| 02:30.34 | *** join/#arm-netbook ManoftheSea (~dlahouss@72.22.81.52) |
| 02:36.57 | *** join/#arm-netbook mikey_w (~mike@c-71-63-115-202.hsd1.va.comcast.net) |
| 02:37.16 | *** join/#arm-netbook mike_ (~mike@c-71-63-115-202.hsd1.va.comcast.net) |
| 03:31.01 | *** join/#arm-netbook CaCtus492 (~Kent@123-243-197-152.static.tpgi.com.au) |
| 03:36.32 | steev | A0Sheds: have a link to the netbook? |
| 04:02.49 | A0Sheds | http://sunlike.en.alibaba.com/product/342355631-212226943/3G_Built_in_Multitouch_MID_tablet_PC_with_android2_2_8GB_to_32GB.html |
| 04:03.42 | A0Sheds | http://www.cnx-software.com/2012/04/16/sunlike-umpc-1021-85-usd-allwinner-a10-based-10-2-android-4-0-netbook/ |
| 04:04.04 | A0Sheds | steev ^^ |
| 04:05.46 | *** join/#arm-netbook mikey_w (~mike@c-71-63-115-202.hsd1.va.comcast.net) |
| 04:06.03 | *** join/#arm-netbook mike_ (~mike@c-71-63-115-202.hsd1.va.comcast.net) |
| 04:07.34 | steev | ... holy shiz, those look like our case design for our unreleased mx53 |
| 04:07.44 | steev | well now i know where our missing prototype is |
| 04:08.58 | steev | though we scrapped it because us devs said the design looked like junk compared to our "old" product |
| 07:57.28 | CIA-158 | rhombus-tech: Gregor master * rfd417918534c /allwinner_a10/orders/zombig__64__gmail.com.html: |
| 08:19.07 | CIA-158 | rhombus-tech: Slavon master * ra081f87d7266 /allwinner_a10/orders/slavon.mdwn: |
| 08:28.05 | CIA-158 | rhombus-tech: Henk master * r8d118eaef63f /allwinner_a10/orders/.mdwn: |
| 08:38.31 | CIA-158 | rhombus-tech: dod master * r01f59809f200 /allwinner_a10/a10_image.mdwn: alternative: use kpartx instead of offset |
| 08:42.15 | CIA-158 | rhombus-tech: dod master * r4dbfa359c2a8 /allwinner_a10/a10_image.mdwn: |
| 09:01.39 | *** join/#arm-netbook QingPei (~qingpei@61.149.86.182) |
| 09:13.04 | CIA-158 | rhombus-tech: Dawid master * rc5cdf30a14f2 /allwinner_a10/a10_image.mdwn: Moved comment to where it belongs |
| 09:31.59 | CIA-158 | rhombus-tech: Zoltn master * r62e09ac45add /allwinner_a10/orders/14.mdwn: |
| 09:32.03 | *** join/#arm-netbook ManoftheSea (~dlahouss@72.22.81.52) |
| 09:32.21 | *** join/#arm-netbook traeak (~bolsen@75-151-91-165-Colorado.hfc.comcastbusiness.net) |
| 10:49.50 | *** join/#arm-netbook penguin42 (~dg@tu006.demon.co.uk) |
| 10:52.16 | mnemoc | what's the driver for the ethernet port of the mele?:q! |
| 10:52.21 | mnemoc | err |
| 10:54.41 | mnemoc | [ 0.530000] wemac Ethernet Driver, V1.00 in file:drivers/net/sun4i/sun4i_wemac.c |
| 10:54.45 | mnemoc | [ 43.890000] wemac wemac.0: enabling eth0 |
| 10:54.49 | mnemoc | uhm.... |
| 10:55.46 | mnemoc | then what's the rtl8201cp realtek chip doing?? |
| 10:55.51 | mnemoc | .oO |
| 11:01.17 | zenitraM | oh, wait, i though rtl8201 was the wireless |
| 11:01.40 | zenitraM | which wireless module does it have then? |
| 11:06.52 | CIA-158 | rhombus-tech: Ivan master * rf6f3376225c5 /allwinner_a10/orders/vaaano.mdwn: |
| 11:08.58 | CIA-158 | rhombus-tech: Ivan master * r16482110cdd9 /allwinner_a10/orders/vaaano.mdwn: |
| 11:09.50 | A0Sheds | http://www.hksunlike.com/ for ~$80 I'm interested in the hack they did for this design. I can't tell if it's that horrible case they used for the AUGEN netbooks or not. But I am wondering why they are including the Augen pics in the description. |
| 11:11.13 | A0Sheds | For a few $ more they could have a case easily on par with any of the Asus or HP netbooks. |
| 11:13.24 | A0Sheds | http://www.cnx-software.com/2012/04/16/sunlike-umpc-1021-85-usd-allwinner-a10-based-10-2-android-4-0-netbook/ |
| 11:15.27 | A0Sheds | in the video it looks like the awful Augen case was reused. Look at the giant bezel around the display! |
| 11:28.35 | *** join/#arm-netbook Byan (byan@gateway/shell/mtu-lug/x-tcvlrfgasmcrcoil) |
| 11:34.54 | CIA-158 | rhombus-tech: ron93 master * r9e4a0c58913e /allwinner_a10/orders/.mdwn: |
| 11:38.34 | mnemoc | zenitraM: I believe the wifi is the other realtek chip next to the power button |
| 11:38.59 | mnemoc | rtl8188cus |
| 11:39.10 | mnemoc | using rtl8192cu driver |
| 11:41.36 | mnemoc | in 2.6.36 wemac enables the port just fine (obviusly) |
| 11:41.38 | mnemoc | [ 43.890000] wemac wemac.0: enabling eth0 |
| 11:41.38 | mnemoc | [ 43.930000] eth0: link up, 100Mbps, full-duplex, lpa 0xFFFF |
| 11:42.05 | mnemoc | but in 3.0.28 doesn't |
| 11:42.11 | mnemoc | <7>wemac wemac.0: enabling eth0 |
| 11:42.11 | mnemoc | <4>wemac wemac.0: WARNING: no IRQ resource flags set. |
| 11:42.11 | mnemoc | <7>wemac wemac.0: resetting device |
| 11:42.11 | mnemoc | <6>wemac wemac.0: eth0: link down |
| 11:42.28 | *** join/#arm-netbook cromo (~cromo@90-255-251-94.net.stream.pl) |
| 11:43.09 | mnemoc | so fixing the disp driver will have to keep waiting :-/ |
| 11:43.14 | cromo | hi! I'm Dawid Wróbel, you may know me from some recent activity on gmane.comp.hardware.netbook.arm |
| 11:43.42 | cromo | I received mele a1000 on Friday and started some hacking |
| 11:44.06 | cromo | first of all, I am trying to build armhf kernel and boot Ubuntu Precise armhf root with it |
| 11:44.33 | cromo | I just built 2.6.36 kernel with gnueabihf toolchain, but it won't boot |
| 11:44.54 | specing | meh ubuntu, /ignore/ |
| 11:45.03 | cromo | uboot does not say anything at all, just loops |
| 11:45.38 | CIA-158 | rhombus-tech: dbilunov master * r67a442c8f8ac /allwinner_a10/orders/DmitryBilunov.mdwn: I am going to support the product by testing the alpha version |
| 11:45.47 | mnemoc | cromo: if you are booting from uboot over SD be sure to preload the script.bin file |
| 11:46.54 | cromo | I forgot to mention I have a pre-made sd card image with 3.0.8 kernel that otherwise boots fine |
| 11:47.18 | cromo | I just replaced the kernel with my own and that's where I am stuck |
| 11:48.26 | cromo | I was thinking that maybe u-boot has to be compiled with gnueabihf, too? |
| 11:48.51 | cromo | sun4i#version says it's compiled with arm-none-linux-gnueabi-gcc |
| 11:48.52 | mnemoc | possible, but no idea |
| 11:49.32 | mnemoc | try reducing the greed a bit and test a hand made 2.6.36 not hf |
| 11:49.50 | cromo | yeah, I am doing it ATM |
| 11:50.42 | A0Sheds | whats the latest on how the cpu card will handle the IO cards ID and configure it self accordingly? |
| 11:51.50 | cromo | BTW. I measured power usage of mele when running with Android - it's only 2W! |
| 11:52.28 | mnemoc | A0Sheds: the idea was to read a devicetree chunk from somewhere standard |
| 11:52.28 | specing | what did you expect? |
| 11:52.51 | cromo | specing: you ask me? ~ 5-7W |
| 11:53.05 | A0Sheds | specing: expect for what? |
| 11:53.31 | specing | cromo: why so high? |
| 11:53.51 | specing | cromo: Not even my embedded Geode computer consumes that! |
| 11:54.08 | specing | and Geode is 500MHz x86 |
| 11:54.10 | cromo | dunno, I have a Asus RT-N16 router that I run with debwrt |
| 11:54.13 | cromo | it's a MIPS cpu |
| 11:54.18 | cromo | and it consumes 10W |
| 11:54.28 | A0Sheds | < mind reading skills poor today > |
| 11:54.39 | specing | cromo: Does it have a harddrive? |
| 11:54.59 | cromo | USB connected 2,5, but disconnected it consumes ~7W |
| 11:55.02 | cromo | hence my guess |
| 11:55.52 | specing | ethernet consumes quite a bit |
| 11:56.02 | cromo | I had it connected over ethernet |
| 11:56.04 | cromo | with wifi on |
| 11:56.09 | cromo | 2W |
| 11:56.22 | cromo | but yeah, the router has 5 ethernet ports |
| 11:56.35 | cromo | and 2x2 mimo 802.11n |
| 11:56.43 | cromo | or even 3x3, don't remember now |
| 11:56.54 | specing | the 10Mbps microcontroller ethernet interface I have here consumes ~0.7W |
| 11:57.53 | cromo | OK, I gotta go for now, have some shower and do some cleaning, haven't done anything since I picked up the parcel yesterday ;) addictive! |
| 11:58.16 | cromo | I'll get back to you once I test the armfp kernel |
| 11:59.08 | cromo | *softfp |
| 12:02.21 | cromo | OK, got the kernel and it still doesn't boot |
| 12:08.23 | zenitraM | mmm could it be because of the mess with the kernel params location? |
| 12:08.51 | zenitraM | try to compile it with hardcoded boot parameters |
| 12:10.11 | CIA-158 | rhombus-tech: Adrian_Bulat master * r80c5bc48c3d7 /allwinner_a10/orders/AdrianB.mdwn: |
| 12:11.48 | mnemoc | cromo: did you preload script.bin? |
| 12:13.01 | zenitraM | if he just replaced the uimage on the ubuntu image.. it should have loaded it |
| 12:13.17 | cromo | mnemoc: as I said I have everything in place, just replacing the image |
| 12:13.37 | cromo | zenitraM: I will try, thanks |
| 12:13.43 | mnemoc | maybe the gpl-violating 3.0.8+ image doesn't have the same reliance on script.bin the published tree has |
| 12:15.54 | zenitraM | it has several script.bin files at the fat partition |
| 12:16.10 | mnemoc | and? |
| 12:16.25 | mnemoc | it could be the linux image who loads it |
| 12:16.45 | mnemoc | while in the public source it's bootloader's job |
| 12:16.54 | cromo | there's no script.bin file present at the fat partition |
| 12:17.11 | cromo | evb.bin, mele.bin, sys_config1.mele_mod.bin |
| 12:17.16 | zenitraM | those are script.bin files |
| 12:17.26 | zenitraM | mm idk. will have to look at the uboot env |
| 12:17.30 | zenitraM | to see which one is being loaded |
| 12:18.15 | cromo | too bad there's no error message at all |
| 12:18.19 | cromo | when booting |
| 12:18.49 | mnemoc | cromo: btw, what exact tree are you building? |
| 12:21.09 | cromo | mnemoc: lychee/2.6.36 |
| 12:23.22 | mnemoc | you could also try the official 2.6.36 from allwinner on github |
| 12:24.49 | cromo | didn't know about that one, will try, thanks |
| 12:25.33 | cromo | is that the one? https://github.com/amery/linux-allwinner |
| 12:26.04 | mnemoc | nope, that's mine, the official is https://github.com/allwinner/linux-2.6.36 |
| 12:26.46 | mnemoc | it's also important you take a look into the env of the uboot you are using to understand how it's booting |
| 12:27.25 | cromo | I saw the wiki page but was hoping I could be agile and skip this until later |
| 12:28.00 | cromo | i.e. with all the information gathered so far I could get it running quickly and then start playing around |
| 12:28.25 | mnemoc | all the source kernels avaiable need the scrupt.bin file preloaded to work. if your uboot doesn't do that, it won't boot anything you compile yourself |
| 12:29.20 | cromo | reading evb.bin; 42052 bytes read; reading uImage (...) |
| 12:29.41 | cromo | not sure if that's the same but it looks like the script.bin is being loaded |
| 12:30.10 | mnemoc | yes, that smells good |
| 12:30.35 | mnemoc | named evb.bin, but it's loaded |
| 12:30.47 | mnemoc | have to go, bbl |
| 12:30.47 | cromo | the kernel is then read, information about kernel is printed, then it says "Starting kernel ..." |
| 12:30.49 | cromo | and reboots |
| 12:31.05 | cromo | OK, thanks for help |
| 12:46.02 | zenitraM | okay. the wifi module is 8192cu |
| 12:56.01 | *** join/#arm-netbook eebrah (~Ibrahim_K@212.49.88.65) |
| 12:56.10 | *** join/#arm-netbook cromo (~cromo@90-255-251-94.net.stream.pl) |
| 12:56.24 | *** join/#arm-netbook cromo (~cromo@90-255-251-94.net.stream.pl) |
| 13:02.14 | mnemoc | zenitraM: I still fail to understand the purpose of the rtl8201cp chip |
| 13:09.47 | *** part/#arm-netbook QingPei (~qingpei@61.149.86.182) |
| 13:10.07 | zenitraM | mnemoc: seems like ethernet to me |
| 13:10.23 | zenitraM | maybe for some reason they couldn't get the original module to work and decided to make it from scratch to their needs |
| 13:13.00 | cromo | ./scripts/build_sun4i.sh: line 74: mkbootimg: command not found |
| 13:13.07 | cromo | anybody knows where do I find it? |
| 13:13.36 | cromo | is this the same as mkimage? |
| 13:15.43 | mnemoc | zenitraM: but even in the default image it's wemac and not rtl8150 who takes care of eth0 |
| 13:17.37 | mnemoc | cromo: I don't use those helpers at all, but that mkbootimg seems to be from the android build env |
| 13:17.51 | cromo | yeah, just meant to write that uBoot got built anyway |
| 13:17.52 | cromo | OK |
| 13:18.02 | cromo | I got it booting with allwinner official kernel |
| 13:18.03 | mnemoc | http://www.netmite.com/android/mydroid/system/core/mkbootimg/mkbootimg.c |
| 13:19.07 | zenitraM | mnemoc: maybe they never got to use the rtl8150 driver |
| 13:19.10 | cromo | uBoot, lol - uImage that is |
| 13:23.59 | cromo | frankly, it appears much slower then the kernel that comes in a pre-built SD card image linked on rhombus |
| 13:24.13 | cromo | and ethernet still not working as expected |
| 13:29.36 | mnemoc | weirdly enough that's the same problem I have on my 3.0 tree :) |
| 13:30.20 | mnemoc | I mean your mail |
| 13:32.29 | mnemoc | but that's a good thing. it means it's not my fault :p |
| 13:33.23 | cromo | yeah, it's a good sign |
| 13:33.45 | *** join/#arm-netbook QingPei (~qingpei@114.241.176.76) |
| 13:33.51 | cromo | got the wifi module to load |
| 13:34.22 | cromo | the shown in Paul's email (not sure who's who here) |
| 13:35.23 | mnemoc | i don't think that "paul kench" is one of those who come to irc |
| 13:36.20 | CIA-158 | rhombus-tech: Max master * r886ed290247f /allwinner_a10/orders/elessar71.mdwn: |
| 13:37.34 | cromo | see |
| 13:41.27 | cromo | do you know from which source was that 3.0.8 kernel in sdcard image built? |
| 13:43.00 | cromo | * built with |
| 13:43.06 | cromo | or which branch |
| 13:44.52 | mnemoc | from tom's image? the internal gpl-violating tree of allwinner |
| 13:45.37 | mnemoc | we don't have access to that source base |
| 13:46.13 | cromo | so both the kernels that I can run are from allwinner's sources |
| 13:46.44 | cromo | I thought allwinner is gpl-compilant ;) |
| 13:47.23 | mnemoc | nope, they are only sigtly less-hostile that others to the idea |
| 13:48.05 | mnemoc | the tree in alioth was made from a 2.6.36 patch leaked by ainol |
| 13:48.24 | mnemoc | later tom published the official 2.6.36 tree on github |
| 13:48.44 | mnemoc | and from that hno's and my trees evolved |
| 13:49.32 | cromo | since the sd card image is prepared by tom, I assume the ethernet issue is coming straight from allwinner then |
| 13:50.54 | cromo | also does it matter if kernel is built using hf or softp toolchain? I want to use ubuntu 12.04 armhf base and I was convinced that for that I need a armhf compiled kernel, too, but now I am not so sure |
| 13:52.24 | mnemoc | the ABI should only affect userspace consistency |
| 13:53.07 | mnemoc | so afaik it shouldn't affect if your kernel is hf and userspace soft |
| 13:53.28 | cromo | sure, but what about other way |
| 13:53.36 | cromo | e.g. kernel soft, userspace hf |
| 13:54.01 | mnemoc | should work fine too as syscalls shouldn't be affected by the difference |
| 13:54.46 | cromo | that's what I thought. still hf kernel would probably increase performance a bit, but what I need now is to get started with userspace hacking |
| 13:55.06 | ManoftheSea | the kernel doesn't use a lot of FP? |
| 13:55.09 | cromo | ok, thanks for that, I'm off to play with it |
| 13:55.32 | cromo | ManoftheSea: hence the "a bit" ;) although some drivers could |
| 13:55.48 | cromo | I can imagine DVB or 802.11 to use it |
| 13:55.55 | cromo | MAC layer maybe |
| 13:56.00 | mnemoc | more likely to use double than float |
| 13:56.02 | ManoftheSea | for what? |
| 13:56.17 | ManoftheSea | isn't a MAC an integer? |
| 13:56.33 | cromo | mac layer as in software phy |
| 13:56.39 | ManoftheSea | Also... I read that as Mandatory Access Control at first |
| 13:56.59 | ManoftheSea | like a software defined radio? |
| 13:57.06 | ManoftheSea | I'm honestly interested and confused |
| 13:57.18 | cromo | http://linuxwireless.org/en/developers/Documentation/mac80211 |
| 13:57.30 | ManoftheSea | clickypop |
| 13:58.23 | cromo | SoftMAC wireless devices relay on kernel/userspace to do most of the job |
| 13:58.35 | cromo | e.g. these things: http://linuxwireless.org/en/developers/Documentation/Glossary#MLME |
| 13:59.29 | cromo | what I wrote above was supposed to say "mac layer as in softMAC" |
| 13:59.37 | cromo | not "software PHY" |
| 14:00.50 | cromo | I'm off to play with hf userspace |
| 14:00.51 | cromo | bbl |
| 14:13.44 | vgrade | cromo, what was your workflow to get a uImage of your own to boot? |
| 14:14.39 | vgrade | I don't see any serial output from my uImages which have replaced the ubuntu image one? |
| 14:15.39 | vgrade | which source did you use? can you pastie your .config? what toolchain? |
| 14:16.13 | cromo | vgrade: I had the same issue with lychee kernels, they would not show any serial output and reboot |
| 14:16.17 | cromo | in a loop |
| 14:17.04 | cromo | I didn't alter .config and I used the gcc-4.5-arm-linux-gnueabihf/gcc-4.5-arm-linux-gnueabi package from ubuntu for compilation (depending on if I was building softp or hp) |
| 14:17.25 | cromo | the only kernel I could compile and boot successfully was from allwinner's branch |
| 14:17.30 | vgrade | that does not matter kernels don't care |
| 14:18.05 | vgrade | cromo, ok i tried that also without sucess, did you hardcode the commandline? |
| 14:19.09 | cromo | no, I did absolutely nothing except for changing the CROSS-COMPILE variable |
| 14:19.21 | cromo | it just worked |
| 14:19.27 | cromo | allwinner's 2.6.36 that is |
| 14:19.38 | cromo | because all the other kernel I would compile wouldn't boot |
| 14:19.40 | cromo | no error shown |
| 14:19.54 | cromo | most likely I need to alter the commandline |
| 14:22.16 | vgrade | ok, I'm on 4.6.1 toolchain that must be the issue then, thanks |
| 14:22.23 | cromo | yes |
| 14:22.27 | cromo | forgot to mention |
| 14:22.31 | cromo | I couldn't compile with 4.6 |
| 14:22.35 | cromo | had to revert to 4.5 |
| 14:22.50 | vgrade | cheers :) |
| 14:23.17 | cromo | where did you get your toolchain from? |
| 14:23.34 | mnemoc | I use linaro's 4.6 without issues |
| 14:23.50 | cromo | if from ubuntu's repository there might be an issue with using 4.5 toolchain that does not symlink properly to gcc command |
| 14:24.07 | cromo | so you won't be affected I think |
| 14:24.27 | mnemoc | to get the symlink you need to install the unversioned package, gcc-arm-linux-gnueabihf in your case |
| 14:24.43 | cromo | sure, but that also installs 4.6 toolchain |
| 14:24.45 | cromo | and links to it |
| 14:25.17 | cromo | I filed a bug against it in ubuntu's launchpad and worked around by doing symlink myself |
| 14:25.48 | mnemoc | --no-install-recommends ? |
| 14:25.57 | cromo | hmm |
| 14:26.01 | cromo | lemme try |
| 14:26.42 | mnemoc | stupid upstart :( root 15595 2.0 0.0 4135892 2820 ? Ss 16:22 0:00 /usr/bin/qemu-arm-static /usr/sbin/cron |
| 14:26.48 | cromo | nope, still wants to install cpp-4.6-arm-linux-gnueabi cpp-arm-linux-gnueabi gcc-4.6-arm-linux-gnueabi gcc-arm-linux-gnueabi |
| 14:26.57 | mnemoc | took over a service from the emulator :-/ |
| 14:27.11 | mnemoc | and now I can't umount the SD |
| 14:27.33 | mnemoc | it's odd you couldn't build with 4.6 |
| 14:27.54 | cromo | it failed for both lychee's branches |
| 14:28.13 | mnemoc | try my 2.6.36 ... I have some small fixes in there |
| 14:28.34 | cromo | I will, I'm now trying to boot something newer than 10.10 |
| 14:28.56 | cromo | apart for hacking for fun I want to replace my old openwrt router with mele |
| 14:29.12 | mnemoc | cromo: might be this: https://github.com/amery/linux-allwinner/commit/a8fb3d2c2fc353e72aab8c1f5319cb7c523b4092 ? |
| 14:29.27 | cromo | most likely, yes |
| 14:29.30 | vgrade | mnemoc, I seem to have multiple versions of gcc installed here (11.10) will clean up a bit and have another go |
| 14:29.56 | cromo | yes, pretty sure, same module, same errors failing on warnings |
| 14:30.11 | cromo | will try later on |
| 14:30.21 | mnemoc | cromo: then it's just about removing that -Werror :p |
| 14:30.41 | cromo | well, I tried editing the sources manually and removing the unused variable gcc complained about |
| 14:31.00 | cromo | and it passed, but seconds later it failed with THAT particular variable missing |
| 14:31.16 | cromo | so I assumed this wasn't really about missing variable warning but something more than that |
| 14:31.51 | cromo | I didn't put to much thinking into it, I just eagerly wanted to get it working asap ;) |
| 14:31.57 | cromo | *too |
| 14:32.13 | cromo | will try again later with your source |
| 14:33.47 | mnemoc | the first time I also invested lots of time trying to clean the source of that driver, but really not worthy |
| 14:34.02 | cromo | that's the wifi used by mele, isn't it? |
| 14:35.00 | mnemoc | no, the mele uses an rtl8188cus |
| 14:35.24 | mnemoc | it's in the corner of the power button |
| 14:35.54 | mnemoc | rtl8192cu is the driver |
| 14:37.02 | mnemoc | but allwinner's service is to provide software for all those using their chip, and so their tree includes all the hacks "needed" for all those devices |
| 14:37.54 | mnemoc | that's why they actually don't care about specs or documentation, all _interesting_ development happens in-house |
| 14:42.43 | cromo | yeah, I read someone's post on gmane explaining the peculiar way of software and hardware development in China ;) |
| 14:43.56 | mikey_w | I am thinking of ordering a mele a1000 from "deal extreme" but hear they are very slow shipping. Has anyone dealt with them? |
| 14:45.28 | cromo | I did |
| 14:45.33 | cromo | ordered some usb cables in feb |
| 14:45.44 | mikey_w | Still waiting? |
| 14:45.47 | cromo | and that's it |
| 14:45.50 | cromo | still waiting |
| 14:46.05 | mikey_w | bummer. |
| 14:46.11 | cromo | mikey_w: Tom Cubie sells it cheaper on aliexpres |
| 14:46.21 | cromo | he's one of Allwinner employees, so it's a sure thing |
| 14:46.29 | cromo | pretty much everyone here ordered with him |
| 14:46.35 | cromo | it's ~70usd |
| 14:46.39 | mikey_w | ok |
| 14:46.41 | cromo | and takes 2 weeks tops |
| 14:46.46 | MaDMaLKaV | yeah, almost everyone here bought it from tom |
| 14:47.11 | mikey_w | What's with the remote control and disk box? |
| 14:47.16 | cromo | yes |
| 14:47.20 | cromo | errr |
| 14:47.29 | MaDMaLKaV | dunno, I'm like the only one that didn't got it from him |
| 14:47.31 | cromo | what do you mean? |
| 14:47.42 | cromo | I got them |
| 14:47.49 | cromo | but the disk box is a piece of cr*p |
| 14:47.54 | cromo | couldn't get it to open |
| 14:47.58 | mikey_w | Extra stuff with the deal extreme one. |
| 14:48.01 | cromo | and once the drive's in the slot, it resonates |
| 14:48.05 | cromo | so there's a lot of noise |
| 14:48.15 | mikey_w | ok |
| 14:48.37 | cromo | I literally had to break some of the box latches to open it |
| 14:48.49 | mnemoc | o_O |
| 14:48.50 | MaDMaLKaV | still no word from tom about video decoder library license? |
| 14:49.34 | mnemoc | MaDMaLKaV: I haven't seen anyone formally asking them about the cedarx library license |
| 14:49.53 | MaDMaLKaV | time to write to the mailing list about |
| 14:51.20 | MaDMaLKaV | done |
| 14:51.57 | MaDMaLKaV | hope the mesage is understanble, perhaps I had more coffee loquor with my father in law than recommended for writing in foreign languages |
| 14:52.13 | mnemoc | cromo: using the origin script.bin fixes the ethernet problem |
| 14:52.46 | cromo | wow! |
| 14:52.59 | cromo | the one from nand that is? |
| 14:53.03 | mnemoc | yes |
| 14:53.06 | cromo | great! |
| 14:53.07 | mnemoc | nanda |
| 14:54.09 | cromo | now how do I pull it? afik the only way to get mele to work with adb is through internal usb connector, am I right? |
| 14:54.13 | mnemoc | cromo: note that you need to pass a mac_address argument or eth0 will have a random MAC |
| 14:54.21 | mnemoc | 1m |
| 14:54.39 | mikey_w | Does the mele a1000 do wifi? |
| 14:54.44 | mnemoc | mikey_w: yes |
| 14:54.51 | mikey_w | great. |
| 14:55.01 | cromo | mikey_w: rtl8188cus |
| 14:55.13 | mnemoc | mikey_w: uses the chassis as antenna, no idea about the quality of the signal |
| 14:55.25 | cromo | mnemoc: I was just wondering where's the antenna |
| 14:56.15 | mikey_w | Beats the raspberry pi considering it has a case. |
| 14:56.43 | mnemoc | cromo: https://www.box.com/s/c0946877c47f174fd250 .... the stock script.bin |
| 14:56.47 | cromo | has much more, better cpu, gpu, more usb ports, etc |
| 14:56.50 | cromo | mnemoc: thanks! |
| 14:57.00 | MaDMaLKaV | beats the raspberry pi in almost all aspects, cpu , gpu, connections, casing, ... if it we just had a complete accelerated unix running on it, it will piss on raspberrry pi |
| 14:57.02 | mnemoc | cromo: you'll need to rename it to evb.bin or change the env of yuour uboot |
| 14:57.20 | MaDMaLKaV | comparing prices, the difference is just ridiculous for all the advantages |
| 14:57.32 | cromo | I know, I already did some reading about boot process |
| 14:57.33 | mnemoc | MaDMaLKaV: rpi is an arduino competitor, just that |
| 14:57.37 | cromo | but thanks for heads up |
| 14:58.14 | mikey_w | How open are the drivers? |
| 14:58.43 | cromo | mnemoc: I have a feeling that this will also make other kernels boot fine that I couldn't get to so far |
| 14:59.35 | MaDMaLKaV | mnemoc, if we could just run accelerated X on the mele, most people would not touch a rpi |
| 14:59.36 | mnemoc | mikey_w: gpu and video decoders are closed |
| 14:59.55 | MaDMaLKaV | people that want to use it as an arduino have to pay the rpi, that is not as cheap as they said, plus the io board |
| 15:00.20 | MaDMaLKaV | there are better arm devices oriented for mcu use than the pi |
| 15:00.36 | MaDMaLKaV | and for unix useage, the rpi is quite weak, sans the video acceleration stuff |
| 15:00.47 | mnemoc | MaDMaLKaV: in the mean time I would be happy been able to see the nand and with the lcd driver not crashing :p |
| 15:00.49 | MaDMaLKaV | that is not better than the one on mele, but it is working on linux |
| 15:01.01 | cromo | mnemoc mikey_w but there's a http://limadriver.org/, so there's a hope |
| 15:01.12 | MaDMaLKaV | by the way, ics for the mele will pop up more sooner than later |
| 15:01.13 | mikey_w | pk |
| 15:01.18 | MaDMaLKaV | I can confirm it |
| 15:01.25 | MaDMaLKaV | but I don'¡t want to use android on my tv |
| 15:01.29 | MaDMaLKaV | I want XBMC |
| 15:01.42 | mnemoc | cromo: that's only about the 3D part... people expects to watch movies at decent quality on devices like these too |
| 15:01.59 | MaDMaLKaV | cromo, I'm afraid that lima driver will delay some time on being useable. also, that doesn't solve the video decoder issue |
| 15:02.36 | mnemoc | aligning with linaro releases we will have mali drivers anyway |
| 15:02.43 | cromo | right |
| 15:02.58 | MaDMaLKaV | yeah, but for hf mali driver, we probably will have to wait a full quarter |
| 15:03.11 | MaDMaLKaV | for the linaro hf ubuntu build |
| 15:03.16 | cromo | MaDMaLKaV: I also want to run XBMC, I thought it's doable with closed source video decorder? |
| 15:03.26 | cromo | then one that got posted some time ago on gmane |
| 15:03.35 | MaDMaLKaV | cromo, licensing issues |
| 15:03.45 | MaDMaLKaV | you just can't link xbmc to that libraries legally |
| 15:03.55 | MaDMaLKaV | at least until we know the license |
| 15:03.58 | mnemoc | cromo: also the binary relies on certain ABI and so won't run on hf |
| 15:04.02 | MaDMaLKaV | that is not specified in the files we got |
| 15:04.30 | penguin42 | does the closed source video decoder rely on any kernel drivers? |
| 15:04.36 | cromo | arrrrr that's bad |
| 15:05.20 | cromo | is video decoding done by mali or some other part on SoC? |
| 15:05.47 | mnemoc | cromo: other part of the SoC |
| 15:05.54 | mnemoc | cromo: it's called "cedarX" |
| 15:06.20 | cromo | is this licensed or allwinner's own? |
| 15:06.34 | mnemoc | i believe it's 3rd party |
| 15:06.38 | cromo | google appears to not know much about it |
| 15:09.05 | MaDMaLKaV | but seems there is something called cedarx that works really good for allergies |
| 15:09.05 | MaDMaLKaV | XD |
| 15:09.20 | mnemoc | haha |
| 15:09.32 | cromo | hah |
| 15:09.49 | cromo | been sneezing all day today so it might be worth a try |
| 15:10.25 | MaDMaLKaV | if after taking it you get both cured from your allergies and able to read mkv 1080p files from an hex editor just tell us |
| 15:10.42 | MaDMaLKaV | XD |
| 15:10.55 | cromo | I doubt I would, I'd rebrand it and sell to Intel |
| 15:11.21 | penguin42 | cromo: yes, the hayfever has been bad this year :-( |
| 15:11.40 | cromo | I am not usually prone to this |
| 15:11.52 | cromo | but today's really bad |
| 15:11.58 | cromo | anyway, I gotta go |
| 15:12.04 | cromo | have some rest aside from PC |
| 15:12.04 | MaDMaLKaV | seeya |
| 15:12.14 | cromo | cu |
| 15:15.50 | vgrade | mnemoc, script.bin renamed to evb.bin on the ubuntu image with my buld kernel did the trick |
| 15:16.20 | vgrade | mnemoc, thanks, |
| 15:16.31 | mnemoc | vgrade: yw :) |
| 15:17.06 | vgrade | wonder what the difference is there, is there a way of decompiling the .bin files? |
| 15:17.33 | mnemoc | zenitraM wrote a "decompiler" |
| 15:18.02 | mnemoc | https://github.com/zenitraM/a10tools |
| 15:21.03 | mnemoc | but I refuse to install ruby |
| 15:21.37 | vgrade | cheers, no HDMI output but I guess I need the 2.6.36 modules in the ubuntu filesystem |
| 15:21.41 | MaDMaLKaV | have to leave, later |
| 15:24.57 | mnemoc | cu MaDMaLKaV |
| 15:52.29 | *** part/#arm-netbook QingPei (~qingpei@114.241.176.76) |
| 16:18.45 | mnemoc | uh |
| 16:19.14 | mnemoc | on the mele ./drivers/net/wireless/rtl8192cu/8192cu.ko works but ./drivers/net/wireless/rtlwifi/rtl8192cu/rtl8192cu.ko doesn't |
| 16:19.50 | mnemoc | I mean, both compile and load, but the later doesn't detect the hardware or even try to do anything |
| 16:20.19 | mnemoc | (3.0.28-android+) |
| 16:21.29 | mnemoc | 8192cu.ko comes fro the origin 2.6.36 patch, the other is upstream |
| 16:22.10 | mnemoc | wonders what evilness are they doing in there |
| 16:46.22 | mnemoc | f* nand driver. /me off |
| 17:08.05 | *** join/#arm-netbook cromo (~cromo@90-255-251-94.net.stream.pl) |
| 17:08.53 | cromo | mnemoc: a friend of mine who also got the Mele says that the script.bin breaks X's |
| 17:09.35 | mnemoc | cromo: diff both |
| 17:10.07 | mnemoc | i doubt we can't have both, wired network and X |
| 17:10.33 | mnemoc | something is wrong on each script.bin |
| 17:11.18 | *** join/#arm-netbook vgrade (~martinbro@cpc2-nrte22-2-0-cust128.8-4.cable.virginmedia.com) |
| 17:11.55 | mnemoc | but for me, currently eth0 is far more important than X |
| 17:13.13 | zenitraM | mnemoc: where did you get the jumper wires from? |
| 17:14.08 | mnemoc | zenitraM: the 2.00mm to 2.54mm? ebay.... but I bought 80... I can mail you some |
| 17:15.33 | mnemoc | http://www.ebay.com/itm/280839459714 |
| 17:15.38 | cromo | just to make sure - the rtc on mele is not battery supported, is it? |
| 17:16.17 | cromo | mnemoc: you mean binary diff? |
| 17:16.27 | cromo | don't get it ;) |
| 17:16.53 | mnemoc | cromo: those .bin are compiled from .fex format. zenitraM wrote a decompiler |
| 17:17.14 | cromo | ok, thanks, I will try it a bit later on |
| 17:17.17 | mnemoc | cromo: so the idea is to find what on Tom's .fex breaks the ethernet |
| 17:18.21 | mnemoc | zenitraM: you may want http://www.ebay.com/itm/180822626764 too ;-) |
| 17:21.50 | cromo | is there a way to change the command line without recompiling the kernel? possibly by using uboot setenv? |
| 17:22.08 | mnemoc | sure |
| 17:22.14 | cromo | yeah |
| 17:22.16 | cromo | I just got it |
| 17:22.19 | cromo | with printenv |
| 17:22.23 | mnemoc | :) |
| 17:22.29 | cromo | nice |
| 17:22.29 | mnemoc | don't forget to saveenv :) |
| 17:22.32 | cromo | I like it more and more |
| 17:22.37 | cromo | sure |
| 17:23.11 | cromo | mnemoc: how is the mac address supposed to be passed to mac_address parameter? with colons or without? |
| 17:23.28 | mnemoc | with |
| 17:23.51 | mnemoc | $ cat /proc/cmdline |
| 17:23.51 | mnemoc | console=ttyS0,115200 root=/dev/mmcblk0p2 rootwait init=/sbin/init loglevel=8 mac_addr=00:CE:39:B6:63:F8 |
| 17:24.43 | cromo | thanks |
| 17:26.47 | mnemoc | ./drivers/video/sun4i/lcd/lcd.ko |
| 17:26.47 | mnemoc | ./drivers/video/sun4i/disp/disp.ko |
| 17:26.49 | mnemoc | :) |
| 17:26.54 | mnemoc | finally |
| 17:27.58 | cromo | is it correct that video decoding libraries are only available for non-hf env? and what kind of API do they provide? |
| 17:28.06 | cromo | openmax? |
| 17:28.53 | mnemoc | cromo: https://github.com/amery/allwinner-a10-video <--- look at both .pdf files there |
| 17:29.00 | cromo | great |
| 17:30.20 | penguin42 | has never heard of MELIS |
| 17:32.35 | cromo | not so great |
| 17:32.57 | cromo | looks to me that we won't see any xbmc support anytime soon? |
| 17:33.05 | cromo | or myttv or any other htpc software |
| 17:33.46 | cromo | *mythtv |
| 17:34.10 | penguin42 | doesn't see any FP being passed across the interfaces there - I'd think it could be forced to work on hf? |
| 17:34.28 | mnemoc | possibly |
| 17:34.43 | zenitraM | mnemoc: how much do you want for 5-10 wires + postage? :) |
| 17:35.16 | mnemoc | forget it, only the address |
| 17:36.37 | zenitraM | okay then, i owe you a beer/coffee if you come down to Madrid |
| 17:37.03 | mnemoc | deal ;-) |
| 17:37.23 | ManoftheSea | wasn't someone talking about NEON support? |
| 17:39.56 | penguin42 | if A10 has it I'd assume it would just work |
| 17:40.27 | mnemoc | $ cat /proc/cpuinfo |
| 17:40.28 | mnemoc | Processor: ARMv7 Processor rev 2 (v7l) |
| 17:40.28 | mnemoc | BogoMIPS: 59.63 |
| 17:40.29 | mnemoc | Features: swp half thumb fastmult vfp edsp neon vfpv3 |
| 17:40.48 | penguin42 | looks good (although abysmal BogoMIPS?!) |
| 17:41.17 | mnemoc | ^^ |
| 17:41.38 | specing | Only 60 BogoMIPS on A10? |
| 17:41.50 | ManoftheSea | so... when do we get CPUCards? |
| 17:42.02 | penguin42 | mnemoc: You got the caches off or the clock right down? |
| 17:43.28 | mnemoc | I didn't do anything... it starts at that freq by default |
| 17:44.54 | mnemoc | cpufreq/cpuinfo_cur_freq:60000 |
| 17:44.54 | mnemoc | cpufreq/cpuinfo_max_freq:1008000 |
| 17:45.05 | penguin42 | ah fair enough |
| 17:45.12 | penguin42 | it's like me, I wake up slowly :-) |
| 17:45.14 | specing | 60 kHz? |
| 17:45.15 | specing | cool |
| 17:45.20 | specing | err |
| 17:45.22 | specing | Mhz |
| 17:45.43 | specing | Isn't A10 1.2 GHz though? |
| 17:45.58 | penguin42 | specing: Which is kind of what the max is saying |
| 17:46.20 | specing | max is saying 1.008 Ghz? |
| 17:46.29 | penguin42 | specing: probably magic with the cpu governors and power config to get it going that fast |
| 17:47.58 | zenitraM | although it is more probably only 1GHz and the additional 200mhz or 500mhz (depending where you get the data from) are for the GPU |
| 17:48.24 | ManoftheSea | that's what I've been saying for months. |
| 17:48.37 | ManoftheSea | The code only claims a max of 1.00GHz |
| 17:48.47 | ManoftheSea | 1.008, sorry. |
| 17:49.04 | mnemoc | marketing adds up the Hz of all cores and gpu |
| 17:49.16 | specing | Hehe |
| 17:49.22 | ManoftheSea | Megahurts myth |
| 17:50.09 | specing | People in marketing are dumb |
| 17:50.35 | specing | There was the folowing advert in my country for several months for some laptop: |
| 17:50.53 | specing | top right corner: Open to anyone |
| 17:51.19 | specing | bottom left: Unauthorized entry disabled |
| 17:51.25 | specing | I was like... WTF? |
| 17:51.44 | specing | s/Unauthorized entry/break in/ |
| 17:54.03 | ManoftheSea | um, if open to everyone, then no one can unauthorized entry |
| 17:54.12 | specing | logic. |
| 18:07.49 | *** join/#arm-netbook mikey_w (~mike@c-71-63-115-202.hsd1.va.comcast.net) |
| 18:08.07 | *** join/#arm-netbook mike_ (~mike@c-71-63-115-202.hsd1.va.comcast.net) |
| 18:11.16 | *** join/#arm-netbook mikey_w (~mike@c-71-63-115-202.hsd1.va.comcast.net) |
| 18:12.20 | mnemoc | <PROTECTED> |
| 18:14.01 | *** join/#arm-netbook mikey_w (~mike@c-71-63-115-202.hsd1.va.comcast.net) |
| 18:14.14 | mnemoc | Only allow certain groups to create sockets (ANDROID_PARANOID_NETWORK) |
| 18:15.48 | *** join/#arm-netbook cromo (~cromo@90-255-251-94.net.stream.pl) |
| 18:16.14 | zenitraM | mnemoc: that's for the app permission system |
| 18:16.32 | mnemoc | right |
| 18:16.33 | zenitraM | apps that don't have the network access permission aren't allowed to.. well, do anything network-related |
| 18:18.39 | mnemoc | I understand the restriction, but it should be launcher's job to remove capabilities from the processes instead of hardcoding a blessed group id on the kernel |
| 18:19.52 | cromo | are mali drivers available for hf, too? |
| 18:20.27 | mnemoc | not yet |
| 18:20.39 | cromo | damn |
| 18:20.49 | mnemoc | linaro's 12.03 is based on ubuntu 11.10, soft float |
| 18:20.57 | mnemoc | ubuntu 12.04 is hf |
| 18:21.12 | mnemoc | so the next linaro release should come with hf mali |
| 18:22.01 | mnemoc | all this idea of year/month versioning is awfully confusing |
| 18:22.25 | cromo | hmm |
| 18:22.30 | cromo | apparently there is linaro 12.04? |
| 18:22.42 | cromo | http://releases.linaro.org/12.04/ |
| 18:22.47 | cromo | dir created today |
| 18:22.49 | mnemoc | yy.mm |
| 18:22.53 | cromo | although empty |
| 18:23.05 | cromo | err, no, not empty |
| 18:23.17 | mnemoc | but it can't be based on ubuntu 12.04 before it's released |
| 18:23.45 | cromo | it's to be released on april 26th |
| 18:23.55 | *** join/#arm-netbook vgrade (~martinbro@cpc2-nrte22-2-0-cust128.8-4.cable.virginmedia.com) |
| 18:23.57 | cromo | and I think ubuntu 12.04 is, too |
| 18:24.13 | cromo | nope |
| 18:24.18 | cromo | based on oneiric |
| 18:24.30 | mnemoc | told you :) |
| 18:24.42 | cromo | yeah, I was naive |
| 18:24.58 | cromo | any chances A10 could be supported by linaro out of the box? |
| 18:25.28 | mnemoc | linaro charges $1M the chip maker |
| 18:25.39 | mnemoc | allwinner won't pay tha |
| 18:25.39 | mnemoc | t |
| 18:25.48 | mnemoc | BUT |
| 18:25.58 | cromo | what if I pay? ;) |
| 18:26.34 | mnemoc | if we provide the kernel .deb and mainstream uboot support we can add generate linaro compatible images |
| 18:27.03 | cromo | too bad allwinner isn't interested that much in it |
| 18:27.13 | mnemoc | https://wiki.linaro.org/Resources/HowTo/AddingNewHardwareSupport |
| 18:27.37 | mnemoc | allwinner is focused in getting out sun6i based products asap |
| 18:27.44 | mnemoc | we are noise |
| 18:29.18 | cromo | any specs available on sun6i? |
| 18:29.40 | mnemoc | only that it's a multicore cortex-a9 |
| 18:30.29 | mnemoc | weirdly called allwinner a15 :p |
| 18:30.36 | mnemoc | marketing people is funny |
| 18:32.09 | cromo | I bought A10 tablet |
| 18:32.12 | cromo | then I bought mele |
| 18:32.17 | cromo | and I was super convinces it's dual core |
| 18:32.25 | cromo | until I found out it's not |
| 18:32.36 | mnemoc | no cortex-a8 is multicore |
| 18:32.41 | cromo | only because I read somewhere that it's dual core (someone's interpretation of multicore) |
| 18:33.12 | penguin42 | well, dual is more than 1! |
| 18:33.21 | mnemoc | :) |
| 18:33.27 | zenitraM | well, it has a core for the CPU and a core for the GPU |
| 18:33.29 | zenitraM | there you go, two cores |
| 18:33.36 | zenitraM | </china-marketer> |
| 18:33.37 | mnemoc | hehe |
| 18:34.03 | penguin42 | actually, more specifically a multicore ARM is probably ARMv7M rather than ARMv7 - there are some prefetch instructions that are only in the multicore version of the core |
| 18:34.10 | mnemoc | and you add both speeds together and becomes 2x1.5GHz |
| 18:36.38 | penguin42 | are these the guys who used to print the specs for amplifiers? |
| 18:36.54 | cromo | hehe |
| 18:36.59 | cromo | 1500 PMPO |
| 18:44.02 | zenitraM | just figured a way to get serial output with 2.54mm jumper cables |
| 18:44.21 | zenitraM | tx+rx fit together... and gnd to the usb gnd |
| 18:53.03 | cromo | I tried to boot the ubuntu precise core armel |
| 18:53.16 | cromo | stuck at "failed to execute /init. attempting defaults..." |
| 18:53.17 | cromo | hmm |
| 18:53.19 | cromo | any idea? |
| 18:58.59 | mnemoc | you are passing init=/init and /init doesn't exist in the rootfs |
| 18:59.18 | mnemoc | "attempting defaults..." means it will try /sbin/init anyway |
| 18:59.26 | cromo | sure, but it as like this with previous rootfs |
| 18:59.35 | cromo | apparently the defaults worked for it |
| 18:59.37 | cromo | and here don't |
| 18:59.59 | cromo | ok, it seems it actually booted |
| 19:00.07 | cromo | but didn't print to serial console |
| 19:00.17 | cromo | since when I removed the SD card udevd started to complain |
| 19:01.09 | mnemoc | by default systems run getty (to get a prompt) on tty1, tty2, .... but not on ttyS0 or console |
| 19:01.24 | mnemoc | so you may have to add the setup for that |
| 19:01.35 | cromo | thanks, that's gonna be it |
| 19:02.31 | mnemoc | in ubuntu, copy /etc/init/tty1.conf into ttyS0.conf and change the getty call |
| 19:03.32 | cromo | thanks |
| 19:04.28 | cromo | not enough I think |
| 19:07.01 | mnemoc | how are you calling getty? |
| 19:08.56 | cromo | I edited the tty0.conf file, changed to tty0 |
| 19:09.06 | cromo | however it seems it's something with upstart |
| 19:09.15 | cromo | https://bugs.launchpad.net/ubuntu/+source/upstart/+bug/980917 |
| 19:09.19 | cromo | https://bugs.launchpad.net/ubuntu/+source/upstart/+bug/936667 |
| 19:09.20 | mnemoc | ttyS0 |
| 19:09.35 | cromo | I tried the --no-log, but it's still stuck |
| 19:10.23 | cromo | hah |
| 19:10.27 | cromo | mnemoc: sure, I know |
| 19:10.41 | cromo | however I did that on my own rootfs instead on the SD card ;) |
| 19:10.54 | mnemoc | hahaha |
| 19:11.38 | cromo | let's see now |
| 19:12.45 | cromo | hmm |
| 19:13.28 | cromo | have to change the baudrate I think |
| 19:14.27 | mnemoc | I use: exec /sbin/getty -L 115200 ttyS0 vt102 |
| 19:14.33 | cromo | yep |
| 19:14.35 | cromo | that worked |
| 19:14.37 | cromo | great! |
| 19:30.47 | zenitraM | mnemoc: trying to compile a 2.6.36, which of your branches should i use? |
| 19:32.23 | mnemoc | for 2.6.36 please pick allwinner-2.6.36-android |
| 19:32.45 | mnemoc | but allwinner-3.0-android is more fun :) |
| 19:34.08 | zenitraM | i know :) but wanna try something stable first |
| 19:34.18 | mnemoc | ^^ |
| 19:35.29 | zenitraM | that sort of explains why linux-2.6.36-sun4i didn't recognize the mmc |
| 19:39.19 | mnemoc | :) |
| 20:18.06 | zenitraM | soo the disp module crashes horrendously when trying to start X11 on 2.6.36 |
| 20:22.11 | zenitraM | http://pastie.org/3830099 |
| 20:23.51 | mnemoc | zenitraM: that is what I think Tom fixed on his .fex |
| 20:24.16 | zenitraM | which one of them? |
| 20:26.06 | zenitraM | i'll try them... |
| 20:26.18 | mnemoc | https://www.box.com/s/937cad1464e0ad98cb7e |
| 20:26.43 | mnemoc | ^--- fex, with documentation! |
| 20:27.00 | mnemoc | https://www.box.com/s/41a9cc83b03b45881c78 <--- and .bin |
| 20:27.09 | mnemoc | but that one kills eth0 :-/ |
| 20:27.50 | mnemoc | we need to diff both and fix how to fix eth in tom's fex |
| 20:27.58 | mnemoc | s/fix/find/ |
| 20:30.22 | zenitraM | mm nope |
| 20:30.34 | zenitraM | that's the .bin i'm using (md5 says it's the same as on the ubuntu image) |
| 20:30.44 | zenitraM | booting using the mele one yields the same result |
| 20:31.02 | mnemoc | uhm |
| 20:31.06 | mnemoc | not nice |
| 21:02.50 | mnemoc | even the simple code you pasted on the mailing list to change resolution makes my disp crash badly |
| 21:03.51 | mnemoc | I suppose we have to start fixing that :) |
| 21:06.12 | mnemoc | https://gist.github.com/2439596 |
| 21:07.11 | mnemoc | that seems equivalent to the code on the mails |
| 21:11.22 | mnemoc | oh, now worked |
| 21:11.34 | mnemoc | niiiice |
| 21:11.36 | zenitraM | what did you do? |
| 21:11.44 | mnemoc | modprobe lcd instead of modprobe disp |
| 21:12.36 | zenitraM | rebooting... |
| 21:14.45 | zenitraM | mmm mine still crashes, maybe because of gdm autostart |
| 21:14.55 | zenitraM | disabling gdm and removing the line from /etc/modules... |
| 21:16.04 | mnemoc | I don't have X installed (yet) |
| 21:16.56 | zenitraM | yup. it's X what makes it crash |
| 21:17.34 | mnemoc | baby steps ;-) |
| 21:19.30 | CIA-158 | rhombus-tech: lkcl master * r126a7cb0c286 /allwinner_a10/orders/ (5 files): order processing cleanup |
| 21:55.40 | mnemoc | found, info->fbops->fb_imageblit(info, image); |
| 21:57.09 | mnemoc | drivers/video/sun4i/disp/dev_fb.c is AWFUL |
| 22:04.02 | zenitraM | let's see |
| 22:04.33 | zenitraM | why do they use a windows .bat to push the kernel modules using adb? |
| 22:04.34 | zenitraM | *facepalm* |
| 22:04.52 | mnemoc | adb.exe ;-) |
| 22:06.06 | zenitraM | i know, but i mean, why not push them directly from linux instead of copying the modules over to a windows machine |
| 22:06.23 | zenitraM | unless they have a smb'd compiling server... |
| 22:08.39 | zenitraM | now, that's a *lot* of uncommented code. |
| 22:09.49 | mnemoc | http://lxr.linux.no/linux+v3.0.28/include/linux/fb.h#L662 there are quite a few callbacks to be implemented :< |
| 22:34.32 | mnemoc | <PROTECTED> |
| 22:34.32 | mnemoc | <PROTECTED> |
| 22:34.36 | mnemoc | nice |
| 22:36.44 | penguin42 | yeuch |
| 23:22.14 | *** join/#arm-netbook A0Sheds (~ly@75.49.211.164) |
| 23:22.14 | *** join/#arm-netbook A0Sheds (~ly@unaffiliated/l84supper) |
| 23:45.37 | penguin42 | hmm describing something as 'no license' so free to use is rather dangrous |
| 23:58.31 | mnemoc | as I understood it he means 'no fee' |
| 23:58.49 | mnemoc | no licensing cost for using it |