| 00:00.26 | penguin42 | I don't think it's that dense is it ? |
| 00:02.04 | destinal | penguin42: http://mazzoo.de/pics/com-on-air/typeIII_pcmcia_controller.jpg |
| 00:02.07 | penguin42 | although something like http://parts.digikey.com/1/parts/303586-conn-header-pcmcia-r-a-5v-pcb-92140-010lf.html would be easier if it was available at a sane price |
| 00:03.09 | destinal | penguin42: I believe that we're having someone source connectors for dev boards / breakout boards from china for us much cheaper and hopefully we can just get them from the rhombus page or something |
| 00:03.31 | penguin42 | yeh that would be nice |
| 00:06.38 | destinal | penguin42: I kind of want to make a cluster IO board that accepts multiple EOMA68's (or maybe EOMACF instead) and put an ethernet switch IC on the board itself, not have to bother with an RJ45 connector for every board |
| 00:08.05 | destinal | I haven't been able to find ethernet switch chips on digikey though, may have to harvest one from a switch and figure out how to use it |
| 00:08.45 | penguin42 | destinal: Yeh that would be kind of nice |
| 00:11.11 | hno | mnemoc, ok, have something I think will work. Please review and merge. Do not have compilers etc right now so can not verify it compiles. |
| 00:19.53 | *** join/#arm-netbook arokux (~roman@HSI-KBW-046-005-103-006.hsi8.kabel-badenwuerttemberg.de) |
| 00:25.25 | A2Sheds | there will be some basic breakout boards for hacking/DIY/embedded |
| 00:27.59 | destinal | A2Sheds: cool. but available around the same time as the A10 board or some time later? |
| 00:28.15 | destinal | is also eduprey btw but decided to switch to using the same nick everywhere |
| 00:30.00 | A2Sheds | destinal, same time since we need test/dev platforms |
| 00:33.14 | penguin42 | destinal: I wonder whether you would be better for that using USB for clustering than 100mbps ether |
| 00:33.55 | destinal | penguin42: can use gigabit with ethernet though |
| 00:35.04 | penguin42 | destinal: Yeh but the a10's are only 100mbps |
| 00:35.13 | A2Sheds | depends on the type of cluster applications you are running, how much actual processing is there going on if you are transferring data to other cpu's all the time? |
| 00:36.50 | destinal | hmmm a10 is sun4i crane? |
| 00:37.08 | destinal | sounds like a sparc |
| 00:37.22 | penguin42 | hehe I said that before :-) |
| 00:39.46 | penguin42 | oh that's funny: http://arstechnica.com/business/news/2012/03/the-pwn-plug-is-a-little-white-box-that-can-hack-your-network.ars |
| 00:40.04 | destinal | yeah, heard about this earlier somewhere |
| 00:40.55 | lkcl | destinal: first version i'd suggest just cracking open a gigabit ethernet switch. |
| 00:41.16 | lkcl | or hell just get a box big enough to fit it in, as-is |
| 00:41.35 | destinal | lkcl: yeah a big enough switch you could just fit the eoma68 into :) |
| 00:41.41 | penguin42 | does the eoma board expect the ether to be wired to some magnetics or can it go straight on the wire? |
| 00:42.20 | destinal | penguin42: I think the magnetics are only needed if you're going to be doing long runs over cat5 type cable or something but I could be wrong |
| 00:43.30 | destinal | ie point to point wire for very short connections doesn't need anything special, maybe someone can correct me if I am wrong |
| 00:45.13 | penguin42 | yeh I think short wire/on board you can avoid it |
| 00:45.51 | Gumboot | I don't think there's enough room for the isolation in an EOMA board, is there? |
| 00:46.09 | Gumboot | Although, I think there's also a trend to not bother trying. |
| 00:46.11 | penguin42 | Gumboot: Well there were PCMCIA ether cards |
| 00:46.37 | Gumboot | They had fat ends, though, right? |
| 00:46.57 | destinal | Gumboot: well that's just part of the RJ45 side stuff that would go on the IO board |
| 00:47.02 | destinal | not on the EOMA68 card |
| 00:47.06 | penguin42 | hmm some did - but some had pop out sockets - were those fat? |
| 00:47.45 | Gumboot | You could, presumably, put the isolation in the socket. |
| 00:48.23 | penguin42 | Gumboot: Yeh I know you can get rj45s with built in magnetics |
| 00:48.39 | destinal | Gumboot: but I mean, EOMA68 is a PCMCIA typr connector, you can put anything you want on the board it plugs *into* |
| 00:49.33 | destinal | ethernet in this case is just being connected directly from the chip to the PCMCIA pins |
| 00:49.36 | penguin42 | destinal: http://search.digikey.com/uk/en/cat/integrated-circuits-ics/specialized-ics/2556114?k=ksz8999 9 port ether switch - didn't come up in a search for an ether switch though! |
| 00:49.55 | penguin42 | (and more expensive than buying the switch) |
| 00:50.29 | Gumboot | destinal: Yes, I was just pointing out that you probably wouldn't ever put isolation inside the EOMA board beacuse it's so damned big. |
| 00:50.29 | destinal | penguin42: ouch, yeah, can find whole switches cheaper these days and harvest lots of goodies |
| 00:50.38 | destinal | Gumboot: yeah agreed |
| 00:51.25 | penguin42 | destinal: Still, Micrel seem to do some interesting stuff |
| 01:04.34 | destinal | penguin42: found one. RTL8309 http://www.futureelectronics.com/en/Technologies/Product.aspx?ProductID=RTL8309SBLFREALTEKSEMICONDUCTOR7787054&IM=1 http://realtek.info/pdf/rtl8309sb.pdf |
| 01:04.43 | destinal | $5.33 |
| 01:04.50 | penguin42 | that's kind of ok |
| 01:07.14 | destinal | yeah only 10/100 alas |
| 01:07.28 | destinal | so was other one |
| 01:09.41 | lkcl | guys - put this stuff on the wiki |
| 01:10.27 | lkcl | http://rhombus-tech.net/community_ideas/cluster_server/ |
| 01:11.02 | lkcl | don't lose it in random conversations |
| 01:11.39 | CIA-46 | rhombus-tech: lkcl master * rac620ba9d91b /community_ideas/cluster_server.mdwn: |
| 01:14.55 | A2Sheds | a cluster is sort of a non brainer application |
| 01:15.13 | A2Sheds | sort of like listing a netbook :) |
| 01:20.28 | *** join/#arm-netbook A2Sheds (~ly@unaffiliated/l84supper) |
| 01:23.50 | A2Sheds | http://dl.dropbox.com/u/9959499/a10/A10_datasheet_2011.8.22.pdf |
| 01:23.59 | A2Sheds | http://dl.dropbox.com/u/9959499/a10/A10_development_board_user_manual--2011.9.23_English.pdf |
| 01:30.55 | penguin42 | Do they have versions that aren't peppered with confidential? |
| 01:39.18 | *** part/#arm-netbook hno (~hno@squid/developer/hno) |
| 01:39.25 | *** join/#arm-netbook hno (~hno@squid/developer/hno) |
| 01:40.28 | penguin42 | (Still, nice set of stuff on there) |
| 01:42.34 | hno | The EVB manual isn't peppered with confidental. Only Copyright. |
| 01:43.07 | penguin42 | the datasheet is marked copyright |
| 01:43.09 | penguin42 | sorry |
| 01:43.12 | penguin42 | marked confidential |
| 01:43.17 | hno | yes. |
| 01:43.35 | hno | so is the P10 Controller document |
| 01:45.50 | penguin42 | good night |
| 01:57.50 | A2Sheds | did someone leak those? http://download.csdn.net/detail/elite_hjh/4045265 |
| 02:03.31 | hno | The datasheet seem to have many errors in pin assignment. Miscellaneous Signal Description ball # do not match at all with Pin Characteristics ball #. |
| 02:06.10 | hno | lkcl, have you triple-verified pin assignment in the design with Allwinner? It's scary to see pin mismatches like this in the data scheet. |
| 02:06.23 | hno | can't speel. Time for bed. |
| 02:07.13 | A2Sheds | hno: we have schematics from working boards |
| 02:38.46 | CIA-46 | rhombus-tech: Chris master * r1e49bf8cd434 /allwinner_a10/orders/CompGeek812.mdwn: |
| 02:42.28 | destinal | A2Sheds: nice, schematics from working boards for layout and I know for a while kernel source was considered basically the documentation (which, hey, if it works, awesome) |
| 02:43.13 | A2Sheds | destinal: it's evolved a bit since then |
| 03:01.01 | destinal | that realtek switch chip is pretty capable actually, looks like it supports VLANs and can trunk two ports together to do > 100mb/s uplink |
| 03:06.17 | CIA-46 | rhombus-tech: Eric master * r025ba36a76a6 /community_ideas/cluster_server.mdwn: added info on switch |
| 03:11.00 | destinal | and looks like there is some dd-wrt code for it |
| 04:21.10 | *** join/#arm-netbook QingPei1 (~qingpei@61.149.87.53) |
| 05:27.21 | *** join/#arm-netbook QingPei (~qingpei@114.241.195.161) |
| 08:14.56 | CIA-46 | rhombus-tech: castrol master * r2b60e33b4747 /allwinner_a10/orders/Castrol.mdwn: |
| 08:16.55 | *** join/#arm-netbook arokux (~roman@HSI-KBW-046-005-103-006.hsi8.kabel-badenwuerttemberg.de) |
| 08:29.34 | CIA-46 | rhombus-tech: Eric master * r58bef798a45e /allwinner_a10/orders/destinal.mdwn: |
| 08:32.36 | CIA-46 | rhombus-tech: Eric master * re710ac882612 /allwinner_a10/orders/destinal.mdwn: updated preorder request |
| 10:31.01 | mnemoc | hno: thanks! |
| 10:35.09 | mnemoc | hno: not sure how familiar you are with `git` but as I cherrypicked+altered the commit you will need to rebase against mine (eg. `git fetch amery; git rebase amery/allwinner-v3.0-android` and `git rebase --skip` the conflict |
| 10:36.01 | Gumboot | So I just got around to reading that Pwn Plug article... BoA are willingly connecting their corporate networks to mobile phone networks? |
| 10:36.34 | zenitraM | mornin' everyone |
| 10:36.48 | mnemoc | moin zenitraM, Gumboot |
| 10:37.28 | zenitraM | Gumboot: not read the article, but my father works in a bank here in Spain and their corporate smartphones are directly connected to their intranet |
| 10:38.01 | zenitraM | mnemoc: trying to compile your kernel with SND_SOC disabled.. let's see how far it gets |
| 10:38.29 | MaDMaLKaV | gonna be out of the city one week, see you people |
| 10:40.28 | mnemoc | zenitraM: the last commit from hno should enable early printk() over a chosen serial |
| 10:40.48 | Gumboot | If it's worth worrying about the security of a network, and buying equipment to test it, then surely it's worth _not_ opening your back door onto mobile phone networks. |
| 10:42.01 | Gumboot | If you just don't care, and let smartphones on the network, then one might assume (hah!) that the sensitive data is protected in other ways. |
| 10:42.07 | zenitraM | assume, that's the word |
| 10:42.37 | Gumboot | We're all doomed. |
| 10:44.14 | zenitraM | mnemoc: nice, pulling the changes and will start compiling again |
| 10:46.35 | mnemoc | zenitraM: I still have to merge a part of usb gadget from 2.6.36 .... not sure how will this affect you |
| 10:47.01 | mnemoc | https://github.com/amery/linux-allwinner/commit/17a6256ed5e17ccff31ac35161a65b9587d38b86 <-- this one I didn't merge, trying it now |
| 10:47.38 | zenitraM | well probably adb won't work yet, but I use the uSD serial so i don't mind for now |
| 10:49.14 | zenitraM | nice, it even has a conf option for the uart to use |
| 10:52.34 | mnemoc | yes, it was a very nice and clean solution |
| 10:59.53 | zenitraM | mnemoc: well, it seems it is needed |
| 11:00.00 | zenitraM | drivers/built-in.o: In function `insmod_device_driver':/home/zenitram/ainol/amery/linux-allwinner/drivers/usb/sun4i_usb/manager/usb_msg_center.c:237: undefined reference to `sw_usb_device_enable |
| 11:00.06 | zenitraM | drivers/built-in.o: In function `rmmod_device_driver':/home/zenitram/ainol/amery/linux-allwinner/drivers/usb/sun4i_usb/manager/usb_msg_center.c:265: undefined reference to `sw_usb_device_disable' |
| 11:02.28 | mnemoc | zenitraM: yes, I'm working on it |
| 11:10.39 | arokux | mnemoc, don't you fear that you will change too much and then, if there is a bug somewhere it will be impossible to find. since you are not testing it now. |
| 11:11.21 | mnemoc | arokux: that's why I'm not really changing anything beside obvious API changes |
| 11:13.47 | arokux | mnemoc, I see. will the old code run on our yet to be produced board without problems? |
| 11:14.07 | mnemoc | in theory... |
| 11:18.20 | mnemoc | <PROTECTED> |
| 11:18.24 | mnemoc | :( |
| 11:18.25 | mnemoc | err |
| 11:18.29 | mnemoc | :) |
| 11:21.32 | arokux | mnemoc, wonderful news, what now? :) |
| 11:22.21 | mnemoc | I'll fix some warnings and then try to port the sound drivers to the new system.... that part changed a lot in 2.6.37 |
| 11:23.10 | mnemoc | if someone can test and fix runtime problems in the while, it would be more than awesome |
| 11:23.31 | arokux | how could it be done? |
| 11:23.34 | mnemoc | no whitespace or not-essential cleaning yet |
| 11:23.46 | arokux | using some emulation? |
| 11:24.15 | mnemoc | tablets |
| 11:25.41 | arokux | I do not have any hardware ( |
| 11:26.36 | mnemoc | there are plenty inexpensive a10-based tablets available |
| 11:26.57 | mnemoc | I got an onda vi10 elite shipped from HK today |
| 11:28.26 | arokux | can you give a link to the website you bought it? |
| 11:28.31 | mnemoc | translation of chinese comments in the kernel tree are also very welcomed |
| 11:31.09 | mnemoc | I bought it on aliexpress.com, but you can also find devices on myefox.com, tomtop.com, dx.com, etc etc |
| 11:32.17 | mnemoc | dx is usually very slow shipping, the others keep local stock so shipping is fast |
| 11:33.54 | mnemoc | zenitraM even found an a10/android a psp "clone" on DX |
| 11:35.08 | arokux | $155.. it could be relatively cheap, but it's not cheap.. ( |
| 11:35.49 | mnemoc | you can find models for $105 up to $250 |
| 11:36.13 | mnemoc | the one I ordered was $134+shipping |
| 11:36.16 | mnemoc | err |
| 11:36.20 | mnemoc | with shipping |
| 11:36.50 | mnemoc | 1G/8G 1024x600 |
| 11:37.05 | arokux | mnemoc, is there just a board a-la raspberry pi? |
| 11:37.42 | mnemoc | the devkit is $1000, that's why most of us ordered a mele a1000 instead |
| 11:38.34 | arokux | could I also have one, or it's already too late? |
| 11:38.51 | mnemoc | for the first batch from Tom, yes |
| 11:38.58 | mnemoc | but you can buy it on DX |
| 11:39.10 | mnemoc | or wait for a *possible* second batch from Tom |
| 11:40.24 | mnemoc | http://www.dealextreme.com/p/mele-1080p-android-2-3-internet-tv-set-top-box-w-wifi-optical-3-x-usb-hdmi-av-lan-sd-119913 |
| 11:41.21 | *** join/#arm-netbook gordan (~gordan@78-86-195-86.zone2.bethere.co.uk) |
| 11:41.22 | mnemoc | that has great potentials as devkit |
| 11:45.53 | zenitraM | is back |
| 11:46.03 | zenitraM | mnemoc: got the kernel to compile too |
| 11:46.07 | zenitraM | gonna try it now |
| 11:46.08 | mnemoc | \o/ |
| 11:46.46 | mnemoc | I'll fix the compilation warnings before switching to port the sound |
| 11:47.27 | mnemoc | but I need to leave now :( |
| 11:47.37 | zenitraM | make modules doesn't work though |
| 11:47.51 | mnemoc | please don't hesitate on filing issues at https://github.com/amery/linux-allwinner/issues |
| 11:47.58 | mnemoc | meh, forgot about those :-/ |
| 11:48.13 | zenitraM | i will :) gonna push it and see how I can manage to boot it from u-boot on the ICS firmware |
| 11:49.27 | arokux | zenitraM, what hardware do you have? |
| 11:49.37 | zenitraM | arokux: a Novo7 Advanced tablet |
| 11:51.18 | arokux | zenitraM, where are the instructions how to flash the kernel on it? :) |
| 11:51.41 | zenitraM | arokux: it depends on the firmware |
| 11:51.50 | zenitraM | on the 2.3 (android) based firmwares |
| 11:51.55 | zenitraM | you can follow this http://elinux.org/Hack_A10_devices |
| 11:52.11 | zenitraM | and the bImage is just on the linux folder on /dev/block/nanda partition |
| 11:53.03 | zenitraM | on the 4.0.3 android fw (i have it now installed), it has directly u-boot |
| 11:53.24 | zenitraM | and you can use u-boot over serial (serial over uSD), and make it boot another kernel |
| 12:02.40 | arokux | what is better for our purposes, Ainol Novo 7 Advanced Tablet or http://www.dealextreme.com/p/mele-1080p-android-2-3-internet-tv-set-top-box-w-wifi-optical-3-x-usb-hdmi-av-lan-sd-119913#open%20full%20view ? |
| 12:03.19 | zenitraM | the mele definitely |
| 12:03.44 | zenitraM | you have to solder to get any serial port on the tablet |
| 12:03.56 | zenitraM | or do the SD-redirection trick, but that leaves you without SD |
| 12:04.43 | arokux | and Mele? |
| 12:05.31 | zenitraM | the male has both SATA and USB (gadget, not host, for adb) with headers inside |
| 12:05.33 | zenitraM | see: http://rhombus-tech.net/allwinner_a10/hacking_the_mele_a1000/ |
| 12:06.00 | arokux | the problem is, I haven't watched TV for 10 years.. so the Mele will be useless for me :D So it will be only for hacking. |
| 12:06.16 | zenitraM | i bought it exactly for the same |
| 12:07.29 | zenitraM | i've used my tablet more for hacking than for its common use (even if that wasn't what i bought it for…) |
| 12:10.52 | arokux | zenitraM, have you created the page about Mele? |
| 12:10.59 | zenitraM | nope, i haven't |
| 12:11.39 | zenitraM | i don't have it yet, in fact i think tom shipped it yesterday |
| 12:14.42 | arokux | zenitraM, is there a guarantee that Mele 1080P will have the same internals as Mele A1000 from our wiki? |
| 12:15.02 | zenitraM | the one on DX? |
| 12:15.04 | zenitraM | it is the same |
| 12:15.19 | arokux | mm.. this upper part is different |
| 12:15.38 | arokux | this one you can plug off |
| 12:17.02 | arokux | but external outputs look exactly the same, you are right |
| 12:18.44 | arokux | zenitraM, ah, this is external HDD??? |
| 12:19.50 | *** join/#arm-netbook penguin42 (~dg@tu006.demon.co.uk) |
| 12:22.02 | zenitraM | damn connection |
| 12:22.38 | arokux | zenitraM, have u seen my messages? |
| 12:22.53 | zenitraM | the last one I saw is: <arokux> zenitraM, is there a guarantee that Mele 1080P will have the same internals as Mele A1000 from our wiki? |
| 12:23.07 | arokux | mm.. this upper part is different |
| 12:23.18 | arokux | what is this? |
| 12:23.20 | arokux | HDD? |
| 12:23.33 | zenitraM | SATA |
| 12:23.47 | zenitraM | with a box to put a 2.5 HDD in |
| 12:24.54 | arokux | zenitraM, but it's shipped without HDD right? |
| 12:24.59 | zenitraM | exactly |
| 12:25.47 | zenitraM | mnemoc: still here? |
| 12:30.02 | zenitraM | okay, went down again, if you said anything more, copy&paste… my bouncer doesn't seem to replay me anything i've missed |
| 12:33.06 | arokux | not yet. I'm very new to this field.. so my next question: we are looking for A10 based device to hack on, because out board will be based on it? |
| 12:33.15 | arokux | zenitraM, |
| 12:33.42 | zenitraM | exactly, at least the first board |
| 12:39.50 | *** join/#arm-netbook QingPei (~qingpei@114.241.195.161) |
| 12:42.18 | arokux | zenitraM, why it's sooo hard to find somebody selling it?! |
| 12:42.54 | zenitraM | arokux: its a new product and so far it was only released in china... |
| 12:43.18 | arokux | after long googling I've found a russian website where it's sold for $188!! |
| 12:43.24 | zenitraM | and the seller doesn't seem to be much interested on international selling, in fact i think everything on the box and the manual is in chinese |
| 12:43.47 | zenitraM | Tom got us some for $65 + shipping, keeping 6$ for himself (he had to pick a ferry to buy them) |
| 12:43.52 | zenitraM | so it is ridiculously cheap there |
| 12:44.12 | zenitraM | but of course exporters will try to earn as much as they can |
| 12:45.00 | *** join/#arm-netbook zenitraM[away] (~zenitraM@2607:f358:1:fed5:11:0:1018:14) |
| 12:45.25 | arokux | zenitraM, he lives in China? |
| 12:45.40 | zenitraM | arokux: yup. he is in fact an engineer who works in Allwinner |
| 12:46.21 | arokux | how much will the shipping cost? |
| 12:46.46 | zenitraM | it depends on where. for me it was 8$ for registered china post airmail |
| 12:47.12 | zenitraM | he used aliexpress site to sell them to us (it provides a payment processor), the cost was calculated by aliexpress |
| 12:47.43 | zenitraM | (also, there was a 10$ discount out there for Aliexpress that we used) |
| 12:47.54 | arokux | in which country do you live? |
| 12:47.57 | zenitraM | Spain |
| 12:48.16 | arokux | aha.. I'm in Germany, so it will be the same.. |
| 12:48.20 | zenitraM | probably |
| 12:48.45 | arokux | 65+20 = 85. (without discount and with 10 for shipping) |
| 12:49.11 | arokux | I could directly by from ex for $100 now. |
| 12:49.43 | zenitraM | 65+10 would be |
| 12:49.48 | zenitraM | not 65+20 |
| 12:50.00 | arokux | maybe there will be not discount anymore |
| 12:50.11 | zenitraM | no, no, 65+10 before discount |
| 12:50.34 | zenitraM | with discount.. 65 i guess |
| 12:51.38 | arokux | so.. it's seems to be best just to buy from ex for 100, with free shipping |
| 12:52.20 | zenitraM | if you don't wanna wait for tom to buy more… well, it's 40$ more expensive... |
| 12:52.33 | zenitraM | i'd wait, probably it will pop up in more stores / alibaba sellers |
| 12:52.40 | zenitraM | s/alibaba/aliexpress/ |
| 12:56.59 | arokux | zenitraM, why do you think it will be cheaper in aliexpress? |
| 12:57.27 | zenitraM | because DX is usually more expensive than aliexpress sellers |
| 12:57.48 | zenitraM | and if the cost price is around 65$.. i can surely expect to see people selling it for 80$+shipping |
| 12:58.16 | arokux | zenitraM, so... it'll be around $90, I'll save time and order from ex. |
| 12:58.40 | zenitraM | ok, if you don't wanna wait go and do that |
| 12:59.11 | arokux | zenitraM, I'm tepmted to run my kernel on it! |
| 13:01.09 | zenitraM | if you've got a kernel i can try to run it for you on my tablet |
| 13:02.23 | arokux | zenitraM, no, I'm going to help mnemoc to port 2.6 to 3.0. but I need hardware, otherwise it seems useless for me. |
| 13:03.00 | zenitraM | oh ok then :) i'm in fact trying to run mnemoc kernel run |
| 13:03.03 | zenitraM | but stuck here: |
| 13:03.04 | zenitraM | https://github.com/amery/linux-allwinner/issues/3 |
| 13:10.22 | arokux | zenitraM, who's hipboi? |
| 13:10.31 | zenitraM | hipboi = tom |
| 13:12.37 | arokux | zenitraM, ah, ok |
| 13:14.03 | arokux | zenitraM, apparently he can speak Chinese :) |
| 13:14.26 | zenitraM | he's chinese, so i'd expect that :) |
| 13:15.24 | arokux | zenitraM, ahhhh... Tom Cubie seems to be an European name |
| 13:15.43 | zenitraM | i'm not sure if that's even his real name |
| 13:15.50 | arokux | :D |
| 13:16.06 | zenitraM | chinese people with difficult names usually tend to invent names we can pronounce :) |
| 13:16.16 | arokux | yeah, I now |
| 13:21.48 | arokux | It's maybe irrelevant for this channel but I've found out the following. recently I've bought an EBook in Ukraine and tried to update the firmware. there were some names in the files of the firmware, and I googled for them. I've found another company in UK which is also producing EBooks. The Ukrainians seem to derive from there firmware. I wonder if Britains know it. |
| 13:24.54 | arokux | by the way, both companies do not publish any source code |
| 13:29.19 | zenitraM | arokux: both britains and ukranians just sell chinese ebooks |
| 13:29.30 | zenitraM | probably |
| 13:30.01 | zenitraM | here in Spain we got a model of eBook that is sold by different "spanish" companies with different names, and it is just the same rebranded chinese ebook |
| 13:30.26 | arokux | but then they should deliver the same firmware |
| 13:31.19 | zenitraM | they don't have to… they might have different versions with different firmware features |
| 13:31.34 | arokux | I see |
| 13:31.39 | zenitraM | usually you can get firmwares from one working on another without much hassle |
| 13:31.52 | CIA-46 | rhombus-tech: Danni master * r28211d9b820a /allwinner_a10/orders/Hello.mdwn: Hello |
| 13:32.21 | zenitraM | look at the A10 tablets.. there are like 5 manufacturers producing A10 based tablets which are essentially the same hardware |
| 13:32.28 | zenitraM | they all have different firmwares, some include the market, some not |
| 13:32.47 | arokux | yes, but in this case the firmware is *the same* |
| 13:33.13 | zenitraM | in which case? |
| 13:33.19 | arokux | with Ebook |
| 13:33.56 | zenitraM | then probably they come from the same factory |
| 13:34.20 | arokux | can I find out which one and push them to publish sources? |
| 13:34.39 | zenitraM | probably, but i doubt they will publish anything |
| 13:35.01 | *** part/#arm-netbook QingPei (~qingpei@114.241.195.161) |
| 13:35.12 | arokux | zenitraM, it was exactly the same interface until recently ukrainians updated it slightly |
| 13:36.43 | arokux | zenitraM, compare http://pokupon.bigmir.net/uploaded/ckeditor_pictures/7438/data/content/black-text1.jpg?1330436569 vs http://www.mye-reader.co.uk/ereader.php# |
| 13:40.36 | arokux | zenitraM, how do you think updates are done? by chinese or ukrainians? |
| 13:50.50 | CIA-46 | rhombus-tech: Danni master * rb54f3a63f75c /allwinner_a10/orders/OXYS.mdwn: |
| 13:59.44 | CIA-46 | rhombus-tech: Danni master * r2d8ec1d30377 /allwinner_a10/orders/Hello.mdwn: |
| 14:14.41 | mnemoc | zenitraM: as the log jumps from [0.0000] to [2.02] seems like it's set to boot quietly... can you try to make it verbose? |
| 14:14.51 | mnemoc | zenitraM: what uart number is the uSD? |
| 14:15.05 | *** join/#arm-netbook arokux (~roman@HSI-KBW-046-005-103-006.hsi8.kabel-badenwuerttemberg.de) |
| 14:39.50 | mnemoc | sweden post... uhm |
| 14:42.22 | zenitraM | back |
| 14:42.56 | zenitraM | arokux: usually by chineses.. but not sure |
| 14:43.02 | zenitraM | mnemoc: how to make it verbose? |
| 14:43.10 | zenitraM | uSD is UART0 afaik |
| 14:43.16 | zenitraM | it is set on uart0 on script.bin |
| 14:43.23 | mnemoc | usually by removing `quiet` from the command line |
| 14:43.30 | mnemoc | no idea how that works on the bootloader |
| 14:44.24 | zenitraM | mhm.. on uboot: |
| 14:44.27 | zenitraM | setargs=setenv bootargs console=${console} root=${nand_root}init=${init} loglevel=${loglevel} |
| 14:44.35 | zenitraM | could be loglevel? |
| 14:45.05 | zenitraM | you're right, if the other kernel is booting quietly then there's something making it boot quietly.. |
| 14:45.19 | mnemoc | yes, change loglevel |
| 14:45.25 | zenitraM | currently set to 8 |
| 14:47.08 | mnemoc | odd, the loglevel of printk goes fro 0 to 7 |
| 14:47.27 | mnemoc | so 8 would mean everything... |
| 14:48.16 | mnemoc | is console set to something useful? |
| 14:48.31 | zenitraM | i guess so… mmm i think there's something wrong with the args |
| 14:48.36 | zenitraM | console=ttyS0,115200 |
| 14:48.41 | zenitraM | it has no space between root and init |
| 14:50.26 | mnemoc | the messages since [2.0] seem to come from initrd |
| 14:50.33 | mnemoc | so the kernel is booting |
| 14:50.50 | mnemoc | just too silent |
| 14:52.10 | penguin42 | it's important to get the command to uboot exactly right - I can't remember exactly how but I put an extra space somewhere and lost a day trying to figure out what uboot had done |
| 14:56.48 | zenitraM | mmm what could be setting the kernel to silent? |
| 14:57.00 | zenitraM | (damn internet) |
| 14:57.24 | mnemoc | it could be something in the kernel config itself |
| 14:58.06 | zenitraM | cat .config |grep CMD |
| 14:58.06 | zenitraM | CONFIG_CMDLINE="console=ttyS0,115200 init=/bin/init rw mem=256M@0x40000000" |
| 14:58.06 | zenitraM | CONFIG_CMDLINE_FROM_BOOTLOADER=y |
| 15:07.16 | penguin42 | you could try adding earlyprintk |
| 15:07.41 | penguin42 | if the port supports it |
| 15:10.50 | *** join/#arm-netbook QingPei1 (~qingpei@123.118.109.170) |
| 15:12.47 | zenitraM | it is enabled i think |
| 15:18.53 | zenitraM | just tried encapsulating the zImage on a boot.img with the cmdline set and booting using the boota command in u-boot |
| 15:19.08 | zenitraM | no dice, nothing is output at all |
| 15:24.06 | penguin42 | zenitraM: Are you sure your serial is ttyS0? Not all serial devices register as ttyS |
| 15:24.43 | zenitraM | penguin42: it should, it is ttyS0 on the 2.6.36 kernel |
| 15:25.32 | penguin42 | is there a git tree for this kernel somewhere? |
| 15:26.00 | zenitraM | https://github.com/amery/linux-allwinner/ |
| 15:26.35 | penguin42 | clones |
| 15:30.17 | *** join/#arm-netbook lkcl (~lkcl@host81-129-112-135.range81-129.btcentralplus.com) |
| 15:31.28 | penguin42 | __slow__ |
| 15:35.04 | mnemoc | not slow, fat :p |
| 15:37.03 | penguin42 | yeh I guess, there's probably a way for me to wire it into one of my existing git trees somewhere to speed the download? |
| 15:37.27 | mnemoc | i know you can, i don't know how |
| 15:38.00 | mnemoc | I usually just `cp -a` another kernel checkout and add a new remote |
| 15:42.18 | zenitraM | just read the mail - sweden airmail?? |
| 15:42.27 | mnemoc | that was weird |
| 15:42.45 | mnemoc | I think it goes bulk to .se and then gets forwarded from there |
| 15:42.51 | zenitraM | might be |
| 15:43.13 | zenitraM | but never heard you could do that |
| 15:48.15 | mnemoc | hipboi is not using CNP directly but a post "reseller", espeedpost.com |
| 15:48.35 | mnemoc | it's probably them who do the bulk+reposting |
| 15:51.40 | zenitraM | http://translate.googleusercontent.com/translate_c?act=url&hl=en&ie=UTF8&prev=_t&rurl=translate.google.com&sl=zh-CN&tl=en&twu=1&u=http://www.espeedpost.com/html-news-service-newsid_2255.html&usg=ALkJrhhTGFpNYmvUQgzrVC_XSuVZ10ywBA |
| 15:52.07 | zenitraM | i guess that means we won't have tracking info until it arrives to sweden |
| 15:53.29 | zenitraM | wait, not sweden |
| 15:53.30 | zenitraM | swiss |
| 15:54.06 | mnemoc | the tracking number can be assigned before posting |
| 15:54.40 | zenitraM | yup, but it is usually blank |
| 15:54.47 | zenitraM | i.e amazon uses mrw to ship to spain |
| 15:55.14 | zenitraM | but they ship in bulk to madrid using another courier and until it doesn't arrive to madrid the package is shown as "En espera de llegada a MRW" in the tracking |
| 15:55.19 | penguin42 | zenitraM: Which config are you using with that kernel? |
| 15:55.45 | mnemoc | i always receive amazon over normal post (correos.es) |
| 15:56.03 | mnemoc | penguin42: sun4i_defconfig + disable SND_SOC |
| 15:56.08 | zenitraM | + enable early printk |
| 15:56.56 | zenitraM | mnemoc: idk, amazon.es always uses mrw |
| 15:57.10 | mnemoc | uh, didn't know amazon.es :< |
| 15:57.19 | zenitraM | you didn't? :o |
| 15:57.36 | mnemoc | I always buy from amazon.com .... or thebookdepository |
| 15:57.44 | mnemoc | how old is amazon.es? |
| 15:57.46 | zenitraM | oh well, then it comes from USA |
| 15:57.48 | zenitraM | a few months |
| 15:58.02 | zenitraM | i meant amazon UK or ES,they come using mrw |
| 15:58.30 | zenitraM | cheap electronics and books.. and 15€/year for unlimited free 3-day shipping |
| 15:58.45 | mnemoc | nice deal |
| 15:59.17 | zenitraM | well, it seems swiss post does indeed have an office in HK |
| 15:59.18 | zenitraM | http://www.swisspost.com/hongkong |
| 16:00.00 | mnemoc | swiss or sweden? |
| 16:00.29 | zenitraM | on the speed post page i only found swiss |
| 16:01.13 | zenitraM | tom might have confused the names (even if us spanish ppl do, imagine the chinese..) |
| 16:01.29 | penguin42 | grrr it doesn't cleanly cross build |
| 16:02.01 | mnemoc | penguin42: disabling sound it cross build fine here |
| 16:02.44 | mnemoc | in uImage sound fails, in modules, video and vibrator |
| 16:02.44 | penguin42 | mnemoc: I'm getting unrecognised command line options for -mlittle-endian and -mno-thumb-interwork in scripts/mod/empty.c |
| 16:03.04 | mnemoc | make ARCH=arm CROSS_COMPILE=arm-linux-gnueabi- uImage |
| 16:03.20 | penguin42 | ah ok; I was assuming that was a host built one in scripts/ |
| 16:03.28 | *** part/#arm-netbook QingPei1 (~qingpei@123.118.109.170) |
| 16:03.57 | penguin42 | right yeh that's happier |
| 16:03.58 | mnemoc | I use linaro's toolchain on ubuntu |
| 16:03.59 | penguin42 | pats -j10 |
| 16:04.21 | penguin42 | yeh, just works out of the box |
| 16:04.37 | mnemoc | still needs lots of work |
| 16:04.56 | mnemoc | I'm only fixing API changes |
| 16:06.13 | mnemoc | but hopefully others (you) will jump on board :p |
| 16:07.27 | penguin42 | API issues where? |
| 16:07.57 | mnemoc | there are many API changes between 2.6.36 and 3.0 |
| 16:08.05 | penguin42 | oh yeh, kernel API changes - sure |
| 16:09.42 | zenitraM | mmm what's exactly the difference between bImage and zImage? |
| 16:09.57 | arokux | compression? |
| 16:10.04 | zenitraM | i'm gonna try to boot using boot0 directly instead of u-boot |
| 16:10.10 | zenitraM | bImage = ungzipped zImage? |
| 16:10.10 | arokux | i.e. different compression? |
| 16:10.19 | zenitraM | or.. bzipped? |
| 16:10.26 | arokux | Image in uncompressed, zImage is compressed. |
| 16:10.45 | zenitraM | bImage should be.. bzipped? |
| 16:10.59 | arokux | run make help |
| 16:11.04 | arokux | in kernel tree |
| 16:11.13 | arokux | and find bImage there |
| 16:11.25 | penguin42 | doesn't seem to list bImage |
| 16:11.31 | zenitraM | just bzImage |
| 16:11.34 | penguin42 | just zImage, Image, xipImage, uImage and bootpImage |
| 16:11.58 | zenitraM | mmm let's see how does the allwinner build script create the bImage |
| 16:12.35 | penguin42 | If you're coming from uboot I'd agree with uImage mnemoc suggested above |
| 16:13.01 | zenitraM | when booting from uBoot the kernel boots silently |
| 16:13.09 | zenitraM | and i haven't figured out why |
| 16:13.44 | penguin42 | zenitraM: It's worth using printenv in uBoot to make sure that what you've set has been stored as you expected |
| 16:14.32 | zenitraM | i've done so, the bootargs parameter doesn't have the quiet options |
| 16:14.40 | zenitraM | and has bootlevel=8 which should output everything |
| 16:15.20 | zenitraM | okay.. found out where does bImage come from |
| 16:15.26 | zenitraM | <PROTECTED> |
| 16:15.41 | zenitraM | (from scripts/build_sun4i_crane.sh) |
| 16:16.15 | zenitraM | mmmm… how about using the sun4i_crane config instead of the sun4i one? |
| 16:20.14 | zenitraM | compiles |
| 16:20.39 | mnemoc | sun4i-vibrator fixed |
| 16:30.20 | Gumboot | If I order that A1000 from dealextreme, and ship to the UK, I'm going to end up paying a huge admin fee to the courier, aren't I? |
| 16:30.53 | mnemoc | to parcelforce |
| 16:31.04 | zenitraM | if it's stopped |
| 16:31.21 | mnemoc | in uk everything is delivered by parcelforce |
| 16:31.27 | mnemoc | and they always stab you |
| 16:31.31 | penguin42 | haha |
| 16:31.47 | zenitraM | oh what a pity |
| 16:31.58 | mnemoc | brits are rich, who cares :p |
| 16:32.44 | penguin42 | Gumboot: I wonder if you get stung by import duty as well |
| 16:32.50 | Gumboot | Well, if I can't get it shipped by DHL I guess I won't bother. |
| 16:33.01 | Gumboot | I wondered if it was coming from within EU. |
| 16:33.25 | mnemoc | we will know in two weeks :p |
| 16:34.05 | Gumboot | I might get someone to post it from China. |
| 16:36.25 | mnemoc | afaik parcelforce likes to delay shippings, but for a $100 purchase they don't charge much |
| 16:37.40 | Gumboot | They've always charged me more in admin fees than in tax. |
| 16:38.10 | penguin42 | Gumboot: Yeh as soon as it hits import tax etc all the admin fees on top get silly |
| 16:38.14 | Gumboot | DHL are the only courier I know of who have reasonable fees, but they probably cost more for the sender. |
| 16:40.19 | mnemoc | in spain dhl is very unreasonable |
| 16:40.28 | mnemoc | and they also force packages to be taxed |
| 16:40.49 | mnemoc | every country is different :) |
| 16:40.55 | Gumboot | In the UK it's something like max(1.25, 0.02 * (VAT+duty)) |
| 16:41.50 | zenitraM | mnemoc: people from Forocoches are getting tablets shipped by DHL and not stopped by customs... |
| 16:41.56 | Gumboot | For all I know they're shipped from EU anyway. Only, unless I know if for sure I won't place the order. |
| 16:42.10 | zenitraM | …the seller labels them as "RMA RETURNS" |
| 16:42.31 | mnemoc | zenitraM: interesting |
| 16:43.19 | zenitraM | it seems that way they won't bother investigating as supposingly for a return you already paid the taxes when you first bought it |
| 16:44.14 | mnemoc | didn't know that trick |
| 16:44.26 | Gumboot | You can still pay double or triple admin fees when returning faulty products. The courier will tax you, and tell you it's your problem to claim the tax back from HMRC, but they say nothing of the admin fee. |
| 16:44.41 | Gumboot | They just keep it. |
| 16:45.06 | mnemoc | thieves |
| 16:45.12 | Gumboot | Pirates, strictly. |
| 16:45.19 | penguin42 | Gumboot: I guess I'd want to be paid a heck of a lot to have to deal with the HMRC |
| 16:46.12 | Gumboot | I don't know if they do deal with them at all. I've been "taxed" on completely ficticious numbers by FedEx. |
| 16:46.43 | Gumboot | They just made up some numbers, and said I owed tax on them. Since there was no such transaction there's no reason they should have to pay that to HMRC. |
| 16:47.16 | Gumboot | But also no reason I should have to pay them, so they just got a serious of angry and progressively more abusive phone calls until they wrote it off. |
| 16:51.14 | *** join/#arm-netbook gordan (~gordan@78-86-195-86.zone2.bethere.co.uk) |
| 16:54.15 | zenitraM | okay, just got the uImage for the kernel compiled with the sun4i_crane config |
| 16:54.41 | zenitraM | had to disable something related with usb video for it to compile |
| 16:55.39 | mnemoc | in sun4i_defconfig drivers/media/video/sun4i_csi is module, not built-in... fighting that currently |
| 16:57.32 | zenitraM | no, it was on UVC drivers |
| 16:57.34 | zenitraM | drivers/media/video/uvc/uvc_queue.c: En la función ‘uvc_alloc_buffers’: |
| 16:57.34 | zenitraM | drivers/media/video/uvc/uvc_queue.c:140:2: aviso: ISO C90 prohíbe mezclar declaraciones y código [-Wdeclaration-after-statement] |
| 16:57.34 | zenitraM | drivers/media/video/uvc/uvc_queue.c:182:2: aviso: ISO C90 prohíbe mezclar declaraciones y código [-Wdeclaration-after-statement] |
| 16:57.34 | zenitraM | drivers/media/video/uvc/uvc_queue.c: En el nivel principal: |
| 16:57.34 | zenitraM | drivers/media/video/uvc/uvc_queue.c:213:5: error: redefinición de ‘uvc_free_buffers’ |
| 16:57.34 | zenitraM | drivers/media/video/uvc/uvc_queue.c:116:5: nota: la definición previa de ‘uvc_free_buffers’ estaba aquí |
| 16:59.10 | zenitraM | nothing - i still get no output |
| 17:11.32 | penguin42 | zenitraM: So the thing is a heck of a lot of stuff has to work right to get any output at all |
| 17:13.09 | zenitraM | i know, and i'm not entirely sure if it is my fault |
| 17:13.54 | zenitraM | but the thing is, the other working kernel, the 2.6.36, when booted from u-boot, boots in quiet mode but there's no quiet flag anywhere |
| 17:14.43 | zenitraM | so i'm not sure if the problem is that the output doesn't work or if the output works but it isn't outputting anything (because quiet mode) and crashing then |
| 17:16.52 | CIA-46 | rhombus-tech: lkcl master * rd56028c9344a /evaluated_cpus.mdwn: |
| 17:19.02 | *** join/#arm-netbook lkcl (~lkcl@host86-131-177-61.range86-131.btcentralplus.com) |
| 17:20.01 | penguin42 | zenitraM: I guess you could add some code really on in the boot to write a character to serial or enable an led/gpio etc |
| 17:21.22 | CIA-46 | rhombus-tech: Tim master * r151a1b5ac812 /allwinner_a10/orders/thewrongstuff.mdwn: |
| 17:21.31 | zenitraM | apart from not having the enough kernel skills… the tablet has no LEDs at all or "accessible" GPIO |
| 17:22.14 | penguin42 | hmm that sucks |
| 17:22.59 | zenitraM | the only accessible way of debugging is serial, and as i'm using serial redirected using script.bin to the SD slot, i'm not sure if i'd be writing to the "real" serial or not |
| 17:26.28 | penguin42 | zenitraM: You could try uncommenting the #define DEBUG_UART at the top of sw-uart.c and then try each uart in DEBUG_NO a few lines below |
| 17:37.13 | *** join/#arm-netbook arokux (~arokux@HSI-KBW-046-005-103-006.hsi8.kabel-badenwuerttemberg.de) |
| 18:45.48 | CIA-46 | rhombus-tech: Tomas master * r10fb9d4148c1 /allwinner_a10/orders/small_server_for_the_school.mdwn: |
| 18:46.14 | CIA-46 | rhombus-tech: Tomas master * rd8a873b3223e /allwinner_a10/orders/small_server_for_the_school.mdwn: |
| 18:47.39 | CIA-46 | rhombus-tech: Tomas master * rbe6f22582f80 /allwinner_a10/orders/small_server_for_the_school.mdwn: |
| 19:20.39 | arokux | returning to my question today. how ca I find original manufacture of some EBooks? |
| 19:21.55 | mnemoc | many rebrand devices made by Jinke for example |
| 19:22.24 | mnemoc | ww.jinke.com.cn |
| 19:22.50 | mnemoc | www.jinke.com.cn |
| 19:23.50 | mnemoc | arokux: look at this list of devices:http://openinkpot.org/wiki/Hardware |
| 19:41.26 | arokux | what do you think if you find something like this in the firmware: WifiPassword |
| 19:41.26 | arokux | 1234567654321 |
| 19:44.22 | CIA-46 | rhombus-tech: Neeraj master * rc1ceba14b862 /allwinner_a10/orders/Contact_Details.mdwn: |
| 19:46.19 | penguin42 | arokux: Well, it's possible that the user is forced to set the password when they 1st enable it |
| 19:48.02 | arokux | penguin42, what do you think, if in the /etc of some ebook I've found WifiEssid |
| 19:48.03 | arokux | TT |
| 19:48.03 | arokux | WifiApmode |
| 19:48.03 | arokux | WifiPassword |
| 19:48.03 | arokux | 1234567654321 then, it could have wifi module? |
| 19:48.05 | hno | zenitraM[away], you need to enable a bunch of kernel options for early printk support, and also give the right boot arguments to enable it. |
| 19:49.06 | penguin42 | arokux: Or could just be the same root image they use for a dozen models |
| 19:49.28 | arokux | yeah... so to really know it I should look at the board? |
| 19:50.04 | arokux | I want to track EOM. I've also found something like this: AccountEmail |
| 19:50.05 | arokux | ibook@lenovo.com |
| 19:51.17 | penguin42 | EOM? |
| 19:51.19 | hno | arokux, the Mele A1000 has that upper HDD dock. |
| 19:51.38 | arokux | penguin42, sorry OEM |
| 19:51.50 | arokux | hno, yes, so? |
| 19:52.06 | penguin42 | arokux: Well that sounds like it could be lenovo.com I guess - although they're normally quite high end |
| 19:52.20 | arokux | yes |
| 19:54.33 | arokux | how are the chances, that I'll find the OEM website? are they actually publishing such products? |
| 19:56.16 | arokux | another interesting infor |
| 19:56.35 | arokux | # WPA2-EAP/CCMP using EAP-TLS |
| 19:56.35 | arokux | #ctrl_interface=/var/run/wpa_supplicant |
| 19:56.35 | arokux | network={ |
| 19:56.35 | arokux | <PROTECTED> |
| 19:56.35 | arokux | <PROTECTED> |
| 19:56.35 | arokux | <PROTECTED> |
| 19:56.37 | arokux | <PROTECTED> |
| 19:56.41 | arokux | <PROTECTED> |
| 19:56.43 | arokux | <PROTECTED> |
| 19:56.45 | arokux | <PROTECTED> |
| 19:56.47 | arokux | <PROTECTED> |
| 19:56.49 | arokux | <PROTECTED> |
| 19:56.51 | arokux | <PROTECTED> |
| 19:56.53 | arokux | <PROTECTED> |
| 19:56.55 | arokux | <PROTECTED> |
| 19:56.57 | arokux | } |
| 19:57.44 | mnemoc | arokux: use a paste service for that, never to the channels directly |
| 19:57.56 | arokux | ok, sorry |
| 20:18.19 | arokux | ok, I've found it. it is: http://www.hanwang.com.cn/ |
| 20:18.30 | arokux | do you know it guys? |
| 20:18.39 | CIA-46 | rhombus-tech: Mozello master * rd20e7055d8a5 /allwinner_a10/orders/whant_ubuntu_on_that.mdwn: |
| 20:41.08 | A2Sheds | arokux: what about them? |
| 20:42.39 | arokux | A2Sheds, they produce the ebook I bought... the firmware that the distributor shipped contains same date 3 times, and there are manuals of other distributors also.. Also I would like them to publish sources |
| 20:43.01 | A2Sheds | arokux: you want to track makers (OEM's) of tablets by the Linux images they are using? |
| 20:43.36 | arokux | A2Sheds, I've already tracked them out. it is: http://www.hanwang.com.cn/ |
| 20:43.53 | A2Sheds | arokux: you could ask them to publish the source, but that doesn't mean that they will comply |
| 20:45.08 | A2Sheds | arokux: why not just spend your money on products that comply with the GPL? |
| 20:46.25 | arokux | A2Sheds, I didn't know it beforehand, now I'm just interested to find things out. I know already 3 distributors: ukrainian, russian and some other company :D |
| 20:46.30 | arokux | just curious |
| 20:56.22 | A2Sheds | http://www.google-melange.com/gsoc/homepage/google/gsoc2012 |
| 20:56.38 | A2Sheds | GSOC 2012 application deadline is in 5 days |
| 20:58.37 | mnemoc | what sort of gsoc project can you imagine related to eoma? |
| 20:59.51 | A2Sheds | maybe firmware for the STM32 controllers? |
| 21:00.54 | A2Sheds | we have the open libs and a few open RTOS, but no working code/support |
| 21:01.28 | A2Sheds | the ARM Linux distros can apply on their own |
| 21:01.48 | A2Sheds | lkcl: ping ^^ |
| 22:20.11 | arokux | what kind of GPU is inside of Mele A1000 if any? |
| 22:20.58 | arokux | ah, it's built in the allwinner a10, so mali 400 |
| 22:24.19 | CIA-46 | rhombus-tech: Tomas master * r2fe8a52f4213 /allwinner_a10/orders/small_server_for_the_school.mdwn: |
| 22:30.07 | arokux | have you seen guys, that our board already has 3.0.8 kernel supporting it? http://tinyurl.com/7j5ooqp |
| 22:32.30 | arokux | so they should have published the sources?! |
| 22:36.13 | hno | arokux, maybe. |
| 22:36.55 | arokux | hno, shoudn't we hunt them instead of doing the port by overselves? |
| 22:37.35 | hno | Sure we can ask Tom. But we also need to be familiar with the A10 kernel. |
| 22:38.30 | hno | I do not really want an Android kernel. |
| 22:39.37 | arokux | but is it android kernel? it says Linux 3.0.8 |
| 22:40.18 | arokux | hm.. but then Android... it cannot run with Linux kernel, right? |
| 22:42.45 | hno | Android can run on a standard Linux kernel nowdays they say. Just not as efficiently. |
| 22:43.17 | hno | but the A10 is very targeted to Android, so the BSP is for Android. |
| 22:44.05 | arokux | hno, BSP? (please ping me if you send a message) |
| 22:46.01 | hno | Board Support Package, Lichee Linux BSP. Basically the krernel. |
| 22:47.00 | hno | arokux, ^ I usually only ping on stuff that matters. |
| 22:50.38 | arokux | I do not know much, but I think staying with upstream kernel is good. maybe we'll have luck pushing changes upstream, then they'll be supported by kernel maintainers once API changes |
| 23:00.41 | arokux | which FPU Allwinner A10 implements? there are: vfpv3, vfpv3-d16 and neon. |
| 23:05.09 | hno | Kernel config says CONFIG_VFP=y CONFIG_VFPv3=y CONFIG_NEON=y |
| 23:08.02 | hno | and http://forum.xda-developers.com/showpost.php?p=20991207&postcount=138 |
| 23:12.02 | arokux | I'm currently building a toolchain with crosstool-ng using latest linaro-gcc. there is useful blog post here. |
| 23:12.09 | arokux | http://randomsplat.com/id192-building-a-hard-float-arm-toolchain.html |
| 23:14.20 | hno | which GCC version? |
| 23:14.44 | arokux | 4.6.2 |
| 23:15.02 | arokux | but it's linaro-gcc-4.6.2, latest |
| 23:18.11 | hno | ok. Linaro also have a 4.7 branch, but may be a little on the edge. |
| 23:27.35 | *** join/#arm-netbook toxicpsion (~toxicpsio@out-ab-165.wireless.telus.com) |
| 23:42.06 | xvilka | 4.7.0 is now frozen for release, so not much on the edge |
| 23:47.11 | arokux | It's likely that the performance benefits are much larger on Cortex-A8 CPUs than on Cortex-A9 CPUs which have a faster VFP design and more conventional pipeline. |
| 23:47.19 | arokux | it's a citation. |
| 23:48.09 | arokux | about binaries generated with hardware powered float point operations |
| 23:54.42 | hno | arokux, citation about what? |
| 23:55.17 | arokux | about the fact that A8 is better then A9 in term of float point |
| 23:57.35 | hno | Where? |
| 23:59.44 | arokux | http://wiki.debian.org/ArmHardFloatPort/ |
| 23:59.58 | arokux | (you could have googled :) ) |
