| 00:16.30 | *** join/#arm-netbook sepp_ (~seb@port-92-206-55-126.dynamic.qsc.de) |
| 00:33.19 | *** join/#arm-netbook uwe_mobile__ (~uwe@static.88-198-8-117.clients.your-server.de) |
| 00:33.31 | *** join/#arm-netbook Jacmet_ (~peko@stolen.plutonium.dk) |
| 00:38.00 | *** join/#arm-netbook hno (~hno@vps1.hno.se) |
| 00:38.00 | *** join/#arm-netbook ssvb (~ssvb@a88-114-220-213.elisa-laajakaista.fi) |
| 00:38.01 | *** join/#arm-netbook specing_ (~specing@93-103-140-50.dynamic.t-2.net) |
| 00:42.29 | *** join/#arm-netbook lundman (~Accident@194.109.206.190) |
| 00:42.50 | *** join/#arm-netbook t0dbld|work (~t0dbld@unaffiliated/t0dbld) |
| 00:42.50 | *** join/#arm-netbook A0Sheds (~ly@unaffiliated/l84supper) |
| 00:44.16 | *** join/#arm-netbook gsilvis (~almostsix@50.12.163.241) |
| 00:44.16 | *** join/#arm-netbook mtd_ (~martin@chop.xades.com) |
| 00:44.16 | *** join/#arm-netbook Gumboot_ (~gumboot@rev.bovine.muck.net.nz) |
| 00:44.16 | *** join/#arm-netbook sepp_ (~seb@port-92-206-55-126.dynamic.qsc.de) |
| 00:44.16 | *** join/#arm-netbook hp_|2 (~kvirc@220-136-110-98.dynamic.hinet.net) |
| 00:44.16 | *** join/#arm-netbook Faqtotum (ad0b7da2@gateway/web/freenode/ip.173.11.125.162) |
| 00:44.16 | *** join/#arm-netbook eflatun (~eflatun@46.1.121.226) |
| 00:44.16 | *** join/#arm-netbook DonkeyHotei (~danielg4@50-0-149-176.dsl.dynamic.sonic.net) |
| 00:44.17 | *** join/#arm-netbook nikk_|zzz (~somebody@nikk.starlink.ru) |
| 00:44.17 | *** join/#arm-netbook burndtjamb (~rapefruit@CPE-72-131-120-166.wi.res.rr.com) |
| 00:44.17 | *** join/#arm-netbook eFfeM_work (~frans@D4B268BA.static.ziggozakelijk.nl) |
| 00:44.17 | *** join/#arm-netbook arokux (~arokux@merkur178.inf.uni-konstanz.de) |
| 00:44.17 | *** join/#arm-netbook lerc (~quassel@121-74-230-7.telstraclear.net) |
| 00:44.17 | *** join/#arm-netbook Charlie_ (~quassel@64.31.59.70) |
| 00:44.17 | *** join/#arm-netbook fjfalcon (~fjfalcon@fjfalcon.ru) |
| 00:44.18 | *** join/#arm-netbook t0dbld (~t0dbld@unaffiliated/t0dbld) |
| 00:44.18 | *** join/#arm-netbook zewelor (~x@89-78-93-36.dynamic.chello.pl) |
| 00:44.18 | *** join/#arm-netbook marcan (marcan@marcansoft.com) |
| 00:44.18 | *** join/#arm-netbook mnemoc (~amery@shell.opensde.net) |
| 00:44.18 | *** join/#arm-netbook Darkwing (~quassel@ubuntu/member/davidwonderly) |
| 00:44.18 | *** join/#arm-netbook CIA-91 (~CIA@cia.atheme.org) |
| 00:50.57 | *** join/#arm-netbook t0dbld|work (~t0dbld@unaffiliated/t0dbld) |
| 00:50.57 | *** join/#arm-netbook A0Sheds (~ly@unaffiliated/l84supper) |
| 00:51.17 | *** join/#arm-netbook ZaEarl (~malmrose@173-11-125-162-SFBA.hfc.comcastbusiness.net) |
| 00:54.37 | *** join/#arm-netbook Hexxeh (u1532@gateway/web/irccloud.com/x-yweadjadiklteuob) |
| 02:21.21 | *** join/#arm-netbook mikey_w (~mike@va-76-6-223-110.dhcp.embarqhsd.net) |
| 02:21.32 | *** join/#arm-netbook mike_ (~mike@va-76-6-223-110.dhcp.embarqhsd.net) |
| 03:02.08 | *** join/#arm-netbook QingPei (~qingpei@221.220.107.169) |
| 03:32.19 | *** join/#arm-netbook kaspter (~kaspter@116.231.122.26) |
| 04:16.29 | *** join/#arm-netbook eflatun (~eflatun@46.1.121.226) |
| 05:02.43 | *** join/#arm-netbook hno (~hno@squid/developer/hno) |
| 05:16.57 | eFfeM_work | gm |
| 05:17.23 | lundman | an |
| 05:17.31 | *** join/#arm-netbook ZaEarl (~malmrose@173-11-125-162-SFBA.hfc.comcastbusiness.net) |
| 06:52.59 | *** join/#arm-netbook rell01 (~chatzilla@p5B07815A.dip0.t-ipconnect.de) |
| 06:55.04 | *** join/#arm-netbook RITRedbeard (~RITReadbe@c-68-37-165-37.hsd1.nj.comcast.net) |
| 07:21.46 | *** join/#arm-netbook vgrade (~martinbro@cpc2-nrte22-2-0-cust128.8-4.cable.virginmedia.com) |
| 07:23.53 | *** join/#arm-netbook hipboi (~hipboi@221.4.213.95) |
| 07:24.50 | t0dbld | good morning hipboi :-) |
| 07:25.00 | hipboi | hi |
| 07:27.08 | hipboi | does anyone know how to compile the x11 driver for mali? |
| 07:28.01 | t0dbld | well we dont have full stack .... but instructions in the packaed of released parts seemed straight forward |
| 07:28.15 | t0dbld | btw hipboi i circiled you on g+ :-) |
| 07:29.13 | hipboi | t0dbld: ok |
| 07:29.22 | t0dbld | without full stack we get no HW acell right now any how so im using jsut fbdev |
| 08:28.44 | zenitraM | hipboi: to compile the X11 driver the mali DDK is needed, which we don't have |
| 08:28.50 | zenitraM | for some ump.h library |
| 08:33.06 | hipboi | i have the mali ddk source code |
| 08:33.33 | hipboi | but i don't know how to build the x driver |
| 08:35.38 | zenitraM | mm let me see |
| 08:36.03 | zenitraM | there is a build.sh |
| 08:36.16 | zenitraM | you have to edit and add the paths to the ddk |
| 08:38.10 | hipboi | is this the x driver for mali? |
| 08:38.11 | hipboi | http://git.linaro.org/gitweb?p=people/chunsangjeong/xf86-video-mali.git |
| 08:39.07 | zenitraM | it seems like it |
| 08:43.21 | zenitraM | http://git.linaro.org/gitweb?p=people/chunsangjeong/xf86-video-mali.git;a=blob;f=build.sh;h=e1b7f1d92d120265e0ecf974babde416bac0119d;hb=85ea5ad85c48db280424ca028ce276f07fb407ed |
| 08:43.48 | zenitraM | you have to add the paths there, i tried it on the a10 itself setting rootfs to /usr/local itself |
| 08:45.29 | hipboi | so you build on A10? |
| 08:48.31 | zenitraM | yes |
| 08:49.06 | zenitraM | it's easier to just apt-get the dependencies |
| 09:04.25 | *** join/#arm-netbook nikk (~somebody@nikk.starlink.ru) |
| 09:33.27 | *** join/#arm-netbook zoobab (zoobab@vic.ffii.org) |
| 09:33.31 | zoobab | hi |
| 09:49.46 | *** join/#arm-netbook Charlie_ (~quassel@64.31.59.70) |
| 10:24.06 | mnemoc | hno: hi, when you have some time can you take a look into sun4i/pm/standby ? we are the *only* mach building a .bin |
| 10:26.58 | mnemoc | so I assume it's the wrong way |
| 10:34.49 | CIA-91 | rhombus-tech: giuseppe master * r713687b5a603 /allwinner_a10/orders/Quantity:_1_Max_Unit_Price:___36__60_Max_Budget:__36__100_Contact:_giuseppe.broglia__64__gmail.com_Purpose:_mediaserver:_stable_Status:_preorder.mdwn: |
| 10:51.36 | hno | mnemoc, no bin in sun5i? |
| 10:53.34 | mnemoc | hno: the 3 sun?i do it |
| 10:54.12 | mnemoc | hno: the original code requires a separated `make` and that is changed for sun4i |
| 10:54.25 | mnemoc | hno: but the basic idea remains, embed a .S |
| 10:54.37 | mnemoc | and no one else does something similar |
| 10:54.47 | hno | what that is is a small standalone program loaded into sram to handle DRAM refresh and resume from suspend. |
| 10:56.00 | mnemoc | so it's ok and doesn't need to be reworked in a different way? |
| 10:56.26 | mnemoc | I find so weird no other mach-* does so... |
| 10:58.43 | mnemoc | because handling DRAM refresh to resume from suspend seems like a normal thing to do |
| 11:01.52 | *** join/#arm-netbook eflatun_ (~eflatun@95.173.235.109) |
| 11:02.08 | zenitraM | this is china, you can't assume normal things to do |
| 11:02.26 | mnemoc | :) |
| 11:06.20 | *** join/#arm-netbook eflatun_ (~eflatun@95.173.230.46) |
| 11:06.23 | hno | mnemoc, not sure there is no other doing similar things. But I guess most simply uses careful alignment in the l2 cache and not dedicated sram. |
| 11:06.44 | mnemoc | i see |
| 11:08.12 | hno | there is quite likely a bit of legacy involved here as weill. many of the functions in standby is no more than stubs. |
| 11:09.43 | hno | Hmm... looking at a number of other arches they all resume from bootloader. |
| 11:19.24 | hno | Interesting. in standby the core is running at 32Khz. |
| 11:53.14 | nikk | guys, don't you know, is initlogo.rle showing on android boot is a kernel or init thing? and why in case of A10 it was not in RLE format? |
| 11:55.53 | nikk | strange thing is grep -R "ANDROID_BOOTLOGO" in root of ICS source gives me almost nothing, only init.rc references. looks like that variable not used at all ;-) |
| 11:58.32 | lundman | uboot can also show logos |
| 12:05.59 | nikk | In case of A10, I think there is 2 logos, one in u-boot (it actually showing), and second one was initlogo.rle (not in rle format), which was showing on Android boot before boot animation starts. Now I see only 1st (u-boot) one. |
| 12:10.54 | nikk | Allwinner also provides windows tool BMPconvert for making that image, which converts graphics to ARGB format, and naming it initlogo.rle... |
| 12:19.26 | *** join/#arm-netbook specing_ (~specing@unaffiliated/specing) |
| 12:26.16 | *** join/#arm-netbook kaspter (~kaspter@116.237.61.223) |
| 12:33.27 | A0Sheds | mnemoc: IMHO there will be fragmentation in the hardware standard, especially to one without patents and more flexibility for modern interfaces |
| 12:41.07 | *** join/#arm-netbook Mazon (~Mazon@95.166.104.217) |
| 12:58.44 | *** join/#arm-netbook leojrfs (~leojrfs@bl6-36-5.dsl.telepac.pt) |
| 13:04.11 | *** join/#arm-netbook eebrah (~Ibrahim_K@41-139-199-130.safaricombusiness.co.ke) |
| 13:14.59 | mnemoc | A0Sheds: I hate patents and licensing involved in free software with all my guts, but while I do love the idea of standarized interfaces and form factors I also believe the legal restrictions (even if merely psycological) are a problem to make anything "standard" used by more than one vendor |
| 13:18.28 | mnemoc | as long as something worthy of been called a standard interface/formfactor for core boards comes out, and there is one with A10 on it, I'm happy :) |
| 13:19.22 | A0Sheds | mnemoc: I've heard the same from others |
| 13:32.46 | *** join/#arm-netbook leojrfs (~leojrfs@233.163.108.93.rev.vodafone.pt) |
| 13:45.56 | hno | the boot logo on android booting is from the boot loader I think. |
| 13:48.08 | hno | wonder if there is an english version of the Allwinner boot loader manual. Only seen chineese version which takes some effort to digest. |
| 13:48.34 | hno | esp when not really trying.. as my goals is more in u-boot. |
| 13:51.50 | mnemoc | hno: is moving the dram refreshing to u-boot in your goals list? |
| 13:52.20 | hno | no |
| 13:52.29 | mnemoc | :) |
| 13:53.00 | hno | uboot have no right to existence after kenel have booted. |
| 13:53.22 | hno | it's just a means to load the kernel. |
| 13:54.40 | hno | which reminds me that I have not announced that spl will be able to load a kernel directly, for those who thinks uboot is crap. |
| 13:54.56 | mnemoc | :o |
| 13:55.45 | hno | all code is there, just need some slight reshuffling to make it available to more arches. |
| 13:56.44 | *** join/#arm-netbook leojrfs (~leojrfs@233.163.108.93.rev.vodafone.pt) |
| 13:58.16 | hno | current spl code is generally a good example of crap category. Loads of very nice generic code all spread out in different arches. |
| 13:59.15 | hno | with most in omap. |
| 14:04.17 | hno | Google translated version of the bootloader manual. https://docs.google.com/document/d/1TdDk7jBtHAEoqbj6q7a2sY9aPp6BxTkkmdlLMUZdInQ/edit |
| 14:08.35 | leojrfs | anyone got adb working with mele a1000? |
| 14:11.46 | mnemoc | hno: thanks I'll try to help improving the translation while reading it this weekend |
| 14:20.52 | mnemoc | hno: do you have the original (in chinese) up there too? |
| 14:28.41 | zenitraM | leojrfs: get adbWireless from the market |
| 14:28.56 | leojrfs | zenitraM, cool tnx |
| 14:32.24 | zenitraM | mnemoc: the original is at buildroot, a .doc with chinese name |
| 14:34.15 | zenitraM | 固件配置，打包，烧写指南.doc to be exact |
| 14:50.44 | hno | mnemoc, yes I have the original there to. |
| 14:51.04 | hno | or actually, a 0.001% translated original. |
| 14:52.29 | hno | the "original" is at https://docs.google.com/document/d/1yvDSXxtGDzUjjhS7jRUlv4l3wRWA-gap89mlqHddK88/edit |
| 14:57.09 | zenitraM | SPRis a thespriteshorthand, translated into Chinese, is the meaning of elf. Use this name, in the hope that the function of the system like the story of the Wizard, the powerful,rich. |
| 15:12.58 | hno | Redoign the translation. Google translate does much better job translating a sentence at a time than whole document. |
| 15:24.41 | *** join/#arm-netbook revident (~scott@fw.iplink.net) |
| 15:39.06 | A0Sheds | whats the status on the boot sections of the A10? do we have a register spec or just the example code for the SD card boot? |
| 15:41.54 | hno | A0Sheds, we have some register specs. And a lot of code. |
| 15:42.16 | hno | A0Sheds, any specifics you wonder about? |
| 15:43.11 | A0Sheds | hno: does the A10 have a self scan to check for strings at a preset order for boot devices? |
| 15:43.25 | A0Sheds | or pin-straps? |
| 15:51.12 | hno | A0Sheds, it has a predefined boot order. sdcard0, nand, sdcard2, spi I think. Not 100% on priority of spi. |
| 15:52.57 | A0Sheds | hno: thanks was wondering if it was TI OMAP like in any way |
| 15:53.55 | A0Sheds | OMAP has lots of boot order options, x86 is usually by pin strap, other ARM's are somewhere in between |
| 16:07.59 | *** join/#arm-netbook leojrfs (~leojrfs@bl6-36-5.dsl.telepac.pt) |
| 16:13.01 | *** join/#arm-netbook tom_say (~quassel@cpe-68-203-248-184.stx.res.rr.com) |
| 16:45.52 | *** join/#arm-netbook Charlie (~quassel@64.31.59.70) |
| 16:51.29 | *** join/#arm-netbook Charlie_ (~quassel@64.31.59.70) |
| 16:57.22 | *** join/#arm-netbook nikk (~somebody@nikk.starlink.ru) |
| 17:20.54 | *** join/#arm-netbook vgrade (~martinbro@cpc2-nrte22-2-0-cust128.8-4.cable.virginmedia.com) |
| 17:22.40 | vgrade | evening |
| 17:23.06 | vgrade | missed the comments after hipbio said he had the mali ddk |
| 17:23.43 | *** part/#arm-netbook QingPei (~qingpei@221.220.107.169) |
| 17:23.44 | vgrade | did anyone offer to help him comple things |
| 17:30.11 | mnemoc | I supose he went to sleep |
| 17:30.21 | mnemoc | btw, http://lightake.com/detail.do/sku.Pre_order_AK802_Android_4.0_Mini_PC_512MB_4GB_A10_Cortex_A8_Google_TV_Player_Black-54358 |
| 17:30.48 | mnemoc | "pre-order" |
| 17:34.38 | vgrade | mnemoc, if you see him before me let him know I'm here to help |
| 17:36.02 | mnemoc | ok, but you need to be here by the same time he appeared today :) |
| 17:42.36 | vgrade | nod |
| 17:45.34 | *** join/#arm-netbook Charlie (~quassel@64.31.59.70) |
| 18:19.58 | *** join/#arm-netbook leojrfs (~leojrfs@bl6-36-5.dsl.telepac.pt) |
| 18:27.49 | Mazon | forgot to bring popcorn before reading the mailinlist |
| 18:29.09 | mnemoc | doesn't like to trigger discussions like that :| |
| 18:29.10 | ZaEarl | I built my first (semi-)working rom! Check out the interface I ended up with: http://bit.ly/Kz41E1 |
| 18:32.20 | *** join/#arm-netbook leojrfs (~leojrfs@bl6-36-5.dsl.telepac.pt) |
| 18:32.36 | mnemoc | ZaEarl: set the launcher to 1 row? |
| 18:32.49 | ZaEarl | mnemoc, not on purpose. ;) |
| 18:33.05 | mnemoc | ^^ |
| 18:33.32 | mnemoc | or it's the docking that has too many options? |
| 19:02.48 | *** join/#arm-netbook frantisheq (~frantishe@kapusany.3log.sk) |
| 19:32.27 | *** join/#arm-netbook ZaEarl (~malmrose@173-11-125-162-SFBA.hfc.comcastbusiness.net) |
| 19:47.47 | *** join/#arm-netbook rell01 (~chatzilla@p4FE563AC.dip0.t-ipconnect.de) |
| 20:31.18 | CIA-91 | rhombus-tech: angelo master * r2163e1ebbe02 /allwinner_a10/orders/.mdwn: |
| 20:33.19 | *** join/#arm-netbook PhantomsDad (~PhantomsD@konversation/developer/PhantomsDad) |
| 20:35.13 | PhantomsDad | Greetings. I've built uboot-allwinner according to instructions at https://github.com/hno/uboot-allwinner/wiki |
| 20:35.56 | PhantomsDad | In the root folder, I now have a file named "u-boot" and another named "u-boot.bin". Which do I write to the SD card? |
| 20:37.13 | mnemoc | the spl.bin and u-boot.bin |
| 20:37.57 | PhantomsDad | ok, thanks |
| 20:37.58 | mnemoc | see the "How to make a bootable sunxi SD-card" section on that wiki |
| 20:38.04 | mnemoc | it's pretty simple |
| 20:38.34 | PhantomsDad | The instructions currently say dd if=uboot.bin of=/dev/sdc bs=1024 seek=32 |
| 20:38.45 | PhantomsDad | I'll correct the uboot.bin to u-boot.bin. |
| 20:39.11 | mnemoc | ok, that's a typo :) |
| 20:39.47 | mnemoc | PhantomsDad: can you fix it? |
| 20:40.07 | PhantomsDad | Trying to figure out how to do that right now. lol |
| 20:40.23 | mnemoc | https://github.com/hno/uboot-allwinner/wiki/Home/_edit |
| 20:41.02 | mnemoc | ruby style :< |
| 20:43.18 | PhantomsDad | Updated. |
| 20:43.33 | mnemoc | thanks :) |
| 20:45.14 | PhantomsDad | I've also successfully built linux-allwinner and copied arch/arm/boot/uImage to 1st partition on SD card. |
| 20:45.47 | PhantomsDad | What else do I need to do to make this work? |
| 20:46.01 | mnemoc | the script.bin corresponding to your device |
| 20:46.13 | *** join/#arm-netbook DonkeyHotei (~danielg4@50-0-149-176.dsl.dynamic.sonic.net) |
| 20:50.47 | *** join/#arm-netbook MaDMaLKaV (~IceChat9@188.86.253.69) |
| 20:51.20 | PhantomsDad | Sorry for stupid ?s. Where do I get the script.bin? I already have some .bin files in the 1st partition. Not sure now where I got them. Think it was following the instructions at http://www.cnx-software.com/2012/04/28/how-to-create-your-own-debian-ubuntu-image-for-mele-a1000-allwinner-a10-based-stb/ |
| 20:52.33 | mnemoc | script.bin is in the nanda, it's around ~40K (max 64K) |
| 20:53.23 | mnemoc | iirc in the ubuntu image it's called evb.bin |
| 20:53.43 | mnemoc | but the uboot you built will expect it been called script.bin |
| 20:53.54 | mnemoc | unless to modify it's env |
| 20:54.17 | PhantomsDad | yes. I tried copying the evb.bin to script.bin on the SD card. But still doesn't boot. |
| 20:55.16 | mnemoc | "doesn't boot" is pretty broad.... connect a serial console and look |
| 20:58.03 | PhantomsDad | Hooked up via HDMI to tv. Power on mele. tv screen goes blank (black), then blue screen. tv reports "720x480 @60Hz", then nothing. |
| 20:58.54 | PhantomsDad | So I'll see if I can get serial console working. tom cubie sent me an adapter I think I can use? |
| 21:00.02 | mnemoc | yes, it's quite simple |
| 21:00.11 | DonkeyHotei | mnemoc: how do you modify env, anyway? |
| 21:00.38 | mnemoc | once uboot boots, interrupt it, printenv, setenv ...., saveenv |
| 21:00.44 | mnemoc | boot |
| 21:01.21 | revident | DonkeyHotei, it needs to be done from the serial console, just as mnemoc just outlined |
| 21:08.13 | DonkeyHotei | but you can't get to the serial console unless you change the env |
| 21:08.34 | DonkeyHotei | bootdelay=0 prevents it |
| 21:08.57 | mnemoc | hno's tree has that set?? |
| 21:09.24 | DonkeyHotei | tree? |
| 21:09.44 | mnemoc | https://github.com/hno/uboot-allwinner |
| 21:10.53 | DonkeyHotei | i used the lichee-dev from hands.com, which has bootdelay=3, but that's ignored due to the existing env |
| 21:11.10 | DonkeyHotei | is hno's tree now preferable to that one? |
| 21:11.41 | steev | mnemoc: wouldn't it be nice to have display in uboot? |
| 21:11.41 | mnemoc | yes |
| 21:11.56 | mnemoc | steev: sure |
| 21:12.21 | steev | so hno's uboot-allwinner, not amery/uboot-allwinner? |
| 21:12.46 | mnemoc | hno's |
| 21:12.49 | steev | okay |
| 21:12.53 | steev | what about linux-allwinner |
| 21:12.58 | mnemoc | he maintains it, I don't :< |
| 21:13.19 | DonkeyHotei | wait, hno's github says it mirrors hands.com |
| 21:13.36 | mnemoc | DonkeyHotei: the description came from mine |
| 21:13.53 | DonkeyHotei | but branches shows lichee-dev is way behind sun4i |
| 21:14.15 | mnemoc | steev: I like to believe linux-allwinner is best from mine :p |
| 21:14.24 | steev | okay :) |
| 21:14.53 | mnemoc | amery/uboot-allwinner was initially just a mirror of hipboi's and denx, but hno forked it and started working on it |
| 21:14.55 | DonkeyHotei | hno: does the sun4i branch of uboot support mmc boot, nand boot, serial over mmc? |
| 21:18.48 | steev | DonkeyHotei: look through the git log |
| 21:18.56 | mnemoc | it's mmc-boot only for now |
| 21:19.02 | steev | "git log origin/sun4i" |
| 21:19.23 | DonkeyHotei | ok, so for nand boot, still lichee-dev? |
| 21:19.41 | mnemoc | yes |
| 21:19.44 | DonkeyHotei | ok |
| 21:20.03 | DonkeyHotei | but the question of the env remains |
| 21:20.37 | steev | you can change it in include/configs/sun4i_config (or whatever the file name is) |
| 21:22.28 | DonkeyHotei | that does not help me |
| 21:22.45 | DonkeyHotei | it's bootdelay=3 in the source and bootdelay=0 in the env |
| 21:22.56 | mnemoc | I suppose you can trash the current env using dd if=/dev/zero |
| 21:26.12 | DonkeyHotei | i want to preserve all but bootdelay |
| 21:26.34 | DonkeyHotei | and there seems to be stuff in the block besides the env |
| 22:09.41 | mnemoc | hno: spl only has access to 24K of memory?? and how on earth do you load linux? |
| 22:10.47 | mnemoc | hno: I knew you had limited bin size, but not that your access to memory was that constrained :( |
| 22:12.15 | DonkeyHotei | doesn't sound right |
| 22:16.06 | PhantomsDad | OK, got serial console working. Turns out that debian is booting! And I can SSH into it. How do I get boot messages to appear on tv? |
| 22:18.39 | mnemoc | the a10's kernel doesn't support fbcon, so no messages |
| 22:24.05 | PhantomsDad | setting CONFIG_FRAMEBUFFER_CONSOLE=y and compiling LYCHEE modules into kernel won't work? |
| 22:24.48 | mnemoc | no, it will panic |
| 22:25.25 | mnemoc | the fb driver (disp) lacks of several callbacks and tables needed by fbcon |
| 22:47.35 | mnemoc | PhantomsDad: the callbacks part is easily solvable using FB_CFB_*, but they rely in tables we don't have, and their implementation is not easy because the disp driver is a wrapper on top of a sort of ugly multiplatform library they wrote |
| 22:49.43 | PhantomsDad | ok, well ATM that is a bit over my knowledge level, but thanks for the info. |
| 22:50.49 | PhantomsDad | Dumb ?. Am I supposed to copy linux-allwinner/output/lib/modules/3.0.31+ to my rootfs? |
| 22:51.22 | hno | mnemoc, spl configures dram before it loads linux. If it is to read dram parameters from system.bin then it need to do that before configuring dram. |
| 22:52.52 | hno | DonkeyHotei, see the uboot-allwinner github wiki page. |
| 22:53.31 | hno | regarding the different branches. |
| 22:55.03 | DonkeyHotei | url? |
| 22:56.51 | hno | DonkeyHotei, https://github.com/hno/uboot-allwinner/wiki |
| 22:57.07 | mnemoc | hno: what about a sort of mmap? reading the .bin without loading it |
| 22:57.58 | mnemoc | script.bin is ~40K, but it could be up to 64K |
| 22:59.49 | hno | Probably better to encode the dram parameters in spl, like what is done in boot0. |
| 23:00.16 | hno | or have an spl configuration block. |
| 23:00.24 | mnemoc | but that means having different spl.bin for each... |
| 23:00.43 | hno | or a tool to copy what is needed from system.bin/fex. |
| 23:01.11 | hno | I don't really like relying on fat. |
| 23:01.48 | hno | boot0 have a file header where this stuff gets configured. |
| 23:02.08 | hno | should be possible to add the same header to spl. |
| 23:02.23 | hno | just need to figure out the structure of that header.. |
| 23:02.31 | mnemoc | what about reading that header from spl directly? |
| 23:02.40 | hno | from where? |
| 23:03.00 | mnemoc | boot0 |
| 23:03.08 | hno | there is no boot0 |
| 23:03.18 | hno | boot0 == spl |
| 23:03.26 | hno | s/=/~/ |
| 23:03.28 | mnemoc | but on the rom of the board |
| 23:03.43 | mnemoc | or it's out of reach? |
| 23:04.30 | hno | pretty much yes. allwinner NAND support is fairly large piece of code, and add FAT to that.. |
| 23:05.43 | hno | plus that I do not want to require a FAT boot partition. I am perfectly fine with block loading spl + uboot + environment. |
| 23:07.14 | hno | [henrik@home sun4i_nand]$ size built-in.o |
| 23:07.14 | hno | <PROTECTED> |
| 23:07.14 | hno | <PROTECTED> |
| 23:07.57 | mnemoc | ouch |
| 23:08.03 | mnemoc | ok, discarded |
| 23:08.52 | hno | FAT is ~10K |
| 23:10.25 | mnemoc | so dram info goes somewhere between the spl and uboot |
| 23:10.48 | mnemoc | a 3rd `dd` or build-time? |
| 23:10.54 | hno | more likely the file header of spl. |
| 23:11.04 | hno | so brom loads it for us. |
| 23:11.10 | mnemoc | nice |
| 23:11.53 | hno | only that we don't know the file header structure used by boot0. But we can always make our own. |
| 23:12.06 | A0Sheds | http://japan.renesas.com/media/products/soc/assp/mobile/emma_mobile/emma_mobile_ev/R18PF0000EJ0000_EMEV2_E.pdf |
| 23:12.24 | hno | reusing boot0 header means we can reuse existing tools for configuring it. |
| 23:12.51 | hno | and a better understanding of the basic system parameters allwinner considers important. |
| 23:12.53 | A0Sheds | ends up renesas has a dual ARM cortex A9 1GHz with all the bells and whistles |
| 23:13.56 | A0Sheds | only with more peripherals than the EV2 above ^^ |
| 23:14.09 | A0Sheds | also includes SATA |
| 23:14.15 | mnemoc | A0Sheds: got the samples? |
| 23:14.49 | A0Sheds | the quad core is >$40 even at 1M pcs |
| 23:15.28 | mnemoc | 1M pcs???? |
| 23:15.31 | hno | A0Sheds, don't see SATA mentioned? |
| 23:15.35 | A0Sheds | so they are offering a dual core arm cortex a9 1GHz with USB, 10/100, SPI, SATA, PCIe |
| 23:15.46 | A0Sheds | oh, it not that exact part |
| 23:15.54 | A0Sheds | it not on their website |
| 23:16.16 | A0Sheds | it similar to the one in the link but with higher speeds SATA and PCIe |
| 23:17.16 | A0Sheds | http://am.renesas.com/products/soc/assp/av/enc_dec/r_home/s1/index.jsp no I found it |
| 23:17.22 | mnemoc | A0Sheds: perfect for something like via's apc.io #fail done right |
| 23:17.25 | *** join/#arm-netbook ssvb (~ssvb@a88-114-220-213.elisa-laajakaista.fi) |
| 23:18.09 | A0Sheds | yeah, who is going to want an arm11 board for $50? |
| 23:18.59 | mnemoc | without PCIe and sata it's absurd to sell it as PCish |
| 23:19.10 | A0Sheds | also offered them Linux support for register specs |
| 23:20.32 | A0Sheds | they are also asking the HQ in Japan about any other parts that would be a good fit |
| 23:20.43 | A0Sheds | so we'll see next week |
| 23:22.02 | A0Sheds | they seemed to like the modular idea |
| 23:23.13 | A0Sheds | not sure about it's low power modes since it's for Home Consumer |
| 23:23.43 | A0Sheds | they might have left out low power modes for mobile/battery powered applications |
| 23:26.10 | A0Sheds | Samples of the R-Home S1 SoC will be available in May 2012, priced at US$30 per unit. |
| 23:26.18 | Mazon | odd, the other day I mentioned that there might be some overlap between olimex rhombus-tech, but was put down - now that Tsvetan mentioned it on the mailing list, luke is all over him *ponder* |
| 23:26.39 | Mazon | oh, well - the more the merrier |
| 23:27.00 | hno | A0Sheds, that CPU have a SGX531 PowerVR GPU, which means chances of something working out well for Linux is pretty slim. |
| 23:27.35 | A0Sheds | hno: yeah, we discussed that, but Renesas seems to have a deal in place with PowerVR |
| 23:28.08 | hno | Mazon, of course there is overlap in interest. |
| 23:28.23 | Mazon | well, that was what I thought |
| 23:28.31 | hno | but quite different goals. |
| 23:28.37 | Mazon | idd |
| 23:28.38 | A0Sheds | they do support Linux, but the PowerVR section would still be closed source, like the TI parts |
| 23:29.23 | hno | PowerVR Linux support do not have a good track record at all, even with closed source drivers from major vendors. |
| 23:29.25 | A0Sheds | whats the current state of linux and PowerVR, i know the libs are still closed, but is accel supported? |
| 23:30.01 | hno | no idea what current status today is, but would not expect it to be great. |
| 23:30.27 | A0Sheds | what works and what doesn't? |
| 23:31.46 | A0Sheds | beagleboards and pandaboards have the OMAP's with powerVR |
| 23:33.40 | hno | SGX531 should be very similar to the SGX530 used in beagleboard so some research on the state there would be recommended. |
| 23:35.30 | A0Sheds | OpenGL ES 2.0 |
| 23:38.26 | A0Sheds | we'll see if they come up with samples and the register spec |
| 23:39.19 | A0Sheds | they might even start lurking here |
| 23:39.43 | A0Sheds | they were impressed with the list of software talent |
| 23:41.44 | A0Sheds | 32 bit DDR3 and PowerVR are the down sides so far for a perfect world |
| 23:57.32 | hno | does some math... fat support most likely will not fit in spl unless there is other crap that can be thrown out. |
| 23:58.50 | mnemoc | hno: just pass the ball to u-boot.bin, not worthy |
| 23:59.21 | hno | will add direct kernel loading anyway. |