rz2k 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 | FOSDEM talks - http://dl.linux-sunxi.org/users/nove/sunxi_at_fosdem2014/
tomboy65 has joined #linux-sunxi
netlynx has quit [Quit: Leaving]
bertrik has quit [Remote host closed the connection]
Net147 has joined #linux-sunxi
techn_ has quit [Ping timeout: 252 seconds]
Net147 has quit [Client Quit]
Net147 has joined #linux-sunxi
leviathanch2 has quit [Ping timeout: 246 seconds]
<libv> oliv3r: this thing has a very awkward android installed on it
<libv> and there seems to be no OTG or device port
<Turl> libv: no otg? wtf :)
<Turl> libv: does it have wifi?
<Turl> you can get adb that way
<libv> yeah, but i stuck in a sunxi installation for mele and just got the script.bin
<libv> and serial needs to simply be soldered on
<libv> now i just need to get the memory layout
<libv> android id strings have been plucked off already
<Turl> libv: a lowest common denominator automagic bringup card would be great
<libv> wasn't oliv3r going to do autodetection?
<Turl> something soc specific you dd on a small card, and it boots and pulls script.bin, stuff from android, maybe eve bootN from nand?
<Turl> even*
<Turl> dell poweredge
<Turl> weird :)
<libv> this is just the machine the kernel was built on
<Turl> I know
<Turl> but I didn't envision chinese makers using dell hw
<libv> this really smells like a standard mele or allwinner evb build
egbert has joined #linux-sunxi
pwhalen has quit [Remote host closed the connection]
egbert_ has quit [Ping timeout: 252 seconds]
arky has joined #linux-sunxi
<libv> yay. meminfo
<HdkR> yay
<libv> patches and pictures, tomorrow
arky has quit [Ping timeout: 246 seconds]
TheSeven has quit [Disconnected by services]
[7] has joined #linux-sunxi
bsdfox has quit [Read error: Operation timed out]
arky has joined #linux-sunxi
arky has quit [Read error: Connection reset by peer]
kuldeepdhaka has joined #linux-sunxi
kuldeepdhaka has quit [Ping timeout: 252 seconds]
kuldeepdhaka has joined #linux-sunxi
kuldeepdhaka has quit [Max SendQ exceeded]
kuldeepdhaka has joined #linux-sunxi
kuldeepdhaka has quit [Max SendQ exceeded]
kuldeepdhaka has joined #linux-sunxi
Net147 has quit [Quit: HydraIRC -> http://www.hydrairc.com <- The professional IRC Client :D]
kuldeepdhaka has quit [Client Quit]
JohnDoe_71Rus has joined #linux-sunxi
arky has joined #linux-sunxi
arky has quit [Client Quit]
JohnDoe_71Rus has quit [Quit: KVIrc Insomnia 4.0.2, revision: 402, sources date: 20100627, built on: 2010-08-02 17:44:33 UTC 402 http://www.kvirc.net/]
bertrik has joined #linux-sunxi
bertrik has quit [Changing host]
bertrik has joined #linux-sunxi
AreaScout has joined #linux-sunxi
diego71 has quit [Ping timeout: 272 seconds]
_massi_ has joined #linux-sunxi
sehraf has joined #linux-sunxi
Seppoz has quit [Remote host closed the connection]
Seppoz has joined #linux-sunxi
Seppoz has quit [Remote host closed the connection]
Seppoz has joined #linux-sunxi
sehraf has quit [Ping timeout: 245 seconds]
sehraf has joined #linux-sunxi
techn_ has joined #linux-sunxi
sehraf has quit [Read error: Connection reset by peer]
sehraf has joined #linux-sunxi
sehraf has quit [Read error: Connection reset by peer]
paulk-collins has joined #linux-sunxi
davi has joined #linux-sunxi
<paulk-collins> hi, where can I find those fedora images I've been told of?
<paulk-collins> my device is the zatab (a10)
<paulk-collins> I'd love to have gnome-shell running if possible
sehraf has joined #linux-sunxi
netlynx has joined #linux-sunxi
paulk-collins has quit [Remote host closed the connection]
pacopad has joined #linux-sunxi
Black_Horseman has quit [Remote host closed the connection]
bbrezillon has joined #linux-sunxi
paulk-collins has joined #linux-sunxi
Nazcafan has joined #linux-sunxi
nabblet has joined #linux-sunxi
gzamboni has quit [Ping timeout: 272 seconds]
gzamboni has joined #linux-sunxi
wigyori has joined #linux-sunxi
Black_Horseman has joined #linux-sunxi
Black_Horseman has joined #linux-sunxi
Black_Horseman has quit [Changing host]
<Wizzup> So, I am reading http://linux-sunxi.org/Installing_to_NAND ... In my book, nand typically means it doesn't do wear levelling and handle faulty blocks
<Wizzup> But the author happily puts ext4 on it
<Wizzup> am I missing something here? :)
<Wizzup> I did see this: There is no real difference between NAND and an SD-card apart from the fact that directly attached flash use the Sunxi NAND controller directly while SD-Cards come with a standard interface and an embedded controller.
<Wizzup> but I am not sure what this implies
<mnemoc> allwinner implements the wear leveling and bad blocks handling as part of their monster nand driver
<mnemoc> which the mtd driver doesn't
<Wizzup> I see, and the monster nand driver is in the 3.4 linux-sunxi kernel
* Wizzup is now starting to understand the problems with mainline nand a bit more
<mnemoc> yup
<mnemoc> same applies to nand support in u-boot
<mnemoc> and the compatibility issues involved
sehraf has quit [Ping timeout: 245 seconds]
FreezingCold has quit [Read error: Operation timed out]
diego71 has joined #linux-sunxi
sehraf has joined #linux-sunxi
nabblet has quit [Ping timeout: 260 seconds]
nabblet has joined #linux-sunxi
cajg has quit [Ping timeout: 272 seconds]
<paulk-collins> wow cool, fedora is starting up
nabblet_ has joined #linux-sunxi
<paulk-collins> ts not working though
bluse has joined #linux-sunxi
lynxis has joined #linux-sunxi
nabblet_ has left #linux-sunxi [#linux-sunxi]
cajg has joined #linux-sunxi
_massi_ has quit [Remote host closed the connection]
<paulk-collins> omg and it's running with llvmpipe
<paulk-collins> awesome
<plaes> what machine?
<paulk-collins> zatab
<paulk-collins> apparently, the ts module gives i2c errors
<Turl> paulk-collins: you have an og one?
<Turl> paulk-collins: with gt801?
<paulk-collins> what's og?
<Turl> original
<paulk-collins> yes it's an original
<paulk-collins> but has ft5x_ts
<Turl> ah
<paulk-collins> I mean, it was bought from zareason
<Turl> mine has a goodix
<paulk-collins> the script.bin handles various cases
<Turl> are you sure it has a ft5x?
<Turl> my script also lists it as #1 but it's a goodix
<paulk-collins> the goodix module wouldn't load, saying ft5x_ts doesn't match its CTP_NAME
<Turl> yeah, the linux-sunxi one won't
<paulk-collins> ahh ok that would make sense
<Turl> paulk-collins: works on my zatab :)
<paulk-collins> because i2c errors likely indicate it's the wrong chip
<paulk-collins> I'll check what android has
<paulk-collins> yeah, runs goodix_touch_3F
<Turl> paulk-collins: the driver there should work
<paulk-collins> alright
<paulk-collins> thanks
<Turl> the one I linked
deasy has joined #linux-sunxi
<Turl> paulk-collins: are you running gnome shell with llvmpipe btw?
<Turl> paulk-collins: last I tried it was unusable, like 0.5fps
Black_Horseman has quit [Ping timeout: 272 seconds]
paulk-collins has quit [Ping timeout: 248 seconds]
paulk-collins has joined #linux-sunxi
<aep> hm, how do i get hdmi working on the cubiebaord? i've got the fex setup, but X is whining about having no display
<aep> how do i debug this?
nove has joined #linux-sunxi
<aep> [ 194.032] (II) FBTURBO(0): No sunxi-g2d hardware detected (check /dev/disp and /dev/g2d)
dlan^ has joined #linux-sunxi
dlan^ has joined #linux-sunxi
<Turl> aep: do you have disp and g2d?
<Turl> (or at least disp?)
<aep> norp
<aep> should the mali driver give me those?
<Turl> then you somehow didn't enable the display stuff on your kernel
<Turl> mali is completely unrelated
<aep> well i have /dev/fb and this worked with lcd
<aep> ah
<Turl> aep: make sure you have CONFIG_FB_SUNXI{,_LCD,_HDMI}=y on your config
<mnemoc> Turl: will you submit the goodix driver to sunxi-3.4?
<mnemoc> just 5806 mails to read :\
<Turl> mnemoc: I wanted to clean it first, it's pretty ugly
<Turl> but if you're not complaining we can push it as is and clean later :)
Seppoz has quit [Remote host closed the connection]
Seppoz has joined #linux-sunxi
<mnemoc> Turl: in the legacy kernels we always import first and clean on the go
<Turl> mnemoc: ok
<Turl> mnemoc: do you want me to push it or ML?
<mnemoc> happy customers first.... quality... for mainlining :p
<mnemoc> Turl: ML please
<Turl> ok
<Turl> after I get my desktop online again I'll send it
<Turl> paulk-collins: if you can confirm it works for you too even better :)
<aep> Turl: yes i have CONFIG_FB_SUNXI_HDMI=y
<aep> and CONFIG_FB_SUNXI_LCD=y
<aep> [ 0.527959] sunxi disp driver loaded (/dev/disp api 1.0)
<aep> and i do have a /dev/disp actually, just not /dev/g2*
<aep> huh. rebooting magically fixed it
<oliv3r> paulk-collins: if you recall my presentation, touchscreen drivers are a big mess on a lot of devices; that said, the zatab should be ok supported; it was one of the earlier sunxi supporters (they kinda left us though)
<oliv3r> paulk-collins: which fedora image did you use in the end? i guess hansg's version?
davi has quit [Ping timeout: 252 seconds]
<paulk-collins> Turl, well I didn't try yet, but I'll try to let you know when I'm to the point of building kernels
<paulk-collins> oliv3r, I used http://fedorapeople.org/~lkundrak/a10-images/
<paulk-collins> and had to place the zatab script.bin
<oliv3r> that's hansg's version
<oliv3r> but if lkundrak's version works, that's fine
<oliv3r> hansg's version is based on xfce
<paulk-collins> this one too
<paulk-collins> dl.cubieboard.org/software/a20-cubieboard/fedora/ => is that supposed to work on A10 too?
jemk has joined #linux-sunxi
<Turl> oliv3r: the touchscreen on the zatab wasn't working until like two weeks ago :p
<oliv3r> Turl: ohh really?
<oliv3r> paulk-collins: that's universal image
<oliv3r> you run setup.sh, which copies the kernel, modules and bootloader to your sd card
<oliv3r> it works really well
<oliv3r> the pdf on that site should give you proper instructions
<paulk-collins> ok, good
<paulk-collins> the other F20 image runs well too
sehraf has quit [Read error: Connection reset by peer]
<oliv3r> i'm a bit puzzled why they don't work together
<mnemoc> damn... >5800 unread mails AGAIN! :(
<mnemoc> shame on marc and hansg for been that active
deasy has quit [Quit: Nom d'un quark, c'est Edmonton !]
shineworld has joined #linux-sunxi
shineworld has left #linux-sunxi [#linux-sunxi]
linkmauve1 has quit [Quit: gateway shutdown]
linkmauve1 has joined #linux-sunxi
deasy has joined #linux-sunxi
sehraf has joined #linux-sunxi
dlan^ has quit [Quit: leaving]
jemk has quit [Remote host closed the connection]
<leviathanch> wens: did the changes I posted to you work for you?
FDCX has joined #linux-sunxi
<leviathanch> wens, Turl: You can checkout the newest revision from here: http://git.o2s.ch/?p=linux-next.git;a=shortlog;h=refs/heads/20140214
<Turl> mnemoc: you should start practicing inbox/unread 0
<Turl> leviathanch: :)
<Turl> leviathanch: nitpick, commas @ 12 The supported chips include a10,a10s,13,a20 and a31,.
<Turl> and spaces 14 would be 0x01c0f000+0x1000=0x01c10000(see example)
<oliv3r> oh! that's funny
<oliv3r> i'm looking at gapps from CM and it has 'install-optional.sh'
<oliv3r> there's a cat /proc/cpuinfo | grep -q sun4i init!
<oliv3r> that's too funn
<Turl> oliv3r: I know right? ;)
<oliv3r> i think it means if you have a sun4i do NOT use face unlock
<oliv3r> but still
<oliv3r> why sun5i and sun7i are ignored, who knows
<oliv3r> Turl: btw, what is 'GMSCore'
<oliv3r> i found that for JB you only need GoogleLoginService.apk, GoogleServicesFramework.apk and Phonesky.apk to get the playstore going
<oliv3r> i thought GMSCore was required too
<Turl> oliv3r: back then there was only sun4i :p
<Turl> it just stuck there
<Turl> oliv3r: I think gmscore is play services
<Turl> it'll download itself from play store I suppose
<oliv3r> then i won't install it as a system app
<oliv3r> (i need to get some apk's via the playstore for work'
<oliv3r> aswell as the banking apps
<leviathanch> Turl: oops, thx! will fix that before I hand in another patchset
<leviathanch> forgot int ret; anyway
<leviathanch> >.>
<Turl> oliv3r: do you have nice banking apps over there?
mkutsevol has joined #linux-sunxi
<Turl> hi mkutsevol, have you fixed your irc client now?
<oliv3r> Turl: yeah; all very much locked up
<oliv3r> but i even prefer the banking app overthe website
<oliv3r> much even
<Turl> oliv3r: here my app looks like a midlet from some old java phone
<Turl> stretched and made android runnable
<Turl> :p
<oliv3r> ah no, the ING app is really well done
geecko has joined #linux-sunxi
kuldeepdhaka has joined #linux-sunxi
sehraf has quit [Read error: Connection reset by peer]
<mnemoc> Turl: trying to catch up with the linux-sunxi/sunxi@linux-arm first.... but it's HARD!
* Turl loves 'mark as read' button
<Turl> :)
<mnemoc> that's cheating
davi has joined #linux-sunxi
kz1 has quit [Ping timeout: 260 seconds]
<paulk-collins> any idea how to enable USB networking on fedora/A10?
rz2k has joined #linux-sunxi
<Turl> paulk-collins: plug a usb to ethernet dongle -> ... -> PROFIT!
<Turl> :)
kz1 has joined #linux-sunxi
<paulk-collins> nah, that sucks
<paulk-collins> I want kernel-provided USB networking
<paulk-collins> not having a chip in between to do it
<mnemoc> you mean use gadget?
<mnemoc> usb*
sehraf has joined #linux-sunxi
<paulk-collins> yes
<paulk-collins> ah nvm, wifi works
<Turl> yeah, wifi is just modprobe 8192cu :)
<mnemoc> nobody made an a10 device with a different wifi IC than the one in aw's reference design?
<Turl> mnemoc: I believe some have a bcm one via sdio
<paulk-collins> also, how about providing a sysfs interface for backlight?
<paulk-collins> IIRC it's done via ioctl, which is not standard at all
<Turl> yeah, disp ioctl
<paulk-collins> doing it through sysfs would make it work out of the box on e.g. fedora
<Turl> I think techn_ worked on a sysfs alternative though
<mnemoc> paulk-collins: there is a very standard answer to that, patches welcomed ;-)
<Turl> paulk-collins: if you want something else to fix, you can tackle the battery :)
bsdfox has joined #linux-sunxi
<Turl> it appears stuff is there, but gets double readings or something
<Turl> once with the right values, and another second time with 0
<Turl> (viewed with upower's monitor)
<paulk-collins> ok
deasy has quit [Quit: Nom d'un quark, c'est Edmonton !]
FDCX has quit [Ping timeout: 248 seconds]
<paulk-collins> I guess I'll look into it when I receive the rest of my allwinner devices
netchip has quit [Read error: Connection reset by peer]
w00tc0d3 has joined #linux-sunxi
w00tc0d3 is now known as Guest28213
ZetaNeta has joined #linux-sunxi
<ZetaNeta> hi
<ZetaNeta> Does the vanilla kernel support booting on allwinner?
<Turl> ZetaNeta: yeah, but with limited hw support
<ZetaNeta> as i wanted to use gentoo's hardened kernel
bsdfox has quit [Ping timeout: 248 seconds]
<ZetaNeta> :P
<ZetaNeta> is A10 fully supported? As it looks like only A20+ got problems with mainlining
<ZetaNeta> oh...
nabblet has quit [Read error: Connection reset by peer]
<ZetaNeta> now i see more clearly
<ZetaNeta> eh...
bsdfox has joined #linux-sunxi
leviathanch2 has joined #linux-sunxi
paulk-a10 has joined #linux-sunxi
<paulk-a10> hi
<paulk-a10> so how do I install the mali stuff on fedora?
FDCX has joined #linux-sunxi
<Turl> paulk-collins: read [[Binary drivers]]
<paulk-a10> yeah but it is apparently debian-oriented
<mnemoc> improve it to cover fedora
<Turl> paulk-collins: cloning a git repo and running make is debian origiented? :p
<paulk-a10> Turl: you know what I mean: I don't want to look up each dependency ;)
<paulk-a10> since their names aren't exactly the same
Seppoz has quit [Ping timeout: 260 seconds]
<paulk-a10> also, it says to modprobe mali & ump, but that crashes Xorg when I do that
leviathanch2 has quit [Ping timeout: 252 seconds]
<Turl> huh
sehraf has quit [Read error: Connection reset by peer]
<Turl> that shouldn't happen
sehraf has joined #linux-sunxi
<paulk-a10> ok, I'll try again shortly
sehraf has quit [Read error: Connection reset by peer]
<paulk-collins> lol
bertrik has quit [Remote host closed the connection]
<Nazcafan> When flashing for the first time my cubietruck, I had to use some weird tool (probably provide by sunxi) this involved building a kernel module
<Nazcafan> I kind of remember coming across a project that aimed to write a replacement tool, probably using libusb
<mnemoc> awutils
<Nazcafan> ah, yes
<mnemoc> it's also possible to use ./fel (from sunxi-tools) to push an open source linux based installer (which no one has written yet)
<mnemoc> but it was awutils who intended to mimic livesuit
sehraf has joined #linux-sunxi
w00tc0d3 has joined #linux-sunxi
Guest28213 has quit [Read error: Connection reset by peer]
w00tc0d3 is now known as Guest99709
nove has quit [Quit: nove]
paulk-a10 has quit [Remote host closed the connection]
<speakman> I'm trying to make a simple Qt app using QGraphicsView with a QGLWidget viewport (for maximum acceleration), but then the app won't start with the following error message "EGLFS: OpenGL windows cannot be mixed with others."
<speakman> Hm. I might have an idea...
<speakman> It's probably because I'm first creating a QWidget which holds the QGraphicsView which in turn has the QGLWidget viewport.
Guest99709 has quit [Ping timeout: 252 seconds]
w00tc0d3_ has joined #linux-sunxi
kuldeepdhaka has quit [Ping timeout: 252 seconds]
JohnDoe_71Rus has joined #linux-sunxi
JohnDoe_71Rus has quit [Changing host]
JohnDoe_71Rus has joined #linux-sunxi
leviathanch2 has joined #linux-sunxi
kuldeepdhaka has joined #linux-sunxi
geecko has quit [Ping timeout: 245 seconds]
leviathanch2_ has joined #linux-sunxi
leviathanch2 has quit [Ping timeout: 260 seconds]
<paulk-collins> so yeah, modprobe mali with X running makes it crash
bertrik has joined #linux-sunxi
<ssvb> paulk-collins: if you have "sunxi_no_mali_mem_reserve" kernel cmdline option used with old kernel, then "modprobe mali" indeed corrupts memory and kills the system
<paulk-collins> ah yes indeed
<paulk-collins> that's the case
<paulk-collins> is it safe to disable it?
<paulk-collins> like just removing the option from cmdline?
<ssvb> yes, this is safe, you will just have 64MB of RAM getting unconditionally reserved for mali at boot time
<paulk-collins> alright
<Nazcafan> I am struggling to build awflash, it seems I need a version of libusb that contains libusb_device_handle, as well as usb_dev_handle, and none of the versions I have on my debian system matches
<Nazcafan> I have 0.1 as well as 1.0 installed
w00tc0d3_ has quit [Read error: Connection reset by peer]
w00tc0d3 has joined #linux-sunxi
w00tc0d3 is now known as Guest6566
tomboy65 is now known as tomboy64
fredy has quit [Excess Flood]
fredy has joined #linux-sunxi
FDCX has quit [Ping timeout: 272 seconds]
leviathanch2_ has quit [Remote host closed the connection]
leviathanch2 has joined #linux-sunxi
<hramrach> Nazcafan: I guess it's just broken
<hramrach> uses a mix of old and new and nonexistent interface
<hramrach> also there were a few versions of awutils floating around iirc so you might have better luck with another repo
hno` is now known as hno
hno has quit [Changing host]
hno has joined #linux-sunxi
<Nazcafan> ...
<ZetaNeta> Does A10 provide anything similar to a TPM?
<ZetaNeta> (Or to be exact, a iNet97-f board)
<ZetaNeta> or a way to edit the bootup in a way preventing booting off anything but the nand?
<ZetaNeta> (No, my dear GNU people, this is meant to enforce the password, not to lock the user)
netlynx has quit [Quit: Leaving]
<mnemoc> A10 can ALWAYS boot from uSD
<mnemoc> thanks to that feature we are where we are
<ZetaNeta> mnemoc, Eh...
<ZetaNeta> Then, i only see "self-destruction" as a security feature
* ZetaNeta no longer trusts encryption... only the explosives
<ZetaNeta> from another side, this may be quite useless, as while there is no way to prove if the "physical" and "logical" owner match, the device will "be still alive"
<ZetaNeta> (I mean like, before its even powered up to show a passwd prompt)
ganbold_ has quit [Ping timeout: 245 seconds]
nabblet has joined #linux-sunxi
breeze7 has joined #linux-sunxi
<breeze7> I have an A13 device. Is it ok to use the toolchain mentionen on AOSP site: /prebuilts/gcc/linux-x86/arm/arm-eabi-4.6/bin/arm-eabi- or is it better to use the toolchain from the wiki /home/user/folder/gcc-linaro-arm-linux-gnueabihf-*_linux/bin/arm-linux-gnueabihf- ?
<breeze7> Which results in a better compilation?
<paulk-collins> breeze7, I suppose both work
<paulk-collins> perhaps linaro has more optimizations?
<atsampson> I just use the Emdebian one...
davi has quit [Ping timeout: 260 seconds]
breeze7_ has joined #linux-sunxi
<breeze7_> And by the way: There exist 2 possible toolchains from the AOSP sources: what is the difference between /prebuilts/gcc/linux-x86/arm/arm-eabi-4.6 AND platform/prebuilts/gcc/linux-x86/arm/arm-linux-androideabi-4.6 ?
<Turl> breeze7_ second one is tuned to build userspace binaries
breeze7 has quit [Ping timeout: 245 seconds]
breeze7_ has quit [Ping timeout: 245 seconds]
ssvb has quit [Quit: Leaving]
ssvb has joined #linux-sunxi
nabblet has quit [Quit: leaving]
leviathanch2 has quit [Read error: Operation timed out]
<paulk-collins> Turl, Goodix-TS 2-0055: gt801 2pulus1 init panel
<paulk-collins> good bet
<paulk-collins> that's on the android image
<paulk-collins> Turl, did you ever manage to open the zatab?
<paulk-collins> I just broke my tools on it…
leviathanch2 has joined #linux-sunxi
<Turl> paulk-collins: yeah, I did open it
<Turl> it wasn't exactly easy though
<Turl> I left some marks on the bezel too
<paulk-collins> ouch ok
<Turl> paulk-collins: aliminium is easy to scratch, esp if you're making force with a screwdriver
<Turl> s/making/applying/
<paulk-collins> I didn't go as far as trying with a screwdriver
<paulk-collins> plastic tools only
<paulk-collins> but they're too weak…
<Turl> paulk-collins: yeah, I first tried with a plastic card
<Turl> no dice
<paulk-collins> that normal?
<Turl> paulk-collins: not really
<Turl> paulk-collins: what are you building?
<paulk-collins> linux-sunxi sunxi-3.4
<Turl> paulk-collins: what toolchain?
<paulk-collins> the android one
<paulk-collins> from 4.2
<Turl> paulk-collins: arm-eabi- ?
<paulk-collins> yes
<Turl> not the androideabi one
<paulk-collins> replicant-4.2/prebuilt/linux-x86/toolchain/arm-eabi-4.4.3/bin/arm-eabi-
<paulk-collins> anyway fixed
<paulk-collins> had to add defined(__KERNEL__)
<Turl> :)
<lioka> on 3.4.79-r0, sun4i (hackberry), it seems only 96M can be reserved for CMA
<lioka> at least for me anything more fails
<lioka> rings any bell ?
<Turl> configuration?
<Turl> ssvb: ^
Black_Horseman has joined #linux-sunxi
Black_Horseman has joined #linux-sunxi
<ssvb> lioka: can you show the initial part of dmesg log with the memory reservation messages?
<ssvb> lioka: have you also tried mostly unmodified sun4i_defconfig?
<lioka> ssvb: well, it fails with 'CMA: failed to reserve xxx MiB' with no additional indications
<ssvb> lioka: there should be a lot of messages about memory reservation for G2D and other things, and also the total amount of RAM in your system and other interesting information
<lioka> ssvb: okee, let me reboot ...
t3st3r has quit [Ping timeout: 240 seconds]
<paulk-collins> Turl, yeah, I confirm the patch works for the zatab :)
<paulk-collins> so cool
<Turl> paulk-collins: great
FreezingCold has joined #linux-sunxi
<ZetaNeta> ls arch/arm/boot/
<ZetaNeta> bootp compressed dts Image install.sh Makefile
<ZetaNeta> Where is my uImage?
<Nyuutwo> ZetaNeta: make uImage
<Nyuutwo> make all doesn't do it
<ZetaNeta> make -j4 ARCH=arm CROSS_COMPILE=arm-linux-gnueabihf- uImage modules
<ZetaNeta> is that good enough?
<Nyuutwo> afaik should be enough
<ZetaNeta> so why it isnt?
<Nyuutwo> ZetaNeta: sunxi-bsp?
<ZetaNeta> linux-sunxi
<Nyuutwo> ZetaNeta: sunxi-bsp?
<ZetaNeta> ?
<Nyuutwo> I have no idea, why it doesn't work
<Nyuutwo> dreadfull up arrow and not changed window
<ssvb> lioka: your kernel config surely enables a lot of stuff, will try to use it and run some tests
<lioka> ssvb: i'm trying with defconfig now
leviathanch2 has quit [Remote host closed the connection]
<ssvb> lioka: but the likely culprit is that something is allocating a chunk of memory between SYS and G2D before CMA can claim it for itself
leviathanch2 has joined #linux-sunxi
hawi has joined #linux-sunxi
<lioka> ssvb: yep, success with defconfig
<ssvb> lioka: it gets handled before dma_contiguous_reserve
<ssvb> lioka: but that's just a guess, this issue still needs to be debugged
<hawi> hi, can anyone explain me why in sun4i_irq_ack() the kernel is or'ing the irq to clear with the previous value of register?
<hawi> is not enough to set only the bit to clear?
<Turl> hawi: link?
<lioka> ssvb: bootm'd without initrd -- cma went ok
<Turl> hawi: this mail and its replies may clarify https://lkml.org/lkml/2013/7/6/59
linkmauve1 has left #linux-sunxi [#linux-sunxi]
linkmauve1 has joined #linux-sunxi
<paulk-collins> is turbofb running without the proprietary mali blobs at all?
kuldeepdhaka has quit [Quit: taataaaaa]
<paulk-collins> when it shows: no 3D acceleration because the driver has been compiled without libUMP
<Turl> mripard: ^ that irq patch looks like never made it
<hawi> Turl: thanks for the reference
<Turl> paulk-collins: fbturbo can run with no blobs, indeed
<paulk-collins> it's quite fast already
<Turl> you just need the blobs if you need GLES
<paulk-collins> so basically, 3D?
<Turl> yeah
<paulk-collins> can I do gnome-shell without the blobs?
<Turl> but just GLES
<Turl> paulk-collins: with mesa llvmpipe thingy, yeah
<Turl> paulk-collins: but be prepared for 0.1FPS shell :)
<paulk-collins> even with fbturbo then?
<paulk-collins> too bad…
<Turl> paulk-collins: the old fallback mode that doesn't need 3D runs fine fwiw
linkmauve1 has left #linux-sunxi ["hg pull -u"]
<Turl> that's shell 3.4 or 3.6, don't recall now
linkmauve1 has joined #linux-sunxi
<Turl> paulk-collins: and even with the blobs you'd need mesa
<Turl> pretty sure shell is GL only
<ssvb> lioka: is your initrd loaded at 0x45000000 as described on http://linux-sunxi.org/Initial_Ramdisk ?
mkutsevol has left #linux-sunxi [#linux-sunxi]
mkutsevol has joined #linux-sunxi
<ssvb> paulk-collins: in theory you can run GLES accelerated gnome-shell, in practice it is still somewhat sluggish
<lioka> ssvb: nope, let me check ..
pacopad_ has joined #linux-sunxi
<Turl> ssvb: oh, there's a GLES version now?
pacopad has quit [Ping timeout: 260 seconds]
pacopad_ is now known as pacopad
<Turl> I bet it's not distro packaged then :)
<ssvb> Turl: I think there is gles accelerated version of gnome-shell in linaro rootfs, at least that's what I tested some time ago
<Turl> ssvb: hm
pacopad_ has joined #linux-sunxi
<ssvb> lioka: anyway, it just looks like initrd grabs a chunk of memory somewhere in the middle between 0x43000000 and 0x50000000
<ssvb> lioka: and because of this CMA can't allocate a large physically contiguous memory area there
pacopad has quit [Ping timeout: 248 seconds]
pacopad_ is now known as pacopad
<ssvb> lioka: we want to have CMA area exactly at that place because CedarX hardware can only work with the first 256MB of physical RAM
pacopad_ has joined #linux-sunxi
deasy has joined #linux-sunxi
pacopad has quit [Ping timeout: 260 seconds]
pacopad has joined #linux-sunxi
<lioka> ssvb: initrd at 50000000 did the trick
<lioka> ssvb: thanks
bertrik has quit [Ping timeout: 252 seconds]
* lioka .oO 2:57am HoC or bed ?
pacopad_ has quit [Ping timeout: 272 seconds]
<ssvb> lioka: you are welcome
pacopad has quit [Ping timeout: 245 seconds]
<ssvb> lioka: now I wonder what would be the best way to resolve this problem, so that nobody hits it again
pacopad has joined #linux-sunxi
<lioka> ssvb: oh, that's simple. document it :]
<ssvb> lioka: :)
AreaScout has quit []
pacopad_ has joined #linux-sunxi
pacopad has quit [Ping timeout: 272 seconds]
pacopad_ is now known as pacopad
pacopad_ has joined #linux-sunxi
pacopad has quit [Ping timeout: 248 seconds]
pacopad_ is now known as pacopad
paulk-collins has quit [Read error: Operation timed out]
hawi has quit [Quit: Leaving]
forcev has joined #linux-sunxi
FunkyPenguin has quit [Ping timeout: 272 seconds]
pacopad_ has joined #linux-sunxi
pacopad has quit [Ping timeout: 272 seconds]
pacopad_ has quit [Ping timeout: 246 seconds]
deasy has quit [Quit: Nom d'un quark, c'est Edmonton !]
pacopad has joined #linux-sunxi
pacopad_ has joined #linux-sunxi
pacopad has quit [Ping timeout: 272 seconds]
pacopad_ is now known as pacopad
pacopad has quit [Ping timeout: 252 seconds]
pacopad has joined #linux-sunxi
sehraf has quit [Read error: Connection reset by peer]
pacopad has quit [Ping timeout: 246 seconds]
pacopad_ has joined #linux-sunxi
pacopad has joined #linux-sunxi
pacopad_ has quit [Ping timeout: 248 seconds]
geecko has joined #linux-sunxi
geecko_ has joined #linux-sunxi
geecko_ has quit [Read error: Connection reset by peer]
geecko has quit [Read error: Connection reset by peer]
bertrik has joined #linux-sunxi
geecko has joined #linux-sunxi
pacopad_ has joined #linux-sunxi
pacopad has quit [Ping timeout: 248 seconds]
pacopad_ is now known as pacopad
pacopad_ has joined #linux-sunxi
pacopad has quit [Ping timeout: 252 seconds]
pacopad_ is now known as pacopad