hno changed the topic of #linux-sunxi to: Allwinner/sunxi development discussion - Don't ask to ask. Just ask! - See | | Logs at
Tartarus_ has joined #linux-sunxi
soul has quit [Ping timeout: 248 seconds]
soul has joined #linux-sunxi
Soru_ has joined #linux-sunxi
soul has quit [Ping timeout: 246 seconds]
Soru_ has quit [Read error: Connection reset by peer]
Soru has joined #linux-sunxi
TheSeven has quit [Quit: No Ping reply in 180 seconds.]
TheSeven has joined #linux-sunxi
egbert has joined #linux-sunxi
egbert_ has quit [Ping timeout: 260 seconds]
Jhon has joined #linux-sunxi
<Jhon> Hi
<Jhon> I'm new to sunxi linux
ZaEarl_ has joined #linux-sunxi
<Jhon> Could somebody tell me how is the status of linux sunxi 3.4 kernel
<Jhon> is it mature enough?
<Jhon> or I should use 3.0 kernel instead
ZaEarl has quit [Ping timeout: 248 seconds]
<Turl> Jhon: yes, 3.4 is recommended
<Jhon> Thank you Turl. Another question: When use 3.4 kernel, do we still need any kind of .fex file? I think .fex file basically doing same work as device tree
<Turl> yes, you need to load script.bin on both 3.0 and 3.4
<Jhon> so this means device tree is not fully supported in sunxi 3.4 kernel right?
<Turl> Jhon: it's not supported at all
<Turl> Jhon: there is sunxi device tree support in mainline, but the hardware support is pretty limited at the moment
<Jhon> ooo, ok. which branch should I checkout?
<Turl> Jhon: for what? 3.4?
<Jhon> yeah
<Turl> sunxi-3.4 if I recall correctly
<Turl> there's also stage/sunxi-3.4 with new changes that are currently being tested, if you want the bleeding edge
<Jhon> Thank you very much, you helped alot
TheSeven has quit [Disconnected by services]
[7] has joined #linux-sunxi
<Turl> you're welcome
<Turl> good night :)
<Jhon> another question Turl: How frequently a new feature in stage/sunxi-3.4 being merged to sunxi-3.4 normally? just curious
<Turl> Jhon: there's no predefined timeframe, usually once they've been sitting there for a while, mnemoc will merge the stage branch back into the non-stage branches, and possibly also merge any patches from mainline stable series
<Jhon> so right now most of the people are working on 3.4 kernel I guess?
<Jhon> good night too
hramrach__ has quit [Ping timeout: 240 seconds]
hramrach__ has joined #linux-sunxi
BJfreeman has quit [Quit: had a good time]
\\Mr_C\\ has quit []
\\Mr_C\\ has joined #linux-sunxi
Jhon has quit [Ping timeout: 250 seconds]
\\Mr_C\\ has quit []
wingrime has joined #linux-sunxi
<wingrime> ssvb: I just trued connect tablet to hdmi
<wingrime> but no success
rm has quit [Quit: ZNC -]
hramrach__ has quit [Remote host closed the connection]
rm has joined #linux-sunxi
hramrach__ has joined #linux-sunxi
\\Mr_C\\ has joined #linux-sunxi
<stekern> I've figured out why fatload from nand doesn't work in aw sun6i u-boot, the internal functions in libnand uses the "wrong" function (they've added some new functions for that) to get the offset to the partition
<stekern> hack to "fix" it:
<rm> wow!!
<rm> VGA, 4 USB
<rm> also seems to have a webcam(?)
<rm> no, that's probably just the IR receiver
eebrah|away is now known as eebrah
rellla has joined #linux-sunxi
n01_away has quit [Quit: leaving]
rellla has quit [Quit: Nettalk6 -]
<oliv3r> rm: most likly; but it's only a10, so how relevant is it still
_BJFreeman has joined #linux-sunxi
_BJFreeman is now known as BJfreeman
rellla has joined #linux-sunxi
BJfreeman has quit [Quit: had a good time]
vicenteH has quit [Ping timeout: 246 seconds]
notmart has joined #linux-sunxi
notmart has joined #linux-sunxi
FunkyPenguin has quit [Read error: Connection reset by peer]
vicenteH has joined #linux-sunxi
Superpelican has joined #linux-sunxi
FunkyPenguin has joined #linux-sunxi
Tsvetan has quit [Ping timeout: 268 seconds]
<hramrach__> oliv3r: there are many a10 devices out there and I don't think a10 is out of stock even if it is no longer manufactured. Don't be like the SoC makers and just stop considering any but the most cutting edge SoC
<oliv3r> heheh, yeah but I wouldn't buy an a10 device unless its REALLY cheap and offers many connections
<oliv3r> i mean, this a10 device is 80 USD
<oliv3r> i can get a20 for less then that :)
<hramrach__> depends on the other goodies, yes
<hramrach__> sound a bit too much for a TV box but if it has some nice wireless keyboard you do not have yet it might be worth it
Black_Horseman has quit [Remote host closed the connection]
<hramrach__> and the TV boxes tend to be more expensive than plain sticks - more connectors, bigger pcb, bigger case, fewer per container so bigger shipping costs
<oliv3r> does it come with the keyboard?
<oliv3r> melee is cheaper
<oliv3r> it doesn't come with a keyboard, just a regular remote
<hramrach__> mele has fewer ports iirc
<hramrach__> and this one has aluminium case frame
<oliv3r> yeah it does look nice
<oliv3r> but a20 for the same price vs a10
<oliv3r> screw the case :)
<oliv3r> mele m5; 87.10 USD shipped
<oliv3r> and that's DX, you can get it cheaper i'm sure from elsewhere
<hramrach__> either of them, sure
<oliv3r> 50 USD for an HDMI stick is not bad
<oliv3r> i might actually even consider one of those :)
<wingrime> hramrach: we have way change video out on the fly?
<hramrach__> wingrime: change video in what sense?
<wingrime> internal tablet VGA to HDMI
<hramrach__> like switch from HDMI to VGA?
<oliv3r> i don't think we can
<wingrime> on android work
<oliv3r> i think we can support 'dual output' with some soc's
<hramrach__> a10 has dual output
<hramrach__> but there are few devices with dual connections
<hramrach__> so not very well tested
<hramrach__> also it might be hard to manage triple/quadruple output. You might need to configure two at boot and then switch between the two
<wingrime> hramrach: a10s lack this?
<wingrime> I think a10/a20/a31 have
<hramrach__> not sure. I only have single output a10
<wingrime> hramrach: my a10 tablet work as screen clone on andoird
<hramrach__> I think a10s has hdmi so it still has dual output but a13 might not
<hramrach__> yes, it's simplest to do that way. both software and user settings side
<hramrach__> but yous hsould be able to get two independent outputs. a10 hardware is capable of that
<wingrime> hramrach: we have any software that can recofigure out on-the-fly
<hramrach__> wingrime: ask somebody with a mele. they have multitude of outputs
<hramrach__> I can't really tell because I did not get to making the hardware for extra outputs
<wingrime> hramrach: without dual output we need simple reconfigure disp registers
<hramrach__> seems it's hardware feasible to have HDMI+LVDS but VGA conflicts with LVDS and TV because they share some pins
<hramrach__> maybe also depends on the pins you have accessible. this is for the pins available on Cubiebaord but not all are on the connectors
<wingrime> VGA+HDMI or LVDS+HDMI are common for tablets
<hramrach__> the internal screen is VGA?!
<hramrach__> or they use HDMI to something convertor to connect the internal screen??
<wingrime> not LVDS defently
<hramrach__> it can be RGB. Probably pre-lvds standard. not differential but also used in laptop screens and supported by a10
<wingrime> wait I check config
<hramrach__> you can fish the cubieboard group for sample of RGB screen. somebody talked about connecting one
<wingrime> screen output type = 1
<wingrime> it strange
<wingrime> but config has lcd0 and lcd1 configs
<oliv3r> vga IS 'tv'
<oliv3r> there's 1 analog output, capable of driving 4 signal pins
<hramrach__> and tv is also tv, presumably
<hramrach__> hdmi seems separate from lcd
<hramrach__> but the lcd sync lines are used for vga
<oliv3r> you can configure them all 4 als composite, 2 svhs, or 1 yuv
<oliv3r> yeah lcd1 sync lines are used for vga
<oliv3r> but if you do 'analog' tv-out, that's not an issue
<oliv3r> that's what the registers/docs etc suggest
<wingrime> hramrach: I make sync code for FPGA
<oliv3r> what really works in the driver is a whole different thing :)
<wingrime> hramrach: it need impulses at and "line" and frame with defened timings
<hramrach__> the cost of the fpga, pff
<hramrach__> better get 2xa10
<wingrime> VGA is R G B analog wires with HS and VS wires for sync (digital)
<wingrime> rgb are same , but signals can be digital
<hramrach__> yes, the sync wires are used from lcd
<hramrach__> yes, and goes over different wires.
forcev has joined #linux-sunxi
<hramrach__> much more of them, too
<wingrime> witch means , you can use it same time if you use same resolution
<hramrach__> for clone, yes
<wingrime> not always clone
<wingrime> but same resolution are mandatory
<hramrach__> heh. the kernel probably does not handle shared timig well
<hramrach__> the drm does not nor does disp. it just happens to work sometimes
<hramrach__> plus feeding two displays off the same sync line is not something that is part of the design so if it breaks you get to keep the pieces
<hramrach__> I guess you want somebody who understands the ee implications comment on that
<wingrime> hramrach: you always can do sync using gpio
<hramrach__> you have a realtime OS?
<hramrach__> linux fails at stuff like this
<oliv3r> linux fails at what? realtime?
<oliv3r> get the RT linux kernel
<wingrime> but if resolution not need big freqency
<hramrach__> it can go oon luch for *seconds*
<oliv3r> duh
<oliv3r> wingrime: what are you trying to do :)
<wingrime> oliv3r: change video out on fly
<hramrach__> understand the mess of the output and create some evil scheme for abusing them I would guess
<hramrach__> oliv3r: does the RT linux kernel support a10?
<wingrime> hramrach: how much speed gpio can do ?
<wingrime> oliv3r: we have timers and irqs
<hramrach__> good enough for sync. iirc it can do in MHz ballpark and the sync lines are in tens of kHZ ballpark so doable
<oliv3r> wingrime: on 3.4? no clue what our status is
<oliv3r> hramrach__: theoretically; sure
<oliv3r> hramrach__: RT is somewhat in sync with ML
<oliv3r> and since we have some ml sunxi stuff; yeah
<hramrach__> yes, in practice you would need hard realtime OS
<oliv3r> wingrime: well the hardware should be capbable, it's up to the drivers
<wingrime> hramrach: we have timers and irqs
<hramrach__> and there is no limit on irq time handling imposed by the kernel
<wingrime> hramrach: 1280x1024 need aporx 100Mhz clock
<hramrach__> which is done with irqs disabled, of course
<hramrach__> wingrime: dotclock, not sync
<oliv3r> i don't think you need to rewrite the disp layer for what you want
<oliv3r> i think the bits and pieces are 'there' allready
<wingrime> hramrach: 100 mhz need to make impluse timings correct
<oliv3r> unless you want to do 1080p analog out via a gpio or two
<oliv3r> then yeha, you'll need to do a lot :)
<hramrach__> wingrime: you want the sync lines which are about 60Hz and 80kHz, respectively
<wingrime> oliv3r: if you have 3channel DAC possible
<hramrach__> and you have blanking around these so there is no need to be pixel-precise
<wingrime> hramrach: it will works but not realy normaly
<hramrach__> but the precision of the sync impacts picture quality, sure
<wingrime> hramrch: we can do video image with gpio and resistor dac even on at mega8
<hramrach__> sure, you can. when the atmega does nothing else
<hramrach__> but you cannot do sound output and ethernet on an atmega at the same time because when an ethernet frame arrives the sound skips
<hramrach__> or so I heard from a very disgruntled hw hacker
<wingrime> hramrach; once I saw device with 8 atmega
<wingrime> 8xatmeaga8
<hramrach__> that's why sound cards have buffers, serial ports have buffers, ethernet cards have buffers, graphics cards have built-in scanout dacs, all independent of the CPU delivering every single bit on time
<hramrach__> when you have 8x atmega you effectively create those independent buffers ;-)
<hramrach__> and yes, the displ layer should be quite capable when dealing with multiple outputs, just don't have any means to test that myself.
<wingrime> anyone tryed make fm-transmitter using single GPIO on a10
rz2k has joined #linux-sunxi
<hramrach__> heh, would be funny to make video sync line in software. every time you write non-trivial amount of data to disk the display blanks :DD
<hramrach__> because Linux MM in recent kernels is junk and it takes 100% cpu time figuring out how to free memory when it's full of dirty buffers and even fails
<wingrime> hramrach: why bliks , you can use timers and non maskable irq
forcev has quit [Remote host closed the connection]
Tartarus_ has quit [Ping timeout: 240 seconds]
\\Mr_C\\ has quit []
jelly-home has quit [Read error: Operation timed out]
FunkyPenguin has quit [Remote host closed the connection]
hurtigbuffer has joined #linux-sunxi
FunkyPenguin has joined #linux-sunxi
<oliv3r> well again, i don't know what you would like to accomplish
<oliv3r> but you can have hdmi out, LCD0 out and analog out simultaniously afaik
\\Mr_C\\ has joined #linux-sunxi
<ssvb> wingrime: you can give a try
<ssvb> wingrime: but I myself only used fex editing to get dual monitor output (vga+hdmi)
Tartarus_ has joined #linux-sunxi
Tartarus_ has quit [Client Quit]
Superpelican has quit [Ping timeout: 276 seconds]
Black_Horseman has joined #linux-sunxi
atiti has joined #linux-sunxi
rellla has quit [Ping timeout: 248 seconds]
Superpelican has joined #linux-sunxi
paulk-desktop has joined #linux-sunxi
vicenteH has quit [Ping timeout: 264 seconds]
hansg has joined #linux-sunxi
<wingrime> oliv3r: a13 have pwm?
hansg has quit [Client Quit]
atiti has quit [Ping timeout: 268 seconds]
rellla has joined #linux-sunxi
<gzamboni> is there something wrong with github? i cant access it ffrom here. im in France
<gzamboni> im getting always: This page is taking way too long to load.
<mripard> gzamboni: I can access it
<mripard> ah
<mripard> well. It's github
<mripard> not very known for its performances and snappiness
<gzamboni> after some refreshes sometimes it loads
<gzamboni> ah, ok: The backend storage is temporarily offline. Usually this means the storage server is undergoing maintenance.
<gzamboni> i will just do it later
<oliv3r> wingrime: yeap, but only 1, a10 has 2
atiti has joined #linux-sunxi
<wingrime> oliv3r: I try make wiki more readable
<wingrime> please not make register guide only for a10
<wingrime> I just try fix it
<wingrime> Controller Pages + Controller Registrty Manual
sud0x3 has joined #linux-sunxi
eebrah is now known as eebrah|away
<oliv3r> wingrime: the wiki needs a lot of work; but
<oliv3r> wingrime: having the register tables is nasty, ideally, we want to have a template, that reads a table of register data, and outputs pretty tables
<oliv3r> wingrime: if you are referring to a13 vs a10 with the PWM, i wrote those pages when we had the a10 user manual, i never read the a13 manual and added those changes
<wingrime> olvi3r: hw very same
<wingrime> olvi3r: we need one page for one deivce with differences in page
<sud0x3> hi folks, my microSD breakout board arrived today. I Have been using it in conjunction with my bus pirate but when i boot my devicce i dont get any console or text comming from the device i have the breakoutboard connected to. I had read a few weeks back that i may need to modify my script.bin to add uart support.
<sud0x3> Could anyone shed some light on editing myu scipt.bin for uart? I have been looking through the fex file myself today and there are allready UART configurations in place
naobsd has joined #linux-sunxi
<wingrime> oliv3r: have we audio hw decoder?
sud0x3 has quit [Quit: Leaving]
<oliv3r> wingrime: if your asking if cedarX can do audio hardware decoding, i don't know. I was hoping we could offload some audio decoding (idct)
<oliv3r> but with neon and thumb, it should be not a lot of burden on the CPU
sud0x3 has joined #linux-sunxi
<oliv3r> wingrime: one page per device also means we have a lot of duplication (in the case of pwm), pwm0 is identical, pwm1 is just 1 additional register
<oliv3r> which means we have to maintain both pages for updates to pwm0
<oliv3r> but we have official manuals now, so the need is lower
<oliv3r> so far i've added each component i'm working on to the wiki
<atiti> did anyone play around with the cedarX h264 decoding?
<wingrime> oliv3r: plese check my fixes
tinti has joined #linux-sunxi
<oliv3r> what happened to the memory map?
<oliv3r> i do like what you did with the disp reg guide
<oliv3r> it's there twice :)
<oliv3r> but the overal memory map remains to be important imo to get an overview of all componetns
<wingrime> memory map in place
<wingrime> strange problem with link
<oliv3r> aye
<wingrime> we have some dups
Tsvetan has joined #linux-sunxi
<rellla> atiti: what do you want to know exactly? there are a few in here, that do some cedarx stuff
Soru has quit [Read error: Operation timed out]
hramrach__ has quit [Ping timeout: 240 seconds]
<drachensun> stekern: So with this are you able to load from an SD card?
<drachensun> or wait this is just so you can use a fat partition on the nand right?
<atiti> rellla: did someone measure the latency of the decode?
<rellla> atiti: don't believe so.
bsdfox has joined #linux-sunxi
bsdfox has joined #linux-sunxi
<stekern> drachensun: right, that enables me to load from a fat partition on the nand
<stekern> which is very useful for me, since I have the uart attached to the sdcard slot
hramrach__ has joined #linux-sunxi
<wingrime> oliv3r: how much current stage fresh
<wingrime> did amely aproved
<wingrime> ?
jemk has joined #linux-sunxi
<wingrime> jemk: I have moved VE page to new place
<wingrime> jemk: And edited
<sud0x3> im having some issues gettin uart out through my micros sd card slot
<sud0x3> I have tried changing variables in the scipt.bin but still i get nothing comming out on the breakoutboard
<stekern> sud0x3: did you edit like it says on this page?
<sud0x3> yes i converted the file to fex and modified the parameters, then i converted the fex back to bin, mounted nanda as a folder on my sdcard, removed old script.bin replaced with new script.bin
<sud0x3> now my device wont even boot i will flash back to stock using livesuit and try again
rellla has quit [Ping timeout: 248 seconds]
_BJFreeman has joined #linux-sunxi
_BJFreeman is now known as BJfreeman
jlj has joined #linux-sunxi
<jlj> I tried to use the latest 3.4 kernel on an A10 with a 1280x1024 monitor and got this message (and a blank screen) disp_clk: Could not find a matching pll-freq for 26150000 pclk
<jlj> is that a known issue?
<jlj> seems like the edid code worked otherwise
<wingrime> you can avoid using edid
<wingrime> just configure cmdline for it
<jlj> would that bypass the disp_clk error?
<jlj> I'm not really sure what it means..
\\Mr_C\\ has quit []
Black_Horseman has quit [Quit: Zwi se logou mou!!!]
<wingrime> jlj: if it related monitor bad edid can help
<wingrime> otherwice I have no idea
<wingrime> you simply can avoid edid
<wingrime> jemk: ping
<jemk> wingrime: pong
paulk-desktop has quit [Quit: Ex-Chat]
<wingrime> jemk:S 0401d118[ 118] 4: 8400300d 0400300d
<wingrime> jemk: 4 means - mpeg
<wingrime> but "8"
<jemk> other engines set this bit too, but i dont know what it does
<wingrime> any idea 8 = 1000
<wingrime> mpeg4 use 0xd at trigger reg for start
<wingrime> jpeg use 0xe to trigger
<wingrime> do you saw differenet at 0x118 first bits?
<jemk> yes, ms-mpeg sometimes uses 0x5 and 0x7, but i had no time to figure out any idea
<wingrime> jemk: I thik this means how much actions do, for example 0xe may be can mean VLD->IQ->IDCT
<wingrime> jemk: VLD->IQ-IDCT->IS
<wingrime> jemk: or how much blocks do
<wingrime> jemk: full frame or just single mc
<jemk> maybe, but do you find any useful relationship?
<jemk> last two lines: don't think so
<wingrime> jemk: now looking for in blob
<jemk> btw, do we have any other name for reg 0x138, it must be some buffer address like dcacaddr for 0x13c
<jemk> i think the vld_offset and vld_len are wrong, someone mixed this up with h264 engine
<wingrime> just second
<wingrime> i take a look
<wingrime> 138 are mp4vldoffset
<wingrime> try look more one if present
<jemk> mp4->h264 engine i'd guess, there it would fit
<jemk> for mpeg engine it must be a buffer address
<wingrime> I see second name
<wingrime> mbhaddr
<jemk> ah, that would make sense
<wingrime> macro bock header address?
<wingrime> jemk: you can add to wiki
<jemk> maybe, 0x138 0x13c and 0x144 are some temporary buffers for mpeg4 decoding
<wingrime> but don't forget update link in page top
<jemk> and now all have some name with addr, so it fits
<wingrime> jemk: I don't know what mean "ncf"
<jemk> i'd guess 0x140 is such an temp buffer too, but didn't see it being used yet
<wingrime> jemk: also add to wiki 13c - dcacaddre
<jemk> doing at the moment...
drachensun has quit [Ping timeout: 264 seconds]
paulk-desktop has joined #linux-sunxi
<wingrime> jemk: also 0x140 not realy used in blob (I think it try but not 100% sure) only only one reference to init procedure blob writes 0 to it
<wingrime> jemk: it looka DC and AC IDCT matrices coefficents
<wingrime> jemk: but you must know that mpeg have different coding methods
<jlj> are you trying to reverse engineer the blob?
<wingrime> jlj: yes
<wingrime> jlj: we have some sucess in it
<wingrime> jlj: working jpeg decoder
<jemk> jlj: but i dont look at the blob, only by tracing
<jlj> cool
<wingrime> jlj: I not write code, I only write manual
<wingrime> jlj: "clean room reserver ingenering"
<jemk> mpeg12 could work too, but i dont have time, writing exams next weeks
<jlj> ah, gotta do it right :)
<jlj> do you have motion jpeg working or single images?
<wingrime> jlj: singe images, but mjmpeg same
<jemk> should be easy to extend to mjpeg
<wingrime> jlj: we should fist decide witch decoder hw interface implement
<wingrime> jlj: beyond this point all this PoC
<wingrime> jlj: also will be perefect find some other guy who will write that interface only using our code examples and manual
<wingrime> jlj: 100% clean room
Superpelican has quit [Read error: Connection reset by peer]
Superpelican has joined #linux-sunxi
<wingrime> jemk: mpeg4 use only 0xf for 118 first bit
<wingrime> jemk: have you tryed vp6 examples?
<jemk> wingrime: no, i have 0xd too and msmpeg v2 also uses 0x5 and 0x7
<jemk> vp6, no, i dont have time ^
<wingrime> vp6 have "5" code number
<wingrime> on mode select
<wingrime> jemk: vp6 according blob vp6 will use 0b1111 for first bit for 0x118
drachensun has joined #linux-sunxi
sud0x3 has quit [Quit: Leaving]
Superpelican has quit [Ping timeout: 252 seconds]
Black_Horseman has joined #linux-sunxi
Black_Horseman has joined #linux-sunxi
<oliv3r> anybody know of any reasonable camera that runs linux (and is hackable?)
<Turl> oliv3r: galaxy zoom? :p
<oliv3r> *facepalm*
<Turl> galaxy s4 zoom*
<oliv3r> i know
<oliv3r> that runs android
<oliv3r> and costs a fortune
<Turl> oliv3r: you didn't ask for cheap :P and technically it runs linux
<oliv3r> ya ya
<Turl> this new debian service rocks
<oliv3r> what about it?
<Turl> oliv3r: you can read the code of all the software packaged in debian
<Turl> with syntax highlighting
<oliv3r> true
<Turl> without downloading anything
<Turl> and it's also searchable :P
<oliv3r> :)
Superpelican has joined #linux-sunxi
<Turl> torvalds started merging for 3.11 :)
<Turl> oliv3r: what's the status of sid driver?
<oliv3r> need to figure out how to do stuff
<wingrime> jmek: ping
<wingrime> jemk: ping
<wingrime> turl: can you block spamer by IP?
Superpelican has quit [Quit: Konversation terminated!]
<wingrime> turl: can you block spamer or browser string
<wingrime> *using
<Turl> wingrime: yes, when you block account you get chance to block that ip
<Turl> wingrime: I dunno about that though
<wingrime> Turl: How about see his version string
<wingrime> *browser
<wingrime> also new spamer registred
<Turl> I don't think wiki registers that
<wingrime> Turl: web server have access log
<wingrime> Turl: just grep it using IP
<wingrime> Turl: apache can block , I think
Tsvetan is now known as TsvetanAway
<Turl> wingrime: we're using nginx
<Turl> wingrime: I cannot see the ip on mediawiki
<wingrime> Turl: strange
<wingrime> just test before do it on production server
<wingrime> Turl: ip must be in mediawiki DB
<Turl> yeah but I'm lazy :)
<Turl> mnemoc: what do you think of using a ip block list?
hurtigbuffer is now known as jelly-home
jlj has quit [Ping timeout: 252 seconds]
<atiti> hm
<atiti> anyone knows if HDMI CEC is supported?
<Turl> atiti: there was some effort on that, but I don't think it has support for any userspace libraries yet
<atiti> oh nice, but theres a kernel driver?
<atiti> sweet thx :)
^hawk^ has quit [Ping timeout: 241 seconds]
jlj has joined #linux-sunxi
^hawk^ has joined #linux-sunxi
dwilkins has quit [Ping timeout: 252 seconds]
wingrime has quit [Ping timeout: 268 seconds]
rellla has joined #linux-sunxi
notmart has quit [Quit: notmart terminated!]
jlj has quit [Ping timeout: 246 seconds]
^hawk^ has quit [Remote host closed the connection]
^hawk^ has joined #linux-sunxi
<mnemoc> Turl: tomorrow is my last day in spain.... not really in conditions to decide anything strategic atm. do whatever you feel better
<Turl> mnemoc: ok
<mnemoc> but next week i'll get a new server where to migrate everything (4x the current and paid by friend ad-eternum). we can discuss how to layout things better then
<Turl> mnemoc: all cleared on that? :)
<mnemoc> hetzner been annoying with paperwork because it seems to be their first .cl customer
<mnemoc> but it's confirmed to happen as soon as hetzner decides to stop acting like a d*ck
<Turl> mnemoc: :P great
<Turl> I have augmented the ip block a bit to block vandals using proxies and enabled some DNSBL filtering on wikimedia
<Turl> let's see how they react to it
jlj has joined #linux-sunxi
^vincenzo^ has joined #linux-sunxi
^hawk^ has quit [Ping timeout: 268 seconds]
<oliv3r> mnemoc: a nother migration?
jlj has quit [Ping timeout: 246 seconds]
<mnemoc> oliv3r: yes. the current server is paid (by me) until july 26. the next server is for the webshop of an school friend, and allowed us to do whatever we want with the hardware as long as it doesn't affect his ~150 product shop with a 100E/M budget
<mnemoc> the machine has 4x1.5TB, i7 980 and 24GB of ram
<mnemoc> plenty for his shop, the sunxi server and a "limited" build server
<mnemoc> and free
<mnemoc> building that shop will all so be first $work$ after leaving .es
<mnemoc> but that's another story
<mnemoc> moving the lxc guests shouldn't be problematic
<mnemoc> and we will be able to get rid of that f* 3xraid1 + btrfs
<mnemoc> in favour of raid10 + reliable fs
<mnemoc> oliv3r: i have to admit i didn't expect to need to migrate it again, but it's a good oportunity to do things better this time
<oliv3r> buildserver :D
<mnemoc> as long as it's not android, sure
<oliv3r> :(
<mnemoc> Turl has a personal android build server monster at home
<oliv3r> actually, i'm building android myself here on my desktop and it works quite well
vincenzo has joined #linux-sunxi
<mnemoc> my amd e-350/4GB/120GB can't do that
<Turl> mnemoc: double your ram and you'll be okay :P
<Turl> 4x even better
<mnemoc> doesn't support more
^vincenzo^ has quit [Ping timeout: 240 seconds]
<oliv3r> yeah i only have 2x4 in my desktop now
<oliv3r> i really should add more
<oliv3r> but then i don't build often on my desktop
<oliv3r> my servers also both each have 8gigs
rellla has quit [Remote host closed the connection]
<Turl> oliv3r: I have 4x4, 8g sticks were/are expensive over here
<oliv3r> but i wanna put a build 'disk' in there soley for android and other source stuff
<oliv3r> yeah
jemk has quit [Ping timeout: 268 seconds]
BJfreeman has quit [Read error: Connection reset by peer]
vincenzo has quit [Remote host closed the connection]
vincenzo has joined #linux-sunxi
deasy has joined #linux-sunxi
deasy has quit [Client Quit]
\\Mr_C\\ has joined #linux-sunxi
Turl has quit [Quit: >.<]
Turl has joined #linux-sunxi
<oliv3r> i bet nobody spotted that!
<Turl> oliv3r: A3X? :o
dwilkins has joined #linux-sunxi
<oliv3r> several revisions
<oliv3r> mripard: i know nothing of pinmuxing, but need some help :)
<oliv3r> i assume that this is done on all sunxi devices, you write some value to some register to setup a few pins?
<oliv3r> this is what i'm really after though
<oliv3r> so if i look at our public 3.4 source, it should be the pinmuxing stuff (if we even have that there, otherwise mainline?)
jlj has joined #linux-sunxi
<oliv3r> i bet that should tell me all i need to know
bsdfox_ has joined #linux-sunxi
bsdfox_ has joined #linux-sunxi
bsdfox has quit [Ping timeout: 256 seconds]
paulk-desktop has quit [Quit: Ex-Chat]
rz2k has quit []
naobsd has quit [Quit: Page closed]
tinti has quit [Remote host closed the connection]
atiti has quit [Ping timeout: 240 seconds]