covers the embedded linux market (also without dev foo in the target)
<bartbes>
(last part being f1 to f4 all work for diff terminals)
<mirko>
(why's everybody starting making his own linux distribution?)
<wpwrak>
mirko: ah yes, right. hmm, a gentoo for small embedded would also be interesting. write everything in C++, for speedier compilation :)
<mirko>
wpwrak: ehrm... c++ is afaik the very worst regarding compilation time
<wpwrak>
mirko: mission accomplished ;-)
<tuxbrain2>
mirko: because they can :)
<tuxbrain2>
alcy: sorry I forguet the ctrl :P
<alcy>
nm, had figured it out anyway, just that spreading the fingers was a lil awkward
<bartbes>
a resurrection!
<tuxbrain2>
hehehe a pseudo briked NN turning to live again?
<bartbes>
I meant qi-bot
<wolfspraul>
bartbes: nah I'm just still not that good with eggdrop, so I restart more often than needed :-)
<wolfspraul>
but I can reproduce mirko's commitlog flooding problem now... good
<tuxbrain2>
I think I have read some report of a succesfull native kernel compilation
<tuxbrain2>
ofcourse it take a bunch of hours :)
<zear>
i once compiled freeciv on jornada 720 (206MHz arm, 32RAM + swap)
<zear>
took ~20min to compile
<SiENcE>
thats acceptable
<SiENcE>
so dingux apps are working on nanonote?
<kristianpaul>
linux? :p
<SiENcE>
zear, is it worth to buy a nanonote to compile dingux apps?
<kristianpaul>
no more?
<zear>
SiENcE, i don't think so, at least you'd have to use the same toolchain, or link them statically
<zear>
but nanonote is fun in general
<kristianpaul>
:d
<zear>
SiENcE, and jlime broke the compatibility with (at least some) dingux binaries
<SiENcE>
jlime ?
<SiENcE>
a mouse dektop for nanonote 8-) ?
<tuxbrain2>
SiENcE: an OpenEmbedded based distribution that also has a version for nn
<tuxbrain2>
kbd controled so don't worry but it runs X :)
<kristianpaul>
and runs vey well btw !!!
<SiENcE>
ok
<tuxbrain2>
indeed
<zear>
SiENcE, jlime, originally a distro for the Jornada handheld pcs (hence the Jornada Linux Mobility Edition name), though recently most of the crew (rafa, kristoffer, me and soon blizzard) got their nanonotes, so we've ported it to the nanonote :)
<alcy>
folks, have run into a bad situation, it seems. it was all fine till I decided to put debian on the nano-note. I am not sure but could using the u-boot provided at the pyneo site (avt2-u-boot) have caused the device to not function ?
<alcy>
software usb boot doesn't work. and the hardware method hasn't been succesful either. I couldn't have possibly bricked the device, could I ?
<kristianpaul>
yes sure you can alcy
<kristianpaul>
you canboot using an alternative method (not U key + power at startup)
<kristianpaul>
You only need boot the ben that way (remenber use the small carbon rubber included in box)
<kristianpaul>
and you can run the update script as always the recover the firmware including uboot and openwrt
<samgee>
I've been wearing out a piece of tin foil for the past 15 minutes, but lsusb shows nothing
<alcy>
kristianpaul: thing is, it isn't powering up ! took out the battery, conencted the usb cable, pressed the power button, no go. I have flashed it a few times with different "official" images, but its the first time I was trying out the debian ones
<alcy>
and that's when this happened
<samgee>
alcy, does yours also tick when you leave out the battery and plug in the usb cable?
<alcy>
samgee, well the tv was on n kids were playing around, i dd think i was hearing something, but not sure
<kristianpaul>
alcy: it is powering up even the screen dint look like it is
<kristianpaul>
alcy: you just may be damaged the uboot
<kristianpaul>
so the Ben cant boot, but dont mean is bricked
<kristianpaul>
just you need boot it using hardware method
<kristianpaul>
and on the computer side as you see the right usb id with lsusb
<kristianpaul>
you can reglash
<kristianpaul>
reflash*
<alcy>
kristianpaul: I know the screen won't show up, but the problem is lsusb doesn't show anything about any new device
<kristianpaul>
it must
<samgee>
same here
<kristianpaul>
how are you shorting the usb boot hardware pins?
<alcy>
kristianpaul: pressing the ruber thingy on the usb pins.
<kristianpaul>
what steps you did before?
<samgee>
wooh! found my rubber thingy again
<samgee>
tries again
<kristianpaul>
let  a watch lsusb running on the computer so can be sure when it appears
<alcy>
kristianpaul: took the battery out, disconencted usb cable, pressed the rubber thing (not sure which side of it though !, help ?) on ths usb boot pins
<samgee>
still no go :(
<kristianpaul>
'watch lsusb' <-- two coommands
<alcy>
kristianpaul: I know, abiut that.
<kristianpaul>
ok
<alcy>
kristianpaul: my real source of apprehenesion is whether I am shorting the usb pins correctly
<kristianpaul>
press the rubeber thing then power on
<kristianpaul>
rubber*
<kristianpaul>
wich side?
<kristianpaul>
just int the4 middle of the two tiny pins
<kristianpaul>
s/int/in
<alcy>
kristianpaul: I will finally ask this, embarassed as much as I might be. which side of the rubebr thing ? the protruded side or the black one ? :(
<kristianpaul>
black one
<kristianpaul>
that touch the two pinds
<alcy>
ok so i was doing that right at least
<kristianpaul>
:)
<kristianpaul>
i remener i did this
<kristianpaul>
unplug usb
<kristianpaul>
push rubber thing
<kristianpaul>
plug usb
<kristianpaul>
then power on
<alcy>
kristianpaul, ok...will try it
<alcy>
kristianpaul: thanks for the help :)
<kristianpaul>
bug me again ig noething happen
<samgee>
holy crap! It finally showed up for some reason.
<kristianpaul>
is really toooo weird usb hardware dint work
<wpwrak>
kristianpaul: now, if you had an idbg, you could: idbg-nnboot usb; idbg-reset; sip a cool beer while you wait for it to boot :)
<samgee>
kristianpaul, thanks a bunch!
<kristianpaul>
samgee: holly good news :D
<samgee>
hugs the channel
<kristianpaul>
wpwrak: i dont but yeah that tiny thing really help
<kristianpaul>
want a beer is friday :)
<kristianpaul>
holy*
<alcy>
samgee , what did you do ?! :D
<samgee>
alcy, I did exactly as kristianpaul described ... again :)
<samgee>
alcy, I switched usb ports a few times, maybe that matters too
<kristianpaul>
just make sure rubber is free of dust also the pins, fingers usually have grase so that may affect
<kristianpaul>
(just thinking waht else can be wrong)
<alcy>
kristianpaul, samgee: the power led blinks once, that's all I notice. it happens when I am plugging in the usb cable.
<kristianpaul>
yeah normal
<alcy>
kristianpaul: but lsusb shows no activity
<alcy>
holy cow it showed :D
<samgee>
Error - can't read bulk data from Ingenic device:-110
<kristianpaul>
:D
<samgee>
:/
<samgee>
alcy, awesome
<kristianpaul>
samgee: what usb cable are you using ?
<samgee>
the one that came with the nanonote
<kristianpaul>
okay
<kristianpaul>
is still flashing or whaterver is doinf aftter that message?
<samgee>
script says ABORTED
<kristianpaul>
if not write to thet list i dont know about that message
<kristianpaul>
ahh
<kristianpaul>
ther is a log
<samgee>
so it seems
<samgee>
dpaste.com/228604
<samgee>
doesn't seem very informative
<kristianpaul>
las version of xburst-tools?
<kristianpaul>
last*
<samgee>
0.0+201004-0.1
<kristianpaul>
same here
<kristianpaul>
try again if not bug to the list and fill a bug i think
<kristianpaul>
if not works
<samgee>
trying another usb port again
<samgee>
doesn't work, I'll post to the list
<alcy>
samgee, have you tried the other method(s) rather than the script ?
<samgee>
no
<alcy>
Earlier when I was trying different kernel images with openwrt, I used only the second method using usbboot, successful every time
<alcy>
so maybe you could try it too, samgee
<samgee>
ok, thanks
<kristianpaul>
samgee: but you got uboot at least?
<kristianpaul>
i gues
<kristianpaul>
thats is flashed before kernel
<samgee>
kristianpaul, reflash.sh says "flashing bootloader...", then "flashing kernel..." and b0rks out
<kristianpaul>
hmm no
<kristianpaul>
the error was in uboot flash
<kristianpaul>
step
<samgee>
I see "flashing kernel..." in the terminal when it's running, when it fails it overwrites that with "fatal error occurred"
<samgee>
I guess that's a bug in the reflash script
<kristianpaul>
hmm
<kristianpaul>
try erase what files in .qi/nanonote/ben/latest
<kristianpaul>
and try again
<kristianpaul>
just to be sure is not a xburst tools bug :/
<samgee>
I'm trying "alternative to using the reflash_ben.sh script" now
<kristianpaul>
ok
<samgee>
wtf? I type "nprog 2048 openblabla.ubi 0 0 -n" and it says "not enough argument."
<kristianpaul>
btw is that the last script?
<kristianpaul>
just exit
<kristianpaul>
and try again
<kristianpaul>
that happens to me i can  cofirm
<samgee>
yes, script freshly downloaded from wiki
<kristianpaul>
using custom openwrt image
<samgee>
ah, it's writing nand now
<kristianpaul>
:)
<kristianpaul>
yeah that happens i dont know why (and forgot repor bug)
<samgee>
so I had regular power on, usb boot, reflash script and manual method failing on me. Good thing I love this little gadget and the underlying philosophy. :)
<kristianpaul>
ahh same error?
<samgee>
no, it's still writing
<samgee>
it's just that every step failed at the first try :)
<kristianpaul>
:p
<kristianpaul>
alcy: how are you doing?
<kristianpaul>
going*
<alcy>
kristianpaul: debian's up and running, thanks a bunch :D
<kristianpaul>
alcy: :)
<bartbes>
lol, launchpad is busy today
<kristianpaul>
ah?
<bartbes>
I submitted a package for my ppa 6 hours ago
<bartbes>
and it's still waiting to build
<bartbes>
(i386 built after 4 hours, I think, amd64 is still pending)
<samgee>
watches nanonote boot and sheds a tear of joy :')
<samgee>
kristianpaul, thanks again for your help
<kristianpaul>
i like help :)
<bartbes>
muahahahahaha >:)
<bartbes>
that deservers an explanation, doesn't it?
<bartbes>
well apparently it doesn't
<tuxbrain2>
samgee: congrats :)
<samgee>
tuxbrain2, oh, you're here. I was just about to reply to you. :)
<samgee>
tuxbrain2, thanks for you help too, btw.
<bartbes>
is there a way to stop the framebuffer?
<bartbes>
see, a fb app just crashed
<bartbes>
I managed to kill it, but I still can't return to the terminal
<kristianpaul>
control + F3
<kristianpaul>
[A
<kristianpaul>
control + F4 ?
<kristianpaul>
but i think if you killed in a not good way if dont allow to swich
<zear>
bartbes, well, in the dingoo you have to reboot it
<zear>
if this happens
<bartbes>
yeah, that's what I did in the end
<bartbes>
but yeah..
<bartbes>
not exactly what I wanted
<zear>
we didn't find a better solution there yet
<bartbes>
what would happen if you'd start another fb app?
<zear>
sdl wouldn't be able to load because of no free fb
<zear>
as for non-sdl games - no idea
<bartbes>
that sucks
<tuxbrain2>
bartbes: thats why some one has invented X :P
<bartbes>
I guess that's true..
<bartbes>
what do you guys know about timer accuracy?
<tuxbrain2>
I think rafa knows about SDL but don't know if he knows about you don't know
<unclouded>
bartbes: what do you want to know about SDL_GetTicks?  is it something simple?
<bartbes>
does it have its normal accuracy on the nn?
<bartbes>
because an algo I use keeps saying no time has passed
<bartbes>
so either that isn't as high
<bartbes>
or there's some rounding error
<unclouded>
SDL_GetTicks returns the number of millisecond since the SDL app started: http://www.libsdl.org/cgi/docwiki.cgi/SDL_GetTicks so if you invoke it more than once each millisecond it will return the same value as it did last time but when a new millisecond begins it will return a new value
<bartbes>
yes, that makes sense
<bartbes>
but even after 10 secs the total diff time is still 0
<bartbes>
I'll just go and add some debug prints
<unclouded>
what's the algo for that you're using?
<bartbes>
and it should have at least a 1 ms diff, first of all because even full-powered computers have
<bartbes>
but also because there's a 1 ms delay (and SDL_Delay isn't even that accurate, so in practice it will be longer)
<bartbes>
but I guess in theory it could be a rounding error somewhere
<bartbes>
(probably when entering lua)
<bartbes>
boots up dev computer
<unclouded>
I can't see anything obviously wrong with that.  dt will be 0.0f sometimes if step() is invoked more than 1000 times a second but I would hope that "fps" would be updated regularly
<bartbes>
oh, it's really dt I'm looking for though
<bartbes>
but yeah, sometimes at most
<bartbes>
I get a consistent 0
<bartbes>
which leads me to believe it's rounding somewhere
<unclouded>
an IEEE 32-bit float can store much smaller numbers than 0.001 though
<bartbes>
unclouded: I checked, and even getTime() returns 0
<unclouded>
bartbes: is this on your workstation or your NN or both?
<bartbes>
only on the nn
<unclouded>
since time_init is never assigned then either SDL_GetTick() is returning 0 or that division results in 0.0f
<bartbes>
well, it is assigned 0
<bartbes>
but I see what you mean
<bartbes>
too bad I can't simply add a cout there
<bartbes>
oh wait
<bartbes>
nvm
<bartbes>
I know why that part failed
<bartbes>
adds a debug print
<bartbes>
right
<bartbes>
SDL_GetTicks works
<unclouded>
SDL_GetTicks() is definitely working on the NN for me
<bartbes>
now let's test the division
<unclouded>
I have an FPS limiter with similar semantics to the one in pygame that limits the FPS to ( say) 30fps and sleeps when the main loop is running faster than 30 Hz
<bartbes>
division is working fine
<bartbes>
then lnum must be screwing me over
<bartbes>
adds a debug print right after insertion into lua
<bartbes>
6.36599e-314
<bartbes>
that.. is weird
<bartbes>
6.36599e-3140.04
<bartbes>
first number being the one in lua, second being the number before insertion
<tuxbrain2>
By default Mobile Atlas Creator uses up to 512 of RAM....
<kristianpaul>
hmm
<kristianpaul>
i dont have all that ram
<kristianpaul>
avalible
<kristianpaul>
well lets forget it ;)
<wpwrak>
tuxbrain2: step 1: make friends at the math department of the local university. step 2: present them with the problem. step 3: get them to assign a bunch of students. step 4: see if they found a space-efficient algorithm. step 5: if not, back to step 3. step 6: celebrate.
<kristianpaul>
heh
<wolfspraul>
wejp: we are making progress on the IRC privacy issue
<wolfspraul>
first I included /irclogs in the robots.txt Disallow section, but then I saw in the irclogs that quite a few people thought that's a stupid idea
<wolfspraul>
so maybe we remove that again, don't know
<wolfspraul>
then I changed the topic of this channel to clearly state that it is being logged and where it's logged
<wolfspraul>
and I also have an idea for the last part, that some people prefer to not have their messages logged
<wolfspraul>
oh, another thing - the qi-bot is now the official owner of this channel and flagged as an operator, so it's clear we have a big brother watching over us :-)
<wolfspraul>
so my idea is that since it is qi-bot anyway who is doing the logging, and qi-bot is supposed to help us (!), qi-bot should have a list of nicks that are excluded from logging
<wolfspraul>
eggdrop is fairly extensible with tcl scripts, and there are many scripts available. maybe something like this even exists already, I need to look into it.
<wolfspraul>
so then qi-bot would maintain this nick list, and you could send commands to the bot to see the list (to check that you are in), or to add and remove nicks.
<wolfspraul>
the bot would then not log any messages from any nick in that list
<wolfspraul>
so basically it's an 'opt-out' model
<wolfspraul>
we don't have this yet, but it's a proposal. if people like it I need to look into it.
<wolfspraul>
just fyi, things are moving...
<roh>
hey wolfspraul
<wolfspraul>
roh: hey
<roh>
how are things?
<wolfspraul>
good, busy, happy
<roh>
nice to hear
<wolfspraul>
yeah, there is lots of good hacking going on everywhere...
<wolfspraul>
you probably know. werner is doing awesome stuff around case scans, and also with his new usrp toy wrt 802.15.4
<roh>
heh
<wolfspraul>
I am working on some gps stuff using a sige 4162 front-end rf ic
<roh>
i still cannot find anybody saying good stuff about 802.15
<roh>
much too complex.
<wolfspraul>
milkymist is moving forward, the xue camera as well
<roh>
thats nice
<wolfspraul>
sie (former sakc) small run
<wolfspraul>
yes, milkymist one is now fully verified!
<wolfspraul>
since yesterday
<wolfspraul>
I am meeting with the guys working on the jtag/serial cable and (other guys) working on the case design in a few hours
<roh>
cool
<wolfspraul>
yeah
<wolfspraul>
we are warming up for another small run to have 20 of them for 27c3 :-)
<wolfspraul>
ideally including reflashing cable, end even including case if we can make it. we work on it...
<wolfspraul>
ben nanonote software also doing very well I think. really some great contributions now, really good stuff.
<wolfspraul>
then lars big success in getting a large part of the Ben NanoNote kernel included in 2.6.36
<wolfspraul>
and of course - the stealth weapon - you have some hoperf modules and will surely emerge soon with the greatest hack ever! :-)
<wolfspraul>
no?
<wolfspraul>
I think it's all cool, bits and pieces... still a way to the free smartphone though
<roh>
wolfspraul no time atm :(
<wolfspraul>
he :-) that's ok - see the new topic of this channel "time is on our side"
<wolfspraul>
if not today, then tomorrow
<roh>
or the other way around: too less paid time, so not much free time to spend
<wolfspraul>
adopted some lessons from Werner
<wolfspraul>
ah I have zero paid time, but I hide in China
<roh>
hrhr
<wolfspraul>
yep, I'm not the only one
<roh>
i would love to hack on such stuff for money, but currently its extremely hard even finding enough work with other stuff
<wolfspraul>
there is a whole colony of westerners that hide here so they can work a few years on whatever project with minimal cash needs
<roh>
uh. some way to get away from taxes also?
<roh>
or rather insurance costs
<wolfspraul>
not really. i still pay my german health insurance and by now it's pretty much the biggest expense I have each month.
<roh>
same here. living in berlin is not that expensive. my flat is cheaper
<wolfspraul>
yes true, Berlin is similar to China in that respect, ha :-)
<wolfspraul>
keep your spirits high, I am 100% certain we will get this off the ground...
<wolfspraul>
I chatted with Harald a bit recently, he is very happy in Taipei :-)
<roh>
and if the letters are all true i will see if the insurance isnt even more expensive than the rest of everything combined. will meet a lawyer about that on monday :(