| 00:04.17 | alcides | more like a chinese ebay |
| 00:05.05 | spm_Draget | So, it is sold/delivered by third-party sellers. And I might get ripped of anytime? |
| 00:06.42 | alcides | yes, but like abay/paypal they have tools to protect you |
| 00:25.37 | spm_Draget | Anyone experience with paying via bank-transfer? I am from germany - no idea where their account is, what fees are added and how easy this works…? |
| 01:33.02 | *** join/#cubieboard gzamboni (~gzamboni@2a02-8426-513b-5100-0224-21ff-fe57-307c.rev.sfr.net) |
| 01:39.16 | *** join/#cubieboard grimes99 (~du2@host-78-147-113-98.as13285.net) |
| 01:39.32 | grimes99 | has anybody got this board yet? |
| 01:43.46 | des2 | points to mnemoc |
| 01:44.36 | grimes99 | 1 person only? |
| 01:45.09 | des2 | no but he's usually here and knows a lot. |
| 01:46.59 | grimes99 | ty |
| 02:45.43 | *** join/#cubieboard stuw_ (~stuw@87.255.14.112) |
| 02:55.15 | *** join/#cubieboard grimes99 (~du2@host-78-147-113-98.as13285.net) |
| 03:38.11 | *** join/#cubieboard Skry (~skry@37-136-205-212.nat.bb.dnainternet.fi) |
| 04:44.42 | avernos | does anyone have the buildroot login password? |
| 05:27.40 | *** join/#cubieboard rellla (~Thunderbi@p5B0793DC.dip0.t-ipconnect.de) |
| 05:44.17 | *** join/#cubieboard Skry (~skry@37-136-122-85.nat.bb.dnainternet.fi) |
| 06:27.15 | *** join/#cubieboard avernos (~avernos@unaffiliated/avernos) |
| 07:02.23 | mnemoc | hno: problem is that I don't know exactly what to suggest as I broke support for it in our tree by removing allwinner boot hacks in 3.4 |
| 07:48.46 | mnemoc | and not sure if i want to revert their removal or advertise an outdated fork... |
| 08:02.54 | *** join/#cubieboard avernos (~avernos@111.192.246.193) |
| 08:02.55 | *** join/#cubieboard avernos (~avernos@unaffiliated/avernos) |
| 09:01.20 | *** join/#cubieboard bfree (~niall@109.255.245.157) |
| 09:38.24 | e-ndy | hno, ping hi |
| 09:40.32 | e-ndy | hno, is planned device tree for cubieboard soon? |
| 09:40.46 | e-ndy | hno, in u-boot (and kernel) |
| 09:41.45 | mnemoc | not soon, but yes |
| 10:29.12 | *** join/#cubieboard mysteryname (~mysteryna@14-202-9-167.tpgi.com.au) |
| 10:29.13 | *** join/#cubieboard uNMd (58470d42@gateway/web/freenode/ip.88.71.13.66) |
| 11:19.42 | *** join/#cubieboard Skry (~skry@37-136-132-188.nat.bb.dnainternet.fi) |
| 11:25.31 | *** join/#cubieboard Anton_ (~aaaa@188.114.56.245) |
| 12:09.25 | *** join/#cubieboard h0cin (~h0cin@unaffiliated/h0cin) |
| 12:40.11 | *** join/#cubieboard h0cin (~h0cin@unaffiliated/h0cin) |
| 12:40.26 | *** join/#cubieboard anton___ (~anton@unaffiliated/anton---/x-4931027) |
| 13:16.07 | *** join/#cubieboard h0cin (~h0cin@unaffiliated/h0cin) |
| 13:18.13 | *** join/#cubieboard mike_sun (~quassel@177.97.65.6) |
| 13:31.12 | hno | mnemoc, not sure I follow |
| 13:31.47 | hno | broke support for what? |
| 13:32.12 | mnemoc | hno: in 3.4 I removed allwinner boot hacks |
| 13:33.20 | mnemoc | hno: that renders it unuable from nand currently |
| 14:01.10 | mnemoc | hno: so I wonder if I should revive the allwinner boot hacks to let people run linux from nand |
| 14:01.17 | mnemoc | like the case of the cubie |
| 14:28.12 | hno | mnemoc, only need an updated u-boot. |
| 14:28.38 | hno | or a boot header setting machtype & atag address. |
| 14:28.55 | hno | allowing to boot directly from boot1. |
| 14:29.10 | mnemoc | so you +1 to keep the boot hacks removed from 3.4+? |
| 14:29.21 | hno | Yes. |
| 14:29.31 | mnemoc | cool :) |
| 14:29.59 | hno | One sec and I'll make that head.S hack into a boot header. |
| 14:49.49 | *** join/#cubieboard lerc_ (~quassel@121-74-226-42.telstraclear.net) |
| 14:56.52 | *** join/#cubieboard trinathon (~deans-and@128.187.215.194) |
| 15:16.32 | hno | mnemoc, boot header in sunxi-tools |
| 15:19.02 | *** join/#cubieboard NAiL (~repvik@nslu2-linux/repvik) |
| 15:19.15 | *** join/#cubieboard trinathon (~deans-and@128.187.215.194) |
| 15:40.34 | *** join/#cubieboard hackkinen (576f10bf@gateway/web/freenode/ip.87.111.16.191) |
| 15:44.27 | *** join/#cubieboard alcides (~alcides@187.23.192.105) |
| 15:44.27 | *** join/#cubieboard alcides (~alcides@unaffiliated/alcides) |
| 15:48.37 | *** join/#cubieboard gzamboni (~gzamboni@2a02-8426-513b-5100-0224-21ff-fe57-307c.rev.sfr.net) |
| 16:18.54 | hno | But in case of cubieboard etc, better to fix the nand u-boot. |
| 16:21.19 | mnemoc | yes |
| 16:23.10 | mnemoc | can we write vfat/nand with the nand-capable uboot? |
| 16:23.47 | mnemoc | that could be an easy procedure to switch from boot0/boot1 to spl |
| 16:25.27 | mnemoc | replace u-boot once, reboot, get data dumped to nanda in a file, use that file to make the board header for spl, reboot again, get boot0/boot1 replaced with free software |
| 16:35.47 | *** join/#cubieboard maciel310 (0c972e82@gateway/web/freenode/ip.12.151.46.130) |
| 16:41.40 | *** join/#cubieboard moravas (c1e029a8@gateway/web/freenode/ip.193.224.41.168) |
| 16:41.54 | moravas | hy |
| 16:42.22 | moravas | can somebody tell me some information about the shipping of boards? |
| 16:42.46 | mnemoc | the second batch is not done yet |
| 16:42.51 | moravas | when can we order those from stock? |
| 16:43.07 | moravas | why? |
| 16:43.35 | mnemoc | lack of factory time i suppose |
| 16:43.47 | mnemoc | factories don't like to make small batches |
| 16:44.53 | moravas | what meens "small batch"? |
| 16:44.57 | moravas | 100 |
| 16:45.00 | moravas | 1000 |
| 16:45.07 | moravas | 10 000 pieces? |
| 16:45.26 | mnemoc | in the 100s. it's still a prototype |
| 16:45.34 | moravas | pff |
| 16:46.09 | moravas | i have hoped, they are stable version |
| 16:46.17 | mnemoc | not yet |
| 16:46.40 | mnemoc | afaik 2012-08-08 is the second prototype rev |
| 16:46.48 | moravas | where can I track this hardware bugs? |
| 16:47.15 | mnemoc | ML i suppose. but there has been none reported |
| 16:47.34 | mnemoc | the first public batch came with the sata problem fixed manually |
| 16:48.19 | mnemoc | the only hw bug i've found on mine in on the case ;-) |
| 16:48.31 | avernos | haha, same here |
| 16:48.34 | moravas | ohh |
| 16:48.39 | moravas | it is a litle bug :) |
| 16:49.16 | avernos | very nice case tho |
| 16:49.33 | moravas | can you offer me some board which can I order from stock? |
| 16:50.05 | moravas | some powerfull and focus on SATA |
| 16:52.11 | mnemoc | none yet |
| 16:54.29 | mnemoc | beside the cubieboard the sabrelite is the only other dev board with sata (and without absurd price) I know |
| 16:55.05 | mnemoc | the wandboard will have sata, but unsoldered... and it's not out yet |
| 16:56.13 | alcides | that is expensive |
| 16:57.58 | mnemoc | yes. but not yet "absurd" :) |
| 16:58.05 | moravas | ere now does not I found some board as cubie :( |
| 16:58.49 | mnemoc | moravas: subscribe to the ML and the [notify me] list and wait some weeks |
| 16:59.28 | moravas | I am subscribed into this notify list... |
| 16:59.35 | *** join/#cubieboard trinathon (~deans-and@128.187.215.194) |
| 17:17.42 | *** join/#cubieboard grimes99 (~du2@host-78-147-113-98.as13285.net) |
| 17:36.04 | *** join/#cubieboard trinathon (~deans-and@128.187.215.194) |
| 18:13.05 | *** join/#cubieboard stuw_ (~stuw@87.255.14.112) |
| 18:19.02 | *** join/#cubieboard infinityplus1 (~infinityp@70-89-160-190-smc-co.hfc.comcastbusiness.net) |
| 18:24.19 | *** join/#cubieboard xemacs5 (~xemacs5@70-89-160-189-Davis-and-Larson-co.hfc.comcastbusiness.net) |
| 18:24.39 | xemacs5 | just got my cubieboard in the mail |
| 18:24.48 | xemacs5 | pretty cool! but now I want a real linux on it |
| 18:24.55 | grimes99 | which country? |
| 18:24.58 | xemacs5 | any recommendations? |
| 18:25.02 | xemacs5 | <-- USA |
| 18:25.08 | grimes99 | 512mb? |
| 18:25.13 | xemacs5 | comes registered mail |
| 18:25.32 | xemacs5 | I think so, though it shows 320MB to linux, maybe some is reserved for graphics ? |
| 18:26.11 | grimes99 | what about a case for it? |
| 18:26.15 | xemacs5 | want to try Asterisk |
| 18:26.17 | xemacs5 | did not get the case |
| 18:26.29 | xemacs5 | right now it is sitting on the foam it came in |
| 18:26.48 | xemacs5 | weighed down by some CD-Rs |
| 18:28.01 | spm_Draget | Careful! |
| 18:28.21 | spm_Draget | I do not know how teh cubieboard is shipped, but some types of antistatic foam is conductive! |
| 18:28.37 | xemacs5 | not this stuff, I don't think |
| 18:28.41 | spm_Draget | *If* it is such a foam, running the cubieboard sitting on it can damage it. |
| 18:28.43 | spm_Draget | Okay. |
| 18:29.16 | xemacs5 | pretty sure it is the cheap stuff |
| 18:30.39 | xemacs5 | just to be sure I put a CD-R under it, the plastic won't conduct anything |
| 18:43.00 | *** join/#cubieboard Anton_ (~aaaa@31.23.108.36) |
| 18:43.56 | *** join/#cubieboard trinathon (~deans-and@128.187.215.194) |
| 18:58.35 | xemacs5 | I just licked the foam, it didn't taste like the special stuff |
| 18:59.47 | *** join/#cubieboard markit (~yaaic@88-149-177-66.v4.ngi.it) |
| 19:01.42 | mnemoc | o.o |
| 19:29.43 | spm_Draget | That is a very non-standard way of testing it. |
| 19:30.39 | spm_Draget | For the record: Using a multimeter and placing two probes 1-2 millimeters appart - if it reaks some kOhm or mOhm, then it is conductive and should not be used under a running PCB |
| 19:32.13 | xemacs5 | ok thanks for the tip |
| 19:35.56 | xemacs5 | I have a multimeter somewhere |
| 19:39.09 | mnemoc | no standoffs? |
| 19:42.24 | Turl | xemacs5: did you just write to the ML about having 320M of RAM? |
| 19:43.24 | mnemoc | 512 - mali - ve - fb - g2d |
| 19:43.47 | mnemoc | ve alone takes 80M |
| 19:44.03 | Turl | if you run dmesg, you should be able to see all the memory reserves around the start of it |
| 19:44.47 | Turl | mnemoc: is there any actual use for g2d? we should disable it by default if it isn't already |
| 19:45.31 | mnemoc | the x11 exa driver |
| 19:45.36 | mnemoc | once someone writes it... |
| 19:45.42 | *** join/#cubieboard mike_sun_ (~quassel@177.97.65.6) |
| 19:46.19 | Turl | mnemoc: will you take a patch to s/default y/default n/ ? |
| 19:46.38 | mnemoc | yes |
| 19:46.45 | mnemoc | on the ML |
| 19:46.53 | Turl | yep, I'll send one in a bit |
| 19:46.59 | mnemoc | against 3.0 please |
| 19:47.04 | mnemoc | and update the defconfigs |
| 20:19.18 | Turl | mnemoc: sent |
| 20:26.09 | Turl | mnemoc: you should be able to apply the first one to 3.4 too |
| 20:26.49 | mnemoc | i'll give people some time to comment first |
| 20:27.02 | mnemoc | it's not critical, and won't reduce whining |
| 20:27.13 | mnemoc | but i have nothing against it |
| 20:27.43 | Turl | :) |
| 20:54.59 | mnemoc | Turl: btw, did you end up fixing the misterious problem in your cpufreq refactoring? |
| 20:58.55 | Turl | nope, haven't touched that code for a while |
| 21:02.35 | mnemoc | ah, ok. i thought the discussion with hno about it had produced some outcome :< |
| 21:07.37 | xemacs5 | mnemoc - thanks |
| 21:08.38 | mnemoc | xemacs5: if you disable mali you can also make fb dynamically allocated |
| 21:08.47 | mnemoc | and save a bunch of memory |
| 21:09.13 | xemacs5 | that sounds good, I am new to ARM so need to figure out how to do that |
| 21:09.57 | mnemoc | while building your kernel |
| 21:11.02 | xemacs5 | OK will keep that in mind ; this is a test board for a new project, |
| 21:11.15 | xemacs5 | so later I would order the 1GB and maybe use stock kernels |
| 21:28.41 | *** join/#cubieboard grimes99 (~du2@host-78-147-113-98.as13285.net) |
| 22:03.31 | *** join/#cubieboard Skry (~skry@37-219-214-193.nat.bb.dnainternet.fi) |
| 22:06.04 | hno | mnemoc, yes you can drop a new u-boot into nanda without problem. But does not really solve the SPL issue. |
| 22:07.08 | hno | we really need to make progress on the low level NAND driver. With that we can even use FEL to extract the stuff needed. |
| 22:07.30 | xemacs5 | hmm, how do I tell the cubie to boot off of uSD ? |
| 22:07.46 | hno | xemacs5, By insterting a bootable uSD and rebooting. |
| 22:08.03 | hno | it always tries uSD first, then NAND. |
| 22:08.05 | xemacs5 | I did a soft-reboot |
| 22:08.16 | xemacs5 | e.g. was logged in via SSH and typed "reboot" |
| 22:08.17 | grimes99 | is it possible to mod the microSD socket to SDHC? |
| 22:08.19 | hno | soft or hard reboot both works. |
| 22:09.12 | hno | grimes99, it is SDHC capable. To use full size SD just use an SD car->uSD socket adapter. |
| 22:09.58 | hno | xemacs5, how did you prepare the uSD card? |
| 22:10.34 | xemacs5 | I downloaded the 33MB armhf image |
| 22:10.39 | xemacs5 | to /tmp |
| 22:10.58 | grimes99 | what courier companies did cubie offer for postage? DHL? |
| 22:10.59 | xemacs5 | then used gunzip -c image.tgz | dd bs=1M /dev/mmcblk0 |
| 22:11.03 | mnemoc | hno: but if we drop a u-boot on nand, can't it dump boot0 and boot1 to the vfat? |
| 22:11.24 | hno | mnemoc, only once u-boot knows to read raw NAND. |
| 22:11.25 | mnemoc | sure `fel` is nicer |
| 22:11.41 | mnemoc | hno: and lichee-dev doesn't? |
| 22:11.46 | hno | it doesn't. |
| 22:11.49 | mnemoc | :( |
| 22:12.17 | hno | well, the driver obviously does. But that's layers below.. |
| 22:13.17 | hno | But it does not look complicated. Just a bit short of time to tackle it at the moment. |
| 22:14.01 | hno | even MTD support should be trivial. |
| 22:15.08 | hno | sunxi_nand/src/ is the driver. The rest is just layers on top and not needed for MTD. |
| 22:15.46 | hno | s/src/nfc/ |
| 22:16.29 | hno | total including headers & license blobs: 2821 lines. |
| 22:16.40 | mnemoc | :) |
| 22:18.24 | xemacs5 | can someone point me to a Linux image that should work on the cubie? I tried the ubuntu 12.04.1 armhf image but it seems it did not work |
| 22:19.18 | mnemoc | xemacs5: you need to use the right bootloader and script.bin |
| 22:20.03 | hno | script.bin is in nanda. |
| 22:20.08 | xemacs5 | ok, so that is the missing ingredient |
| 22:20.42 | mnemoc | and here: https://github.com/linux-sunxi/sunxi-boards/tree/master/sys_config/a10 |
| 22:21.01 | xemacs5 | got it, so |
| 22:21.09 | hno | bootloader is u-boot available from github.com/hno/https://github.com/hno/uboot-allwinner |
| 22:21.16 | xemacs5 | mkdir /testmount ; mount /dev/nanda /testmount |
| 22:21.31 | xemacs5 | gives me access to that |
| 22:21.37 | hno | yes |
| 22:22.55 | xemacs5 | so am I copying something from nanda to the SD card, or modifying a config file in nanda |
| 22:23.58 | hno | xemacs5, script.bin needs to be copied from nanda to /boot on your SD card. |
| 22:24.28 | hno | then you need to build u-boot and install that as well on the sd card. |
| 22:24.43 | hno | instructions at https://github.com/hno/uboot-allwinner/wiki |
| 22:25.50 | xemacs5 | ok this all makes more sense |
| 22:26.11 | xemacs5 | I am coming from a Sun SPARC and PC/x86 world , so this part is new to me |
| 22:27.29 | xemacs5 | is there a binary u-boot somewhere? |
| 22:27.37 | hno | Not yet. |
| 22:32.52 | xemacs5 | so I have to set up a complete cross compiling toolchain under Linux, in order to be able to boot from uSD ? |
| 22:33.15 | hno | I'll push an u-boot image to my web server in a minute. |
| 22:34.44 | xemacs5 | I should confirm, that a 16GB uSD is not a problem? Like, there is not a restriction as there can be on old x86, of less than 4GB size |
| 22:35.05 | hno | The controller is SDHC capable. |
| 22:36.32 | xemacs5 | that is progress1 :-) |
| 22:36.36 | xemacs5 | progress! |
| 22:39.23 | hno | xemacs5, http://www.hno.se/code/A10/u-boot/ |
| 22:40.23 | mnemoc | i use 16GB class8-10 cards only |
| 22:40.33 | xemacs5 | got them both |
| 22:42.04 | xemacs5 | I bought a 16GB card from Costco, SanDisk , class 6 I think |
| 22:42.23 | mnemoc | should work without troubles |
| 22:42.28 | xemacs5 | this is a test board, we would probably buy things in bulk if everything works OK |
| 22:44.12 | xemacs5 | ok I used wget, have it under /tmp |
| 22:44.23 | xemacs5 | so following the above, I think I should run |
| 22:44.24 | xemacs5 | dd if=/tmp/sunxi-spl.bin of=/dev/mmcblk0 bs=1024 seek=8 |
| 22:44.45 | mnemoc | hno: is `[ 0.000000] Ignoring unrecognised tag 0x00000000` normal? |
| 22:45.01 | mnemoc | xemacs5: https://github.com/hno/uboot-allwinner/wiki |
| 22:45.44 | hno | mnemoc, yes from what I can tell. |
| 22:45.44 | xemacs5 | then dd if=/tmp/u-boot.bin of=/dev/mmcblk0 bs=1024 seek=32 |
| 22:46.18 | hno | yes |
| 22:46.41 | xemacs5 | then I use fdisk, create a partition which starts at 2048 |
| 22:47.05 | mnemoc | 2048*512 |
| 22:47.12 | xemacs5 | is that the number of the first cylinder? |
| 22:47.18 | hno | block |
| 22:47.19 | mnemoc | some cards come with weird geometries |
| 22:47.44 | xemacs5 | it says |
| 22:47.50 | hno | xemacs5, if your fdisk defaults to cylinders then use the u command to switch to blocks. |
| 22:47.56 | xemacs5 | Disk /dev/mmcblk0: 15.9 GB, 15931539456 bytes |
| 22:47.56 | xemacs5 | 4 heads, 16 sectors/track, 486192 cylinders |
| 22:47.56 | xemacs5 | Units = cylinders of 64 * 512 = 32768 bytes |
| 22:48.37 | hno | that's an odd one. But not important. Everything uses blocks anyway these days. |
| 22:48.39 | xemacs5 | ok now it is showing 512 byte blocks, I used fdisk -u /dev/mmcblk0 |
| 22:49.25 | hno | The days of CHS addressing is buried deep down with old DOS floppies somewhere. |
| 22:49.57 | xemacs5 | it was still used during IDE days |
| 22:50.16 | hno | Not in many years. IDE also used LBA addressing. |
| 22:50.20 | xemacs5 | as drives got bigger, the disk would lie about C/H/S in order to stay within limits |
| 22:50.44 | hno | xemacs5, all IDE drives lie about C/H/S. |
| 22:51.32 | xemacs5 | yes, it has been 18 years since I had to deal with this, so I am fuzzy on the details |
| 22:52.41 | xemacs5 | so I used dd to write the u-boot and sunxi-spl.bin files, then created a linux partition layout starting at sector 2048 |
| 22:53.00 | hno | Which filesystem? |
| 22:53.29 | hno | You need first partition to be ext3 or vfat. |
| 22:53.40 | f1r3 | to enter custom chs values in the bios set disk type 47 ;-) |
| 22:53.46 | xemacs5 | I did not create a FS yet, it set it as type 83 (ext2 or 3 ) |
| 22:54.57 | f1r3 | i didnt meant the FS type |
| 22:54.59 | xemacs5 | dd wrote the other files "raw" |
| 22:56.07 | xemacs5 | so I think the next step is to figure out where the other files go, then, write the ubuntu-12.04 or other OS files to the disk |
| 23:00.38 | hno | xemacs5, yes. For partitoning you have two choices |
| 23:00.46 | hno | a) Separate /boot |
| 23:00.49 | xemacs5 | would the ubuntu armhf file, that I use dd to write to the uSD card, contain the system.bin etc. or are those other file |
| 23:00.51 | xemacs5 | files |
| 23:01.01 | hno | b) A single filesystem with root + /boot. |
| 23:01.38 | hno | xemacs5, system.bin is unique for each A10 board model. |
| 23:02.06 | hno | I don't know which ubuntu system image you are using. |
| 23:02.38 | xemacs5 | the armhf image from here: http://cdimage.ubuntu.com/ubuntu-core/releases/12.04/release/ubuntu-core-12.04.1-core-armhf.tar.gz |
| 23:02.57 | xemacs5 | of course, I do not know if it will work |
| 23:03.31 | hno | Ok, that's the generic ubuntu rootfs for ARM, in tar.gz format. Contains a generic userspace, but no kernel or modules. |
| 23:05.25 | xemacs5 | no wonder it did not work! so I need to find a kernel and module package |
| 23:08.21 | xemacs5 | can I boot the kernel from nand and have it use the userspace for ubuntu? really I just need something that has a package manager |
| 23:11.53 | *** part/#cubieboard maciel310 (0c972e82@gateway/web/freenode/ip.12.151.46.130) |
| 23:23.03 | hno | xemacs5, maybe. I have not looked at how the nand layout looks like. But there is A10 hwpacks with kernel available. |
| 23:28.23 | hno | http://dl.linux-sunxi.org/nightly/2012.10.09/ is the latest I think. |
| 23:28.59 | hno | the mele one should be ok, just replace u-boot. |
| 23:32.57 | hno | and script.bin |
| 23:35.09 | hno | xemacs5 ^ |
| 23:35.27 | xemacs5 | ok I see what you have typed |
| 23:37.03 | xemacs5 | just trying to make sure I understand where everything should go |
| 23:37.47 | hno | There is a script to automate stuff |
| 23:39.40 | hno | http://www.cnx-software.com/2012/06/13/hardware-packs-for-allwinner-a10-devices-and-easier-method-to-create-a-bootable-ubuntu-12-04-sd-card/ |
| 23:39.57 | hno | http://undergroundmod.com/?p=1244 |
| 23:46.35 | hno | mnemoc, is there a linux-sunxi page about this? Can't seem to find one... |
| 23:47.40 | mnemoc | none yet |
| 23:47.49 | mnemoc | there is a newer post on his blog. 1m |
| 23:48.04 | mnemoc | hno: http://www.cnx-software.com/2012/07/20/nightly-builds-for-allwinner-a10-u-boot-linux-kernel-and-hardware-packs/ |
| 23:48.49 | mnemoc | they are made using https://github.com/cnxsoft/a10-tools |
| 23:48.57 | mnemoc | on a build server cnxsoft got |
| 23:49.30 | mnemoc | and from there (by cron I suppose) he builds and pushes new 'nightly' builds |
| 23:53.40 | xemacs5 | so this just requires a Linux box, not a cross-compiling toolchain |