mnemoc changed the topic of #arm-netbook to: EOMA: Embedded Open Modular Architecture - Don't ask to ask. Just ask! - http://elinux.org/Embedded_Open_Modular_Architecture/EOMA-68 - ML arm-netbook@lists.phcomp.co.uk - Logs http://ibot.rikers.org/%23arm-netbook or http://irclog.whitequark.org/arm-netbook/ - http://rhombus-tech.net/
tinti_ has joined #arm-netbook
rz2k has quit []
aesok has quit [Remote host closed the connection]
XenGi_ is now known as XenGi
tinti_ has quit [Quit: Leaving]
dyoung-away has quit [Ping timeout: 246 seconds]
dyoung-away has joined #arm-netbook
lemonade` has joined #arm-netbook
dyoung-away is now known as dyoung
dyoung has quit [Changing host]
dyoung has joined #arm-netbook
anunnaki has joined #arm-netbook
roric_ has quit [Ping timeout: 252 seconds]
stefanro1 has joined #arm-netbook
stefanro has quit [Ping timeout: 245 seconds]
alcides has quit [Quit: > SELECT * FROM users WHERE clue > 0]
voronaam has quit [Quit: Leaving.]
cnxsoft has joined #arm-netbook
dyoung is now known as dyoung-away
popolon has quit [Quit: Quitte]
cnxsoft has quit [Ping timeout: 260 seconds]
lemonade` has quit []
cnxsoft has joined #arm-netbook
dyoung-away is now known as dyoung
dyoung is now known as dyoung-away
rsalveti has quit [Read error: Connection reset by peer]
rsalveti has joined #arm-netbook
herdingcat has joined #arm-netbook
hipboi has quit [Remote host closed the connection]
ZaEarl_ has joined #arm-netbook
ZaEarl has quit [Read error: Connection reset by peer]
aholler_ has joined #arm-netbook
aholler has quit [Ping timeout: 248 seconds]
KoH__ has joined #arm-netbook
KoH_ has quit [Ping timeout: 245 seconds]
anunnaki has quit [Ping timeout: 248 seconds]
anunnaki has joined #arm-netbook
kaspter has joined #arm-netbook
abesis has quit [Ping timeout: 276 seconds]
cnxsoft has quit [Ping timeout: 260 seconds]
abesis has joined #arm-netbook
abesis has quit [Ping timeout: 240 seconds]
eebrah has joined #arm-netbook
eebrah is now known as eebrah|away
ssvb has quit [Ping timeout: 245 seconds]
gimli has joined #arm-netbook
dfletcher has joined #arm-netbook
ssvb has joined #arm-netbook
mSquare has joined #arm-netbook
XenGi is now known as XenGi_
pcat has quit [Ping timeout: 244 seconds]
eebrah|away is now known as eebrah
rellla has joined #arm-netbook
cnxsoft has joined #arm-netbook
ZaEarl_ has quit [Ping timeout: 264 seconds]
pcat has joined #arm-netbook
rz2k has joined #arm-netbook
roric_ has joined #arm-netbook
sspiff_ has joined #arm-netbook
cnxsoft has quit [Ping timeout: 248 seconds]
sspiff has quit [Ping timeout: 245 seconds]
cnxsoft has joined #arm-netbook
rz2k has quit []
<libv> submodules are quite the nightmare.
<libv> and quite unreliable too
<mnemoc> that's the one thing cvs was great
<libv> one misstep and the whole thing goes haywire, with little hope of returning to normality or the previous situation
<libv> rm -Rf and then a submodule update. great.
<aholler_> ?
aholler_ is now known as aholler
<aholler> no problem here with submodule, just that there should exist some magic to add own patches to them ;)
<libv> ah, yes, i am sure that you can avoid the nightmare by not touching the submodule at all
<libv> as soon as you need to, or if you are trying to split an existing repo into a superproject and a submodule, you immediately run into idiocy and broken concepts
<libv> aholler: so try editing both superproject and submodule, and then get back to me with "no problem here"
sspiff has joined #arm-netbook
<aholler> libv: I'm using them since quiet some years. what's your problem with such?
<libv> git cherry-pick -n, and this translates to something resembling rm subproject
<libv> aholler: basically, all the normal tools are pretty submodule unaware
<libv> this while there is a clear split between main repo and the submodule
<libv> meaning that submodules are fully manual, and pretty fragile
sspiff_ has quit [Ping timeout: 245 seconds]
<libv> then, you better be using a public submodule, and that one better not change ever, as then your repo is f-ed
<libv> it's not tracked over remotes
<libv> but...
<aholler> sorry, I can't follow
<libv> you cannot just clone the main repo and get the submodule separately
gimli has quit [Remote host closed the connection]
<libv> err, s/separately/at the same time/
<libv> you have to manually fetch that one over some special incantations
<libv> why, if the submodule repo is directly recorded in your repo?
<aholler> because the mechanism to get it might be totally different
<libv> and what stopped people from adding submodules as extra remotes?
<libv> it's somewhere in between, which makes it crap
<aholler> that would be even worse
<libv> really?
<aholler> yes, i would want that some remote is able to overwrite my stuff
<libv> ?
<libv> ok, current situation: you have .gitmodules in your top level of your repo
<libv> it has a symbolic name, and it has a path, and then it has a url
<libv> what is the url doing there?
<libv> if the url is there, then git clone should know what to do
<libv> but it doesn't
<libv> you need: git submodule update --init
<libv> what stops us from doing git submodule remote <symbolic name> <url> ?
<libv> it's just halfarsed, and that halfarsedness is not limited to that.
<libv> git commit could also be smart enough to first commit in the submodule locally, and only then in the superproject
<mnemoc> totally agree
<libv> git push should also be smart enough, etc. etc...
<libv> it's halfarsed, across the board.
<rm> git sucks
<aholler> committing in a submodule makes only sense if it is under your control
<aholler> and normally that isn't the case
<libv> well, there are two options there
<libv> either submodules and changes are tracked locally, giving you another remote to pull from cleanly. but that kind of beats the purpose of submodules, would just make pulling/merging easier for the tools to understand
<libv> or... you are forced to have a submodule under your own control somewhere, a fork, and can pull/merge to that one separately
<libv> the second makes the most sense in the git world
<libv> what we have now, is, should i repeat myself once more?
<libv> halfarsed.
<libv> oh, a third option is to have submodules be read-only, that too would make sense
<libv> now we really have something that is in the middle of all three sensible options
<aholler> every of your options has it's own drawbacks
<libv> heck. we could combine option 2 and 3, by adding a read-only option to .gitmodules
<libv> and the top level tools should be smart enough to handle it for you, or complain loudly and refuse to do things
<libv> aholler: sure, but now we have everything and nothing.
<libv> aholler: and it's unusable and fragile.
<aholler> I can handle it just fine.
<libv> aholler: good for you and your apparently much more limited use-cases
<aholler> but I'm sure someone would appreciate work on that topic
* libv goes back to splitting a small git history over a superproject and a submodule
<aholler> libv: yeah, world isn't perfect, unfortunately
<aholler> but at least there are submodules.
<libv> aholler: also, i am supposed to be writing a mesa driver for mali-400
<libv> not redesigning git
<aholler> maybe helping with git would make more sense, than writing a driver which should come from the manufacturer. ;)
<libv> aholler: sure.
<libv> that's how the world works.
<libv> no wonder you are happy with git submodules today.
<mnemoc> *g*
<aholler> libv: there isn't something better
hramrach has quit [Remote host closed the connection]
<libv> aholler: which doesn't make it good
<aholler> you could use rcs if you want
<aholler> nobody forces you to use git(-submodules)
<libv> aholler: i do not get along with keithp much
<libv> aholler: what he did with modesetting when he joined intel in 2006 was shameful
<libv> aholler: but... he stated "now everyone shut up, we now use git for everything"
<libv> the best thing he ever did
<libv> and i was the first to move my drivers to git back in 2006.
<libv> git was really rather broken and halfarsed back then
<aholler> yes, especially the submodule support ;)
<aholler> I know those times too
* mnemoc loves git, but submodule integration is lame
<libv> git does a lot of things well, submodules are a broken concept today.
<libv> and...
<libv> the way submodules are built up today, means that this cannot change either.
<libv> there is simply no way to transition existing repos with submodules.
<libv> really really poor design
<mnemoc> git status/commit/push can be made aware of submodule without breaking compatibility
<mnemoc> because the current state is unusable
<libv> sha-ids are tracked in the repo, not in .git
<libv> ah, no, they aren't... but the url is, which is broken
<libv> but i had that before
<mnemoc> they only need to detect the right .git
<libv> hrm, maybe there is a way to be compatible
<libv> simply by ignoring the url.
<libv> in .gitmodules
<mnemoc> :)
<libv> and warning loudly
<libv> anyway, back to sunxi-mali
<libv> git diff | filterdiff :(
<mnemoc> it's for a good cause :)
<mnemoc> and the submodule won't be touched often
<SPG> Luis master ff0a264 rhombus allwinner_a10/orders/lscalder.html * http://git.hands.com/?p=rhombus.git;a=commitdiff;h=ff0a264
<libv> well, due to way it is built up, there are Makefiles headed for the submodule
<libv> normally cross-compilation shouldn't change anything down there, but still, scary
<ganbold> pcDuino 59$
<mnemoc> wasn't it going to be like $20?
rellla has quit [Ping timeout: 264 seconds]
<libv> who is behind that, and is he working with us sunxi folk?
rellla has joined #arm-netbook
ssvb has quit [Ping timeout: 252 seconds]
<libv> or is this going to be yet another halfarsed fork with part of our trees and part of our wiki copied over?
<libv> my word of the day :)
<libv> is that one known?
<libv> if not, i might pick it up
<aholler> I just browsed that this night too ;)
<libv> 1GB and vga, and smaller than the mele
<libv> which makes it a better machine for demoing
<mnemoc> soc?
<libv> a10
<mnemoc> nice
<libv> mnemoc: so the first time you've seen this then
<mnemoc> libv: yes
<libv> why does hansg not use screen.
<libv> because he has a large collection of devices
<mnemoc> if you don't need sata it seems to be better deal than a mele a2000g... specially been already available in your country
<mnemoc> hansg seems to be traveling
<libv> yeah, not having to waste 3h getting it from the customs office is a plus
<libv> which i might end up needing to do on tuesday, as my mali-400MP4 from the place with the expensive shipping and handling, the one that knows the pain of binary drivers, but was totally to sponsor lima a pair of devices, is on its way.
<libv> +unwilling
ssvb has joined #arm-netbook
* mnemoc feels very lucky to have never needed to go to the customs office. everythign just arrives directly to my home. rarely with a to-pay invoice one pays directly to the postman
<libv> wait and see how it works this time round
<libv> i had to go over for my mele
<libv> and 4y ago, i got a dvi submodule for a via board from norway.
<libv> 3h gone for just a few eur in customs
<mnemoc> :<
<libv> mnemoc: maybe the customs people in portugal are not germans
<mnemoc> libv: spain ;-)
<libv> ah, i was convinced you were in portugal :)
<mnemoc> near .pt, and having to hear a similar language, but still spain
<libv> still, a pan-european customs bureaucrat exchange might be nice, for me :p
<libv> apparently those paperpushers are still smarting from the shengen agreement, 20ys on
<mnemoc> i would have expected a more efficient import taxing charging system in .de than "take this letter and go to that office and wait a bunch of hours only to pay a few euros"
<mnemoc> that's more .it culture than .de culture
<libv> i see that differently
<libv> the germans are being overly proactive and overly correct
<mnemoc> so they don't charge you until you show the invoice
<libv> yeah, and therefor they steal 3h of your life
<mnemoc> here, if they charge you, it's a totally arbitrary amount... without even opening your package
<libv> which is quite criminal
<libv> ah, different kind of criminality
<mnemoc> in this case, your and their time costs more than the tax diff
<libv> but it keeps us all busy and keeps these people employed!
<mnemoc> *g*
<mnemoc> maybe that would help the 6M unemployed people we have (out of a 45M population)
<libv> definitely
<libv> more checking and more paperwork leads to better piecharts to show the rating agencies and the eu :)
<mnemoc> :)
<libv> ah, git mv does not deal with submodules either.
<libv> push submodule changes upstream, rm -Rf submodule, edit .gitmodules manually
<libv> submodule update again.
<libv> *sigh*
<libv> and this process of course makes it impossible to keep stuff in your submodule locally, so you better be very very certain that your changes are good for public consumption
* mnemoc misses a mele-ish device but compatible with vesa mounting, to screw behind a display
<SPG> Luis master 60f552f rhombus allwinner_a10/orders/lscalder.html * http://git.hands.com/?p=rhombus.git;a=commitdiff;h=60f552f
Ershov has joined #arm-netbook
<libv> mnemoc: that indeed would make for a compelling product
<libv> not sure whether there is a sizable market for it though, but it does make sense
<mnemoc> the same device you are going to buy could have been made with the right dimensions and fit that market
<libv> great, cherry-pick a commit which does not touch the submodule, and it complains as well
<mnemoc> o_O
<libv> yeah, full conflict.
<mnemoc> what about git-format-patch+git-am instead of cherry-pick?
<focus> installed ubuntu on to nexus 7 yesterday https://wiki.ubuntu.com/Nexus7/Installation
<libv> ignore the conflict, and commit and all is well.
<libv> focus: important, for me, question: what graphics driver is ubuntu using?
<mnemoc> i received 4.2.2 on my nexus 7 yesterday... can't see the diff :|
<mnemoc> android+
<focus> libv: how do I check - that thing is the beta - not the full tablet os they launched yesterday
<RaYmAn> mnemoc: for one thing, they added secure adb
<libv> focus: ldconfig -v | grep GLES
<mnemoc> RaYmAn: meaning not-exploitable or ssl-ed?
mSquare has quit [Ping timeout: 248 seconds]
<libv> running es2_info or something should also tell you what renderer you are using
<RaYmAn> mnemoc: meaning it requires that you need to authorize the computer on the device first
<RaYmAn> mnemoc: which requires that you unlock the device
<libv> es2_info is in mesa-utils-extra
<mnemoc> RaYmAn: uhm
<RaYmAn> I suspect it's ssl'ed as well, given it's some kind of public key auth type thing
<RaYmAn> mnemoc: the big point here is that it you can't take someone elses locked device and adb to it
mSquare has joined #arm-netbook
<RaYmAn> without knowing the unlock code
<mnemoc> ah, ok
herdingcat has quit [Remote host closed the connection]
sx172 has joined #arm-netbook
<sx172> Is there a downloadable development environment or specific instructions for Debian and/or Ubuntu related distros? The provided BSP is erroring in many instances.
<libv> oh, wow, looks like there are native tegra libs
<focus> wonder how they got those :)
<mnemoc> sx172: context please
<sx172> I try FirstStep page, and apt-get listed dependencies. Execute commands listed. Get Error 1 and or Error 2 depending on which command from the sunxi-bsp options {hwpack etc}
<libv> focus: what is in the dynamic section of objdump -x /usr/lib/nvidia-tegra/libEGL.so.1
<libv> focus: is it referencing X stuff?
<libv> like libX11.so
<mnemoc> sx172: paste (on dpaste.com or similar) the whole error
<sx172> yes, one moment
<sx172> cp: cannot stat `build/sun4i_defconfig-linux/output/lib/modules': No such file or directory then mini-x-1gb_hwpack.tar.xz] Error 1
<mnemoc> the whole output, including what you typed
<libv> focus: what about libnvwsi.so
<libv> looks mighty promising though
<libv> i am telling kusma, the guy doing tegra-re this, nexus7 and ubuntu seems like a proper gnu-linux
nicksydney has joined #arm-netbook
<mnemoc> sx172: :(
<nicksydney> hello all
<sx172> :(
<mnemoc> sx172: I need to see the things BEFORE the error
<nicksydney> based on AllWinner A10
<mnemoc> sx172: to try to find what went wrong
<sx172> From what i try to understand, it seem that the dirs are not made or used correct
<libv> nicksydney: the question is, who is behind that, and is he working with linux-sunxi.org? (i think that the latter is a "no")
popolon has joined #arm-netbook
<sx172> from u-boot things
<mnemoc> sx172: that's a consequence, not a cause
<nicksydney> libv: that is exactly i was thinking that's why i was thinking maybe someone here know about it
<libv> focus: bingo!
<libv> libX11.so.6
<sx172> These are last thing before error: INSTALL sound/soc/sun4i/spdif/sun4i_spdma.ko
<sx172> DEPMOD 3.0.62
<nicksydney> libv: i checked the website there is only image file for Ubuntu and Android no source or link to source referred too
<nicksydney> libv: but the schematics is posted in .pdf format on the website http://www.pcduino.com/?page_id=14
<mnemoc> sx172: please pass J=1 to your make call, and paste the WHOLE output
<libv> focus: kusma is mighty interested now, and will get a nexus 7 soon :)
<libv> focus: thanks a lot :)
<RaYmAn> libv: fwiw, those drivers work on any tegra3 devices with a suitable kernel (this may be obvious, but it isn't always =P)
<focus> libv: libX11.so.6 no such file - where be it?
<libv> focus: in your system already
<focus> libv: if need more, just create a script, with all the outputs dumped to text file
<libv> focus: the goal of this excercise was to find whether the GL stack was proper linux or not
<focus> libv: out of my depth at the moment but glad to help!
<libv> focus: we found that libnvwsi is linking to libX11.so.6
pcat has quit [Ping timeout: 244 seconds]
<libv> libEGL depends on libnvwsi, which depends on libX11
<libv> so a proper X enabled GLES/EGL stack
<libv> which is rather great news
<libv> the nexus7, with ubuntu installed, is then a stellar tegra3 REing target
fragmint has quit [Read error: Connection reset by peer]
<sx172> My terminal not go high enough for total
<sx172> But, I see flaw, try to go to dir that does not exist
<mnemoc> sx172: the thing is to find what the dir isn't there
<sx172> "cd /build/sun4i_defconfig-linux" is not correct because dir does not exist
<mnemoc> s/what/why/
<sx172> that
<ibot> mnemoc meant: sx172: the thing is to find why the dir isn't there
<sx172> it not there obvious not made
<mnemoc> make linux J=1 2>&1 | tee make_linux.txt
<focus> libv: do you want ls -R / of all files in nexus7?
<libv> RaYmAn: nice
<libv> focus: now, all that is needed is now known
<libv> s/now/no/
<ibot> libv meant: focus: no, all that is needed is now known
<focus> ok - glad to help :)
fragmint has joined #arm-netbook
<sx172> mnemoc, it seem after config using def instead of custom ?
<focus> libv: does it mean opengl might be working?
<sx172> variable inside not *_hwpack dir, but also defconfig dir which not made
<libv> focus: yup
<focus> One thing I found between nexus 7 and pengpod that has me worked up is when loading firefox locally and with ssh -X. ssh -X is quicker with large files. The remote rendering is making a big difference - a lot quicker.
<libv> focus: get es2_gears
<focus> libv: thanks - in that case i try to compile gambas3 for it :)
<libv> wtf, nexus10 is 500+EUR
<focus> may be future tablets have two CPUs - one for LCD and second for ssh -X to do rendering more quickly
<libv> with the EUR above 1.30usd
<libv> they do seem to be messing up the math there
<focus> nexus 7 http://www.ebuyer.com - 200 UK pounds
<mnemoc> sx172: http://paste.pm/4mh.c is the relevant part of the Makefile. something weird is happening on your system because your K_O_PATH doesn't include a reasonable $(PWD)
<libv> focus: no, that is the nexus 7
<libv> the 10 has the exynos5 with the mali t604
<focus> hmmm
<sx172> What should $PWD be normal, i can do manually
oliv3r has quit [Quit: Leaving.]
<libv> oh, 400EUR from the google play store.
<libv> but 500+ from amazon.de, that does not make sense
bsdfox_ has joined #arm-netbook
kaspter has quit [Ping timeout: 255 seconds]
<mnemoc> sx172: $PWD is the dir where the makefile is standing. but again, it's a consequence of a bigger problem. not a cause
<sx172> $ $PWD
<sx172> bash: /home/test: Is a directory
<sx172> work fine
bsdfox has quit [Ping timeout: 256 seconds]
<mnemoc> sx172: try changing the $(PWD) references in Makefile to $(shell pwd)
<mnemoc> sx172: or better, $(CURDIR)
<sx172> interest, not sudo it output cp: cannot stat `build/boot.scr': No such file or directory
<sx172> Done
<mnemoc> boot.scr is optional. that should be taken only as a warning
<sx172> How edit to include?
<sx172> Oh, nevermine, kernel opt
<mnemoc> does $(CURDIR) work ok for you?
<focus> libv: for reference sudo ls -R / of nexus7 http://www.gplsquared.com/paste/ls_nexus7.txt.tar.bz2
<sx172> yes seems to be working that way
<sx172> you have to excuse, second language english and eight years age
<mnemoc> eight years old? seriously? nice :)
<sx172> sorry first time try my fault sudo and wrong operating system i think
<mnemoc> don't use `sudo` for this things. it's risky
<sx172> yes i see now, still sudo with git program thought?
<mnemoc> don't sudo at all
<mnemoc> it's only needed when actually making the card
dyoung-away is now known as dyoung
pcat has joined #arm-netbook
<mnemoc> libv: btw, `git pull --rebase` understands submodules
<mnemoc> libv: at least in 1.8.0.3
* mnemoc wasn't expecting the recursion
<libv> ah, 1.7.7 is where i am at
<mnemoc> if ubuntu, sudo apt-add-repository ppa:git-core/ppa might be useful
rz2k has joined #arm-netbook
maethorechannen has joined #arm-netbook
<RaYmAn> libv: tbh, something like the ouya should be just as good a dev platform (though, I guess it's still not quite available for regular users?)
<libv> RaYmAn: proper gnu/linux is the key
<libv> it makes life _a_ _lot_ easier
<RaYmAn> nexus7 used linux for tegra.
<RaYmAn> it's in no way nexus7 specific
<RaYmAn> proper gnu/linux can run just fine on my TF201 as well, or an Ouya
<RaYmAn> porting it is trivial
cnxsoft has quit [Quit: cnxsoft]
<libv> RaYmAn: proper gnu/linux, with working drivers to RE
<libv> RaYmAn: this is unbelievably rare
<RaYmAn> the drivers work on all tegra3 device. (ouya is tegra3)
<libv> yes, this was news to me
<libv> but i am not REing tegra, kusma is doing that
<RaYmAn> the drivers canonical uses are just hte linux4tegra drivers downloadable from nvidia - they aren't specific to canonical or ubuntu or nexus7 :)
<RaYmAn> fair enough
<mnemoc> focus: hi, do you have your kicad files in a repo somewhere? or only downloadable .zip files?
<libv> until very recently, sunxi was the only one with working binary drivers for mali
<RaYmAn> nvidia has actually had working binary drivers for several years (tegra2 first)
<libv> and until half a year or so, all i had was android
<RaYmAn> (for gnu/linux)
<focus> mnemoc: at the moment its all in single zip file, but I move it soon to github soon
<mnemoc> focus: ok :)
<focus> .
<mnemoc> focus: any recommended resource for a not-EE (software ape like me) to learn to do a PCB for a carrier board like your?
<mnemoc> any not-absuredly-overpriced but good book?
<focus> mnemoc: lots of videos in youtube. I also leave links to two good videos. Install KiCAD, and follow the video and repeat 10 times and you be EE!! :)
<libv> so there were about 1300 cubieboards shipped already, and 100 still waiting?
<libv> what was the original goal again, 500?
<mnemoc> i think there are over 2k cubieboards out there already
<mnemoc> but don't remember the exact numbers of each batch
<libv> they are popping up on ebay.de now too
<focus> mnemoc: links in here http://www.gplsquared.com/SoM2/SoM2.html - there is also some Gambas project manager software there that talks through the KiCAD and has links to the PDF files to learn as much as you want.
<mnemoc> focus: so you don't recommend anything in dead-tree form?
<focus> mnemoc: definitely not - the two videos and repeatedly following it x10 is how I did it
<mnemoc> cool :) thanks!
<focus> mnemoc: just saying - I used Firefox extension DownloadHelper with youtube - one click to save the youtube file to disk. No need to constantly download to replay it x10 :)
tzafrir_laptop has quit [Ping timeout: 260 seconds]
abesis has joined #arm-netbook
<mnemoc> focus: tubemate in my case. I don't like to waste computer display space with videos ;-)
<focus> mnemoc: thanks - I can use that with my androids
nicksydney has quit [Read error: Connection reset by peer]
<mnemoc> focus: as it violates youtube's terms of service the app is an outcast, and the name abused by malwared variants. http://m.tubemate.net/ is the official site
maethorechannen has quit [Quit: Leaving]
maethorechannen has joined #arm-netbook
<focus> mnemoc: thanks - good to know - I have throw out so many apps - either malware or resource hogger.
cnxsoft has joined #arm-netbook
<gzamboni> mnemoc i started with gEDA, but kiCAD is much better, im finishing my extension board for cubieboard using kicad, im very happy with it
<gzamboni> i could say it has all the funciontalities needed
<gzamboni> even for high frequency you can make the same distance when routing.
<mnemoc> gzamboni: i trust kicad is fine. what I lack is knowledge :p
<gzamboni> and yes i started with some learning tutos in youtube about one week or 2 ago
<gzamboni> now im able already to make schematic components, footprints, autorouting and almost everything needed
<mnemoc> so I was wondering about a book for hobbyist but teaching you in practice to design extension or carrier boards where theory and complexity is suposed to be low
<mnemoc> nice
<gzamboni> the use of same route lenghs in routing comunication of the arm with the sdram and/or eth phy are essential
<gzamboni> well, i dont have this complication as i am doing a very simple extension board
<mnemoc> that's fortunatelly the part you don't need to care when designing extension or carrier boards ;-)
<gzamboni> but if i decide to do a "main board" its nice to know the functionality is there
<mnemoc> does it deal with same route lengths and that stuff?
<gzamboni> well after there are some electronique knologe to know that some routes cant be so close to others to avoid interferences
<mnemoc> i heared someone ranting you can't even position components in the exact coordinates you want
<gzamboni> Kicad yes
<gzamboni> well, you have a grid in the background, you have to configure the "resolution" to be able to position exactly where you want
<gzamboni> the default resolution is low
<gzamboni> so what he said is not true
popolon has quit [Quit: Quitte]
<mnemoc> great to know :)
<gzamboni> i just received 10 rgb matrix pannels, when i will finish my extension board i will start playing with it :P
<mnemoc> it was a ranting of the sort "open source can't do what my beloved pirated app does"
<gzamboni> you can even have a 3D preview in kiCAD
<mnemoc> sexy
<gzamboni> i thougth the same at the beggining, but know i see kiCad is a very good software
<gzamboni> of course the library is not that big
<gzamboni> i think in the paid softwres they have more components in their libraries
<gzamboni> but its easy to create components and footprints
<gzamboni> for instance: they dont have the 2x24 header
<gzamboni> neither the 2mm 2X24 footprint
<gzamboni> i had to create it to make the extension board for cubieboard
<mnemoc> anything like openclipart for components and footprints?
oliv3r has joined #arm-netbook
<gzamboni> dont know really openclipart, but the apps are easy to use, and you can open a similar component and just change it and save as the one you want
<gzamboni> of course, all the components and footprints you create if you dont associate a 3D model they wont appear in the 3D preview, it will appear only the solder parts in 2D at the board
<mnemoc> ok
<gzamboni> im filling my cart at farnell to know the encapsulations i will use to relate the components with its footprints to do the autorouting to send to the circuit board manufactor.
popolon has joined #arm-netbook
<gzamboni> i will make 10 units, in this extension board i will have a keypad, a FPC connector to a camera sensor, a header to a camera sensor also, an amplifier with speaker, an preamplifier with mic, 4 optocouplers with 4 relays, and headers for all gpios of cubieboard using 2.54mm
<mnemoc> gzamboni: may I /q ?
<gzamboni> what /q means ?
<mnemoc> irc command for query, aka private message
<gzamboni> sure
<mnemoc> :)
destinal has quit [Quit: Bye]
dyoung has quit [Quit: ZNC - http://znc.sourceforge.net]
abesis has quit [Ping timeout: 255 seconds]
ganbold_ has joined #arm-netbook
bfree has quit [Remote host closed the connection]
destinal has joined #arm-netbook
dyoung has joined #arm-netbook
bfree has joined #arm-netbook
merbanan has quit [Ping timeout: 272 seconds]
Ershov has quit [Quit: Ershov]
abesis has joined #arm-netbook
tzafrir_laptop has joined #arm-netbook
merbanan has joined #arm-netbook
popolon has quit [Ping timeout: 248 seconds]
<focus> gzamboni: equilength wires can be automagically made by twisting wires around each other.
<focus> in PCB - each 180 degree twist costs 2 vias however
fragmint has quit [Read error: Connection reset by peer]
popolon has joined #arm-netbook
fragmint has joined #arm-netbook
Jef91 has quit [Remote host closed the connection]
<focus> gzamboni: mnemoc: http://smisioto.no-ip.org/elettronica/kicad/kicad-en.htm - hundreds of kicad models, all of those files bundled into one ziped file http://www.gplsquared.com/SoM2/electronica_jan_2013.tar.bz2
<mnemoc> neat, thanks!
<mnemoc> focus: naive question, when using a self power usb hub, does one need to do something special to feed the device port too?
<mnemoc> pcb-design wise
abesis has quit [Ping timeout: 276 seconds]
<gzamboni> i got some from here: http://www.reniemarquet.cjb.net/bibliotecas.html
<mnemoc> kind of annoying there is no open central library
<focus> mnemoc: nothing special as far as I know - a USB port can draw 500mA from bare +5V supply without having to talk with the controller and demand more power.
<gzamboni> seems like your link is better, apperently he has all the 3D models also
<focus> 3D models and the source files in .wrl format to edit it in wings3D (available in ubuntu). I normally never try to design 3D from scratch - e.g. a 24x2 I would start with 2x2 and then duplicate the existing pins until I get what I want - much easier unless you are expert in 3D design
<gzamboni> woot, nice library he has the fpc 3d connector models
<SPG> Arie master 7cc5d57 rhombus allwinner_a10/orders/arievanwi.txt * http://git.hands.com/?p=rhombus.git;a=commitdiff;h=7cc5d57
cnxsoft has quit [Quit: cnxsoft]
zzshllehs has joined #arm-netbook
<zzshllehs> can anyone advise a good netbook for gf (os: debian, purposes: web, films, texts etc..)?
eebrah has quit [Ping timeout: 252 seconds]
<focus> The Asus brands have been good to me.
<focus> The original ones I bought had no os - now they are stuffed with windoze. Supposedly some have got their money back by returning windoze.
Kraln has quit [Ping timeout: 260 seconds]
Kraln has joined #arm-netbook
<zzshllehs> I know
<sx172> father put ubuntu on great now had few options before
<zzshllehs> I'm from Russia, unfortunately they are not available there
<zzshllehs> I have an asus k53sv but here are a few disadvantages: its weight is about 3 kg and here's a Nvidia Optimus which is too bad with Linux
<sx172> без году неделя :)
<zzshllehs> нужен нетбук с маленьким весом и максимально большим экраном (для нетбука), цена ориентировочно 17K
rellla has quit [Ping timeout: 248 seconds]
<zzshllehs> шиндовс можно законно вернуть, это не проблема
<sx172> Простите, cколько стоит?, MSI X-slim xорошо ?
rellla has joined #arm-netbook
<zzshllehs> что насчет хромбука?
<mnemoc> english please
<sx172> I do not have answer zz
<sx172> I see good result battery and video
<sx172> linux page relating, but english only
<mnemoc> your english is good enough to stick with it and don't exclude all non-russians from the conversation ;-)
<sx172> i thank you, some wordds i have to google change, i think works well
<sx172> mnemoc i mail father he say i did wrong linux version, 3.8 out i did 3.0
<zzshllehs> you're talking about samsung chromebook?
<mnemoc> sx172: 3.8 is out, and has core sunxi support, but it can't be used to run a full system yet. only 3.0 and 3.4 for now.
<sx172> no i make, how you say in english
wingrime has joined #arm-netbook
<sx172> zz, "модель" mini-x-1gb
hg_5 has joined #arm-netbook
<sx172> 3.4 is new version for mini-x-1gb ok i see. i think i see in git i can change checkout
<mnemoc> git co -b sunxi-3.4 origin/sunxi-3.4
<mnemoc> s/co/checkout/
<ibot> mnemoc meant: git checkout -b sunxi-3.4 origin/sunxi-3.4
lemonade` has joined #arm-netbook
<mnemoc> .oO(mental note, not everyone uses the same git aliases)o
mSquare has quit [Ping timeout: 248 seconds]
<sx172> ok yes i see download-ing only new parts this is nice
Brandon15811 has quit [Ping timeout: 264 seconds]
cnxsoft has joined #arm-netbook
<sx172> i see stage/ and origin/, what is not the same
Brandon15811 has joined #arm-netbook
<mnemoc> origin is a remote, sunxi-3.[04] and stage/sunxi-3.[04] are branches
<mnemoc> branches starting with a remote name are remote branches, others are local branches
<mnemoc> sx172: http://git-scm.com/book
dyoung is now known as dyoung-away
<sx172> ok i see i put on google change Thank you very much.
Gumboot has quit [Read error: Connection reset by peer]
Gumboot has joined #arm-netbook
rsalveti has quit [Ping timeout: 248 seconds]
wingrime has quit [Read error: No route to host]
ganbold_ has quit [Remote host closed the connection]
niftylettuce has quit [Ping timeout: 256 seconds]
<Turl> sounds like we got a regression :/
<mnemoc> uhm
<mnemoc> why the f* olimex and miniand never reports anything, just rant around
<mnemoc> Turl: do you know if the slugishness happens happens in 3.4 or also in 3.0?
<Turl> no idea really, I just saw that G+ post
<Turl> I haven't used GPIOs other than to turn the cb's green led
cnxsoft has left #arm-netbook [#arm-netbook]
<mnemoc> 3.4 has some pio handling changes, also available in stage/sunxi-3.0. but not (yet) in sunxi-3.0
pcat has quit [Read error: Connection reset by peer]
rsalveti has joined #arm-netbook
drachensun has quit [Quit: Leaving]
sx172 has left #arm-netbook [#arm-netbook]
rsalveti has quit [Ping timeout: 272 seconds]
ZaEarl_ has joined #arm-netbook
Kraln has quit [Ping timeout: 248 seconds]
rz2k has quit []
Kraln has joined #arm-netbook
Kraln has quit [Read error: Connection reset by peer]
Kraln has joined #arm-netbook
rsalveti has joined #arm-netbook
drachensun has joined #arm-netbook
dyoung-away is now known as dyoung
pcat has joined #arm-netbook
dyoung is now known as dyoung-away
Kraln- has joined #arm-netbook
niftylettuce has joined #arm-netbook
Kraln has quit [Ping timeout: 260 seconds]
FergusL has left #arm-netbook ["Quit"]
popolon has quit [Quit: Quitte]
rellla has quit [Remote host closed the connection]
maethorechannen has quit [Remote host closed the connection]
Gumboot has quit [Quit: I'm just going offline and may be some time.]
Yaku has joined #arm-netbook
eebrah has joined #arm-netbook
Gumboot has joined #arm-netbook
zzshllehs has quit [Quit: Leaving]
eebrah has quit [Ping timeout: 264 seconds]
ZaEarl_ is now known as ZaEarl
<ZaEarl> I have confirmation that Ubuntu Tablet is running on a Jelly Bean kernel. The source will be available with the dev builds tomorrow.
<mnemoc> what is "a Jelly Bean kernel"? 3.0?
<ZaEarl> Android 4.2.1
<ZaEarl> oh, the kernel version
<ZaEarl> yeah, 3.0.x
<mnemoc> android 4.0 runs the very same kernel :p
anunnaki has quit [Ping timeout: 248 seconds]
merbanan has quit [Ping timeout: 252 seconds]
<Turl> the kernel version is mostly irrelevant irrelevant
eFfeM has joined #arm-netbook
eFfeM has left #arm-netbook [#arm-netbook]
focus_it has joined #arm-netbook
Kraln- has quit [Quit: http://quassel-irc.org - Chat comfortably. Anywhere.]
Kraln has joined #arm-netbook
anunnaki has joined #arm-netbook
sky770 has joined #arm-netbook
sky770 has quit [Client Quit]
sky770 has joined #arm-netbook
sky770 has quit [Client Quit]
aesok has joined #arm-netbook
dyoung-away is now known as dyoung
Yaku has quit []
lerc has quit [Ping timeout: 245 seconds]
eebrah has joined #arm-netbook
user has joined #arm-netbook
<user> Hi, I'm trying to port replicant to the All Winner A13B, and I was wondering whether this kernel might work: https://github.com/linux-sunxi/linux-sunxi ?
<user> Otherwise, does anyone know where I can find the device specific kernel?
<Turl> user: A13B? first time I hear of it
<user> whoops, wrong link
<user> sun5i
<Turl> ah, so it's an A13-powered tablet
<Turl> chinese are worse than samsung when naming products >.<
<user> lol
<Turl> user: yes, A13 should work, it's sun5i mach
<user> good stuff.
<Turl> what version of android is replicant working on these days?
<user> I'll ask ...
<user> I don't know myself.
<Gumboot> Speaking of bad naming; I was looking at those Ainol tablets a while back trying to figure out what the hell "manycore" meant. I ended up asking our FAEs at Allwinner.
<Gumboot> ("our" being my ex "our", not current "our")
<Turl> all the things are 'manycore' these days
<Turl> as long as it has a DSP or BP or GPU it qualifies as manycore :p
<Gumboot> Well, this was A10. Cortex-A8. I asked around the office and they said that was, by any practical definition, impossible.
<Gumboot> But yes, it turned out to be that.
<Gumboot> What's a "BP"?
<Turl> marketing speech at its finest
<Turl> Baseband Processor
<Gumboot> Oh yes. I'd stopped hearing about those.
<Gumboot> I got kind of fuzzy about where the boundary was.
<user> Replicant is apparently a fork of CM 9, so ICS.
<Turl> user: might come in handy for you then, https://github.com/allwinner-dev-team/android_device_allwinner_common
<user> ok, thanks :)
<Turl> master branch is ics iirc
<Gumboot> Is there any Android-derivative which sandboxes apps more thoroughly and runs inside of Linux alongside proper Linux apps?
<Turl> Gumboot: what do you mean with 'more thoroughly'?
<jinzo> You're stretching it a bit :D There are Dalvik "players" for Linux/Windows
<Turl> Gumboot: android apps are already heavily sandboxed
<jinzo> but security vise, I don't think anyone adds anything special
<jinzo> then, on this end of the stick you have some interesting patchsets/programs for more in-depth control of permission (OpenPDroid)
<Gumboot> At installation Android says that apps need all sorts of things I don't want them to have. My only option is to not install them at all. So I was wondering if maybe there was something which would run them but deny them that access.
<Turl> pdroid is that iirc
<Turl> CM7 can deny permissions too, but apps don't like to hit hard walls
<Gumboot> And the other thing is, I'd much rather run Linux (proper), but I need access to some Android apps, and I don't want them having rights to make a mess.
<Turl> if an app requests unreasonable permissions, it's usually sign of trouble though
<Gumboot> I don't think the walls need to be hard. They can be elastic.
<Turl> Gumboot: you can run proper linux in a chroot
<Gumboot> No, i don't trust Android as the host.
<Turl> Gumboot: the host is the linux kernel
<Gumboot> I don't trust the Android userland as the host.
<Turl> "ubuntu for android" aims at that
<Turl> Gumboot: no userland is host of anything
<Gumboot> Who decides what goes on the display?
<Turl> whoever writes on the framebuffer device
<Gumboot> Who decides who's allowed to write on the framebuffer device?
<Turl> the kernel
<Gumboot> Who configures the kernel to make these allowances?
<Turl> you
<Gumboot> That's not been my experience with Andriod.
<Gumboot> I used to run Android with this "I don't want anything which makes stupid demands for privileges" attitude, until I realised I wasn't installing _anything_ and that Android was useless to me.
<Turl> there's two additions to that attitude you need to do
<Gumboot> What it boils down to is that I think the whole Android security model is utterly stupid. If Android is the host of my Linux environment then I'm still giving a stupid system top-level rights to make stupid decisions.
<Turl> if it doesn't request internet permission, the rest don't matter much
<Turl> and if it requires root, permissions are irrelevant
<Gumboot> I didn't even bother rooting for months.
<Turl> so you trusted the phone manufacturer
<Gumboot> Then I made a last-ditch effort at turning my device into something useful.
<Gumboot> tablet.
<Gumboot> I don't think I've seen an app that didn't want internet permission.
<Gumboot> And I didn't have an awful lot of faith in the original firmware, either.
<Gumboot> pdroid sounds like half of what I want.
XenGi_ is now known as XenGi
<jinzo> Gumboot, there's also some apps on the android play store (one chinese one is famous, but can't remember the name) that do the same as OpenPDroid
<jinzo> also the MIUI rom has it baked in (the previously mentioned app)
<jinzo> and the stuff at least looks great (dunno how it works)
<jinzo> but no sources afaik.
<Turl> MIUI :/
<Turl> no source of it at all
<Turl> they don't even comply with GPL
<jinzo> MIUI from the "naive" user perspective is great. Polished, good set of apps, themes, quite big community
<Turl> and it's made in china, I don't personally trust closed chinese software
<jinzo> yeah I know, but I don't trust the closed american software either
<jinzo> probably even more than the chinese.
<Gumboot> Oh, and I'd also like to reduce Android apps to being just a window.
<Gumboot> So I can run a proper window manager.
<jinzo> there's some samsung stuff with that, also some open afaik. Nothing major tho.
<jinzo> or at least that I would know
<Turl> jinzo: the thing is, when an american company does bad things and it gets discovered, there's big trouble
<Turl> not so much when it's chinese
<jinzo> WAT?
<Turl> remember Carrier IQ?
<jinzo> you do know the USA is intercepting ALL of the internet traffic that goes thorought it?
<jinzo> even if it's mine for example?
<Turl> we were talking about software companies, what does the USA have to do? :P
<Gumboot> "intercepting"?
<jinzo> USA is only a bit on it's toes when it goes for internal stuff regarding americans
<jinzo> they're quite content with spying on foreigners
<jinzo> Gumboot, yes, they claim that everything that goes thorough their internets is theirs to inspect, EU is disagreeing but nothing major will come
<Gumboot> What if smoeone has citizenship but might still be a terrorist? Surely you have to spy on _everyone_ to be sure.
<jinzo> there are also no laws that would forbid the CIA for example to intercept souch traffic.
<jinzo> Gumboot, yes, but it's a lot easier if you spy on your own people under the umbrella of spying for foreigners
<Gumboot> IP traffic is open for inspection. What of it?
<jinzo> at least, that's what I would doo.
<Gumboot> Once it's out on the internet you have to assume it's being looked at.
<jinzo> should not be legal. You can assume many things
<user> indeed, and we'll never know whos spying on us if things remain closed sourced
<jinzo> but I don't think it should be legal for CIA for example, to look at my data
<jinzo> even if it's cats.
<Gumboot> This is going to descend into a gun control debate. I'm out.
<Gumboot> "if you outlaw IP packet inspection, only outlaws ... "
<Gumboot> etc.
<jinzo> gun control? we don't have souch problems in developed countrys :D
<Gumboot> I'm from New Zealand. In the charts I see it has about the third-highest gun ownership rate in the countries people bother to survey.
<Gumboot> Only, in NZ they have this thing where people JUST DON'T SHOOT EACH OTHER.
<jinzo> not only in NZ
<jinzo> quite some other countries too :D
<jinzo> and ofcourse that's good
<Gumboot> I think in Switzerland they mostly shoot themselves.
<jinzo> but on the other hand, you're next on the list of being the terrorists because of Kim :P
<user> anyways, thanks for the help, I'm off.
<Gumboot> dotcom or jung un
<Gumboot> jong un?
<Turl> dotcom I'm betting
user has quit [Quit: Leaving]
<Gumboot> Two different axes of evil, I guess. Does evil have multiple axes? I bet it does.
tinti has quit [Remote host closed the connection]
bfree_ has joined #arm-netbook
bfree has quit [Ping timeout: 252 seconds]
tinti has joined #arm-netbook
tzafrir_laptop has quit [Ping timeout: 276 seconds]
eebrah has quit [Quit: ChatZilla 0.9.90 [Firefox 18.0.2/20130201065344]]
anunnaki has quit [Read error: Connection reset by peer]
anunnaki has joined #arm-netbook
sspiff has quit [Ping timeout: 245 seconds]
pwhalen has quit [Ping timeout: 248 seconds]
pwhalen has joined #arm-netbook
hg_5 has quit [Read error: Connection reset by peer]
FergusL has joined #arm-netbook
dyoung is now known as dyoung-away
focus_it has quit [Quit: Leaving]
tzafrir_laptop has joined #arm-netbook
lejonet has quit [Ping timeout: 260 seconds]
lejonet has joined #arm-netbook
lejonet has joined #arm-netbook
lejonet has quit [Changing host]
eebrah has joined #arm-netbook
aesok has quit [Remote host closed the connection]
theRat__ has joined #arm-netbook