Turl changed the topic of #linux-sunxi to: Allwinner/sunxi development discussion - Don't ask to ask. Just ask and wait! - See http://linux-sunxi.org | https://github.com/linux-sunxi/ | Logs at http://irclog.whitequark.org/linux-sunxi | #blameOliver
Tartarus has quit [Read error: Connection reset by peer]
Tartarus has joined #linux-sunxi
montjoie[home] has quit [Ping timeout: 260 seconds]
montjoie[home] has joined #linux-sunxi
naobsd has quit [Quit: Page closed]
mcbrick has quit [Remote host closed the connection]
wingrime has quit [Ping timeout: 272 seconds]
\\Mr_C\\ has quit []
ykchavan has joined #linux-sunxi
tinti has quit [Quit: Leaving]
popolon has quit [Quit: Quitte]
egbert has quit [Disconnected by services]
egbert has joined #linux-sunxi
ykchavan has quit [Quit: Leaving]
hipboi has joined #linux-sunxi
[7] has quit [Disconnected by services]
TheSeven has joined #linux-sunxi
servili007 has quit [Quit: Leaving]
kaspter has joined #linux-sunxi
ykchavan has joined #linux-sunxi
Gerwin_J has quit [Quit: Gerwin_J]
rz2k has joined #linux-sunxi
ZetaNeta has quit [Ping timeout: 260 seconds]
JohnDoe_71Rus has joined #linux-sunxi
eebrah has joined #linux-sunxi
eebrah is now known as Guest63107
Guest63107 is now known as eebrah_
ykchavan has quit [Quit: Leaving]
eebrah_ has quit [Ping timeout: 265 seconds]
HeHoPMaJIeH has joined #linux-sunxi
HeHoPMaJIeH has quit [Changing host]
HeHoPMaJIeH has joined #linux-sunxi
hehopmajieh_ has joined #linux-sunxi
eebrah has joined #linux-sunxi
eebrah has quit [Ping timeout: 250 seconds]
rellla has joined #linux-sunxi
n01 has joined #linux-sunxi
BluesBoy has joined #linux-sunxi
panda84kde has joined #linux-sunxi
rellla2 has joined #linux-sunxi
fredy has quit [Excess Flood]
rellla has quit [Ping timeout: 260 seconds]
fredy has joined #linux-sunxi
<libv> pfff... when will this package churn end...
ykchavan has joined #linux-sunxi
<oliv3r> libv: even so, it's very exciting!
<libv> i just ran into a missed version bump in ubuntu...
<libv> meaning that i now require different fbturbo packages for raring and quantal
<libv> 4th change to the debian/changelog, and now each version really does need a different name...
<libv> and here i was hoping i could be bringing up precise already ...
Sonic1 has quit [Ping timeout: 260 seconds]
Sonic1 has joined #linux-sunxi
deasy has joined #linux-sunxi
n01 has quit [Ping timeout: 265 seconds]
vicenteH has quit [Ping timeout: 245 seconds]
<libv> grmbl... somehow raring lost its indexes...
<libv> fixed now...
gzamboni has quit [Read error: No route to host]
n01 has joined #linux-sunxi
gzamboni has joined #linux-sunxi
\\Mr_C\\ has joined #linux-sunxi
notmart has joined #linux-sunxi
cubear has joined #linux-sunxi
\\Mr_C\\ has quit [Read error: Connection reset by peer]
popolon has joined #linux-sunxi
\\Mr_C\\ has joined #linux-sunxi
rz2k has quit []
\\Mr_C\\ has quit []
\\Mr_C\\ has joined #linux-sunxi
FR^2 has joined #linux-sunxi
Black_Horseman has quit [Remote host closed the connection]
hehopmajieh_ has quit [Quit: Leaving]
vicenteH has joined #linux-sunxi
Gerwin_J has joined #linux-sunxi
ZetaNeta has joined #linux-sunxi
rellla2 is now known as rellla
deasy has quit [Quit: Nom d'un quark, c'est Edmonton !]
bamvor has quit [Remote host closed the connection]
deasy has joined #linux-sunxi
Soru has quit [Read error: Connection reset by peer]
Soru has joined #linux-sunxi
iamfrankenstein has joined #linux-sunxi
wens has quit [Quit: Lost terminal]
hipboi has quit [Quit: Leaving]
jinzo has joined #linux-sunxi
jinzo has quit [Changing host]
jinzo has joined #linux-sunxi
<libv> ah, the reprepro db is on the local directory, apparently
Gerwin_J has quit [Quit: Gerwin_J]
Gerwin_J has joined #linux-sunxi
kaspter has quit [Remote host closed the connection]
<libv> so i had 2 dbs all of a sudden... and now i have to redo quantal...
<oliv3r> this all sounds like a painfull learning experience in setting up repo's i'll be very happy when it's done and I can just use yours ;D
<libv> well, the repo stuff is not too painful
<libv> i just hit one of the gotchas
<libv> getting all the niggles with packaging worked out is more troublesome
<libv> it would be nice to see kernel and u-boot packaged up in a useful way, together with a tool to set up script.bin
<libv> on the other hand, kernel is probably the most useful, as that is not too hw dependant
<libv> and the other bits tend not to change
\\Mr_C\\ has quit []
\\Mr_C\\ has joined #linux-sunxi
Black_Horseman has joined #linux-sunxi
Black_Horseman has joined #linux-sunxi
Black_Horseman has quit [Changing host]
Soru has quit [Read error: Connection reset by peer]
Soru has joined #linux-sunxi
<oliv3r> aye
Black_Horseman has quit [Remote host closed the connection]
<oliv3r> well atleast we have daily builds now
JohnDoe_71Rus has quit [Quit: Miranda IM! Smaller, Faster, Easier. http://miranda-im.org]
vicenteH has quit [Ping timeout: 261 seconds]
ganbold_ has joined #linux-sunxi
paulk-collins has joined #linux-sunxi
cubear has quit [Quit: Leaving]
fredy has quit [Excess Flood]
fredy has joined #linux-sunxi
vicenteH has joined #linux-sunxi
pitillo has quit [Ping timeout: 240 seconds]
rz2k has joined #linux-sunxi
pitillo has joined #linux-sunxi
<libv> ssvb: on ubuntu precise, i need to explicitely enable shadowfb, otherwise X seems to restarted endlessly by lightdm
<ssvb> libv: hmm, this should not be happening
<ssvb> libv: how to reproduce this problem?
<libv> you have an odroid right?
<libv> as this is happening on the odroid
<ssvb> yes, but not in active use
<libv> well, i guess that as long as sunxi disp is not found, it should be reproducable
<libv> let me find the precise alip image that i am using
<ssvb> ok, thanks
<ssvb> actually, this sounds like a problem with modesetting
<ssvb> when sunxi disp is found, we are loading the mali kernel modules
<ssvb> and modesetting messes up with them
<libv> standard fbdev is happy
<libv> fbturbo is not, unless i explicitely enable shadowfb
<ssvb> the shadowfb part is hard to explain
<ssvb> this needs some debugging for sure
<ssvb> ok, thanks for the link, I'll try it
<libv> and apt-get build-dep xserver-xorg-video-fbdev
<libv> no, it is not modesetting
<libv> fbturbo is loaded fine
<libv> but Xorg just keeps on getting restarted
<ssvb> no traces of modesetting in /var/log/Xorg.0.log ?
<libv> fbturbo is loaded fine
<libv> that means that is happily passed the probe stage
<libv> and that other drivers are not getting tried.
<ssvb> yes, but the failed initialization of modesetting could cause memory corruption, etc.
<ssvb> that is if modesetting was tried
<libv> _again_ fbturbo is started.
<libv> it is the first to get started, at all, due to an xorg.conf
<libv> as X does not find fbturbo on its own
<libv> it's only when there is no config that the kms driver is attempted.
<ssvb> if grep for modesetting in /var/log/Xorg.0.log does not find anything, then modesetting is surely unrelated
<libv> *facepalm*
<ssvb> that's a simple basic test
<libv> mooooo.
<ssvb> I'll try it myself, don't worry about this :)
<libv> there. is. no. trace. of. kms. driver.
<ssvb> ok
<ssvb> thanks for confirming
<libv> it's all fbturbo.
<libv> some things in X are sane
<libv> and i would not be able to enjoy standard fbdev if i didn't have a working config
<libv> let alone shadowfb
<libv> on fbturbo
<libv> also, this bug is "we need to set a proper config"
naobsd has joined #linux-sunxi
BluesBoy has quit [Quit: HydraIRC -> http://www.hydrairc.com <- \o/]
pitillo has quit [Ping timeout: 260 seconds]
pitillo has joined #linux-sunxi
ykchavan has quit [Read error: No route to host]
ykchavan has joined #linux-sunxi
HeHoPMaJIeH has quit [Remote host closed the connection]
rz2k has quit [Read error: Connection reset by peer]
eebrah has joined #linux-sunxi
stekern has quit [Ping timeout: 272 seconds]
<popolon> does someone have an idea of who manage the usbids now ? gowdy email at stanford univeristy doesn't work anymore :(
<popolon> and there is bugs in the db
<popolon> new names doesn't appear
popolon has quit [Ping timeout: 260 seconds]
stekern has joined #linux-sunxi
pitillo has quit [Changing host]
pitillo has joined #linux-sunxi
rellla has quit [Quit: Nettalk6 - www.ntalk.de]
wingrime has joined #linux-sunxi
pfdm has joined #linux-sunxi
<pfdm> morning
popolon has joined #linux-sunxi
<pfdm> Do someone know about the call function trace for libvecore ?
mcbrick has joined #linux-sunxi
mcbrick has quit [Remote host closed the connection]
mcbrick has joined #linux-sunxi
n01 has quit [Ping timeout: 240 seconds]
Gerwin_J has quit [Quit: Gerwin_J]
\\Mr_C\\ has quit []
rellla has joined #linux-sunxi
notmart has quit [Quit: notmart terminated!]
\\Mr_C\\ has joined #linux-sunxi
pfdm has quit [Ping timeout: 250 seconds]
FR^2 has quit [Quit: Connection reset by peer]
panda84kde has quit [Quit: Konversation terminated!]
hramrach has joined #linux-sunxi
deasy has quit [Quit: Nom d'un quark, c'est Edmonton !]
eebrah has quit [Ping timeout: 264 seconds]
jemk has joined #linux-sunxi
<mnemoc> arokux1: thanks, added the commit to stage 3.4 too
notmart has joined #linux-sunxi
notmart has joined #linux-sunxi
notmart has quit [Changing host]
deasy has joined #linux-sunxi
<libv> hrm... precise has a mesa without automake, but with autoconf
<libv> and it does not like dpkg-buildpackage -j5
<libv> make -j5 is fine... go figure...
<libv> it will take a few years until i have this neatly packaged up
<wingrime> mnemoc: push cedar patch!
* libv boosts the odroids cpu to 2GHz...
Brod_ has quit [Quit: No Ping reply in 180 seconds.]
Brodomir has joined #linux-sunxi
notmart has quit [Quit: notmart terminated!]
notmart has joined #linux-sunxi
notmart has quit [Changing host]
notmart has joined #linux-sunxi
wingrime has quit [Ping timeout: 272 seconds]
<ssvb> libv: I have started setting up linaro, but if you have it up and running, as a quick test you could probably try setting 'Option "AccelMethod" "CPU"' in the device section of xorg.conf
<ssvb> libv: maybe the probing for Raspberry Pi ioctl fails and crashes the system
<libv> ok
<libv> yeah, it's running, and slowly churning through the mesa package
<libv> the system does not crash
<libv> lightdm just restarts X quickly for about 10 minutes
<libv> where is this accelmethod handled btw?
<ssvb> libv: and here is the code which probes it - https://github.com/ssvb/xf86-video-fbturbo/blob/master/src/fb_copyarea.c#L45
<libv> ssvb: i onle see g2d and copyarea
Gerwin_J has joined #linux-sunxi
<libv> not "cpu"
<libv> so g2d and copyarea are then simply not enabled?
<libv> i could type "cheese" as well?
<ssvb> libv: yes :)
<libv> X endless restart
<libv> (due to shadowfb disabled)
<libv> so i guess that that is not it
<ssvb> ok, I hope to reproduce it here shortly
<ssvb> thanks
<libv> perhaps some fb initialization that was required for 1.11 but is no longer required for 1.13 is no longer present in fbturbo?
<libv> fbdev initialization even
<libv> do you remember removing something of the sort?
<ssvb> no, I tried not to remove anything
<ssvb> is the package built for the right xorg abi?
<libv> it wouldn't load otherwise
<ssvb> right
<libv> and yes, it was freshly built
kenny has joined #linux-sunxi
<libv> if you want to reproduce the package build, you can take the quantal .dsc and diff.gz
<libv> next to the orig.tar.gz you can dpkg source -x *.dsc
<libv> and it will check it out for you
<libv> then dpkg-buildpackage will build it, and the resulting .deb can be installed with dpkg -i
<libv> but i think that a plain make will also make it happen
<libv> as the build should have no influence
<libv> err, packaging
<libv> at least not on this bug
<libv> i haven't tried to run sunxifb though
<libv> worth a go, since i have *glares at mesa build* 3 cpus idling
<libv> the sunxifb version i have does work
<hramrach> Debian has something old as well
<libv> ssvb: d147e25fc507 works
<hramrach> is fbturbo newa name for sunxi or something completely different?
<libv> hramrach: it is the sunxifb repo of ssvb, which he renamed recently
<hramrach> debian has 1.12.4
<hramrach> in stable
<libv> ssvb: backingstore smells like a likely candidate.
<hramrach> they changes bs recently
<libv> crap, mesa build just died on me
<hramrach> and changed or plan changing beffer handling for glx
iamfrankenstein has quit [Read error: Connection reset by peer]
<hramrach> *buffer
<hramrach> but before 1.14 release you like had to chack out the world from git to build X server
<hramrach> too much stuff changed at once
deasy has quit [Quit: Nom d'un quark, c'est Edmonton !]
<libv> nope, the backingstore branch point works
iamfrankenstein has joined #linux-sunxi
notmart has quit [Quit: notmart terminated!]
<hramrach> hmm, I kinda have the X server checked out on board where X does not work :s
<WarheadsSE> i'd be happy to see 1.14 compatible..
ykchavan has quit [Quit: Leaving]
eebrah has joined #linux-sunxi
geecko has joined #linux-sunxi
Gerwin_J has quit [Read error: Connection reset by peer]
<hramrach> what is broken?
Gerwin_J has joined #linux-sunxi
<hramrach> I have gnome-shell running with git tip
<hramrach> well, crawling
<hramrach> it went back to fallback shell
<WarheadsSE> if it works, then great
<WarheadsSE> I haven't had time to revisit sunxi stuff ina while
<WarheadsSE> soon to have almost an extra 1hr/day
<hramrach> I have Debian stable 1.12
<hramrach> ln -sf /bin/true /usr/lib/gnome-session/gnome-session-check-accelerated
<hramrach> need to make some pacakge that does that
<hramrach> libv: anyway, mali works as poorly as ever. gnome UI has blue tint and flashes occasionally but crawls at super-snail speeds
<ssvb> hramrach: if you want to try a reasonably good gles compositing manager, that would be kwin
<libv> grmbl, i think i might have been wasting ssvbs time
<ssvb> hramrach: you can also run 'perf top' to see what's going on
<libv> ssvb: i bisected it, and it all worked
<ssvb> libv: ?
<libv> ssvb: i built the package build tree directly, and it worked
<libv> ssvb: i installed the .deb again, and it worked.
<libv> file corruption?
<ssvb> hmm, still something must be fishy
<libv> it really makes no sense
<libv> it's the same .deb i installed earlier, and lightdm just runs
* libv rms the driver again
deasy has joined #linux-sunxi
<libv> and now it nicely loads fbdev (as this predates modesetting)
<libv> bwhaha
<libv> ssvb: i will now bisect with shadowfb disabled :p
<libv> but first i will hit my head against a wall for a bit
<libv> so ignore me.
<ssvb> ok :)
<libv> i am clearly a really big idiot
<libv> luckily, i am doing nothing else than waiting on...
<libv> crap, another broken mesa.
<libv> why, it's exactly the same position as before, so it is not a heat issue
<libv> anyway, now bisecting again, as i disabled the broken driver, and i need another 40 minutes for the new build
<Turl> libv: are you building on target?
<libv> yeah
<Turl> libv: maybe virtualizing your box on x86 will build it faster
<libv> quad a9 @ 2GHz, and a usb disk (which has improved performance massively, even though it is only usb2.0)
<libv> it's actually quite ok, normally
<libv> just that under precise, the mesa tree is a bit shaky and does not handle dpkg-buildpackage -jX well
<Turl> ah, I thought you were building on a C-A8 sunxi :)
<libv> by the time i have that set-up, i hope to be done with mesa
<lunra> Ooh, what quad A9 did you get to 2GHz?
Gerwin_J has quit [Quit: Gerwin_J]
<WarheadsSE> sounds like an imx6q
<Turl> libv: foreign minbase debootstrap + qemu should get you set up rather quickly
<WarheadsSE> maybe an RK ?
<libv> lunra: an odroid-x2 that i got to buy myself, as the hardkernel folks back then were unwilling to support free software projects properly
<libv> so don't buy from them
<lunra> yeah, hardkernel seem a little... uncaring
<libv> after fixing some kernel issues, i got to discover that my authorship got dropped of my patches
<WarheadsSE> I've seen that. It's assanine.
<Turl> o.o
<WarheadsSE> they are getting better
<WarheadsSE> Or at least, we(ALARM devs) have a decent relationship with them
paulk-collins has quit [Remote host closed the connection]
<lunra> ALARM is fantastic, guys. <3
Gerwin_J has joined #linux-sunxi
derethor has joined #linux-sunxi
rellla has quit [Quit: Nettalk6 - www.ntalk.de]
tinti has joined #linux-sunxi
<libv> ssvb: the issue is with the package
<libv> i get a sigill when i dpkg-buildpackage, and a working desktop when i just ./autogen.sh; make -j5 clean; make -j5
<Turl> libv: diff -O or fpu/neon flags?
<Turl> libv: check debian/rules
<libv> debian/rules is quite empty
<libv> might be something to do with how it is stripped
<libv> or yes, optimization
pokus has joined #linux-sunxi
<pokus> hi...i just want ask if exists some toolchain for images which are avaliable on site...or is better make whole system from source codes..
<pokus> because when i have installed arm-linux-gnueabihf cross compiler i got GLIBC mismash on arm board (host is 2.15 and target have 2.13)
<derethor> hey guys, i have a question... some time ago, I compiled the kernel sunxi 3.4. Now I did a git pull, and it has many changes. How can I merge my old .config with the new one? i guess that if I simply copy the file, I wont see the new features, etc
<derethor> well, in fact, now it is a linux x86 kernel, and not an ARM
<libv> derethor: make ARCH=arm oldconfig
<libv> or old-config, or something
<libv> but you probably already overwrote your old .config
<pokus> ...i mean A20 board
<pokus> i think....
<derethor> ah, ok
<derethor> i am on the sunxi-3.4 branch.. .i see that it has no commits since september... which branch should i use?
<derethor> this did the trick : make ARCH=arm oldconfig
<derethor> thank you1!
Black_Horseman has joined #linux-sunxi
Black_Horseman has quit [Changing host]
Black_Horseman has joined #linux-sunxi
eebrah has quit [Ping timeout: 256 seconds]
geecko has quit [Ping timeout: 246 seconds]
jemk has quit [Ping timeout: 272 seconds]
geecko has joined #linux-sunxi
fredy has quit [Excess Flood]
fredy has joined #linux-sunxi
gzamboni has quit [Read error: No route to host]
tinti has quit [Excess Flood]
<libv> it's to do with LDFLAGS
tinti has joined #linux-sunxi
<libv> urgh... no, apparently
<libv> dammit.
<oliv3r> heh, i feel ya; i can't seem to be able to program my arduino :S
<libv> ah, yes, it does seem to point to LDFLAGS
<libv> ssvb: -Wl,-Bsymbolic-functions
<libv> that's what kills fbturbo
<ssvb> who is adding this?
<libv> dh_something
<libv> if i set it in LDFLAGS, it bombs though
<libv> let me try a newer ubuntu
<libv> since this mesa thing isn't working out apparently.
<oliv3r> :(
<libv> ssvb: same thing happens on quantal
<libv> so it's not a compiler or toolchain issue
<libv> could it be that shadowfb disables your assembler stuff?
<libv> and that this linker flags plays merry hell with your assembler code?
<Turl> oliv3r: what's wrong with your arduino? :p
<oliv3r> i don't know :)
<oliv3r> i'm trying to flash some hex via avrdude but its refusing
<oliv3r> so not usign the IDE
<Turl> make sure you have access to the ttyUSB device
jinzo has quit [Quit: Leaving]
<oliv3r> not ttyACM?
<Turl> I think there was a programming window too
<Turl> yeah that :p
<Turl> you need to reboot the board and then program I think
<Turl> the IDE does it for you
<oliv3r> crw-rw---- 1 root uucp 166, 0 Oct 23 22:34 /dev/ttyACM0
<oliv3r> wheel uucp audio cdrom video games qemu cdrw users pulse android kvm scanner oliver
<oliv3r> yeah i was spamming reset
<oliv3r> so gonna see if i can get the ide to compile anyhting
<ssvb> libv: yes, with shadowfb the assembler functions are not called
<libv> ssvb: heh, that should've been suspect from the start then
gzamboni has joined #linux-sunxi
\\Mr_C\\ has quit []
derethor has quit [Quit: Leaving]
\\Mr_C\\ has joined #linux-sunxi
<popolon> ganbold, https://commons.wikimedia.org/wiki/File:Phagspa-Hangul_comparison.svg <= didn't know that the hangul used the sames rules than japanese kanna/chines zhuyin(or bopomofo), but with mongol phagspa :)
ganbold_ has quit [Remote host closed the connection]
ZetaNeta has quit [Read error: Connection reset by peer]
tinti has quit [Ping timeout: 248 seconds]
hramrach has quit [Remote host closed the connection]
tinti has joined #linux-sunxi
deasy has quit [Quit: Nom d'un quark, c'est Edmonton !]
<eagles0513875> hey guys :)
<eagles0513875> hey Turl
Tartarus has quit [Ping timeout: 240 seconds]
Sonic1 has quit [Remote host closed the connection]
Sonic1 has joined #linux-sunxi
Black_Horseman has quit [Quit: Αποχώρησε]
Black_Horseman has joined #linux-sunxi
<ganbold> popolon: interesting
naobsd has quit [Quit: Page closed]