IRC log for #arm-netbook on 20121017

00:01.06lundmanmoaning
00:08.50lundmanahh on the 3.0.8+ kernel boots
00:14.27lundmanbut without nic.. guess some script.bin differences
00:22.14rz2kalcides: 0.5fps on opengl swrast
00:22.49lundmanrz2k: cool, got a binary I can try? or upload to my ftp
00:23.08*** join/#arm-netbook ping__ (~ping__@125.162.229.135)
00:23.16rz2klundman: standard xbmc from linaro and new libs from Tom for oGL ES
00:23.31rz2kI will update wiki/Mali400 soon
00:23.35lundmanoh we got new libs?
00:23.38rz2kyeah
00:23.41lundmanniice
00:23.47rz2khttp://dl.linux-sunxi.org/mali/
01:01.25*** join/#arm-netbook ping__ (~ping__@125.162.229.135)
01:01.27*** join/#arm-netbook specing (~specing@unaffiliated/specing)
01:13.34*** join/#arm-netbook specing (~specing@unaffiliated/specing)
01:21.45*** join/#arm-netbook hp__ (~kvirc@118-168-195-38.dynamic.hinet.net)
01:21.45*** join/#arm-netbook discopig (~Sv@unaffiliated/sv)
01:21.45*** join/#arm-netbook lerc (~quassel@121-74-231-26.telstraclear.net)
01:21.45*** join/#arm-netbook akaizen (~akaizen@c-67-180-86-61.hsd1.ca.comcast.net)
01:22.41lundman#define DMA_CPU_TRRESHOLD 2000
01:22.43lundmanamusing
01:24.31*** join/#arm-netbook gsilvis (~almostsix@50.12.163.241)
01:44.59lundmanbah, can not get the NIC to come up with replacement kernel, there sure is some black magic there
01:54.47lundmanrepalce script.bin and it works
01:56.04lundmanSo, ultimately, even though allwinner A40 looks like good hardware, we would still have the situation that they don't provide linux kernel, nor drivers for anything?
02:19.10*** join/#arm-netbook Gumboot (~gumboot@rev.bovine.muck.net.nz)
02:40.37*** join/#arm-netbook mSquare (~selvan@122.178.210.195)
02:55.40Turlor worse, they provide a arch/arm/mach-sun6i/built-in.o
02:59.25lundmanjust thinking that even with A40, do I want it, when it is a hassle to work with. When there are similar boards with great source support
03:01.41lundmanwell, replaced mele-v1.3 android kernel with mine which has NIC fix, and zfs
03:10.04Turlon this price point?
03:10.24*** join/#arm-netbook tekzilla (~jon@hmbg-4d06a157.pool.mediaWays.net)
03:10.25TurlI mean, we could all get a pandaboard, it has source and all
03:10.31Turlcosts 5x of an A10
03:12.51lundmanfor me at least, the prices is irrelevant if the hardware can't do what I want anyway, right? So what if it is cheap
03:13.23lundmanodroid-x maybe
03:24.45lundmanI wonder which bit in the script.bin is it that makes the NIC from not working
03:25.59Turlemac_used? :P
03:28.41lundmanonly diffs in emac area are:
03:28.56lundman-emac_power =
03:29.04lundmanso one doesn't have it, and one has it empty
03:29.52lundmanhttp://pastebin.com/2FRpfUeJ
03:30.55lundmani changed the display to 1080p, which was a mistake, need to change it back :)
03:31.04lundmantheir android stuff did not like me doing that
03:35.26Turllundman: the power stuff if 0'd https://github.com/linux-sunxi/linux-sunxi/blob/sunxi-3.0/drivers/net/sun4i/sun4i_wemac.c
03:35.33Turlis if0'd*
03:35.50Turl--- mele-1.3-script.fex 2012-10-17 00:48:43.000000000 +0000 +++ mele-a1000.fex      2012-10-17 00:49:06.000000000 +0000
03:35.52Turlwhich one works?
03:35.53lundmanyeah I know, actually tried '1' as well
03:36.21lundmanmele-1.3-script.fex works with *their kernel, but my kernel does not NIC link-up
03:36.35lundmanmele-a1000.fex works with my kernel and NIC link-up
03:37.36Turllundman: does your kernel have the sun*i ugly gpio thingy enabled?
03:37.53lundmanusing default sun4i_crane_defconfig
03:38.26Turllundman: then edit the script and not the fex, do the gpio_para changes on your nonworking stuff
03:38.28TurlI bet it works then
03:39.01lundmanyou say, change the gpio_para in mele-1.3-script ?
03:39.46Turlchange gpio_num to 3 and nuke the line talking about the 4th
03:39.53Turland I bet it works on your kernel too
03:40.12lundmangreat, good hints, will try
03:43.32*** join/#arm-netbook ping__ (~ping__@125.162.229.135)
03:49.43Turlthese FirstSteps on the wiki look wrong
03:49.44TurlNote that the boot partition should start at cluster 2048 (1024MB).  Also note that this partition can have the linux partition type, we will  format it to vfat later. You might want to make your boot partition  bigger though.
03:49.58TurlI doubt I need to start my partition on a 1GB offset :)
03:50.09Turluboot env ain't that big :P
03:50.26Turllundman: any luck?
03:50.52lundmanwill try after I check zfs works ok
03:51.06lundmanzfs dont run without /etc/mtab which is amusing
03:52.50Turl/etc/what?
03:56.56*** join/#arm-netbook avernos (~avernos@222.128.157.156)
03:56.57*** join/#arm-netbook avernos (~avernos@unaffiliated/avernos)
04:06.33*** join/#arm-netbook cheng (~cheng@219.92.77.39)
04:35.18lundmanand of course root is readonly
04:36.12*** join/#arm-netbook cheng (~cheng@219.92.77.39)
05:03.45*** join/#arm-netbook mikey_w (~mike@c-71-63-115-202.hsd1.va.comcast.net)
05:03.59*** join/#arm-netbook n6pfk (~mike@c-71-63-115-202.hsd1.va.comcast.net)
05:05.58lundmanok, I am satisfied with the zfs stability under mele android
05:07.44lundmanheh ok, I should remove the debug prints from the kernel
05:21.26*** join/#arm-netbook ping__ (~ping__@125.162.229.135)
05:37.08lundmanAfter you manually mount something in android, anyone know what is needed to tell the UI that it is there?
05:37.25lundmanmust be some combination of dbus, uevent and vold-fstab
05:46.52*** join/#arm-netbook ping__ (~ping__@125.162.229.135)
05:56.53*** join/#arm-netbook wingrime (~alex@78.109.115.103)
06:04.19Turllundman: you don't just 'mount something manually'
06:04.25Turllundman: you use vold to mount
06:05.32Turland dbus is there just for bluetooth as far as I know
06:06.11lundmanyeah I mean its mounted, but android UI dont know that
06:06.32lundmannormally vold gets the uevent for new device, then mounts it, then sends the event "volume_mounted"
06:06.38lundmanI just dont know what happens after that
06:06.54Turlandroid talks to vold
06:07.17Turlso when vold mounts something, it tells android 'br0 I mounted /sdcard' then android knows that
06:08.11Turlon ics/jb there's an overlayable xml listing the mounts and its characteristics on the framework too
06:08.48lundmansure, both talk over /dev/socket/vold
06:09.02lundmanand the message is literally "volume_mounted: /mnt/extsd"
06:09.13lundmanjust not seeing any tools to make that event by hand
06:09.29Turlwhy don't you just mount it with vold? :P
06:09.34lundmanannoyingly hard to google for android crap, always 5 million morons wondering how to see their SD card
06:09.36Turlyou're not supposed to make random events by hand
06:09.44lundmanvold dont really know zfs
06:09.56Turlmake it know it? :)
06:10.13lundmanstop making my 5 mins project that took 3 days into a 1 week project
06:10.34Turlwhy do you need android to know it's mounted anyway?
06:10.43lundmanso the filemanager can see the pool
06:10.56Turlfilemanagers don't care as long as the permissions are right
06:11.05Turljust navigate to it
06:11.07lundmanfilemanagers only show "mounted volumes"
06:11.17lundmanwhich currently is "/mnt/sdcard"
06:11.31lundmanthe built in one I mean, not those smart enough to go outside that
06:11.37Turllundman: https://play.google.com/store/apps/details?id=org.openintents.filemanager&hl=es
06:11.51lundmanno google market on mele now :)
06:12.02lundmanremember, 5 mins project!
06:12.08Turlhttp://openintents.googlecode.com/files/FileManager-1.2.apk
06:12.12Turlthe power of open sauce
06:12.14lundmansure, IIm on the 3rd hour
06:12.26lundmanbut its actually something I would like to know anyway
06:12.35lundmanI'll just stackoverflow it
06:13.10Turlthe Right Way (TM) is making vold zfs aware
06:13.45Turlhere's ntfs support for example https://github.com/CyanogenMod/android_system_vold/commit/ce7457dfe9706b24bbfe98af82adceab50c2198e
06:20.00*** join/#arm-netbook Quarx (~Quarx@46.233.210.49)
06:22.03lundmanyes, but when you import pools, zfs mounts the filesystems
06:22.17lundmanso its a different way to handle it, there is no "mount" command for zfs, normall
06:22.21lundman+y
06:23.51lundmananyway, its on stackoverflow
06:23.58lundmanright, next was hmm
06:24.10lundmanturl: hey, can you get the apk url for bubbleupnp?
06:24.15lundmanactually, how do you get the apk urls
06:26.23*** join/#arm-netbook ZaEarl (~malmrose@17.host-182-48-148.compassnet.co.nz)
06:26.59lundmancool, your FileManager goes into zfs file, and plays my person.of.interest sample smoothly
06:33.04*** join/#arm-netbook pawel5870 (~pkarpins@pc.193147149.ip.amg.net.pl)
06:36.51Turllundman: bubbleupnp is proprietary, that'd be warez :)
06:37.03oliv3rmornin'
06:37.10Turlnight oliv3r
06:37.13TurlZaEarl: ohai
06:37.19oliv3rsleep well turl :)
06:37.29lundmanreally? even the free ad version?
06:37.42oliv3rany android users here by chance? got 2 extremly android specific versions :)
06:37.54Turllundman: well nobody cares for the 'free ad version' on the 'distribute apks illegally' business :P
06:38.04oliv3rwell 1 android question :p; 1 boot question
06:38.12Turloliv3r: I'm one, lundman is another :P
06:38.16lundmanyeah but I just wanted a quick way to get the upnp app to do streaming tests :)
06:38.30Turllundman: I heard skifta is better btw
06:38.33lundman720p/1080p are playing again anyway
06:38.35lundmanskifta?
06:38.40lundmanok send me that apk :)
06:39.03oliv3ri have a cheap clone 7" tablet, which should pretend it's a galaxy nexus 7, phone.apk IS installed, i know with CM7 (on my phone) the phone apk lets you use SIP to make calls. The tablet has phone.apk, but no icon nowhere
06:39.18*** join/#arm-netbook shivansps (ba3a01a8@gateway/web/freenode/ip.186.58.1.168)
06:39.37Turllundman: http://www.google.com.ar/search?q=skifta+1.56.03
06:39.43Turlchoose your poison :P
06:39.47TurlI mean, distributor :P
06:39.50Turlit's a free app though
06:40.03Turlhttps://play.google.com/store/apps/details?id=com.skifta.android.app
06:40.08lundmanwonder if its by swedes, due to the word skifta
06:40.13shivanspshi
06:40.31Turloliv3r: you menan the dialer? that's contact(s?).apk
06:40.42Turllundman: it's by qualcomm & atheros
06:40.53oliv3ris it the dialer? yeah i am missing dialer.apk
06:41.02oliv3rbut do have phone.apk? weird
06:41.21Turloliv3r: phone.apk isn't the stuff where you punch the numbers and hit call
06:41.28oliv3rthe ability to make sip calls, is that a CM7 thing, or was that in the later original dialers?
06:41.54oliv3ri know i can install several sip clients via f-droid etc
06:41.58Turlmaking sip calls is a system level ability, it has to be enabled during build on the xml overlays
06:42.18oliv3rah, see, that is interesting info :) i'll work that in when i setup my own cm tree
06:42.27shivanspsi wonder why allwinner a10 netbooks does not have hdmi lol
06:42.32TurlCM has it enabled
06:42.46Turloliv3r: there's really good 3rd party clients to do sip though
06:42.55Turlwith more features than the built in one even
06:42.57Turltry sipdroid
06:43.05oliv3rwell this rom doesn't have it included, i recon i can just download any of the CM10 zips and extract it from there?
06:43.28Turlit's not a magic app you can just replace
06:43.56Turloliv3r: https://code.google.com/p/sipdroid/downloads/detail?name=Sipdroid-2.7.apk&can=2&q=
06:43.57oliv3ri did find the quality of calls in dialer abmissal
06:44.08Turlhttps://code.google.com/p/sipdroid/downloads/detail?name=Sipdroid-2.7.apk&can=2&q= *
06:44.09oliv3ryeah, tthere's several in f-droid's repository :)
06:44.21oliv3r(f-droid == opensource market)
06:44.33TurlI somehow fail too much at ~4AM
06:44.33Turlhttps://sipdroid.googlecode.com/files/Sipdroid-2.7.apk
06:44.39oliv3rlol
06:45.01oliv3rbut if there's no dialer installed; might aswell install 3rd party sip client indeed
06:45.17shivanspssomeone knows if there are any know problem of why i cant boot a kernel never than 3.0.36 on a A10 tablet? i hear the problem where on mk802 too
06:45.36Turlshivansps: what are you using to boot it?
06:45.46Turlyou might need to enable the ram hack for old bootloaders
06:46.29lundmanodd, the bootanimation on the mele just started again
06:46.35shivanspsyou mean ATAG_MEM option? i tryied that too
06:46.39lundmankernel has been up 13612 seconds
06:47.11TurlCONFIG_SUNXI_IGNORE_ATAG_MEM = y
06:50.06*** join/#arm-netbook mSquare (~selvan@122.178.210.195)
06:50.25lundmanah skifta is a upnp mediarenderer, not client
06:51.01Turlisn't it the same thing?
06:51.10Turlgets puzzled at UPNP buzzwording
06:51.44lundmanI've not tried skifta
06:52.03lundmanbut bubbleupnp is for finding mediaservers on your network, then playing that media on local device
06:52.23lundmaneyecon, mediarenderer finds other mediaserver, and clients, then you can push videos to remote devices to play
06:52.36lundmanalthough, often you can also play locally
06:52.42lundmanso I guess its a superset
06:53.18Turlnow has brain buzzworditis
06:53.19shivanspsmmm its enabled, but no... 3.0.42 and 3.4.12 does the same thing, they power on USB devices, but there is not signal on HDMI neither it load rootfs
06:53.41Turlshivansps: did you enable disp, hdmi modules?
06:53.49Turlhave you tried the other outputs?
06:53.56TurlI think by default it used vga?
06:54.04Turllundman: probably knows better
06:54.48Turlif it's powering usb devices it should be fine then
06:55.00lundmanthe script.bin file says what display to use and resolution
06:55.06shivanspsits a tablet not the mele, but im very sure it does not load rootfs because if it wheredoing so pendrives should turn on the light because it gets mounted
06:55.50shivanspsalso wifi should be connected too
06:56.06Turlmaybe it's your mali version?
06:56.08lundmanbootanim is soo obnoxious
06:56.27Turlmake sure the kernel one and the one on your android matches
06:56.30lundmangpio changes in script.bin did not fix it :)
06:57.05Turllundman: why don't you use the good script bin then? :P remember, 5 min project! :P
06:57.13Turlit even has faster mali clock if I saw right
06:57.18shivanspsactually i just need some extra modules, there is some way i can build the 3.0.36 ones?
06:57.33lundmani did, but I changed res :) I'll add that emty power thing
06:58.05Turlthe wemac driver doesn't use it lundman, it's #if0'd
06:58.11shivanspswell ill like to use 3.4 kernel as well :S
06:58.13Maqsdoes anybody know whether a cp2102 with rx/tx/gnd connected works as uart-to-usb for the a10? ;) i'm not quite sure about the levels
06:58.25lundmani know i know.. but we also left the realm of sanity 2 hours ago
06:58.28Turlshivansps: you should be able to git reset to the .36 hash and build whatever you need
06:59.41shivanspsok ill check it tomorrow them, im too tired now lol
06:59.50TurlMaqs: apparently it's 3.3v so it should work I'm guessing
07:00.05Turlconfirm with the others though :P it's 4AM over here and I need sleep
07:00.09*** join/#arm-netbook sspiff (~quassel@unaffiliated/yukito)
07:00.19Turlnight guys
07:00.34lundmancheers
07:01.07Maqsthx turl and good night
07:01.29*** join/#arm-netbook sspiff (~quassel@unaffiliated/yukito)
07:02.05*** join/#arm-netbook dfletcher_ (~fletch@108-196-222-251.lightspeed.sntcca.sbcglobal.net)
07:05.01mnemocthere are sunxi-$version-r$rev tags in the kernel tree...
07:05.29mnemocthe latest -r$rev is just before the version jump
07:05.52mnemocand -r0 right after
07:06.11*** join/#arm-netbook ssspiff (~quassel@unaffiliated/yukito)
07:11.53relllawith the newest kernel from git master branch i get the following error, when doing a modprobe mali:
07:12.36relllamali: use config clk_div 3
07:12.37relllamali: clk_div 3
07:12.37relllaMali: mali clock set completed, clock is 128000000 Mhz
07:12.51relllaafter that system freeze
07:13.26mnemocuhm
07:13.37relllacould ther be a problem with the clock.c sata fix?
07:13.37mnemocand when was the last it worked?
07:13.53mnemocit shouldn't affect, but...
07:14.09mnemoccan you test once with that change reverted?
07:14.26relllai will build one with the old clock.c and report
07:14.37mnemocthanks
07:14.38Maqsokay, that's what i got with the non-booting kernel: http://dl.dropbox.com/u/11244516/a10-boot.txt
07:15.35Maqscp2102 works fine :-)
07:15.52mnemockernel BUG at mm/bootmem.c:275! .... wth
07:18.14Maqsthat's what i'm thinking ;)
07:19.08mnemocuSD + latest u-boot from https://github.com/linux-sunxi/u-boot-sunxi ?
07:19.58mnemocor nand or older u-boot?
07:20.18Maqsuboot from https://github.com/hno/uboot-allwinner.git
07:20.29Maqsand uSD
07:20.44*** join/#arm-netbook ping__ (~ping__@125.162.229.135)
07:21.56mnemocMaqs: what device?
07:22.16Maqsmk802
07:22.56shivanspshttp://www.ebay.com/itm/10-Android-4-0-A10-1GHZ-4GB-4G-Tablets-mini-Laptop-10-Inch-Netbook-Notebook-/300767889620?pt=LH_DefaultDomain_0&var=&hash=item460729c8d4 I wonder why they dont use the full potencial of the A10, hdmi, sata... al wasted...
07:24.03mnemocMaqs: there is no board definition in u-boot for it.... when was the last it worked?
07:24.36shivanspsoh, btw, i should use a newer u-boot if i have problems to boot kernels never than 3.0.36?, even if the ATAG MEM optin is enabled?
07:24.57Maqsthe previous image worked, but i did not compile it, it came with the image
07:25.57mnemocMaqs: U-Boot 2012.04.01 .... can you try a newer one?
07:26.14mnemocshivansps: a paste of the boot log?
07:26.45Maqsi'll try it with uboot from https://github.com/linux-sunxi/u-boot-sunxi
07:27.05mnemocMaqs: you'll need a board definition
07:27.21Turlshould really sleep
07:27.27mnemocMaqs: example: https://github.com/linux-sunxi/u-boot-sunxi/commit/0c38834e76b99446970f14bf0e23a1b374b0a109
07:27.34mnemocTurl: indeed
07:27.36Turlif you get random failures it's probably a misconfigured clock
07:28.02TurlI saw stuff crash on the weirdest places when playing with clocks :P
07:28.13Turlnow for real, good night :)
07:28.18mnemocsleep well Turl
07:28.54Maqsgn8 Turl
07:28.57relllamnemoc: ok. there's possibly a bug in that commit escpecially on the mele a2000
07:29.18relllawithout that commit, mali driver loads correctly
07:29.26mnemocrellla: f* :<
07:29.59mnemoccan you reply the corresponding thread on the ML?
07:30.11relllai'll do
07:30.14shivanspssorry no uart for now... i still waiting to arrive, i need it to debug and figureout linux boot on a JZ4770 device with it too
07:30.33Maqswhat is this fex2bin-stuff all about? i found https://github.com/Miniand/mk802-u-boot-config
07:31.24mnemocMaqs: script.fex describes your board. like ... DT
07:31.51mnemocMaqs: unfortunatelly that file from miniand is useless for u-boot. https://github.com/Miniand/mk802-u-boot-config/blob/mele/script.fex#L67    their [dram_para] is empty
07:32.26Maqsoh, okay
07:32.54mnemocotoh it's good enough for kernel purposes (enabling pins and features)
07:33.01shivanspsbut i use miniand images on my A10, because im lazy :P 3.0.36 works, but anything never, like 3.0.42 or 3.4.12 it gets stucks somewhere in the boot process... USB devices turns on but thats it, no hdmi signal and the system is stuck because no wifi or usb storage gets mounted
07:33.17mnemocMaqs: please boot from nand and stop by u-boot
07:33.24mnemocMaqs: to make a dump of boot1
07:33.59*** join/#arm-netbook popolon (~popolon@og-free.planet-service.fr)
07:34.22mnemocshivansps: miniand has never cared to collaborate. no idea what problems they have or what changes they've done
07:35.12mnemocshivansps: without a serial console it's hard to find what's going wrong
07:36.43mnemocrellla: can you try with the original patch that started that thread?
07:38.16shivanspsyeah i know, ill have to wait until my usb uart arrive, still it will not hurt to try with a never u-boot...
07:39.03shivanspsi guess ill need a newer evb.bin, bah scritp.bin
07:39.18relllamnemoc: i will do, but something later ....
07:39.38mnemocshivansps: to use a newer u-boot you need a board definition on it
07:39.49mnemocshivansps: what device you said?
07:40.02shivanspsnovo 7 aurora
07:40.14shivanspsi think there is one for the elf araound
07:40.58mnemocfor u-boot needs, it would only work if the have exactly the same dram
07:41.20mnemocshivansps: also, when replacing u-boot it's always useful to clear the env too
07:41.53rz2krellla: what revision is your A10?
07:42.31relllaon the mele there is a little paper with V1.3. is this correct?
07:42.48rz2knope, first letter of code printed on A10
07:43.19rz2kthere is A, B and C revisions of silicone, code from A doesnt work on C (sata pll) and seems like vice versa for Mali
07:43.42relllarz2k: B!
07:44.05rz2kpokes mnemoc to do CONFIG_ for selection of revision
07:44.33mnemocbut revision is supposed to be detected programatically
07:44.34*** join/#arm-netbook dfletcher__ (~fletch@108-196-222-251.lightspeed.sntcca.sbcglobal.net)
07:44.55mnemocmy mele a1000 has B too, and boots fine :<
07:48.30relllamnemoc: with mali enabled? mine boots fine, too. sata... ev is working
07:49.12relllas/enabled/loaded/
07:49.44Maqsso i have to set dram_size and otherwise this mk802 script.fex can be compiled into script.bin, placed on the boot fat partition and it should work?
07:51.44mnemocMaqs: u-boot doesn't read the script.bin file. it goes to the source
07:51.58mnemocMaqs: see https://github.com/linux-sunxi/u-boot-sunxi/commit/0c38834e76b99446970f14bf0e23a1b374b0a109 as example
07:52.13Maqsok
07:52.32mnemocMaqs: the dram.c file is generated from script.fex using `fexc -I fex -O uboot`
07:52.52mnemoc(fex2bin is a friendly alias of fexc)
07:52.57Maqsjust found this with dram_size set: http://webcache.googleusercontent.com/search?q=cache:QWenv7PYvs0J:pastebin.com/NZhHLQFZ+&cd=5&hl=de&ct=clnk
07:53.06Maqsok
07:53.30mnemocdiff both, it can have other differences
07:54.52mnemoconce you have a good and _verified_ script.fex, it would be awesome if you could add it to sunxi-boards
07:55.19mnemoceventually u-boot will need the pmu info from it too
07:55.25rz2krellla: dumb idea, but could you please try playing with mali_div @ script.bin so it will go around 400mhz?
07:55.38*** join/#arm-netbook leszcz (~marek@193.138.208.6)
07:55.44rz2kif pll is stable enough, you can still live with that
07:55.55Maqsscreen*_output_mode and _type differs
07:56.11rz2kmy mele by default has divisor of 4 instead of 3 for some reason
07:56.42Maqsis there a way to dump the config from another image that is working?
07:58.34Maqsi guess one problem might be that there are mk802 with 512mb and 1gb of ram, so there have to be different versions
07:59.37shivanspsone question, CPU clock is set on u-boot or in the scritp file?
07:59.46mnemocMaqs: from nand's u-boot you can dump boot1
07:59.58mnemocMaqs: and then we can extract the real dram_para from it
08:00.30mnemocshivansps: the initial is set by u-boot, once running linux cpufreq takes control
08:00.59mnemocshivansps: latest u-boot starts at 1GHz
08:01.48Maqsok.. how do i dump it? there is an android image on nand
08:02.09mnemocget into u-boot's console and...
08:02.22Maqsthe working kernel says "DRAM: 1024<6>Total Detected Memory: 1024MB with 2 banks"
08:02.23mnemocmd.b 0x42400000 0x82d0
08:03.20Maqsok, i'll try that
08:03.40relllabtw: i actually successfully play a 1080p hd-trailer @1280*1024vga smoothly with xbmc on mele :-) constant 24fps, cpu~60%
08:07.07oliv3rdid you get libcedarx to work properly?
08:07.25oliv3ri know empat0 (or who was it again) was having issues with 1080p h264 stuff
08:07.41Maqsmnemoc: wow, this gives me a lot of 0x00 :D
08:07.49mnemocMaqs: :<
08:07.55mnemocMaqs: that is BAD
08:09.25Maqsthere's something at the very beginning
08:09.56mnemochno: ---^
08:10.11relllaoliv3r: i forked empat0's repo and did some min changes. sys is debian armel because of armel cedarx libs. kernel sunxi-master branch. everything natively compiled on mele/a10
08:10.11relllai tried the bourne ultimatum trailer from http://www.h264info.com/clips.html
08:10.46Maqsone moment please, i'll upload the dump
08:15.07Maqsmnemoc: the dump: http://dl.dropbox.com/u/11244516/boot1.txt
08:16.44*** join/#arm-netbook hipboi (~tom@123.89.251.41)
08:17.49oliv3rrellla: ahh, nice one. Good to see vpu support imporving/working. Should be soon time to buy a mele as mediacenter then :)
08:17.58oliv3rand sell m 1 day old pivos :)
08:18.13oliv3rmornin' hipboi
08:18.32*** join/#arm-netbook hubert_ (c7b6e8b9@gateway/web/freenode/ip.199.182.232.185)
08:19.28hipboimorning
08:19.32mnemocwb hipboi!
08:19.45hipboimele a2000 stopped manufacturing
08:19.45mnemochno: isn't this boot1 kind of... empty? http://dl.dropbox.com/u/11244516/boot1.txt
08:20.36mnemochipboi: so only the fancy models with antenna and expensive remote are available now or they are moving away of the A10?
08:20.58hipboithey are not moving away from a10
08:21.45hubert_anyone know if the A10 vpu support RGB pixel format? or we need convert rgb to yuv
08:22.24Maqshow do i get that script.fex/script.bin into the uboot image?
08:22.25hipboiyou need to look at the [disp] part in sys_config
08:22.35hipboihubert_
08:22.56mnemocMaqs: fexc, from sunxi-tools, can generate the dram.c for you
08:23.15mnemocMaqs: using -O uboot
08:24.00hubert_I want encode rgb format input video,not display
08:24.56mnemochipboi: nice to see you around :) busy preparing the second batch of cubieboards? or something else?
08:25.41lundmanheh, I havent got zfs to show as mounted device, but it scanned all the media on it, and plays fine in TvdPlayer
08:26.53oliv3rhipboi: what about A1000? they still make those?
08:27.25hipboimnemoc: yes. the second batch of cubieboards are back and needs testing
08:27.41*** join/#arm-netbook leszcz (~marek@193.138.208.6)
08:27.51hipboioliv3r: they stopped a1000, too. but mele has many in stock
08:29.51RaYmAnhipboi: are they all sold out or?
08:30.16hipboiRaYmAn: you means mele or on aliexpress?
08:30.34RaYmAn<PROTECTED>
08:30.45hipboisigh
08:31.00mnemochipboi: please reserve one for RaYmAn ... please
08:31.07hipboi150 is even not enough for those already paid on aliexpress
08:31.12mnemocouch
08:31.19RaYmAnah, fair enough
08:31.21hipboiok, i can reserve some
08:31.21RaYmAn:)
08:31.29oliv3rhipboi: so if I order a mele a1000 in the next few weeks from your webstore, you should be able to deliver?
08:31.31lundmananyone know where I would supply kernel boot argument when it is Android booting?
08:31.55hipboilundman: in android makefile
08:31.59lundmanheh
08:32.10hipboimkbootimg
08:32.11lundmanoh wait, you are serious
08:32.21*** join/#arm-netbook merbzt (~benjamin@c-94-255-220-30.cust.bredband2.com)
08:32.22hipboi: )
08:32.23mnemochipboi: and as i'm in I-wish mode, and considering you can't get armhf cedarx libs, can you get us mali r3p0/r3p1 armel libs?
08:32.25oliv3rhipboi: selling a A40 mele soon :D
08:32.30lundman<PROTECTED>
08:32.51hipboioliv3r: i can reserve on cubieboard for you
08:33.14RaYmAnhipboi: how does such reserving work? :) Just order on aliexpress and tell you? :)
08:33.25jelly-homewhat's A40!?
08:33.25oliv3rhipboi: I'll be getting a cubieboard 2 i guess, if it has an a40 :D
08:33.25hipboiwrite a mail to me
08:33.50hipboioliv3r: mail me your address
08:33.53oliv3rjelly-home: new Allwinner soc, quadcore i belive with 8-core GPU
08:33.54mnemocjelly-home: the name someone decided to give on a random banner on the latest .hk fairs to the next allwinner soc
08:34.13oliv3rhipboi: as soon as I have some money, I will absolutly!
08:34.16lundmanmnemoc; to be fair, the banner did say A40
08:34.20jelly-homeoh, the "next big thing!"
08:34.35mnemoclundman: other banners say A15X :|
08:34.38hipboihave you heard of Action's new soc?
08:34.39lundmanbut in general, allwinners view on source releases, even of kernel, is sucky
08:34.45oliv3rallwinner hasn't announced anything yet have they
08:34.57mnemochipboi: action?
08:35.11hipboiyes, another ic company in Zhuhai
08:35.28hipboithe founders of Allwinner are from Actions
08:35.29lundmanbut if everyone here buys odriod-x, we'll set a trend! :)
08:36.41mnemocoliv3r: eva (who seems to answer all mails directed to allwinner) refered to the next soc simply as "the Quad-core"
08:36.43oliv3rbut are action's a) oss friendly, b) have an awesome VPU :)
08:36.53oliv3r'the quad' lol
08:37.12RaYmAnlundman: it does seem so cool, but people seems to mmm, hate exynos - atleast the android part of it , unsure if linux support is any better
08:37.13oliv3rit's sucha shame they haven't done any source dumps way before the release of their soc
08:37.32hipboiATM7029
08:37.41RaYmAnhipboi: sent :)
08:37.49*** join/#arm-netbook Gumboot (~gumboot@rev.bovine.muck.net.nz)
08:38.16mnemochipboi: so http://www.actions-semi.com is the parent company of allwinner?
08:38.22hipboiyes
08:38.25lundmanheh odd, setting --cmdline made the uart console just die
08:38.42lundmanit all booted, but no console
08:38.48hipboithey left Actions and founded Allwinner
08:38.54hipboinow they made money
08:39.16hipboiand planning to build a new big building beside Actions
08:39.22oliv3rlol!
08:40.01jelly-homenice
08:40.01rman A10 box with a camera https://dx.com/p/h300-intelligent-android-4-0-google-tv-box-white-158743
08:40.04rmbut too expensive
08:40.29jinzothe firesale of MK802's is interesting (~30USD + shipping)
08:41.00rmand how much is the shipping?
08:41.40oliv3rjinzo where is it selling
08:42.54oliv3rhmm, maybe i should wait for a cubieboard/mediabox with an A40 :D
08:42.57jinzoit's selling on hotmid: http://hotmid.com/uhost-android-tv-stick-allwinner-a10.html (39USD) and unknown shipping (but my friend got the phone he ordered there)
08:43.02jinzoand on aliexpress, lemme find the link
08:43.10rmhttp://www.aliexpress.com/wholesale/wholesale-mk802.html?SearchText=mk802&CatId=0&SortType=price_asc&filterCat=70803003%2C4402%2C5092303&SortType=y
08:43.16rmI wouldn't say it's a fire sale
08:43.21rmjust the new regular price for it
08:43.34oliv3rmaybe trying to dump their A10 based devices
08:43.36lundmanhuh how weird, setting cmdline stops it using uart..
08:43.38jinzohttp://www.aliexpress.com/item/Factory-sell-directly-MK802-DDR3-1GB-RAM-4GB-ROM-android4-0-allwinner-a10-Mk802-Hdmi-Android/653302734.html (30USD + 9USD shipping here)
08:43.39rm$30+$9, or same, $39 free shipping
08:44.22oliv3r$39.83 + freeshipping i see
08:44.22jinzouhost is allready promoting uhost 2 stick
08:44.38jinzoso I don't know - looks like they want to get rid of them
08:44.46oliv3rwow, i'd still get a cubieboard to support hipboi, but those prices, damn
08:44.48jinzoit's quite a good deal newertheless
08:44.59oliv3ryeah, I might get one just because it's cool
08:45.05jinzoI'll buy cubieboard as it's more appropriate to what I'm trying to do
08:45.24jinzobut indeed - this looks cheap. Maybe hipboi can get his hands on some?
08:45.29oliv3rbut same still holds, what if there's an mk804, with a new A40 :D
08:45.47jinzoand uhost2
08:45.56jinzoimo there is. but didn't have time to dig much info
08:46.08rmcubieboard has a number of things MK802 doesn't
08:46.17mnemocthe new A10s is more "hdmi dongle" oriented
08:46.22jinzoi.mx6 is killing Allwinner
08:46.23rmSATA; Ethernet; proper UART
08:46.51rm(with actual pins)
08:46.54jinzorm, ofcourse - way better if you plan to do something new with it. But for 40USD shipped it's still a nice deal.
08:46.59rmyes
08:47.11rmdefinitely get that one too :)
08:47.36mnemocbut that's the model that needs an hdmi cable to connect to the tv...
08:48.45lundmanoh i see, the default cmdline is console=ttyS0,115200 root=/dev/nandb rw init=/init loglevel=8, and I replace it
08:48.45discopigi like both the hackberry and cubieboard
08:51.04mnemoca "hacker board" without pins, come on
08:51.06hipboihackberry can never has arduino shield but cubieboard will
08:51.35mnemoceven the cubieboard has too few pins considering the capabilities of the soc
08:51.56mnemocbut the hackberry has NONE
08:53.24lundmanthere booted
08:55.32jinzomnemoc, then I guess it requires some hardcore hacking :P
09:00.20discopigwell even the pandaboard and beagleboards dont have many pins
09:00.37discopigbut i do tend to prefer cubieboard
09:01.04mnemoccubieboard for world domination! :D
09:01.58discopigyes
09:02.39andomadoes the cubieboard have the CEC pin in the HDMI connector wired?
09:03.02mnemocbut you could still add two more expansion headers and still leave parts of A10's greatness out
09:03.40mnemocandoma: don't know. check the schematics. http://dl.cubieboard.org/hardware/cubieboard_schematic_2012-08-08.pdf
09:04.00andomajust did
09:04.03andomaseems it's hooked up
09:04.47Maqshow do i install the new uboot on the uSD card? I patched uboot, compiled it for mk802 and copied uImage, but I think this is not enough, as the version did not change
09:05.32mnemocMaqs: https://github.com/linux-sunxi/u-boot-sunxi/wiki
09:05.44Maqsthx
09:07.11*** join/#arm-netbook cheng_ (~cheng@219.92.77.39)
09:07.16andomai wonder if the mk802 have CEC wired as well
09:07.36hipboihas anyone noticed that cubieboard.org had a new theme?
09:07.53RaYmAnandoma: no one seems to know how to access the CEC part of a10 though :P
09:08.33*** join/#arm-netbook tzafrir_laptop (~tzafrir@local.xorcom.com)
09:08.53andomaRaYmAn: aww....
09:11.26mnemochipboi: yes, since last nice. looks nice :)
09:11.35mnemocs/nice/night/
09:11.41*** join/#arm-netbook mysteryname (~mysteryna@14-202-9-167.tpgi.com.au)
09:11.56hipboithe Indian designer doesn't mess this time
09:13.54mnemochipboi: it only misses the shop to pre-order 2012-08-08, "next" and to buy power bricks and cases :)
09:14.03Maqsuboot now just shows "U-Boot SPL 2012.10-g0c38834-dirty (Oct 17 2012 - 10:55:39)".. i don't have a system.bin, as i have no idea what it is or where to get it :)
09:14.18mnemocsystem.bin???
09:14.28mnemocMaqs: clean the env in the card
09:14.36*** join/#arm-netbook ZaEarl_ (~malmrose@17.host-182-48-148.compassnet.co.nz)
09:14.38mnemocMaqs: so u-boot can use it's default
09:15.02mnemocMaqs: dd if=/dev/zero of=${card} bs=1024 seek=544 count=128
09:15.14Maqsthe wiki page says something about system.bin on boot
09:15.34mnemocmust be a typo. it's script.bin
09:15.44Maqsok
09:17.14Maqsit's the same, only that one line
09:18.04mnemoctypo corrected
09:18.05Maqsok, i got an idea
09:18.31*** join/#arm-netbook leszcz (~marek@193.138.208.6)
09:20.05ping__hy how to instal linux on ainol aurora 7 ? type allwiner
09:20.34Maqswow, it works
09:20.55Maqsi had to dd u-boot-spl.bin instead of sunxi-spl.bin
09:21.04*** join/#arm-netbook newbie1 (~kvirc@118-168-195-38.dynamic.hinet.net)
09:21.07Maqsno, didn't work
09:21.18Maqsit seems to have booted from nand again
09:21.24mnemocit's spl/sunxi-spl.bin
09:22.04Maqsthat's what i did in the first place.. where it stopped after the first line
09:22.06mnemocMaqs: unless you decided to use "sun4i" target instead of one with SPL enabled
09:23.24Maqsi added a line with mk802 [...] sun4i:SPL to uboot
09:23.50mnemocok
09:24.15mnemocMaqs: sunxi branch?
09:24.28Maqsand created a directory for mk802, added the generated dram.c and copied the rest from mele1000
09:24.33Maqsyes
09:24.35mnemocif spl/u-boot-spl.bin isn't there something else went wrong, hunt that down first
09:24.48Maqsit's there
09:25.03Maqsit just does not boot
09:25.17mnemoctry a different card, and zero the first 1M
09:25.31mnemocback in 30m
09:25.40Maqsok
09:28.35rmhttp://www.cnx-software.com/2012/10/17/89-hiapad-hi802-android-mini-pc-based-on-freescale-i-mx6-quad-processor/
09:29.11rm"but does it run GNU/Linux"
09:29.17*** join/#arm-netbook ping__ (~ping__@125.162.229.135)
09:30.22libvrmmod hdmi is interesting.
09:33.10andomadefine interesting
09:34.04libvhung task :)
09:35.03andomayeah that happens for me aswell
09:35.11andomajust wonderd if you saw something else
09:35.39libvthat's already interesting enough in my book
09:36.16libvpfff, is there no corner that doesn't need massive amounts of work :)
09:36.40*** join/#arm-netbook newbie12 (~kvirc@118-168-195-38.dynamic.hinet.net)
09:36.51*** join/#arm-netbook hipboi (~tom@123.89.251.41)
09:36.53relllalibv: same for me
09:37.24relllamnemoc: you meant the patch in the first mail, which is smaller than the commit?
09:40.07*** join/#arm-netbook Gumboot (~gumboot@rev.bovine.muck.net.nz)
09:41.15*** join/#arm-netbook Maqs (~marcus@188.190.99.49)
09:45.25mnemocrellla: yes
09:49.42oliv3rRaYmAn: i can't find any reference to CEC in the manual, and the HDMI references are extremly limited aswell.
09:51.48RaYmAnoliv3r: indeed
09:52.03RaYmAnCEC would be quite awesome on these boards
09:52.36oliv3rCEC should be quite aweomse in general :)
09:52.45RaYmAnyeah
09:52.48RaYmAnloves it
09:52.53oliv3rdoes the stock mele rom do CEC?
09:52.57oliv3ri like the idea of it :)
09:53.05oliv3rnever used it and my TV is so old, it may not do cec :p
09:53.16RaYmAnthat's the issue - I've yet to see any a10 software that supports CEC
09:53.19RaYmAnso nothing to reverse it from
09:53.38RaYmAnIn all likelyhood, it's extremely simple, essentially just providing a way to read and write data to it, and then rest is userspace
09:53.57RaYmAnso we "just" need to figure out how to write/read data from CEC and then we can hack up libcec to do the rest
09:54.46oliv3ri think that it may be even just connected to 'a' GPIO
09:55.02libvhas anyone gone through drivers/video/sunxi and pulled the chinese through google translate?
09:55.03oliv3rwhich means bitbanging it in software
09:55.26*** join/#arm-netbook mSquare (~selvan@122.178.210.195)
09:55.29oliv3rlibv this all for fosdem next year?
09:55.40libvoliv3r: nope
09:55.56libvis there a native chinese around who has some time to burn?
09:56.02RaYmAnoliv3r: it's two-way though?
09:56.26oliv3rso is a gp I O, Input Output :D
09:56.50oliv3rlemme find how someone did that on an arduino
09:57.59oliv3rRaYmAn: http://code.google.com/p/cec-arduino/wiki/ElectricalInterface
09:58.35oliv3rthere, they actually split the CEC signal into an input and an output
09:58.45RaYmAnyeah
09:58.55*** join/#arm-netbook rellla (~Thunderbi@p5B0789A0.dip0.t-ipconnect.de)
09:59.00oliv3rso A10 could still do it that way :p
09:59.10oliv3rand the circuit is really really simple; so that's not the issue
09:59.13RaYmAnwell, assuming you can access the gpio
09:59.13RaYmAn:P
09:59.31oliv3rBut! if the A10 does that in hardware, and exposes 1 CEC pin, that would be better.
09:59.32RaYmAnI guess one could design the board so CEC pin is hooked up to something else
09:59.36RaYmAndefinitely
09:59.46oliv3rRaYmAn: taht's what I ment yea :)
09:59.53lundmanthen make TCPIP over CEC!
10:00.10oliv3rhipboi: could give _some_ insight, as he did connect CEC pin to P23 i think
10:00.57andomaon the cubieboard the CEC pin on the HDMI connceter is wired to CEC_HDMI on the A10
10:01.03andomaso there's got to be some support in the chip
10:01.25oliv3ri can't find CEC_HDMI in the manual though
10:01.51RaYmAnyeah, there should be support, but without a clue on how to access it, it's rather hard :/
10:01.58andomayeah
10:03.01lundmanusually they are usb devices, but occasionally /dev/ nodes for ioctl
10:03.39RaYmAnon tegra3, it's just a /dev node with ioctls, which then very simply just puts the raw data into hdmi-cec registers
10:04.15oliv3rCEC buss a USB device? in the A10, i doubt it.
10:07.43lundmanmerely commenting on 80% of the device support in libcec, is for usb device s :)
10:13.42oliv3rlundman: that's probably because pulse-8 has an USB -> cec bridge and wrote libcec i think
10:13.49*** join/#arm-netbook Maqs (~marcus@marcus.eunomia.de)
10:20.06lundmanstill, i agree a10 most likely doesnt have it on usb bus
10:20.10lundmanalthough wifi is there
10:20.18lundmanmaybe a lsusb would quickly rule it out
10:24.59oliv3ri think i did that on my tablet, and didn't find anything other then wifi
10:25.12oliv3rwhich only showed up when it was active
10:29.19lundmantablet has hdmi?
10:31.14oliv3rmini-hdmi out yeah
10:31.45RaYmAntbh, given it seems to be connected to a dedicated CEC pin on the SoC, it won't be USB
10:31.54RaYmAnlike, connected straight from cable to soc
10:31.59RaYmAnwifi is external to the SoC
10:32.30oliv3rto bad there's nothing in the manual (that i've found yet)
10:32.40oliv3rcould be named other then CEC or hdmi though
10:32.53RaYmAndefinitely
10:33.49mnemocthere might be something hidden within the hdmi driver
10:34.11mnemocat least there is partial edid support
10:34.20mnemocwhich iirc also uses i2c
10:34.35oliv3rif the time is right, i'll dig some into that
10:34.52mnemocyou are the registers master ;-)
10:35.03oliv3rLOL, right :p
10:35.19oliv3rPin 13 is CEC and Pin 15/16 is I2C
10:36.18RaYmAnDDC/EDID seems to be implemented as a seperate i2c controller accessed through hdmi regs - so it's quite likely cec is similar (Except not i2c obviously)
10:36.36oliv3ronce i'll get to that section, i'll write about it
10:36.46oliv3rnow, after aa week hyatus, it's back to documenting the CCM :p
10:36.54oliv3rstupid tablet cost me way to much time :)
10:41.55mnemocoliv3r: go by importance to hackers, not document order :)
10:42.39oliv3ri thought you said CCM was the first crucial bit :p
10:43.35*** join/#arm-netbook Quarx (~Quarx@46.233.210.49)
10:47.04oliv3rnew cheap nexus 7 coming from google, they'll be using WM8950 (wondermedia). I don't get it. Last I heard about wondermedia, was they weren't so opensource friendly ... :S
10:47.26oliv3rgoogle should have used the A10 and used their leverage for open cedarX :D
10:47.29oliv3r*dream*
10:50.25mnemocgoogle needs to show they love all members of the aliance
10:51.02*** join/#arm-netbook hp__ (~kvirc@118-168-195-38.dynamic.hinet.net)
10:53.02*** join/#arm-netbook mnemoc (~amery@geeks.cl)
10:55.29cat1doing mali update is painful.. really tempted to build it outside tree..
10:57.57mnemocif we don't get r3p0/p1 armel libs from hipboi by the weekend I'll also +1 that
10:58.47hipboihope i can get the cedarx hf lib
10:59.02mnemocthat would be even better :)
10:59.30mnemocwe will be able to forget about armel and r2p4 entirely
10:59.52cat1and switch to r3p1 completely?
11:00.19mnemocr3p0 first, we don't have p1 libs for android :|
11:00.50ZaEarl_are the libs for android and linux different?
11:00.55mnemocyes
11:01.13*** join/#arm-netbook alcides (~alcides@187.23.195.197)
11:01.13*** join/#arm-netbook alcides (~alcides@unaffiliated/alcides)
11:02.09mnemocby "linux" we don't only refer to a kernel, but also to a certain common userspace shared by mosly all those things called "linux distributions"
11:02.52mnemocZaEarl_: the libraries are built for a very specific... environment, not only a kernel api
11:02.54RaYmAnmnemoc: isn't that, usually GNU/Linux when userspace is included :P
11:03.00RaYmAntechnically speaking, linux is the kernel :P
11:03.13mnemocsure
11:03.38libvwhat is up with all the files in linux-sunxi/include/config ?
11:04.19libvcan this whole directory not be removed?
11:04.27mnemocRaYmAn: but when you say I want a linux laptop you don't mean android even when android uses the linux kernel :p
11:04.55libvall the files in there are empty anyway
11:06.02mnemoclibv: include/config is not part of .git, it's a build time thing
11:06.15libvah, hrm, how stupid of me then :)
11:06.20mnemoc:)
11:06.29RaYmAnmnemoc: sure, but it's still wrong :P
11:07.29mnemocRaYmAn: human communication relies on what people understand by the terms, not on what the terms are defined in the dictionary ;-)
11:07.55RaYmAn:P
11:08.26mnemocand sure, it is wrong to call the whole thing "linux", and it's also wrong to call it "gnu/linux"
11:08.44mnemocbut well....
11:08.47cat1mnemoc: can p0 libs for android work with p1 kernel counterpart?
11:09.00mnemoccat1: i don't think so
11:09.16mnemoccat1: but baby steps ;-)
11:09.51cat1mnemoc: i can't do it, i am gigantic jumper :)
11:09.58mnemoc"D
11:09.59mnemoc:D
11:13.52cat1mnemoc: can we keep 3.4 branch somewhat in sync with mali official releases? it should be safe enough as none is using it for graphics/video atm. Or I am completely wrong?
11:14.17cat1is constantly loosing strategic line..
11:15.04mnemoccat1: i might agree on that. I already removed allwinner boot hacks from 3.4 and that kicked out everyone booting from nand :|
11:15.46RaYmAnimo that's OK - goal is to make it work with open u-boot on nand, isn't it?
11:15.55mnemocyes
11:16.28RaYmAnffs, I hate corporate proxies *sigh*
11:22.23libvcat1: not this again
11:47.22*** join/#arm-netbook avernos (~avernos@222.128.157.156)
11:47.23*** join/#arm-netbook avernos (~avernos@unaffiliated/avernos)
11:51.45cat1libv: why, i am simply seeking for the baseline from which i can start doing something. sunxi-3.4 is a good candidate. Moreover, in my ideal world i dream of working process like this: patch -- review/test -- ack/nack -- push/pull into main branch, w/o having branch multiplications.. Well, sorry for being idealistic :)
11:53.15mnemoccat1: there are very little limitation on what you can do on 3.4
11:57.16libvcat1: with binary blobs, the world is not that simplistic
11:58.14mnemocand there is tons of work to do on the rest of the kernel too, not only mali
11:58.35cat1libv: right, but afaiu we consider binary blobs being temporary while progressing with kernel not.
11:58.40cat1mnemoc: exactly
11:59.23relllamnemoc: still have a freeze with minor patch
11:59.25oliv3rwhat is AHB and APB in the A10? dont' quickly see it described
11:59.54mnemocrellla: please reply that one too, commenting the issue
12:01.03RaYmAnoliv3r: it's a standar ARM thing - different bus'es of some sort (I'm not really entirely sure how it works, but it should give a hint I guess)
12:01.26mnemoccat1: work on mali is obviously slower, can need to keep userspace working.
12:01.39RaYmAnAdvanced High-performance Bus and Advanced Peripheral Bus
12:01.57oliv3rsnap, I just figured that out whilst not looking at IRC
12:02.05RaYmAn:)
12:02.13oliv3rthough I thought it was Allwinner Host bus and Allwinner Peripheral bus
12:02.15oliv3rclose enough
12:02.35mnemoccat1: on the other edge we have the testing limitation, rellllla now has troubles with a clock change which was supposed to be tested my a large group of people
12:02.40oliv3rThey should have done some marketing spin on that and named it as such :p
12:02.49RaYmAnwouldn't have been too surprising ;) But it does seem to be fairly standard :P
12:03.28mnemoccat1: and know we have a monster patch (usb unification) which also needs review/testing. and I can't do it myself because of lack of time.
12:03.55oliv3rRaYmAn: anyway, that makes perfect sense :)
12:04.18mnemoccat1: please try to target your energy where it helps more, and not were it only helps to waste time discussing things again and again :<
12:04.22*** join/#arm-netbook ping__ (~ping__@125.162.229.135)
12:04.30libvcat1: binary blobs are not as temporary as you'd think
12:04.59libvcat1: reverse engineered graphis drivers do not fall out of the sky
12:05.27lundmanare you sure?
12:05.45cat1mnemoc: i am trying but desperately, too many branches too many things to follow..
12:06.42mnemoccat1: no, only two main branches, the stable sunxi-3.0, and the development sunxi-3.4
12:07.47cat1mnemoc: what about fixes that are in next_mali?
12:08.15mnemocfixes in next_mali aren't relevant to either sunxi-3.0 or sunxi-3.4
12:08.44cat1mnemoc: how come, fb does not work if i do not apply patch from there
12:09.16mnemocwhich patch?
12:09.36cat1video: sunxi: disp-ump: Add support for buf2
12:09.53cat1well, not fb, but rather x
12:10.28mnemocr2p4 doesn't use buf2
12:11.50cat1mnemoc: this is what i am talking about.. for 3.4 r2p4 is not relevant but the former is kinda broken because of that.
12:12.36mnemoccat1: please
12:12.45cat1mnemoc: ok, ok, i'd rather stop nagging and shut up :)
12:12.49oliv3rShall i start holding my hands out for those drivers falling? :D
12:13.05lundmanwow, uh, interesting reply from github
12:13.18lundman"I suggest you pull in upstream changes first, then create a new topic branch from the default branch, do your change, push, and create the pull request from that topic branch instead of the commit directly."
12:13.22lundmanwhat?!
12:13.53oliv3rhuh, why?
12:14.17jinzothat's preatty common?
12:15.16lundmanwhat, you guysunderstand that? thats unfair:)
12:15.40oliv3roh, yes, after re-reading yeah
12:16.48lundmanto me, it sopunds like it is not worth sending a 1 line fix for allwinner :)
12:16.49oliv3rclone, branch, pull, merge, push (i push from branch to remote/master though) pull-request!
12:17.03oliv3rin that case, just send the patch to the ML
12:17.23oliv3rsomeone will put it in their branch, credit you and push it
12:17.26lundmanmeh, nobody cares about making android kernels for mele here
12:19.31oliv3rif it's a fix to one of the linux-sunxi branches, it's a fix nevertheless :p
12:20.25*** join/#arm-netbook wingrime (~alex@78.109.115.103)
12:24.35*** join/#arm-netbook mnemoc (~amery@geeks.cl)
12:25.08*** join/#arm-netbook mnemoc (~amery@geeks.cl)
12:25.56mnemocznc is getting on my nerves :<
12:28.12*** join/#arm-netbook mysteryname (~mysteryna@14-202-9-167.tpgi.com.au)
12:35.50oliv3rdoes the wiki support fancy math stuff?
12:35.59oliv3rlatex-style etc?
12:36.08lundmanboth addition AND subtraction
12:37.31jellyhow does what-if.xkcd.com do their fancy math stuff?
12:38.14mnemociirc he use to work for nasa...
12:38.16jellythey seem to use something called MathJax
12:38.19mnemocused*
12:39.11mnemoci think he draws the stuff himself
12:39.23oliv3rwikipedia, or rather, mediawiki, uses something math/latex-y
12:39.38mnemocoliv3r: need it for linux-sunxi.org
12:39.54mnemoc?
12:39.55oliv3rwell the PLL's use some simple math
12:40.01oliv3rand looks prettier with math stuffs :)
12:40.14oliv3ri'll make do with regular ascii math
12:40.50mnemocchoose an extension with few deps and I can install it
12:41.23oliv3rhttp://www.mediawiki.org/wiki/Manual:Math
12:42.01mnemocand the deps?
12:42.11oliv3ri think texvc
12:42.16RaYmAndo we really need latex? It's uhm, usually not recommended for not encessarily being very safe :P
12:42.21RaYmAnbut it might have improved
12:42.31oliv3rwell wikipedia uses it :p
12:42.53rz2k<mnemoc> [16:08:17] r2p4 doesn't use buf2
12:43.03mnemocrz2k: it does?
12:43.06rz2kI believe techn_ wrote patch that way that it is compatible with r2p4
12:43.13*** join/#arm-netbook sspiff (~quassel@unaffiliated/yukito)
12:43.24rz2kioctl for standard secure id is buf1
12:43.30oliv3rmnemoc: texvc should be part of the extension: math
12:43.41mnemocrz2k: but needed before merging r3p0?
12:43.42rz2kbuf2 is just a int counter and multiply of framebuffer
12:43.50oliv3rmnemoc: deps should be ocaml, gnu make; latex and dvips, imagemagick and ghostscript
12:44.26rz2kcould someone with android test buf2 patch on r2p4 libs/kernel ? Turl?
12:44.49rz2kmnemoc: if buf2 patch will be in kernel, we can compile r3 externally for now
12:47.14mnemocrz2k: fair enough. if it doesn't harm and all agree it's good. give me the hash I can push it
12:47.25rz2klets test first
12:47.35rz2kandroid guys can use it
12:48.38mnemocok
12:55.39*** join/#arm-netbook wingrime (~alex@78.109.115.103)
12:57.04mnemocoliv3r: will poke you after the ext. is installed
13:00.08*** join/#arm-netbook sspiff (~quassel@unaffiliated/yukito)
13:01.09*** join/#arm-netbook hipboi (~tom@123.89.251.41)
13:02.29wingrimewhat status have 3.4 kernel?
13:04.01relllamnemoc: problem fixed for me. send patch to ml.
13:05.14oliv3rmnemoc: rgr
13:05.48*** join/#arm-netbook drachensun (~drachensu@142.196.83.182)
13:09.40oliv3rmnemoc: did a save for now so my data should be in the db :)
13:17.33*** join/#arm-netbook pawel5870 (~pkarpins@ip-91-205-72-3.iwacom.net.pl)
13:38.06oliv3ri have to take my VM offline for a short while, i'm not editing anything atm and have saved everything.
13:38.09oliv3rwill check back later
13:59.28relllais there a reason, why the following lines are in cubieboard.fex but not in my mele_a2000.fex:
13:59.28rellla[target]
13:59.34relllapll4_freq = 960
13:59.35relllapll6_freq = 960
13:59.35relllapower_start = 0
14:00.31RaYmAna guess would be that one is 'decoded' from script.bin (mele) while the other is the original "source" file (cubieboard)
14:00.55RaYmAnWe already know that some values are coded into e.g. boot0/boot1 instead of written to script.bin
14:03.03relllaok. so could there be a relation for my issue with the mentioned sata fix?
14:03.25*** join/#arm-netbook pawel5870 (~pkarpins@ip-91-205-72-3.iwacom.net.pl)
14:11.19wingrimemnemoc: I send fist patch
14:19.55mnemocwingrime: thanks!
14:24.52*** join/#arm-netbook arokux (~arokux@merkur178.inf.uni-konstanz.de)
14:37.46libvso there is no chinese speaker with time to burn here?
14:50.54*** join/#arm-netbook arokux (~arokux@merkur178.inf.uni-konstanz.de)
14:51.07mnemoctry #chinalug
14:55.08mnemoclibv: ----^
14:59.29libvthat'll be a tough sell :)
15:01.00mnemoclong text?
15:01.17mnemocgoogle translate does a pretty decent job for single sentences
15:07.09*** join/#arm-netbook rz2k (~rzk@128-69-95-50.broadband.corbina.ru)
15:12.34*** join/#arm-netbook arokux (~arokux@merkur178.inf.uni-konstanz.de)
15:33.01libvjust many bits of it
15:55.29wingrimehow about to make own SPL (boot.axf) ?
15:55.39wingrimeGLP based ofcouse
15:56.01L84Supperlibv: is it all highly technical?
15:56.25libvL84Supper: how would i know, i do not speak chinese :p
15:57.01libvL84Supper: but i guess so, yes, but i am going to be working with $chinese_volunteer on making sense from it
15:58.00L84Supperlibv: ah I thought you needed it translated into Mandarin vs out of
16:05.21wingrimeI have dram_clk = 360 , if I set 480 will it work ?
16:16.16techn_rz2k: you got xbmc running? :D
16:16.27rz2ktechn_: yep
16:16.35techn_what was the trick.. or is there some bug fixed to r3p1
16:16.36rz2kwith libUMP recompiled with your fixes
16:17.08techn_I'm using r3p0 and it wont start :(
16:17.40techn_chrashes with that callstack I paster yesterday
16:18.40techn_*pasted
16:18.55rz2kI have unity and plasma-active crashing though
16:19.07rz2kso there is some missing stuff there
16:19.50techn_oh so you are using other version of *ubuntu
16:20.02techn_I'm using linaros ALIP version
16:20.17techn_Or do you?
16:20.33rz2kno, linaro-alip
16:20.46techn_and my ALIP is from july
16:20.46rz2kI've just installed plasma-active & untiy from repo
16:20.58rz2kmine is updated to latest packages
16:21.36techn_will it update with upgrade/dist-upgrade?
16:22.15techn_Atleast for ubuntu there is some hackery needed for to get newer version
16:25.09rz2kyep
16:25.14rz2kit will update
16:25.50techn_oh.. then we must have same ALIP then.. only difference is that you are using r3p1 and I r3p0?
16:26.12rz2kprobably
16:27.23rz2kalso there was a thing that I needed to do: somehow before I've killed all mesa/mesa-egl swrasts including OpenGL (not ES)  one. in this situation xbmc dies with libGL not found, reinstalled mesa and now it works
16:28.25rz2kalso I have debugging off in my kernel
16:29.04rz2kwith ARM's awesome libraries and code this might be mandatory.
16:35.00*** join/#arm-netbook pawel5870 (~pkarpins@staticline54070.toya.net.pl)
16:41.10traeakhttp://liliputing.com/2012/10/hiapad-hi-802-pc-on-a-stick-with-freescale-quad-core-cpu.html
16:41.19traeak89usd
16:42.09Turlrz2k: pinged me?
16:42.12Turlwhat's buf2?
16:43.04mnemoctraeak: old news ;-)
16:43.26mnemocTurl: mali r3 uses two buffers
16:44.11mnemoctraeak: subscribe charbax' youtube channel and enjoy the feed of sexy gadgets
16:44.35L84Supperany news on an imx6 board with SATA?
16:44.54Turlmnemoc: double buffering?
16:44.59L84Supperhttp://www.wandboard.org/  lists SATA as unpopulated
16:45.38techn_cat1: To use sunxi 3.4 without patches.. Use r2p4 libs.. I gave wrong direction/guindance last time ;)
16:45.40mnemocL84Supper: for ~$300 the sabrelite
16:46.00mnemocTurl: yes
16:46.08Turlmnemoc: what do you need tested?
16:46.17rz2kTurl: test r2p4 with https://github.com/linux-sunxi/linux-sunxi/commit/05b817314dfe0543aba938d496f9ff30a5545134 commit applied with android libraries
16:46.31L84Suppermnemoc, I guess those SATA connectors are pretty darn expensive  :)
16:46.38mnemocL84Supper: :D
16:46.42rz2kif it will run ok, we can apply same to head branch and have mali r3pX compiled externaly for now
16:46.47rz2kand r2p4 as default one.
16:47.35TurlI don't think it uses that ioctl even
16:48.10*** join/#arm-netbook gzamboni (~gzamboni@2a02-8426-513b-5100-0224-21ff-fe57-307c.rev.sfr.net)
16:48.14rz2kin x11 EXA uses for window creation and on driver initialization
16:48.29rz2ks/uses/uses it/
16:48.32cat1techn_: thanks for update, i'll check it a bit later! now preparing some patchset with hci refactoring on top of Andrey's patches..
16:49.56Turl<PROTECTED>
16:49.56Turl<PROTECTED>
16:49.57Turl...
16:50.32mnemoccat1: \o/
16:50.40Turlump_secure_id_get is probably part of libUMP.so
16:50.51TurlI don't see it on the gralloc code
16:51.02rz2kinteresting
16:51.11mnemocwe have disp_ump.ko which is a glue module
16:51.42Turl$ objdump -Tt vendor/allwinner/zatab/proprietary/lib/libUMP.so |grep ump_secure_id_get
16:51.43Turl000014bc g    DF .text000001a0 ump_secure_id_get
16:54.29rz2kTurl: main problem is that we dont know if ump_secure_id_get is exactly same as ump_secure_id_buf1, they should be same, but it needs a test.
16:57.50traeakthanks sorry
17:10.21*** join/#arm-netbook mnemoc (~amery@geeks.cl)
17:13.40mnemocany suggested alternative for znc that doesn't die when you have too many open channels?
17:14.59techn_I have been using screen+irssi
17:15.39mnemocscreen+irssi kind of sucks from the phone :|
17:15.54rz2khave a wide channel, up to date znc ? mine dies only when server crashes. I have 3 networks added with 10 channels per network.
17:16.41mnemoc0.206-1 from ubuntu
17:17.15jinzomnemoc, znc is rock solid for me
17:17.34rz2kI'm sure there is ton of updates from version that is in ubuntu
17:17.49jinzocurrently 5 networks and  ~15 channels - no problems.
17:17.55rz2kcheck their github
17:18.08mnemoc206 is the latest release, but 7M old
17:18.12jinzoand it's easy to compile.
17:19.07mnemocrz2k: what do you mean by "wide channel"?
17:20.08*** join/#arm-netbook gimli (~gimli@xbmc/staff/gimli)
17:20.41rz2kI mean that znc can possibly fail when you have bad connection with many packets dropped
17:21.27mnemocnah, it's by "flood"
17:21.30rz2koh, new znc in archlinux testing by the way :p
17:26.52Turlmnemoc: rock solid znc over here too
17:28.01Turlmnemoc: 3 networks for a total of 14 channels autojoined
17:28.53mnemocmy problem is with freenode in particular, 18 channels.
17:30.18Turlmnemoc: maybe lower MaxJoins config?
17:30.31Turlmine is set to 5
17:30.51mnemocmine is 2 :<
17:32.57Turlmnemoc: do you have ctcpflood enabled?
17:34.37mnemocno module at all... 1m
17:34.37Turlmaybe znc is sending you the ctcp queries and your client feels like answering all of them and flooding :<
17:34.58mnemocthe flood is not on the client side, but on znc itself
17:35.11mnemocznc gets disconnected, not me
17:35.38cat1mnemoc: can we have some place where kernel+modules can be uploaded for testing? it would be good to supply each patch series with binaries for those who does not compile stuff themselves.
17:36.04mnemoccat1: dl.linux-sunxi.org/something ?
17:36.23cat1mnemoc: why not
17:36.35mnemocgive me your pub key
17:36.52cat1mnemoc: like testing/sunxi-3.0 testing/sunxi-3.4
17:38.25mnemochttp://cat1.linux-sunxi.org/ might work better than dl.  i thinkg
17:39.01cat1mnemoc: then i would prefer simply cat, possible?
17:43.03Turlmnemoc: want to participate in an experiment? I can send you CTCP VERSION queries and see if you drop
17:43.23mnemocsure
17:43.25mnemoctry
17:48.31*** join/#arm-netbook rellla1 (~Thunderbi@p4FE56095.dip0.t-ipconnect.de)
17:50.41Turlwell you haven't dropped yet :P
17:51.18Turlmnemoc: I also have route_replies enabled fwiw
17:51.32mnemoc:)
17:51.53Turlboth irssi and andirc like to answer to stuff
17:53.26mnemoci'm connected to znc from andchat and irsii. but as i said, it's znc who gets flooded, not me
17:55.50techn_mnemoc: does znc have some setting to prevent flooding
17:56.11techn_If I remember correctly.. some clients has flood protection
17:56.45techn_but it would be stypid to drop from server in that case :D
17:57.21mnemocwill lower the maxjoin to 1 and see if i don't get kicked out again today
18:01.13*** join/#arm-netbook tmerle (~tmerle@bea31-1-78-226-152-136.fbx.proxad.net)
18:01.42*** join/#arm-netbook arokux (~arokux@merkur178.inf.uni-konstanz.de)
18:08.31*** part/#arm-netbook tmerle (~tmerle@bea31-1-78-226-152-136.fbx.proxad.net)
18:09.23captainiglooi'm trying to enable spdif output on my mele a2000
18:09.34captainigloobut alsa doesn't list this outpu
18:10.31Turlit's a different device iirc
18:10.35Turltry alsamixer -c 1
18:10.47Turl<PROTECTED>
18:11.16captainiglooi'm doing aplay -l
18:11.27*** join/#arm-netbook tmerle (~tmerle@bea31-1-78-226-152-136.fbx.proxad.net)
18:11.37*** part/#arm-netbook tmerle (~tmerle@bea31-1-78-226-152-136.fbx.proxad.net)
18:11.37captainigloohttp://pastebin.ca/2246812
18:11.41captainigloohere the output
18:12.17Turlah, the separate one is hdmi
18:12.26Turlcard 0 should be spdif + rca audio
18:12.59captainiglooTurl: you mean that sound output is enabled for both at the same time ?
18:13.06captainigloofor rca + spdif ?
18:13.10*** join/#arm-netbook tmerle (~tmerle@bea31-1-78-226-152-136.fbx.proxad.net)
18:13.33Turlno, but using alsamixer/amixer you should be able to toggle them
18:15.40captainigloomy amixer output : http://pastebin.ca/2246814
18:17.55captainiglooi didn't see any spdif output :/
18:18.02Turlthe dac one isn't it?
18:18.28Turlmake sure you have CONFIG_SND_SUN4I_SOC_SPDIF on your kernel too
18:19.04captainiglooi loaded sun4i_spdif, sun4i_sndspdif and sndspdif modules
18:19.24captainiglooi missed maybe one module ?
18:19.27*** join/#arm-netbook tmerle (~tmerle@bea31-1-78-226-152-136.fbx.proxad.net)
18:20.09captainiglooah !
18:20.13captainiglooyes i missed one
18:20.21captainigloo<PROTECTED>
18:20.23captainiglooan d now
18:20.46captainiglooaplay -l lists another card
18:20.55captainigloocard 2: sun4isndspdif [sun4i-sndspdif], device 0: SUN4I-SPDIF sndspdif-0 []
18:21.00captainigloothat's cool !
18:21.29captainiglools
18:21.33captainiglooouch sorry
18:22.52*** join/#arm-netbook hg_5 (~chatzilla@91.234.245.245)
18:23.07hg_5hello
18:25.18*** join/#arm-netbook arokux (~arokux@merkur178.inf.uni-konstanz.de)
18:27.27hg_5does this device able to run linux http://www.cnx-software.com/2012/10/10/droid-stick-a2-amlogic-aml8726-mx-dual-core-android-4-1-1-mini-pc/ ?
18:28.38jelly-homehg_5: run linux and use all of the hardware features at the same time, without signing a NDA?
18:29.03hg_5hm, NDA, what is that?
18:29.43*** join/#arm-netbook arokux (~arokux@merkur178.inf.uni-konstanz.de)
18:30.48*** join/#arm-netbook arokux (~arokux@merkur178.inf.uni-konstanz.de)
18:31.22jelly-homehg_5: some vendors want a developer to sign a http://en.wikipedia.org/wiki/Non-disclosure_agreement before providing source code needed to make the system work
18:31.42hg_5oh
18:32.39*** join/#arm-netbook arokux (~arokux@merkur178.inf.uni-konstanz.de)
18:34.40techn_rule
18:34.45techn_upps
18:34.56*** join/#arm-netbook hg_5 (~chatzilla@91.234.245.245)
18:58.51Turlmnemoc: over here :P
18:58.55Turlcat1: ping
18:59.12cat1Turl: pong
18:59.34Turlcat1: did you merge the 4 usb patches on the ML on 3.4?
19:00.17*** join/#arm-netbook functionbeforefo (~functionb@206-248-137-101.dsl.teksavvy.com)
19:01.07cat1nope, sorry , had to do some housekeeping, now free and about to run tests with 3.4
19:10.29*** join/#arm-netbook L84Supper (~ly@unaffiliated/l84supper)
19:13.10Turlhad a quick look, the patches look sensible
19:13.32Turlhaven't tried them yet mnemoc
19:14.56cat1Turl: applied same patches against 3.4 -- not happy, now rtlwifi oopses. will hunt a little bit..
19:16.46mnemocouch
19:16.49cat1ah, that's known to me, seems to be the problem with fw loading code :) need to re-build again
19:17.20mnemocthe allwinnerized driver should work in 3.4 now...
19:17.34mnemocbut oopsing?
19:17.47*** join/#arm-netbook ooo_ (b2009e2a@gateway/web/freenode/ip.178.0.158.42)
19:18.14cat1i am actually using upstream one
19:18.37cat1i think the sw one should be fine
19:19.54Turlcurious diffstats http://sprunge.us/ZRIT
19:21.23Turlmnemoc: do you think cpufreq will be better suited as a single patch, or as several?
19:21.44TurlI won't send my current set as I refactored the refactorization while refactoring refactored code, pretty much :P
19:21.51mnemocif it can be properly understood in one, one
19:22.12Turlhttp://sprunge.us/QeZE?diff I'll let you guys judge
19:22.42mnemoccan't just until 1:40h more
19:23.04Turlnp
19:24.10techn_Turl: It would be nice if cosmetics are in different patch :/
19:24.52techn_or is there some tool to ignore cosmetics from diffs?
19:25.49hnotechn_, cosmetics like?
19:25.50Turlyou can -w on git show/diff
19:27.24Turlhttp://sprunge.us/XjGW?diff this should be it withotu cosmetics
19:27.27Turlwithout*
19:30.28hnoTurl, did you fix that for loop?
19:30.55Turlhno: yeah, I've been running it for some days and didn't notice any issues so far
19:31.27hnoHave a link to the full source?
19:31.35Turlhno: those patches ^
19:32.09Turldrop the ?diff to get the plaintext
19:32.25hnois only too lazy to apply the patch to read the result.
19:33.00hnocpu-freq.c in particular.
19:33.30mnemocTurl: url to the blob on github ;-)
19:33.32Turlhttps://github.com/allwinner-dev-team/linux-allwinner/blob/wip/allwinner-3.0/cpufreq/arch/arm/mach-sun4i/cpu-freq/cpu-freq.c
19:33.59cat1mnemoc: 3.4 looks fine too, at least no regressions. will send another set for 3.4 and let people test it.
19:34.03hnothanks!
19:34.39mnemoccat1: btw, 1623 is the chip-id of A10. so it doesn't fit in the header of a sunxi.c :p
19:34.52Turlcat1: add a link to github so I can cherrypick :)
19:35.36*** join/#arm-netbook merbzt (~benjamin@c-94-255-220-30.cust.bredband2.com)
19:36.23hnomnemoc, which header?
19:37.28cat1mnemoc: what do you mean does not fit?
19:37.58mnemochno: patch 4/4, first lines
19:38.11mnemoccat1: - * drivers/usb/host/sw_hci_sunxi.h
19:38.12mnemoc+ * drivers/usb/host/sw_hci_sunxi.h: header file for AW1623 HCI HCD
19:38.31cat1mnemoc: ah, you mean they now have different chip?
19:38.43mnemocA10 = 1623
19:38.47mnemocA13 = 1626
19:39.11Turlmnemoc: the current sun5i file claims * Description   : Include file for AW1623 HCI Host Controller Driver
19:39.16mnemocAW1623 is like the pro name, and A10 the marketing name
19:39.36mnemocTurl: :)
19:39.46Turland they invented another one for mach
19:39.51cat1mnemoc: i think i need to get rid of this confusion by removing any chip classification :)
19:39.53hnoTurl, cpu-freq changes looks sane to me. I assume you have verified the divisor selection.
19:40.00mnemoccat1: indeed
19:40.50Turlhno: well my chip hasn't burned or crashed yet :) it should choose the same divisors
19:42.22Turlbtw, this comment is interesting https://github.com/allwinner-dev-team/linux-allwinner/blob/wip/allwinner-3.0/cpufreq/arch/arm/mach-sun4i/cpu-freq/cpu-freq.c#L168
19:43.46hnoYes. And divisor table is well below those limits.
19:44.41Turlthe wiki quotes diff values however http://linux-sunxi.org/A10/CCM
19:45.44hnoReduced the divisor table like this in u-boot: https://github.com/hno/uboot-allwinner/blob/sunxi/arch/arm/cpu/armv7/sunxi/clock.c#L122
19:46.03techn_static struct cpufreq_frequency_table *sun4i_freq_tbl = NULL; .. should this be combined to static struct cpufreq_frequency_table *sunxi_freq_tbl = sunxi_frequency_table();  ??
19:47.46*** join/#arm-netbook pawel5870 (~pkarpins@staticline54070.toya.net.pl)
19:47.55Turltechn_: static requires a constant initializer
19:47.56techn_ah.. you have used it some other place..
19:48.09techn_but it's not used everywhere
19:49.20hnoTurl, have not seen those clock ranges before.
19:49.55mnemocafaik wiki values come from the datasheet/manual
19:50.04mnemocwhich not necesarily match reality :|
19:51.22hnodon't copy stuff from the manual please.
20:02.55*** join/#arm-netbook tmerle (~tmerle@bea31-1-78-226-152-136.fbx.proxad.net)
20:04.40*** join/#arm-netbook functionbeforefo (~functionb@206-248-137-101.dsl.teksavvy.com)
20:05.58*** join/#arm-netbook mikey_w (~mike@c-71-63-115-202.hsd1.va.comcast.net)
20:06.01*** join/#arm-netbook n6pfk (~mike@c-71-63-115-202.hsd1.va.comcast.net)
20:11.06techn_Turl: How I'll see effect of your patch?
20:11.52Turltechn_: if you see any effect, it's a bug
20:12.00Turlmake sure you have cpufreq enabled
20:13.40*** join/#arm-netbook lkcl (~lkcl@host81-135-78-226.range81-135.btcentralplus.com)
20:13.53techn_bogomisp while idle 60, bogomips while top/htop 60 (cpu 30%), bogomips while es2gears 1000 (cpu 85%)
20:15.05Turltechn_: this doesn't change anything regarding the cpufreq logic
20:15.12Turljust the magic to switch between them
20:15.43techn_so it should be more responsive?
20:15.49Turlno
20:15.52Turlit should be the same
20:16.30Turlthe logic that makes it check load, notice it's high, therefore go 60->1000 or whatever is unchanged
20:16.42Turlmy patch changes the stuff that actually does the switch
20:16.45Turlit's lower level
20:17.24mnemoctechn_: it's a cleanup
20:18.02Turlthat pretty much summarizes it :P
20:18.32mnemoceverything should work just as always, but with an implementation simpler to maintain, extend and improve in future
20:19.44*** join/#arm-netbook cat1 (~cat_x301@a91-152-66-200.elisa-laajakaista.fi)
20:20.34techn_ok.. no difference seen.. yet :)
20:21.21*** join/#arm-netbook cat_n9 (~communi@a91-152-66-200.elisa-laajakaista.fi)
20:21.53*** join/#arm-netbook popolon (~popolon@og-free.planet-service.fr)
20:23.49hnotechn_, that's good.
21:04.57mnemoccat_n9: does your patchset consider the commit from a13-sdk we had to merge?
21:07.36*** join/#arm-netbook alcides` (~alcides@187.23.195.197)
21:08.31mnemoccat_n9: or that goes after the refactoring?
21:09.32Turlmnemoc: I think he said he thought it was better after cleaning up the current stuff
21:09.45*** join/#arm-netbook gzamboni (~gzamboni@2a02-8426-513b-5100-0224-21ff-fe57-307c.rev.sfr.net)
21:10.18mnemocTurl: thanks :)
21:16.50libvf me.
21:17.07libvi just pulled one of the OSAL files through google translate.
21:17.32TurlOSAL?
21:17.35libv"NO SHIT SHERLOCK" is the phrase that popped up in my mind every few seconds
21:17.37Turlmnemoc: https://github.com/allwinner-dev-team/linux-allwinner/commits/allwinner-3.0-cpufreq
21:17.45libvvideo/sunxi/disp/OSAL
21:17.48libvwhatever it means
21:18.02libvfunction definition example...
21:18.05libvvoid OSAL_Sleep(__u32 Milliseconds);
21:18.23libvcomment in chinese on the line before, and a comment afterwards
21:18.32libvguess what was in the comment afterwards
21:18.42Turlsleeps for x milliseconds?
21:19.02mnemocOperating System Abstraction Layer
21:19.06libvTurl: nono, that would fit in one comment!
21:19.17libvTurl: the comment before was (translated) "sleeping"
21:19.30libvthe one afterwards said "unit: milliseconds"
21:19.35Turllol
21:19.50Turlmnemoc: if you don't see anything obviously wrong with that cleaned up branch I'll push to ML
21:20.15mnemocneither do I, but let's wait
21:20.41Turlbtw github has ?w=1 in case you guys didn't know :)
21:20.43Turlhttps://github.com/allwinner-dev-team/linux-allwinner/commit/86d73a5eebedb961e076177741132e008812df8b?w=1
21:20.50mnemocalso, there is a patch to fix the sata fix to not break mali in A10-revB
21:21.09libvthis will make for a time and nerve wrecking commit, which every one will have a really nice laugh about
21:21.43mnemoclibv: OSAL removal deserves a monster commit
21:22.07libvwhat is OSAL anyway?
21:22.10libvah
21:22.16libvos abstraction layer
21:22.18libvf me.
21:22.48libvno wonder anyone translated this crap yet :)
21:22.56mnemocto use their stuff in their rtos, on bare metal, on u-boot, on linux, on wince, ...
21:23.10*** join/#arm-netbook ibrah (~chatzilla@212.49.88.105)
21:23.40libvyeah
21:23.58Turlthey have a RTOS? o.O
21:24.01libvit hit me as i saw my "what is OSAL anyway?" line hit the channel
21:24.19mnemoc:)
21:25.00mnemocTurl: yes, they have their own closed rtos based on another well known rtos.... it's mentioned in the mail tom sent to the u-boot list
21:25.06mnemocdon't remember the names
21:25.10libvthis is just bad busywork...
21:25.51TurlVxWorks
21:25.51Turl?
21:25.59Turl(is that an RTOS?)
21:26.17libvTurl: well... it might be realtime certified.
21:26.38libvand the mother company might have been one of the companies defining the certification protocol
21:26.59*** join/#arm-netbook leszcz (~marek@host-89-167-63-165.nplay.net.pl)
21:27.39libvdon't go near automotive companies... you lose the will to live.
21:29.05mnemoclooks at his copy of misra c
21:31.32libvmnemoc: osal removal does not get a monster commit, it still needs to be reviewable.
21:31.48lkclosal comes up in the cedarx library.  one of the files
21:32.07lkclthe c source is available for it.
21:32.08mnemocthere are copies of osal on most drivers
21:32.09Turlprobably legacy from sun3i :)
21:32.14libvbut there will be a big "all these functions only return 0 anway" commit
21:32.19mnemoclkcl: we are talking about removing it :)
21:32.23hnoOSAL is operating system abstraction layer or similar acronym.
21:32.24lkcljoy!
21:32.49lkclmnemoc: that wouldn't impact libcedarx from working, would it?
21:32.53mnemocand about the comments on them
21:33.14lkcllibcedarx uses osal to initialise the cedarx hardware (via ioctls)
21:33.34mnemoclkcl: libcedarx has it's own copy. not affected by removing abstractions layer within the drivers
21:33.53lkcls/it's/its
21:34.02libvlkcl: will just be about libc wrapping i guess
21:34.23jelly-homes#s/it's/its#s/it's/its/# </pedant>
21:34.49mnemocok. sorry for the typo
21:35.09lkcli actually saw someone use "its" where "it's" was supposed to be used.  first time, last week, i fell over backwards
21:35.17lkclfirst time in 16 years of being on the internet
21:35.19lkclanyway
21:35.24Turlwha?
21:35.29Turlfirst time? :|
21:35.36libvmnemoc: if only the osal authors knew as much english as you do :)
21:35.52lkclTurl: yeah.  most people get it wrong the other way round.
21:35.56libvwait, then they would've been clued and they wouldn't have written osal
21:36.06lkcllibv: :)
21:36.39Turllibv: it's probably a 'we have this old sun3i code, need to ship sun4i the cheapest, what do we do? REUSE ALL THE THINGS!11'
21:37.05libvTurl: they just did what ATI does for every new radeon revision...
21:37.12Turlcedarx for example, it's hacked into android as a media player
21:37.29libvcopy tree of closest revision, hack it, throw it on the pile, start again
21:37.30Turlwhen the right, future proof way is OMX
21:37.30mnemocTurl: that's the normal way of "the industry"
21:38.04Turlintel rewrites its drivers yearly pretty much :P
21:38.13libvTurl: no, intel is different
21:38.50libvTurl: they have dozens of teams competing against eachother, each with a slightly different focus and absolutely no visibility of what the rest of the company is doing
21:39.06libvthere is no big pile
21:39.25libvjust tons of small piles that are broken in different ways, and noone can help you fix them
21:41.08libvmaybe osal is not the worst thing on the planet :)
21:42.51*** join/#arm-netbook mysteryname (~mysteryna@14-202-9-167.tpgi.com.au)
21:42.55hnothe previous os (which is still running in boot.axf) was appraently uCOS based.
21:44.12hnoand boot.axf, u-boot and kernel all share the same drivers mostly.
21:44.24hnoboot1 also.
21:46.33mnemoclibv: not sure which copy of osal it was, but there is an osal_malloc() that always return NULL
21:47.10Turllol
21:47.14Turlnice implementation of it
21:48.44Turljust noticed cpufreq commit date
21:48.52Turl5 months ago :O
21:49.45mnemoccheck authoriship date
21:50.07mnemoc5 months ago I stoped rebasing the 3.0 branch
21:50.25Turlmnemoc: Date: Thu, 31 May 2012 17:57:20 +0000
21:50.46mnemocscratches his head
21:51.22Turltime flies
21:53.09mnemocit's october and the first public proto of the cubieboard is from august
21:53.47mnemocdoesn't remember when the eoma68 thing started
21:53.57mnemoc2y ago?
22:02.50*** join/#arm-netbook ibrah (~chatzilla@212.49.88.108)
22:08.56*** join/#arm-netbook functionbeforefo (~functionb@206-248-137-101.dsl.teksavvy.com)
22:14.07*** join/#arm-netbook revident (~scott@fw.ss.org)
22:22.50leszczi tried to build u-boot with : make sun4i CROSS_COMPILE=armv7a-hardfp-linux-gnueabi-
22:23.02leszczbut it failed : error: /usr/lib/gcc/armv7a-hardfp-linux-gnueabi/4.7.2/libgcc.a(bpabi.o) uses VFP register arguments, u-boot does not
22:23.14leszczafter : git checkout 10664cbea6575bf77259f4e4db5bb566540b1961
22:23.26leszczcompilation went successfully but there is no sunxi-spl.bin in spl directory
22:23.34mnemochttps://github.com/linux-sunxi/u-boot-sunxi/wiki
22:33.54libvavernos: seems i have been wasting your time
22:34.35libvavernos: i have just cleared video/sunxi of non-ascii, and the OSAL crap can be ignored anyway
22:34.55libvnot a single comment otherwise contained anything useful
22:48.19mnemoclibv: quite expectable for such a library where things are basically only OSAL_foo() calling foo()
22:53.31libvmnemoc: but the few comments outside of osal were also useless
22:53.57libvwho would've thought that the guy implementing a linked list would have comments worth reading?
22:54.13mnemoc:D
22:57.12hnolibv, quite many of the chineese comments in allwinner sources are no more than a translation of the function name or variable being accessed.
23:34.57lundmanyeah the comments are not great
23:37.19*** join/#arm-netbook ZaEarl (~malmrose@17.host-182-48-148.compassnet.co.nz)

Generated by irclog2html.pl Modified by Tim Riker to work with infobot.