| 00:01.26 | slapin | hno: http://ossfans.org/nand_init.txt |
| 00:04.30 | hno | mnemoc, this is odd. PMU looks OK. Issue seems to be with DRAM controller somewhere, or maybe DRAM power. |
| 00:05.32 | hno | slapin, looks familiar. |
| 00:08.46 | slapin | hno: ah |
| 00:09.02 | slapin | hno: I try to figure out init sequence for NFC |
| 00:09.33 | slapin | and I have lots of wtfs there... |
| 00:09.37 | mnemoc | hno: might it be related to libv's halt when probing the uart after rebooting his tablet? |
| 00:10.45 | hno | No idea. missed that. |
| 00:11.10 | libv | not sure whether it is actually a uart issue |
| 00:11.17 | libv | reboots just hang at that point |
| 00:11.20 | libv | first boots are fine |
| 00:11.53 | hno | kernel hangs during hardware setup? |
| 00:12.07 | slapin | 4am here, going to bed now, cu all |
| 00:12.25 | mnemoc | slapin: good night |
| 00:12.34 | libv | hno: looks like it, want a pair of logs (first boot, and then reboot)? |
| 00:12.42 | hno | Not tonight. |
| 00:13.03 | libv | hno: it's what you tell me everytime |
| 00:13.32 | libv | feels rejected |
| 00:13.35 | libv | :p |
| 00:13.40 | mnemoc | :) |
| 00:13.43 | hno | Well there isn |
| 00:13.53 | hno | there isn't much interesting in those logs. |
| 00:14.11 | libv | true |
| 00:14.48 | hno | but if the DRAM controller is slightly misconfigured then odd things will happen. Same if the PMU is not set right. |
| 00:15.43 | libv | it hangs reliably at the same point though |
| 00:15.51 | hno | the reboot issue on cubieboard is clearly DRAM related. Hangs when trying to read from DRAM, which happens to be the MMC initialization. |
| 00:16.27 | hno | libv, had A13 OlinuXIno hang quite reliably at the same point in the kernel when the PMU was misconfigured. |
| 00:16.47 | libv | ok |
| 00:17.03 | libv | due to more power draw when that specific hw was enabled |
| 00:17.41 | hno | It then hang on first call to printk. Could not enter the function. |
| 00:18.59 | libv | ouch, how did you track this issue down? |
| 00:20.00 | hno | JTAG helped a bit. |
| 00:21.04 | hno | but major part was analysis of Allwinner boot procedure. |
| 00:22.22 | hno | Still have not understood why it does not affect Mele v1.3 |
| 00:26.02 | libv | ah, i seem to have a 1.2 |
| 00:27.20 | hno | Right, meant 1.2. |
| 00:35.39 | *** join/#arm-netbook freakazoid0223 (~matt@pool-173-75-233-172.phlapa.fios.verizon.net) |
| 00:47.34 | hno | Uhm.. this do not look right: DRAM: 0 Bytes |
| 00:48.39 | alcides | definitely not |
| 00:56.10 | *** join/#arm-netbook ccssnet (~ccssnet@c-98-216-141-157.hsd1.ma.comcast.net) |
| 00:58.13 | *** join/#arm-netbook rz2k (~rzk@95-24-125-219.broadband.corbina.ru) |
| 00:58.44 | hno | Ok, u-boot want's ram size in Bytes, not MBytes. |
| 00:59.33 | hno | DRAM: 1 GiB |
| 00:59.59 | hno | and it even starts. |
| 01:28.44 | libv | woah |
| 01:28.53 | libv | 1 line patch makes my life bearable! |
| 01:29.20 | libv | name of the patch is going to be: net/wireless/rtl8192cu: STFU |
| 01:30.12 | *** join/#arm-netbook rz2k (~rzk@95-24-125-219.broadband.corbina.ru) |
| 01:30.15 | libv | mnemoc: take this into your next tag, please |
| 01:32.53 | Turl | :p |
| 01:33.13 | Turl | were you able to test the updated wifi libv? |
| 01:33.50 | libv | Turl: ah, right, let me scroll up, it is a bit painful on a7hd though, i have to stop the device and take out the sd card, as i currently do not have working wifi on it |
| 01:33.57 | libv | well, working wpa-psk on wifi |
| 01:34.01 | libv | scanning works |
| 01:42.34 | *** join/#arm-netbook stefanro (~stefan@pD9FF9F43.dip.t-dialin.net) |
| 02:10.26 | drachensun | oh my god, thank you libv |
| 02:10.33 | drachensun | get that patch committed |
| 02:11.06 | Turl | libv: testing scanning is okay |
| 02:11.52 | L84Supper | 13 straight hours of working on this! |
| 02:12.22 | Turl | L84Supper: working on what? |
| 02:14.11 | L84Supper | Turl: is this the same thing you guys were working on earlier today/yesterday? |
| 02:14.29 | Turl | what? :P |
| 02:17.19 | Turl | is lost |
| 02:38.49 | *** join/#arm-netbook t0dbld1 (~t0dbld@50.33.195.166) |
| 02:40.56 | *** join/#arm-netbook tzafrir_laptop (~tzafrir@212.179.75.202) |
| 03:02.41 | *** join/#arm-netbook cheng (~cheng@175.143.109.117) |
| 03:27.11 | *** join/#arm-netbook tekzilla (~jon@hmbg-4d0680e8.pool.mediaWays.net) |
| 03:52.10 | *** join/#arm-netbook alcides (~alcides@177.35.118.65) |
| 03:52.10 | *** join/#arm-netbook alcides (~alcides@unaffiliated/alcides) |
| 04:06.24 | *** join/#arm-netbook RITRedbeard_ (~redbeard@c-68-37-165-37.hsd1.nj.comcast.net) |
| 04:37.58 | *** join/#arm-netbook kop (~kop@c-67-185-30-133.hsd1.wa.comcast.net) |
| 04:41.52 | *** join/#arm-netbook hipboi (~hipboi@199.15.117.211) |
| 04:53.39 | *** join/#arm-netbook tekzilla (~jon@hmbg-4d06dfd2.pool.mediaWays.net) |
| 05:14.08 | *** join/#arm-netbook hipboi (~hipboi@27.23.139.229) |
| 05:42.57 | *** join/#arm-netbook eebrah (~chatzilla@212.49.88.98) |
| 05:51.01 | Turl | hey hipboi :) |
| 05:55.15 | *** join/#arm-netbook eFfeM (~frans@c73189.upc-c.chello.nl) |
| 06:10.19 | rz2k | omg, libv fixed 8192cu spamming in dmesg? |
| 06:10.26 | rz2k | approves that with both hands |
| 06:13.24 | Turl | rz2k: yup |
| 06:13.39 | Turl | rz2k: and I updated 8192cu to latest realtek release if you fancy trying :) |
| 06:14.19 | Turl | rz2k: https://github.com/allwinner-dev-team/linux-allwinner/commits/allwinner-3.0 top 3 commits |
| 06:15.16 | rz2k | cool, any major changes? |
| 06:15.58 | Turl | seemed to not disconnect on suspend |
| 06:16.13 | Turl | realtek's changelog was outdated when I checked though, so no idea :P |
| 06:19.11 | Turl | gnight! :) |
| 06:20.59 | *** join/#arm-netbook Quarx (~Quarx@46.233.226.255) |
| 06:23.48 | *** join/#arm-netbook rellla (~Thunderbi@p5B0784E5.dip0.t-ipconnect.de) |
| 06:52.25 | *** join/#arm-netbook cheng (~cheng@175.143.109.117) |
| 07:01.01 | *** join/#arm-netbook Quarx|2 (~Quarx@46.233.226.255) |
| 07:02.22 | *** join/#arm-netbook Quarx (~Quarx@46.233.226.255) |
| 07:09.42 | *** join/#arm-netbook tzafrir_laptop (~tzafrir@212.179.75.202) |
| 07:15.17 | *** join/#arm-netbook rellla (~Thunderbi@p5B078393.dip0.t-ipconnect.de) |
| 07:24.23 | *** join/#arm-netbook pawel5870 (~pkarpins@pc.193147149.ip.amg.net.pl) |
| 08:15.06 | *** join/#arm-netbook cheng (~cheng@175.143.109.117) |
| 08:19.36 | *** join/#arm-netbook popolon (~popolon@og-free.planet-service.fr) |
| 08:38.00 | *** join/#arm-netbook Sternennebel (~martin@p5494A7A8.dip.t-dialin.net) |
| 09:01.36 | cat_x301 | morning! noticed things are getting better with sunxi :) |
| 09:09.50 | *** join/#arm-netbook kop (~kop@c-67-185-30-133.hsd1.wa.comcast.net) |
| 09:29.03 | *** join/#arm-netbook kop (~kop@c-67-185-30-133.hsd1.wa.comcast.net) |
| 10:08.51 | mnemoc | sunxi-v3.0.42-r2 tagged, before the merge of stage 3.0 |
| 10:09.59 | discopig | yeah lots of progress lately |
| 10:10.18 | discopig | and with cubieboard and other boards launching it should attract even more attention |
| 10:21.40 | *** join/#arm-netbook mnemoc (~amery@geeks.cl) |
| 10:23.13 | rm | Linaro, the not-for-profit engineering organization developing open source software for the ARM architecture, announced today the formation of the Linaro Enterprise Group (LEG) |
| 10:23.17 | rm | this is hilarious |
| 10:23.36 | specing | Haha |
| 10:27.39 | cat_x301 | clever. not-for-profit sounds just right imo, or maybe i am missing something in there. |
| 10:29.54 | mnemoc | been not-for-profit is about what you do with what you earn. not about how you earn it |
| 10:32.43 | cat_x301 | mnemoc: yeah, right, it is just my jealousy that is speaking on my behalf ;) |
| 10:32.54 | mnemoc | :) |
| 10:35.44 | *** join/#arm-netbook merbanan (~benjamin@h217-27-188-82.cust.tyfon.se) |
| 10:51.19 | stefanro | mnemoc, hno: do you know if anyone invested some time to support the A10 ethernet (wemac) in u-boot? |
| 10:54.56 | *** join/#arm-netbook merbanan (~benjamin@h217-27-188-82.cust.tyfon.se) |
| 10:58.34 | mnemoc | stefanro: we would need to pull random PHYs too... |
| 10:58.47 | mnemoc | stefanro: the ethernet controller is MAC only |
| 11:00.02 | stefanro | mnemoc: yes, thats how it is usually done in embedded ethernet controllers |
| 11:00.29 | stefanro | mnemoc: supporting "standard" PHY's shouldn't be a big deal |
| 11:01.01 | stefanro | mnemoc: i was just wondering if somebody else has already invested some time here - i would really love to use ethernet from within u-boot |
| 11:01.21 | stefanro | mnemoc: as this is my usual development cycle |
| 11:02.47 | mnemoc | stefanro: not that I know off. but it may imply making u-boot read the script.bin file to know about the involved pins |
| 11:02.55 | mnemoc | s/off/of/ |
| 11:13.45 | *** join/#arm-netbook slapin_ (~slapin@slapinbuild.ihdev.net) |
| 11:26.01 | RaYmAn | peers at his new chromebook on his desk |
| 11:26.09 | RaYmAn | they aren't making it easy to do actual work :P |
| 11:26.40 | mnemoc | :) |
| 11:29.03 | *** join/#arm-netbook Almamuetya10 (~almamuety@186.134.32.56) |
| 11:30.57 | fredy | hello, I am experimenting with an a10 tablet and I was wondering if it is possible to change init.rc (or other rc files) on the device. I don't mean make a new boot.img and flash it, but change and save changes when the device is on. thanks :) |
| 11:41.01 | specing | fredy: why not? |
| 11:44.06 | *** join/#arm-netbook lkcl (~lkcl@host86-131-178-159.range86-131.btcentralplus.com) |
| 11:44.29 | fredy | specing: I am new on this so not sure how to do this. I tried to change the files on / but after reboot they are all gone, so I guess I have to mount sth and make changes there. So I am aksing more if somebody have done this procedure and can give me any advice :) |
| 11:45.40 | specing | I did some *fixes* and they stuck around |
| 11:46.36 | fredy | hm... maybe I've done sth wrong then... I'll try again :) thanks |
| 11:50.16 | *** join/#arm-netbook slash_random (~david@host238.190-230-245.telecom.net.ar) |
| 11:51.01 | *** join/#arm-netbook merbanan (~benjamin@h217-27-188-82.cust.tyfon.se) |
| 11:52.45 | slapin_ | fredy: on android? |
| 11:53.17 | fredy | slapin_ on an unofficial CM9 |
| 11:53.56 | slapin_ | fredy: if it has busybox, you can edit files with vi |
| 11:55.03 | fredy | slapin_: I have done this but after reboot the changes weren;t there |
| 11:56.13 | fredy | but I'll try again to be sure that this is the case... |
| 11:59.02 | oliv3r | make sure to mount -o remount,rw; edit; sync; mount -o remout,ro; and check if your changes are still there |
| 11:59.38 | oliv3r | actually all files in / are obviously stored in the ramdisk if i'm not mistaken, so you can't edit those permanently |
| 12:00.08 | fredy | ^that is what I think... |
| 12:01.10 | fredy | so I ll check it again, thanks again :) |
| 12:08.29 | slapin_ | oliv3r: that depends on particular device, mine all have normal / and /system |
| 12:10.20 | *** join/#arm-netbook von_fritz (~fritz@host221-176-dynamic.0-87-r.retail.telecomitalia.it) |
| 12:25.40 | *** join/#arm-netbook wingrime (~alex@78.109.115.103) |
| 12:35.18 | slapin_ | hno: ping |
| 12:51.16 | mnemoc | meh, I replace nanda's u-boot and now boot1 halts initializing the usb... WTF |
| 13:03.04 | *** join/#arm-netbook rsalveti (~rsalveti@linaro/rsalveti) |
| 13:04.39 | slapin_ | by the way wtf is boot0 and boot1 and why I just boot u-boot via spl and being fine? |
| 13:05.09 | specing | slapin_: Who boots u-boot? |
| 13:06.30 | mnemoc | boot0/boot1 is allwinner's SPL, used when booting from nand |
| 13:08.59 | libv | lcd driver is no more locally, but i will only push this out on monday |
| 13:09.06 | libv | err, tuesday |
| 13:09.20 | libv | hdmi will meet a similar fate |
| 13:10.12 | mnemoc | :) |
| 13:10.22 | andoma | \o/ |
| 13:11.59 | mnemoc | f* boot1! |
| 13:12.17 | andoma | when on the LCD topic.. how does display refresh work? |
| 13:12.27 | andoma | is it normal fixed refresh |
| 13:12.34 | andoma | or can it be strobed "at will" ? |
| 13:13.34 | libv | ? |
| 13:13.44 | libv | andoma: that does not parse for me |
| 13:14.18 | libv | please rephrase |
| 13:14.32 | andoma | yeah |
| 13:14.33 | andoma | hm |
| 13:15.20 | andoma | does the LCD require the display to refreshed at fixed intervals, just like a normal TV |
| 13:15.34 | andoma | or can you just update the display when you have something new to show |
| 13:15.45 | libv | it is usually a rather stupid piece of solicon which needs to be driven directly at specific timing, yes |
| 13:15.52 | andoma | ah ok |
| 13:16.13 | libv | since this is cheap chinese hw, i am sure that it is just that |
| 13:16.22 | andoma | yeah ok |
| 13:16.41 | libv | on the nokia n9 we had a very intelligent lcd module, which retained its image and only needed to be refreshed as needed, and even had a low power mode |
| 13:16.47 | andoma | i remember when poking around in the omap display driver that there were some kind of display strobe mode |
| 13:17.26 | libv | which is what is used to display the time on the n9 constantly |
| 13:18.54 | andoma | ah ok |
| 13:19.24 | andoma | i thought this technique standard for all LCD displays, but apparently not |
| 13:19.39 | libv | with the price of our soc, i doubt we will see intelligent panels like that attached to it |
| 13:19.49 | andoma | right |
| 13:29.06 | slapin_ | libv: this is more tribute to module (which is external framebuffer with own VRAM) than to omap cpu (re n9). |
| 13:29.33 | slapin_ | regarding n9, have anybody managed to flash custom built kernel to it? |
| 13:31.07 | Turl | slapin_: I think there's even an android port for it? I assume running custom built kernels is not an issue |
| 13:38.13 | *** part/#arm-netbook hipboi (~hipboi@27.23.139.229) |
| 13:42.17 | *** join/#arm-netbook RITRedbeard (~redbeard@c-68-37-165-37.hsd1.nj.comcast.net) |
| 13:46.54 | *** join/#arm-netbook drachensun (~drachensu@142.196.83.182) |
| 13:47.26 | *** join/#arm-netbook drachensun (~drachensu@142.196.83.182) |
| 13:56.14 | drachensun | has any one else tried to run the plasma-active from the linaro repository? |
| 13:56.51 | drachensun | I've got some quirks but I'm not familiar with plasma and how its supposed to work |
| 14:00.10 | rz2k | I've tried |
| 14:00.25 | rz2k | it doesnt even open UMP session (no EGL initialization) |
| 14:05.19 | *** join/#arm-netbook t0dbld1 (~t0dbld@50.33.195.166) |
| 14:08.48 | *** part/#arm-netbook Guest67144 (~t0dbld@50.33.195.166) |
| 14:49.42 | rz2k | any sata users with rev.c A10? |
| 14:49.59 | rz2k | seems like lowering pll6 brings back the problems :/ |
| 14:55.03 | mnemoc | f* |
| 14:55.10 | mnemoc | cries |
| 14:55.29 | rz2k | yeah |
| 14:57.37 | mnemoc | i hate this thing |
| 14:57.54 | slapin_ | rz2k: what is so bad about pll6? |
| 14:59.19 | mnemoc | we keep killing either mali or sata with that f* pile of cra^Wcode |
| 14:59.56 | mnemoc | !!!! |
| 14:59.59 | mnemoc | hate |
| 15:04.24 | Marex | uaaa |
| 15:04.30 | Marex | murder death kill |
| 15:05.00 | stefanro | now these are words for Marex to jump in... ;) |
| 15:06.59 | Marex | :D |
| 15:09.09 | Marex | stefanro: btw do you happen to have any idea where dzu might have gone ? |
| 15:15.27 | mnemoc | Turl: can you test reverting your PLL6 mali fix and changing the mali_div to 4? |
| 15:26.26 | *** join/#arm-netbook bsdfox (~Bob@unaffiliated/bsdfox) |
| 15:26.53 | mnemoc | http://dl.linux-sunxi.org/amery/sunxi-3.0/2012-11-02_15%3a26%3a10/ has the pll6 mali fix reverted |
| 15:26.59 | mnemoc | cries |
| 15:29.11 | arete74 | QHTr[MJ]14101110328438276700000326191161304461758 |
| 15:29.11 | arete74 | Qhve[MJ]242118881429461000005441549223610528 |
| 15:29.11 | arete74 | QHSol[MJ]903129021621266000008269867988231 |
| 15:29.11 | arete74 | QH[MJ]14450105856640158100000240185611331357531 |
| 15:29.13 | arete74 | Q'h401129271811421000006572366369615929 |
| 15:29.16 | arete74 | QH,D,Po |
| 15:29.18 | arete74 | QH,D,PoPri162147162780000089157162957 |
| 15:29.21 | mnemoc | ? |
| 15:29.21 | arete74 | QWGnOut2462222462382462382462462382462382462893 |
| 15:29.28 | arete74 | sure, bad paste |
| 15:29.46 | slapin_ | Marex: hi, btw |
| 15:30.25 | slapin_ | gone for vodka & NAND digging |
| 15:30.41 | Marex | slapin_: hi |
| 15:31.39 | slapin_ | Marex: samsung, samsung, samsung |
| 15:31.44 | slapin_ | gone |
| 15:36.22 | *** join/#arm-netbook RITRedbeard (~redbeard@c-68-37-165-37.hsd1.nj.comcast.net) |
| 15:42.13 | Turl | mnemoc: regardless of mali, libv was getting full mele lockups with the high pll6 |
| 15:42.51 | mnemoc | Turl: and with mali_div 4? |
| 15:42.58 | Turl | can't they move sata to one of the other clocks? |
| 15:43.19 | mnemoc | oliv3r is the clocks expert |
| 15:43.33 | Turl | mnemoc: mali_div 4 + big pll6 makes it a 300Mhz clock, loses performance :< |
| 15:44.31 | mnemoc | if that is the price for having both mali and sata support, c'est la vie |
| 15:46.49 | Turl | mnemoc: I'm more worried of libv's crashes |
| 15:47.04 | Turl | my mali stalls rendered the screen unusable but system operated normally |
| 15:47.18 | Turl | on the other hand libv's mele stalled completely |
| 15:47.37 | mnemoc | increasing mali_div might fix those |
| 15:47.40 | Turl | ("normally" as linux didn't crash, usb worked) |
| 15:48.41 | mnemoc | "works and doesn't crash" has far higher priority than performance |
| 15:52.25 | traeak | mnemoc: reliability tends to trump performance |
| 15:52.34 | traeak | fast and crap doesn't deliver very well |
| 15:52.41 | traeak | fast and crap or fast and crash |
| 15:53.53 | Turl | anyone who can reliably reproduce the sata issue here? |
| 15:55.43 | mnemoc | sata, me |
| 15:57.55 | Turl | mnemoc: can you test something now? |
| 15:58.23 | mnemoc | yes |
| 16:01.26 | mnemoc | hno: dropped u-boot.bin from the new lichee-dev branch on my cubie's nand and doesn't even greet :< |
| 16:10.25 | bsdfox | mnemoc, my sata issues have been gone for a couple weeks |
| 16:10.36 | bsdfox | a2000 v1.7 |
| 16:12.13 | mnemoc | bsdfox: they probably returned today |
| 16:12.24 | mnemoc | bsdfox: after I merged the stage branch |
| 16:12.47 | bsdfox | thanks? |
| 16:12.48 | mnemoc | naively thought people was testing it.... |
| 16:13.06 | mnemoc | a fix for mali |
| 16:13.12 | mnemoc | on the same pll |
| 16:13.24 | bsdfox | Linux linaro-ubuntu-desktop 3.0.42+ #4 PREEMPT Mon Oct 22 17:32:10 PDT 2012 armv7l armv7l armv7l GNU/Linux |
| 16:13.24 | bsdfox | h36sa@linaro-ubuntu-desktop:~$ uptime |
| 16:13.24 | bsdfox | <PROTECTED> |
| 16:13.48 | bsdfox | my system load is always around there with nothing running.. but otherwise things function |
| 16:13.52 | bsdfox | and that's without X |
| 16:13.52 | mnemoc | i very rarely connect the sata.... will have to do it more often |
| 16:14.06 | bsdfox | I'm running my OS off sata |
| 16:14.12 | bsdfox | performance is much better than sd card |
| 16:16.03 | mnemoc | that's why we need a fix asap |
| 16:17.16 | techn | mnemoc: can we detect somehow that it's that specific chip with sata problem? |
| 16:17.34 | Turl | mnemoc: http://sprunge.us/DHUf |
| 16:17.36 | mnemoc | yes, it's revC |
| 16:17.39 | Turl | try that please |
| 16:18.05 | techn | mnemoc: and revC chips has maliclock already at 4? |
| 16:18.18 | techn | *mali div |
| 16:18.30 | mnemoc | most .fex I have use div 3 |
| 16:18.46 | mnemoc | but in general they don't use sata |
| 16:19.26 | techn | has those revC stock .fex have div 4 for? :/ |
| 16:22.23 | techn | bsdfox: what's your mali clock? you can check that from dmesg |
| 16:23.15 | bsdfox | https://gist.github.com/4002413 |
| 16:23.18 | bsdfox | looks like 4 |
| 16:24.01 | techn | mnemoc: we need to make separate hwpack for revC :/ |
| 16:24.26 | mnemoc | we need to fix the kernel |
| 16:24.41 | bsdfox | am I revC? |
| 16:25.08 | mnemoc | bsdfox: open the device and see the first letter of the serial number of the A10 chip |
| 16:25.28 | mnemoc | in 3.4 I printk() it |
| 16:25.29 | bsdfox | yup C |
| 16:25.33 | mnemoc | need to add the same to 3.0 |
| 16:25.51 | bsdfox | c4045ca |
| 16:26.08 | mnemoc | yes, A10 revC |
| 16:26.24 | bsdfox | is my system load due to improper clocking? |
| 16:26.31 | mnemoc | no |
| 16:26.42 | mnemoc | by `sleep` abuse in the kernel drivers |
| 16:26.58 | techn | mnemoc: Could it be done so that for revC we increase sata div? |
| 16:27.10 | bsdfox | <PROTECTED> |
| 16:27.16 | bsdfox | <PROTECTED> |
| 16:27.22 | mnemoc | techn: yes, but sounds evil |
| 16:27.28 | bsdfox | love the descriptions in /proc/interrupts |
| 16:27.34 | techn | since reducing mali clock could be showstopper for xbmc |
| 16:27.43 | mnemoc | techn: we can otoh change the .fex we distribute in the hwpacks |
| 16:28.31 | mnemoc | oh, you said sata_div |
| 16:32.23 | bsdfox | I wonder if the power light can be controller by /sys/devices/platform/sun4i-i2c.0/i2c-0/0-0034/axp20-gpio.29/ |
| 16:42.11 | Turl | mnemoc: can you try http://sprunge.us/EKfL and see if you get sata? |
| 16:42.39 | mnemoc | Turl: will reboot to it now |
| 16:45.30 | mnemoc | Turl: <6>ata1: SATA link down (SStatus 0 SControl 300) |
| 16:46.06 | Turl | mnemoc: notice that sprunge is a different one than the one I linked earlier |
| 16:46.24 | Turl | (either that or you were able to build, push and reboot in 20s :P) |
| 16:47.21 | mnemoc | Turl: ok... :| thought it was the same |
| 16:48.13 | mnemoc | Turl: where are you taking these numbers from? |
| 16:49.36 | Turl | first paste were a long shot, second one I took from older code revisions |
| 16:52.19 | mnemoc | i don't learn to not `reboot` :< |
| 16:54.24 | mnemoc | Turl: congrats :) |
| 16:54.39 | mnemoc | and mali still works? |
| 16:54.47 | Turl | idk :P I'm testing on mele :< |
| 16:55.23 | Turl | $ dmesg|grep A10 |
| 16:55.23 | Turl | [ 0.000000] chip-id: A10 (AW1623 revision B) |
| 16:55.28 | Turl | mnemoc: what revision do you have? |
| 16:55.49 | mnemoc | uh. i didn't know you were using 3.4 |
| 16:56.03 | mnemoc | [ 0.000000] chip-id: A10 (AW1623 revision C) |
| 16:56.42 | mnemoc | and mali is loaded.... dev 3 |
| 16:56.44 | mnemoc | div* |
| 16:56.51 | mnemoc | don't have a display at hand |
| 16:57.56 | mnemoc | hdparm -t made it puke :< |
| 16:58.29 | Turl | mnemoc: how many Mhz on mali? check dmesg |
| 16:58.36 | mnemoc | 320 |
| 16:58.58 | mnemoc | but sata is not usable |
| 16:59.10 | mnemoc | [ 20.230000] Mali: mali clock set completed, clock is 200000000 Mhz |
| 16:59.22 | mnemoc | 200MHz o.o |
| 16:59.47 | Turl | clocks are so much fun, ain't them? :) |
| 17:00.06 | Turl | I cannot test hdparm because my HD is full of bad sectors :( |
| 17:00.08 | mnemoc | not enjoying this |
| 17:02.36 | Turl | [ 1395.930000] mali: use config clk_div 3 |
| 17:02.37 | Turl | [ 1395.950000] mali: clk_div 3 |
| 17:02.37 | Turl | [ 1395.950000] Mali: mali clock set completed, clock is 320000000 Mhz |
| 17:02.38 | Turl | on my mele |
| 17:02.52 | bsdfox | my mali at 300MHz works with sata that's reasonably fast 419430400 bytes (419 MB) copied, 12.8662 s, 32.6 MB/s |
| 17:03.24 | *** join/#arm-netbook t0dbld|work (~t0dbld@unaffiliated/t0dbld) |
| 17:04.08 | mnemoc | bsdfox: unless you built your kernel today, you aren't facing the problem we are trying to fix :) |
| 17:04.16 | Turl | try this one mnemoc http://sprunge.us/idCh |
| 17:04.33 | mnemoc | but you might have the problem on mali Turl tried to fix when killing sata |
| 17:05.06 | bsdfox | which problem is that? |
| 17:05.16 | mnemoc | Turl: seriously? |
| 17:05.21 | bsdfox | my performance in X is awful. not great in console either |
| 17:06.20 | Turl | mnemoc: yeah |
| 17:06.48 | Turl | mnemoc: it doesn't make sense otherwise that they'd check for model *and* have an ifdef option to toggle it |
| 17:06.52 | rz2k | bsdfox: what drivers do you use? |
| 17:07.21 | rz2k | fbdev should be ok, mali freezes with pll6 being too high only after some use by 3d apps |
| 17:08.03 | bsdfox | I've got mali |
| 17:08.12 | mnemoc | Turl: div 4 sounds far more reasonable imo |
| 17:08.29 | Turl | mnemoc: but the whole mele stalled for libv, not just mali |
| 17:08.40 | mnemoc | uhm... |
| 17:08.55 | Turl | and lowering it stopped the stalls |
| 17:08.56 | mnemoc | but mali can be causing it |
| 17:09.11 | Turl | or wemac :< he connects to ssh to trigger it |
| 17:09.19 | rz2k | Turl: as I said in ML, my mele a2000 with default script bin already had div=4 |
| 17:09.30 | rz2k | rev.C a10 |
| 17:09.36 | bsdfox | rz2k, same here |
| 17:11.35 | bsdfox | rz2k, what's your system load idling in console? |
| 17:12.02 | rz2k | [ 6.650000] mali: clk_div 4 [ 6.650000] Mali: mali clock set completed, clock is 300000000 Mhz |
| 17:12.12 | rz2k | thats default kernel on mele a2000 |
| 17:12.20 | rz2k | android from nand |
| 17:24.30 | mnemoc | Turl: Mali: mali clock set completed, clock is 320000000 Mhz |
| 17:24.51 | Turl | mnemoc: what about sata? |
| 17:24.51 | mnemoc | Turl: hdparm didn't crash |
| 17:25.07 | mnemoc | performance is a shame |
| 17:26.08 | rm | finally put together a debian server image, http://romanrm.ru/en/a10/debian |
| 17:26.19 | mnemoc | Turl: increased cpufreq, and crash :< |
| 17:27.39 | Turl | mnemoc: what crashed? |
| 17:27.43 | mnemoc | http://dpaste.com/823082/ |
| 17:27.45 | mnemoc | sata |
| 17:28.13 | mnemoc | used to get 120MB/s... and no crash obviusly |
| 17:28.26 | mnemoc | that's with the cpu at 1GHz |
| 17:28.52 | bsdfox | wow that's very fast for a 2.5" drive |
| 17:29.09 | bsdfox | not sure I've ever seen those speeds on a spinner |
| 17:29.11 | mnemoc | external 2x3.5" over esata |
| 17:29.29 | bsdfox | oh :) |
| 17:29.49 | rz2k | wait, Matson Hall pushed something to kernel for cubieboard some time ago about sata, or, with default kernel cubieboard has dead mali/sata? |
| 17:29.59 | rz2k | that would be surprising |
| 17:30.13 | mnemoc | rz2k: all this started from his patch |
| 17:30.49 | *** join/#arm-netbook pawel5870 (~pkarpins@ip-91-205-72-3.iwacom.net.pl) |
| 17:31.58 | rz2k | :/ |
| 17:32.17 | rz2k | so how nobody checked crashes after EGL demos and sata? |
| 17:32.34 | rz2k | we need hipboi I guess. |
| 17:32.59 | mnemoc | i rarely connect sata to my test devices.... it's the driver I use for $work$ |
| 17:33.03 | mnemoc | drive* |
| 17:33.19 | rz2k | I want to use my old 2.5" 120gb as rootfs |
| 17:33.20 | mnemoc | had the naive hope that people was actually using the stage kernel |
| 17:33.48 | rz2k | I've tested disp and mali :D |
| 17:33.54 | mnemoc | :) |
| 17:35.58 | rz2k | also techn's commit fixed color spaces finally |
| 17:36.05 | rz2k | I have normal htop colors |
| 17:36.11 | rz2k | happy |
| 17:36.39 | techn | rz2k: what kind problem you had? |
| 17:36.41 | mnemoc | rz2k: not fixed, he just forced RGB |
| 17:37.08 | rz2k | green was blue and other weird stuff |
| 17:37.17 | rz2k | probably yuv/rgb mismatch |
| 17:37.41 | techn | and only visible with htop? |
| 17:37.57 | rz2k | no, everywhere |
| 17:38.10 | rz2k | but everything was fine in X |
| 17:38.12 | Turl | mnemoc: this one? http://sprunge.us/EicP |
| 17:38.22 | techn | rz2k: that's weird :/ |
| 17:38.23 | rz2k | only fbcon had wrong colors |
| 17:38.24 | rz2k | yeah |
| 17:38.31 | rz2k | also I use VGA |
| 17:38.36 | techn | lol |
| 17:38.43 | Turl | mnemoc: (basically revert my fix/breakage + disable pll6&4 replacement) |
| 17:39.08 | techn | That change didn't effect to VGA.. are you sure it was that fix? :p |
| 17:40.13 | rz2k | I built kernel twice this week, so first build had the bug and second (now) doesnt have :p |
| 17:40.25 | rz2k | might be libv's stuff changed something |
| 17:40.44 | techn | I can see only userspace problem.. or maybe .fex file change :/ |
| 17:41.13 | mnemoc | Turl: and libv's inestability? |
| 17:41.28 | techn | there is some colorspace settings in .fex... but anyway great that it was fixed :) |
| 17:41.29 | rz2k | I didnt change .fex :) |
| 17:41.37 | rz2k | yeah |
| 17:41.57 | Turl | mnemoc: I bet libv has a revC |
| 17:43.24 | Turl | the instability is because on revC pll4 is == pll6 with the #if as it was, and many stuff doesn't like the 25% faster clock :< |
| 17:43.39 | mnemoc | 10MB/s @ 60MHz |
| 17:44.10 | mnemoc | 116MB/s @ 1GHz |
| 17:44.12 | mnemoc | no crash |
| 17:44.25 | Turl | mnemoc: with /EicP sprunge right? |
| 17:44.45 | Turl | can you test mali and see what clock does it get? |
| 17:45.04 | mnemoc | [ 14.130000] mali: clk_div 3 |
| 17:45.04 | mnemoc | [ 14.130000] Mali: mali clock set completed, clock is 320000000 Mhz |
| 17:45.13 | techn | whee :) |
| 17:45.44 | mnemoc | rz2k: can you test that too? |
| 17:45.59 | mnemoc | to see how it affects mali in real life |
| 17:47.02 | mnemoc | libv: around? |
| 17:48.04 | mnemoc | Turl: so I revert your fix and then you send the other patch as new fix? |
| 17:49.21 | Turl | let me try on zatab first |
| 17:51.27 | *** join/#arm-netbook cheng (~cheng@56.48.48.60.kmr03-home.tm.net.my) |
| 17:56.37 | techn | rz2k: thanks for mali stuff :) |
| 17:58.40 | Turl | mnemoc: well zatab boots, mali is clocked 320 like before all this sata mess :P |
| 17:59.03 | mnemoc | and your mele is happy too? |
| 17:59.29 | Turl | apparently, yeah |
| 17:59.31 | Turl | it's rev B |
| 17:59.36 | techn | Turl: how about those .fex files which has mali div as 4 |
| 18:01.33 | techn | nevermind.. they can change that to 3 :p |
| 18:01.44 | Turl | techn: no idea :< |
| 18:02.33 | Turl | mnemoc: two passes of antutu 2D+3D and it hasn't crashed yet, it used to crash nearing the end of the first one |
| 18:03.21 | techn | hmm.. now I remember something.. Tom said that cedar also had some stability problems with some clocks :/ |
| 18:04.00 | Turl | yeah cedar uses PLL4 |
| 18:04.25 | techn | "there is a bug in some of the A10 chip, only in the C version, that |
| 18:04.25 | techn | the PLL4 is not stable, can cause the system die when playing video." |
| 18:05.51 | Turl | well I'm playing video and it works fine so far |
| 18:06.09 | techn | Turl: are you using cedar? |
| 18:06.18 | techn | and revC? |
| 18:06.42 | Turl | yeah, cedar |
| 18:06.57 | Turl | I suppose this one is revC, 3.0 doesn't print it to be sure |
| 18:09.18 | mnemoc | will "fix" that now |
| 18:10.21 | Turl | if libv's system doesn't crash with this, I guess it's our best shot until we can figure out how not to use PLL4 on revC |
| 18:10.57 | Turl | after all it's 'the possiblity of' cedar crashing |
| 18:11.20 | Turl | 6 minute video playback and no crash yet |
| 18:12.55 | jlj | there is cedar support on linux? or is this for anderoid? |
| 18:13.27 | techn | jlj: yes there is.. check XBMC project |
| 18:13.45 | *** join/#arm-netbook cheng (~cheng@118.101.201.208) |
| 18:14.08 | jlj | cool. is the driver open source or a blob? |
| 18:14.44 | techn | blob |
| 18:15.41 | jlj | oh okay. not as cool but still neat. |
| 18:16.52 | mnemoc | Turl: can you try http://sprunge.us/AdFT ? |
| 18:17.57 | mnemoc | Turl: http://sprunge.us/ZTUN .... had a typo |
| 18:18.51 | Turl | .. and patch -R saved the day :) |
| 18:18.56 | Turl | :P |
| 18:23.49 | Turl | mnemoc: <6>[ 0.000000] chip-id: A10 (AW1623 revision C) |
| 18:23.51 | Turl | indeed |
| 18:24.38 | techn | mnemoc: what is needed to fix /proc/cpuinfo ? :) |
| 18:25.15 | Turl | is it broken? o.O |
| 18:25.34 | techn | Turl: it is not outputting all values |
| 18:25.48 | mnemoc | techn: so you want cpu info in /proc/cpuinfo???? how random.... |
| 18:26.32 | techn | mnemoc: also mhz etc.. but that's not priority |
| 18:26.41 | mnemoc | techn: file a ticket :) |
| 18:27.16 | Turl | none of my arm devices print "mhz etc" |
| 18:27.54 | mnemoc | but adding the cpu-id sounds a fair request |
| 18:30.01 | techn | mnemoc: naah.. other things are more important things.. :p |
| 18:30.16 | techn | but fyi.. my router has following values |
| 18:30.18 | techn | system type : Broadcom BCM47XX |
| 18:30.18 | techn | processor : 0 |
| 18:30.18 | techn | cpu model : MIPS 74Kc V4.0 |
| 18:30.33 | mnemoc | a ticket as reminder doesn't harm |
| 18:34.56 | rz2k | <mnemoc> rz2k: can you test that too? - only tomorrow evening :( |
| 18:35.07 | rz2k | sorry |
| 18:35.42 | mnemoc | rz2k: np |
| 18:38.16 | Turl | mnemoc: do you want me to send patch to ML, or can you commit it directly? |
| 18:38.44 | rz2k | <techn> rz2k: thanks for mali stuff :) - we should now intergrate it into BSP, any suggestions? |
| 18:39.21 | mnemoc | Turl: paste me the commit (not the revert) with decent text and I merge it directly |
| 18:39.45 | techn | rz2k: I have been thinking about that.. solutions: crosscompilation, binary or deb.. none of them are really good solutions |
| 18:40.06 | mnemoc | packages.linux-sunxi.org is waiting... |
| 18:40.20 | techn | mnemoc: CPU revision : 2, is this same as cpu-id? |
| 18:40.28 | techn | *revision |
| 18:40.44 | mnemoc | might mean C |
| 18:41.00 | mnemoc | 'A'+2 |
| 18:44.17 | techn | rz2k: best solution would be to get drivers to distro upkeepers.. but linaro wont be option since we dont have any money? :/ |
| 18:44.51 | techn | or allwinner hasn't paid for them :p |
| 18:46.29 | techn | but maybe we should move fully top of ALARM.. as we have maitainer here? :) |
| 18:46.35 | techn | *maintainer |
| 18:47.08 | *** join/#arm-netbook benjamin__ (~benjamin@c-94-255-220-30.cust.bredband2.com) |
| 18:52.31 | techn | or could we get those directly to debian.. or how these things works? |
| 18:53.21 | techn | hmm.. or could we create repository which delivers those binaries :/ |
| 18:53.33 | techn | launchpad? |
| 18:54.16 | mnemoc | luanchpad doesn't compile arm |
| 18:56.11 | mnemoc | my idea was to host .deb/.rpm/.opkg ... at http://packages.linux-sunxi.org |
| 18:56.23 | Turl | mnemoc: http://sprunge.us/OEif |
| 18:57.10 | mnemoc | Turl: thank you |
| 18:57.43 | Turl | yw |
| 19:03.41 | techn | rz2k: .deb's to http://packages.linux-sunxi.org |
| 19:04.06 | Turl | mnemoc: I updated the realtek driver from their website btw https://github.com/allwinner-dev-team/linux-allwinner/commits/allwinner-3.0 |
| 19:04.06 | techn | mnemoc: could that url be modified so that is apt-getable |
| 19:04.16 | Turl | seems to work ok |
| 19:04.28 | mnemoc | techn: of course. it's empty currently |
| 19:04.52 | techn | Turl: is that breaking connecting in power save? |
| 19:05.33 | Turl | techn: idk, give it a try :) |
| 19:06.06 | techn | Turl: ok :) |
| 19:06.48 | Turl | https://github.com/allwinner-dev-team/linux-allwinner/compare/allwinner-3.0~3...allwinner-3.0 if you want to suffer and look at its diff :P |
| 19:07.15 | mnemoc | techn: the idea is to make it apt/yum/etc friendly |
| 19:07.47 | mnemoc | so users don't need to care |
| 19:08.12 | techn | mnemoc: how about kernel releases via that ? :D |
| 19:08.20 | mnemoc | yes, also |
| 19:08.27 | mnemoc | we only need package maintainers :p |
| 19:08.39 | mnemoc | i have no idea how to make good .deb/.rpm/etc files |
| 19:08.52 | techn | mnemoc: same |
| 19:08.57 | mnemoc | if someone gives me scripts, I can do the ape job too |
| 19:09.16 | Turl | mnemoc: I heard you can make deb-pkg or so on the kernel tree |
| 19:09.49 | jelly-home | yeah, it generates decent .debs on i386 or amd64 |
| 19:10.04 | mnemoc | :) |
| 19:10.20 | jelly-home | can't vouch for crosscompiles or anything else tho |
| 19:11.16 | techn | Damn.. what a waste.. I have now hobbyed with linux for 4months.. It's been 10 years since last time.. :( |
| 19:11.32 | jlj | this makes packages http://asic-linux.com.mx/~izto/checkinstall/ |
| 19:11.42 | jlj | maybe not good ones..but they work usually |
| 19:11.54 | techn | What would I know if I had worked with linux that time.. instead I spend that time with symbian :D |
| 19:12.23 | Turl | checkinstall is a ugly package generator by definition |
| 19:13.29 | drachensun | well I was actually starting a project to do all this myself |
| 19:13.48 | drachensun | glad others are interested in having a repository |
| 19:14.09 | techn | mnemoc: maintainer => drachensun :) |
| 19:14.38 | drachensun | can do, I was actually going to do it anyway |
| 19:15.12 | drachensun | its gonna be a little bit of a slow start though, I'm pretty distracted with my big project to fund it |
| 19:15.13 | mnemoc | drachensun: ssh public key? :) |
| 19:15.17 | drachensun | which just went online |
| 19:15.39 | drachensun | www.indiegogo.com/pengpod |
| 19:16.33 | mnemoc | drachensun: 800x480? come on :( |
| 19:16.50 | drachensun | wait, for the 7"? |
| 19:17.09 | drachensun | thats how they come, at least from my current supplier |
| 19:17.15 | drachensun | it looks pretty good |
| 19:17.19 | jelly-home | 1024x768 is the humane minimum! |
| 19:17.25 | drachensun | lol |
| 19:17.49 | mnemoc | and the pengbook doesn't have sata? |
| 19:17.54 | rm | AW COME ON |
| 19:18.05 | rm | they didn't even bother to paint some other pic than just the MK802 |
| 19:18.10 | drachensun | I haven't managed to find an A10 net book with sata |
| 19:18.23 | rm | or is that you |
| 19:18.27 | drachensun | its me |
| 19:18.44 | drachensun | and yes I didn't bother to manipulate the photos |
| 19:18.55 | techn | btw. cubieboard has raised 47k now :D |
| 19:19.00 | drachensun | well beyond getting rid of the glare |
| 19:19.38 | rm | so as you can tell I am not sure what's the innovation deserving to be funded here |
| 19:20.07 | rm | basically you plan to resell A10 tablets and sticks, with some software on SD card |
| 19:20.11 | jlj | how about an indiegogo to get the cedar reverse engineered lol |
| 19:20.23 | drachensun | basically fund the cost to work on this full time and finish up the tablet features that the android has but dont work on the tablet |
| 19:20.59 | drachensun | I've noticed there are a lot of people on the forums who are intimidated setting up there own cards and compiling and all so I'm hoping I can make it all a little smoother |
| 19:21.21 | mnemoc | .oO |
| 19:22.24 | jlj | I'm pretty sure that tablet thing won't fly, so how about that deb repository? :) |
| 19:22.47 | drachensun | heh well if it doesn't I'm gonna be forced back to paying work with my time |
| 19:23.04 | jlj | aww |
| 19:23.07 | drachensun | but I'm going to be doing it anyway until I find out how this is going |
| 19:23.19 | drachensun | so I'll share it all as I do it |
| 19:23.46 | Turl | do your MK802 have AXP? :P |
| 19:24.48 | drachensun | The AXP209? yeah, I'm pretty sure |
| 19:25.00 | Turl | drachensun: you got a typo on the 10" keyboard case perk |
| 19:25.08 | mnemoc | drachensun: big news! we aren't paid for our sunxi related work either |
| 19:25.51 | Turl | why PengPod btw? your logo actually reads "PengPad" (and it makes more sense that way imo :P) |
| 19:26.22 | techn | mnemoc: we sould add donate button to wiki ;) |
| 19:26.28 | drachensun | Turl: I thought so too but I'm not the graphics guy ;) gotta roll with it sometimes |
| 19:26.47 | mnemoc | techn: nah |
| 19:27.10 | Turl | techn: adding donate buttons makes stuff messy |
| 19:27.22 | drachensun | mnemoc: I know and I still plan to contribute regardless |
| 19:27.28 | Turl | techn: and donators often feel entitled afterwards too |
| 19:27.30 | rz2k | techn: alarm could be an option but manuals for compiling/creating packages are missing |
| 19:28.03 | rz2k | so we will hardly depend on alarm maintainers, which is not good. |
| 19:28.10 | techn | Turl: yep.. that was joke :p |
| 19:28.13 | drachensun | mnemoc: I just dont think realistically I will have time to do everything as maintainer the way its growing |
| 19:28.17 | rm | drachensun, you plan to spend 15 minutes on packaging what others work for months not being paid |
| 19:28.22 | rm | then collect the money |
| 19:28.29 | rz2k | lol |
| 19:28.32 | rm | suppose if you obey the corresponding licenses nothing wrong with that |
| 19:28.35 | rz2k | rm: welcome to real life |
| 19:28.58 | rm | (legal-wise, but not morally) |
| 19:29.00 | rz2k | tons of such stuff on XDA/any tablet forum |
| 19:29.08 | Turl | rm will be watching closely to make sure you do ;) |
| 19:29.08 | techn | Atleast I'm doing this as hobby.. I want to learn how things are working (referring to 10 years wasted with symbian) |
| 19:29.13 | drachensun | well it takes a lot more than 15 minutes to make it all run smooth |
| 19:29.26 | mnemoc | ROTFL |
| 19:30.05 | RaYmAn | 85$ for PengStick? Seems a bit steep considering an identical device sells for around 50$ everywhere else. |
| 19:30.16 | rz2k | still has nokia 6630 as backup phone |
| 19:30.25 | RaYmAn | lol |
| 19:30.31 | mnemoc | it's to pay his engineering hours making sunxi's software work perfect |
| 19:30.44 | drachensun | shipping from china, shipping to customers, fees for the kickstarter, fallout rates, it adds up pretty quick |
| 19:31.17 | RaYmAn | the devices are just rebranded standard a10 devices, aren't they? |
| 19:31.26 | drachensun | pretty much |
| 19:32.19 | techn | rz2k: I think maybe WarheadsSE could do the maintaining.. he was one of ALARM maintainers? :/ |
| 19:32.26 | rz2k | yeah |
| 19:32.30 | RaYmAn | "As a bonus, all these devices already run a high quality, stable version of Android they can easily be made to run both systems." |
| 19:32.38 | RaYmAn | a10 devices and stable android? :P |
| 19:32.46 | WarheadsSE | ? |
| 19:33.12 | rz2k | WarheadsSE: we're discussing what to do with our BSP and what base we should choose |
| 19:33.31 | WarheadsSE | Well, I can tell you how to make packages |
| 19:33.35 | WarheadsSE | its damned easy. |
| 19:34.02 | Turl | RaYmAn: my CM port is stable :P |
| 19:34.03 | WarheadsSE | And you don't have to rely on us, our github is public |
| 19:34.45 | WarheadsSE | our packages are pretty much Arch PKGBUILDs with some minor additions for the build system & updated arches |
| 19:34.55 | RaYmAn | Turl: yeah, fair enough :) But that's proper open source |
| 19:35.09 | rz2k | we should try then |
| 19:35.15 | Turl | RaYmAn: yeah the chinese builds aren't too good :P |
| 19:35.29 | RaYmAn | hacky-hacky-hacky!:P |
| 19:35.31 | rz2k | goal is "one command"=one rootfs image ready for sdcard or nand (livesuit image) |
| 19:35.43 | WarheadsSE | rz2k: I was trying to get the kernel sunxi-3.0 to build, but I kept getting either missing dirs or git failures. |
| 19:36.17 | WarheadsSE | yeah, pretty straight forward in ALARM :P just have to know which to run lol |
| 19:37.53 | WarheadsSE | Also, pretty much the same as upstream, with minor tweaks again. devtools-alarm |
| 19:38.35 | techn | drachensun: If you can make allwinner to code publicly.. maybe then I could fund your project :) |
| 19:39.15 | techn | Atleast jinx got opportunity get sources if he ordered 1000 pcs of some device |
| 19:39.29 | rz2k | g2d drivers would be enough for me to buy mk802II for $80 |
| 19:40.53 | Turl | allwinner probably doesn't have g2d drivers :P |
| 19:40.54 | drachensun | techn: heh yeah, thats the dream, get them to open it all. Most of the chinese companies seem to be open to concessions at 10k but I dont think I going to be near the volume |
| 19:40.56 | Turl | android doesn't use it |
| 19:41.02 | WarheadsSE | anyways, you guys know where I am when you need me. just ask. It's not hard. |
| 19:41.41 | drachensun | techn: one of my suppliers was talking that way about the rockchip, which almost had me take this project that way but I'm pretty sure he was pulling a bait and switch on me |
| 19:44.33 | drachensun | techn: "just wire the 30% down and we can make available" ugh |
| 19:51.16 | drachensun | turl: I looked, they don't have the AXP 209 but now that I am staring it them again and remember how big the IC is I should have known right off |
| 19:55.26 | mnemoc | CB at 97%... |
| 19:55.49 | Turl | just 1500$ to go? :) |
| 19:56.56 | mnemoc | drachensun: what about a crowdfunding selling mtk6577 phones with sources? :) |
| 19:57.34 | drachensun | mnemoc: Oh man, I would love to. I' |
| 19:58.02 | drachensun | I've been bringing in 6575 & 6577s and they are awesome for the price |
| 20:00.25 | techn | mnemoc: when did cubieboard campain start? |
| 20:00.29 | techn | 10 days ago? |
| 20:01.19 | mnemoc | it's a 45d campaign |
| 20:01.59 | techn | 36 days left.. maybe there could be 200k at the end of campaign :p |
| 20:02.20 | mnemoc | drachensun: find the factory of a good 720p mtk6577 quad band device and negociate the realase of the code |
| 20:02.29 | mnemoc | drachensun: that will bring you the funding for sure |
| 20:02.33 | Turl | did they get the paypal issue sorted mnemoc? |
| 20:02.48 | mnemoc | drachensun: rebranding A10 devices won't get you far |
| 20:03.08 | mnemoc | techn: not yet. but the ball is on their side |
| 20:03.09 | drachensun | mnemoc: I'll start the conversation |
| 20:03.35 | mnemoc | drachensun: excellent :) |
| 20:03.48 | drachensun | mnemoc: I was kinda inspired by the vivaldi, there he was rebranding the ZT C71 and got a ton of attention but just seemed to drop the ball |
| 20:04.58 | mnemoc | drachensun: but picked the wrong platform :) .... with the A10 we already have all you can get, and more |
| 20:05.16 | mnemoc | drachensun: we have more sources than what the factories get |
| 20:06.11 | mnemoc | drachensun: and none of these can be worked by a single person. you *need* a community and volunteers |
| 20:07.29 | drachensun | mnemoc: for sure |
| 20:07.57 | RITRedbeard_ | what is the right platform? |
| 20:08.09 | drachensun | <PROTECTED> |
| 20:08.15 | mnemoc | one for which there are no public sources yer |
| 20:08.17 | mnemoc | yet* |
| 20:08.44 | mnemoc | that will catch attention, and the community will be created around |
| 20:08.55 | RITRedbeard_ | I think Cubieboard is a good dev board idea pulling out SATA, AC97, all that stuff |
| 20:09.00 | *** join/#arm-netbook ZaEarl (~malmrose@62.host-182-48-149.compassnet.co.nz) |
| 20:09.25 | mnemoc | RITRedbeard_: yes, it's an awesome device |
| 20:09.32 | Turl | AC97? :O |
| 20:09.43 | drachensun | yeah, the sata is such a great feature, I wish more of the devices would start using it. |
| 20:09.45 | RITRedbeard_ | even if A10 dies it doesn't mean the mali work has to die along with it, as far as I understand it puts up excellent fight against tegra and snapdragon |
| 20:10.11 | RITRedbeard_ | I wish I could draw out the SATA on my MK802, begging for 30GB SSD |
| 20:10.54 | mnemoc | drachensun: you can talk with ZaEarl to learn about the business of selling devices with code |
| 20:11.36 | ZaEarl | mnemoc, more businesses need to do that. |
| 20:12.00 | drachensun | Zaearl: have you gotten other suppliers to open up? |
| 20:12.34 | ZaEarl | No. They change one line of code out of millions and think that they own the whole thing. |
| 20:12.52 | ZaEarl | Own, as in the GPL no longer applies. |
| 20:13.09 | drachensun | yup |
| 20:13.27 | Turl | ZaEarl: any new goodix drivers? :) |
| 20:13.52 | Turl | the last one had an interesting @goodix email address, I was thinking about emailing it |
| 20:14.02 | ZaEarl | Turl, I'll ask. |
| 20:14.52 | mnemoc | rm: I just pushed another sata fix you might want for your server image |
| 20:17.01 | ZaEarl | The funny thing, all these component vendors, Goodix, Bosch, etc. are GPLing their drivers, but they don't seem to care that everyone fails to comply with the GPL when distributing the binaries. |
| 20:17.39 | *** join/#arm-netbook t0dbld1 (~t0dbld@50.33.195.166) |
| 20:17.51 | mnemoc | and what's their reason to not make those drivers publicly downloadable? |
| 20:18.00 | mnemoc | as they do consider them GPL |
| 20:18.29 | ZaEarl | they haven't said |
| 20:18.53 | Turl | bosch is a pretty big, international company, I'd be surprised if they didn't GPL stuff properly |
| 20:19.21 | Turl | mnemoc: it's probably just that the GPL doesn't require you to make it publicly downloadable |
| 20:19.39 | WarheadsSE | depends on the choice of gpl |
| 20:19.50 | mnemoc | Turl: sure it doesn't, but if you mail them they don't reply :< |
| 20:19.53 | WarheadsSE | and yeah, a lot of re-users suck. |
| 20:20.35 | Turl | mnemoc: yeah, that's the illegal bit |
| 20:20.56 | Turl | wonder what'd happen if you mailed them a physical letter with a legal threat :P |
| 20:21.19 | ZaEarl | Turl, bosch? they're not in uncompliance |
| 20:21.51 | Turl | ZaEarl: I never said they were, in fact I'd be surprised if they were |
| 20:22.03 | Turl | they're a big company |
| 20:22.34 | bfree | just fyi "make deb-pkg" has some drawbacks :-/ |
| 20:22.36 | bfree | 1) it makes it hard to keep configs in sync for multiple architectures/flavours |
| 20:22.43 | bfree | 2) it doesn't give you a source package (e.g. for apt-get source) so you would need to tar up the used kernel sources to dump alongside the debs just for license compliance, which is a bit alien for Debian |
| 20:22.48 | bfree | 3) no patch management so you need to pre-apply them all, and so it's harder to rebase it to new kernels |
| 20:22.54 | bfree | 3.5) you can't build it in a proper/ clean buildd environment |
| 20:24.09 | Turl | 2 is a nonissue with the public git repo, isn't it? as long as you have the commit hash on the kernel version |
| 20:24.20 | Turl | same with 3, any patches will be merged on git |
| 20:24.50 | mnemoc | bfree: so, can you get me an script to make .deb kernels out of our tree? :) |
| 20:24.51 | bfree | not really, to comply with GPLv2 (the easy way) you need to offer the sources with the binary |
| 20:25.38 | mnemoc | bfree: I can then integrate it in the bsp and keep an apt repo |
| 20:25.50 | Turl | bfree: git checkout $(thingy you see on kernel version) not enough? :( |
| 20:26.19 | bfree | Turl: I don't think it is really but IANAL ;-) |
| 20:27.12 | Turl | you'd have to be a pretty nasty person to sue us while you have a public git repo tracking it all though :P |
| 20:28.27 | techn | bfree: how about binary only delivery (mali egl) ? |
| 20:28.48 | mnemoc | considering all big brands assume putting the sources online is good enough, i wouldn'y fear |
| 20:29.11 | mnemoc | that's not gpl |
| 20:29.14 | Turl | mnemoc: at times the sources don't even match the binary >.< |
| 20:29.27 | ZaEarl | A git repo would fulfill the requirement to distribute the sources. |
| 20:30.18 | drachensun | I think a git repo is fine, it just has to be available, I mean they considered mailing CDs upon request fine didn't they? |
| 20:30.59 | ZaEarl | CDs are one option, they are not a requirement |
| 20:31.49 | drachensun | I just mean as an example |
| 20:31.49 | bfree | mnemoc: I didn't want to volunteer myself :-p I'll try and find some time though (sooner rather then later) to see if I am able to adapt something as I'd like nice packages myself anyway ;-) |
| 20:32.31 | drachensun | I mean cost wise it would be nuts these days, but before the internet was so common lots of linux source cds were being mailed around |
| 20:32.39 | mnemoc | bfree: http://packages.linux-sunxi.org/ looks maintainer :p |
| 20:32.44 | bfree | I think CDs on request is one of the not nice ways, it means providing a written offer valid for 3 years to do it for anyone (for a sane cost to only match distribution costs) |
| 20:33.35 | ZaEarl | The CD offer is often used to delay the source distribution. |
| 20:33.44 | drachensun | oops, I managed to derail things |
| 20:34.07 | drachensun | I just meant as legal precedent something that bad was acceptable, so git must be |
| 20:34.23 | drachensun | but of course no one here would want something like that now :) |
| 20:34.24 | ZaEarl | definitely |
| 20:34.47 | bfree | techn: the binary only drivers ... no comment ;) I actually don't know enough about them to know how icky they might be |
| 20:35.34 | mnemoc | CB @ 98%... |
| 20:35.35 | techn | bfree: atleast ubuntu makes user to accept terms when using binary only stuff :/ |
| 20:35.50 | techn | (I never read them) |
| 20:36.32 | bfree | GPLv2 (i.e. the kernels) "If distribution of executable or object code is made by offering access to copy from a designated place, then offering equivalent access to copy the source code from the same place counts as distribution of the source code, even though third parties are not compelled to copy the source along with the object code." |
| 20:38.06 | Turl | upload kernel binaries to github downloads? :P |
| 20:38.39 | bfree | afaic that means pointing to a git repo tag is not "equivalent access to copy the source code from the same place" ... but you could just tar up the sources used for make deb-pkg and throw them in the same dir ... it's not really very nice or right (i.e. apt-get source doesn't work) but probably meets the letter and spirit of the license sufficiently |
| 20:39.16 | bfree | and yes, probably only a complete #!@*&% would cause trouble about it anyway as long as the true source of the kernel was available |
| 20:39.48 | Turl | yeah |
| 20:39.48 | *** join/#arm-netbook L84Supper (~ly@unaffiliated/l84supper) |
| 20:42.06 | Turl | bfree: I guess we could take the "written offer" path too and put a txt on the dir saying something like 'you can download the source here <link/>' |
| 20:43.08 | mnemoc | a gpl 2.5 would be nice :< |
| 20:44.24 | bfree | Turl: icky ... cause then you have to make sure to keep that valid for at least 3 years. imho not the sort of hassle a volunteer project wants. make it available via "equivalent access" and then you can yank it when the binary is gone and forget about it |
| 20:45.33 | drachensun | well that quote says "counts as distribution of the source code" that doesn't imply its the only method that counts, I think its just one way that qualifies |
| 20:46.13 | ZaEarl | I love the github method. They're virtually assured to be around for 3 years. |
| 20:46.18 | drachensun | true |
| 20:46.47 | bfree | sorry to dump that and run, but I'm late now already ;-) hoping to be at a Debain BSP tomorrow and then maybe I'll try and have a first stab at something on Sunday if I can motivate myself enough to dive into the pain without having any hardware to run it on yet ;-) |
| 20:48.12 | bfree | drachensun: http://www.gnu.org/licenses/old-licenses/gpl-2.0.html section 3 if you want to see the other ways and the full context of the quote ;-) |
| 20:48.29 | drachensun | cool thanks |
| 20:49.33 | drachensun | hmmm I see its either or |
| 20:49.50 | ZaEarl | If someone asks me to give them the source, I say sure, do "git clone ...". |
| 20:50.09 | ZaEarl | they then say, "Thanks!" |
| 20:50.13 | drachensun | 3 years however you'd like or source+binary and you can stop hosting whenever |
| 20:50.52 | ZaEarl | For example, when Bruce Perens bought a ZaTab, he asked for the source and I pointed him to our github. |
| 20:52.33 | ZaEarl | Software Freedom Conservancy was also happy with github, and they're GPL-enthusiasts. |
| 20:52.40 | techn | 800$ missing of goal :) |
| 20:53.19 | techn | drinks bear for that (and for friday) |
| 20:53.31 | techn | :DDD |
| 20:53.39 | techn | drinks beer for that (and for friday) |
| 20:54.04 | jelly-home | the AM8726-M3 liberation indiegogo might not fare that well |
| 21:01.03 | ZaEarl | 500 units is a good deal. Most factories have 1000 minimum order. |
| 21:07.47 | jelly-home | yeah, but I fear j1nx hasn't done enough promotion for it to have a chance to work |
| 21:25.41 | *** join/#arm-netbook slash_random (~david@host238.190-230-245.telecom.net.ar) |
| 21:29.28 | Turl | jelly-home: has it hit cnxsoft yet? :) |
| 21:29.57 | mnemoc | nope |
| 21:29.59 | jelly-home | Turl: that's where I read about it |
| 21:30.06 | mnemoc | uhm |
| 21:30.56 | jelly-home | not just a Halloween prank http://www.cnx-software.com/2012/10/31/crowfunding-initiative-to-open-amlogic-aml8726-m3-stb-source-code/ |
| 21:32.52 | techn | Turl: works with power save off.. now I'll try without |
| 21:35.23 | mnemoc | i saw it on j1nx's feed.... no idea why I didn't notice cnxsoft's post about it |
| 21:35.39 | mnemoc | wonders what else did he miss |
| 21:44.51 | *** join/#arm-netbook slash_random (~david@host238.190-230-245.telecom.net.ar) |
| 21:46.32 | slapin | hno: ping |
| 21:50.07 | techn | Turl: power save mode problem still exists :( |
| 21:51.07 | Marex | slapin: pong ;-) |
| 21:52.19 | slapin | Marex: uh-uh |
| 21:54.43 | Turl | techn: must be your router then :) |
| 21:54.50 | Turl | what's the issue more specifically? |
| 21:55.21 | Marex | slapin: how's your harem doing ? |
| 21:55.24 | techn | Turl: high amount of following messages: |
| 21:55.24 | techn | [ 453.950000] rtw_set_ps_mode(): Enter 802.11 power save mode... |
| 21:55.24 | techn | [ 453.960000] rtl8192c_dm_RF_Saving(): RF_Normal |
| 21:55.24 | techn | [ 453.970000] rtl8192c_set_FwPwrMode_cmd(): Mode = 1, SmartPS = 2 |
| 21:56.03 | techn | network works ok with load.. but when idle it breaks every now and then |
| 21:56.19 | techn | .. ssh connection timeouts |
| 21:57.01 | Marex | techn: ssh shouldn't timeout that easily ... unless misconfigured |
| 21:57.35 | techn | Marex: yep.. no problem with other hosts |
| 21:58.24 | *** join/#arm-netbook slapin_ (~slapin@slapinbuild.ihdev.net) |
| 21:58.52 | techn | I think this is general problem since I have seen workaround(disable power save) at many places |
| 21:59.45 | *** join/#arm-netbook jeremb (uid2617@gateway/web/irccloud.com/x-pjjekdsvyehuqtws) |
| 22:01.37 | cat_x301 | wonders what is wrong with fetching from git://github.com/linux-sunxi/linux-sunxi : http://sprunge.us/bUHW |
| 22:01.54 | Turl | techn: the messages are normal |
| 22:02.16 | Turl | techn: your router might not like the card on low power mode |
| 22:02.19 | Turl | some do, some don't |
| 22:03.24 | *** join/#arm-netbook destinal (destinal@unaffiliated/destinal) |
| 22:03.44 | Turl | cat_x301: weird error |
| 22:03.50 | cat_x301 | techn: you may try to increase DTIM Interval on your router if such option exists |
| 22:03.52 | Turl | cat_x301: try another protocol |
| 22:05.53 | cat_x301 | Turl: same error.. |
| 22:06.09 | cat_x301 | perhaps cloning anew should cure.. |
| 22:06.11 | Turl | cat_x301: if you have fast interwebz try a fresh clone |
| 22:06.30 | cat_x301 | Turl: will do ;) |
| 22:12.08 | mnemoc | lkcl: watch charbax's interview to gary smith. at ~16m he says he makes a point for eoma68 on cars ;-) |
| 22:13.35 | mnemoc | lkcl: 17:00 |
| 22:19.16 | *** join/#arm-netbook mawell (51398256@gateway/web/freenode/ip.81.57.130.86) |
| 22:20.01 | techn | cat_x301: Infact DTIM interval increasing should make things worse |
| 22:23.21 | cat_x301 | techn: maybe, but in my case, when i changed it from 1 to 3 it helped to get rid of random disconnects on all computers i have conntected via router |
| 22:24.46 | cat_x301 | techn: ... including mobile devices |
| 22:25.22 | *** part/#arm-netbook mawell (51398256@gateway/web/freenode/ip.81.57.130.86) |
| 22:25.47 | techn | My network works ok with n9, n900, sgsII, n8, PS3, wii.. what else.. ;) |
| 22:26.29 | techn | but sure.. I can't deny that router could have some config problem |
| 22:27.47 | techn | there is so many different parameters to adjust |
| 22:28.07 | cat_x301 | techn: i just recalled that i played with another option too, WMM. Got it disabled. Maybe you are right and DTIMM did not really work out ;) |
| 22:29.04 | cat_x301 | techn: each experiment took me around 40 minutes and now it is difficult to remember change of which parameter actually made things stable. |
| 22:30.51 | Gumboot | So there's this Arndale board thing, but it comes with 100mbit ethernet. Is that for real? |
| 22:31.38 | rm | even raspberry pi comes with 100 mbit ethernet, what's unreal about that? |
| 22:31.40 | Gumboot | Can proper ethernet be attached somehow? |
| 22:31.54 | xenoxaos | nope |
| 22:31.56 | Gumboot | rm: rpi is about as bottom-end as you can get, though, isn't it. |
| 22:32.03 | rm | hence "even" :p |
| 22:32.20 | Gumboot | I want gigabit. |
| 22:32.40 | Gumboot | I'd seriously consider it for its A15s, but it has no gigabit. |
| 22:33.26 | xenoxaos | not many mobile processors have ethernet |
| 22:34.42 | xenoxaos | although i would prefer the asix over the smsc that's in most dev boards |
| 22:36.23 | Turl | barely 3 cubieboards to go! \o/ |
| 22:36.50 | Gumboot | ... until nothing happens! |
| 22:37.12 | Gumboot | Or has there been word back from the website on turning around the first $50k quickly? |
| 22:37.39 | Turl | no idea |
| 22:37.40 | Gumboot | xenoxaos: i.MX6 has gigabit. I'm full of hope for that. |
| 22:37.46 | *** join/#arm-netbook freakazoid0223 (~matt@pool-173-75-233-172.phlapa.fios.verizon.net) |
| 22:37.49 | Turl | but it's a milestone we can celebrate :P |
| 22:38.07 | Gumboot | It could be the most terrible MAC in the world, to th point of being worse than an A15 with 100mbit, but that'd take a lot of bad design. |
| 22:38.17 | cat_x301 | techn: jfyi: http://www.dd-wrt.com/wiki/index.php/Advanced_wireless_settings#DTIM_Interval -- -The higher the DTIM period, the longer a client device may sleep and therefore the more power that particular client device may potentially save. -- I interpreted this in the way that when device is sleeping it cannot response on every beacon frame casuing some buffering issues.. but this is just my guess. |
| 22:38.24 | xenoxaos | Gumboot: i know, we have a couple of imx6's in our builder |
| 22:38.27 | Gumboot | There goes another. |
| 22:38.43 | Gumboot | Oh. Cool. Tell me about that ethernet. Is it good? |
| 22:39.00 | xenoxaos | we dont really care about the ethernet....we liked the sata |
| 22:40.21 | Gumboot | Does this Exynos 5 at least have external PCIe so that proper hardware can be attached on the board? |
| 22:40.48 | techn | cat_x301: Anyway I can't adjust DTIM or WMM with my current firmware :/ |
| 22:41.02 | cat_x301 | techn: oops then.. |
| 22:41.16 | Turl | openwrt ftw :) |
| 22:41.31 | techn | Turl: debwrt :) |
| 22:41.55 | Turl | dewhat? |
| 22:51.47 | techn | 64$ missing :p |
| 22:51.57 | *** join/#arm-netbook Sternennebel (~martin@p5494A7A8.dip.t-dialin.net) |
| 22:54.48 | cat_x301 | Turl: btw, cloning fails too.. http://sprunge.us/GEQY |
| 22:55.36 | techn | cat_x301: It's not first time that github is broken |
| 22:56.16 | mnemoc | they do some odd things there. you can access the commits of any related fork from any other |
| 22:56.35 | Turl | cat_x301: https://github.com/linux-sunxi/linux-sunxi/commits/sunxi-3.4 404 on first commit |
| 22:56.43 | Turl | mnemoc: can you repush that branch? |
| 22:57.15 | Turl | mnemoc: well duh it's git :) |
| 22:57.35 | mnemoc | https://github.com/linux-sunxi/linux-sunxi/commits/sunxi-3.4 works fine for me... |
| 22:57.46 | Turl | click to see first commit |
| 22:57.49 | Turl | it 404s |
| 22:57.51 | mnemoc | oh |
| 22:57.53 | mnemoc | wth |
| 22:57.56 | Gumboot | $64 to go! |
| 22:58.04 | Turl | it probably got corrupted :< |
| 22:58.07 | Turl | push it again mnemoc |
| 22:58.36 | mnemoc | not sure if it will let me |
| 22:58.46 | mnemoc | i mean. it will ignore the push |
| 22:59.00 | Turl | try -f to an older hash then again to fastforward? |
| 23:01.10 | Gumboot | And we're over. |
| 23:01.40 | Turl | by almost one full unit :) |
| 23:03.50 | techn | I hope that paypall worked :p |
| 23:05.00 | *** join/#arm-netbook slash_random (~david@host238.190-230-245.telecom.net.ar) |
| 23:06.05 | mnemoc | Turl: running fsck locally. it's not letting me push |
| 23:07.54 | mnemoc | didn't help |
| 23:07.59 | mnemoc | <PROTECTED> |
| 23:08.27 | mnemoc | <PROTECTED> |
| 23:09.04 | techn | git has checksum in itself.. so it should corrupt |
| 23:09.10 | techn | shouldn't |
| 23:09.32 | cat_x301 | mnemoc: seems to work now |
| 23:09.59 | mnemoc | :) |
| 23:29.38 | Marex | cat_x301: clone linux from git://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux-2.6.git |
| 23:29.49 | Marex | cat_x301: then git remote add github <github url> |
| 23:30.00 | Marex | git fetch and git branch -t -b sunxi github/master |
| 23:30.03 | Marex | or something |
| 23:30.48 | cat_x301 | Marex: mnemoc actually has fixed the issue already :P |
| 23:36.04 | Marex | good |
| 23:36.21 | Marex | it's still good to pull the objects from proper location and only pull the remnants from other repos |
| 23:37.29 | *** join/#arm-netbook bsdfox (~Bob@c-71-197-83-152.hsd1.ca.comcast.net) |
| 23:37.30 | *** join/#arm-netbook bsdfox (~Bob@unaffiliated/bsdfox) |
| 23:38.17 | slapin_ | Gumboot: for gigabit you could use one of kirkwood based things like sheevaplug/dreamplug, later does have quite capable esata to attach your raid box... |
| 23:39.52 | slapin_ | do anybody knows how to make markup like http://ossfans.org/shot.png with LaTeX? some package maybe? |
| 23:42.57 | mnemoc | what about svg and add it to the wiki? :) |
| 23:43.29 | slapin_ | mnemoc: noooooo |
| 23:43.57 | slapin_ | mnemoc: I'm still sane regardless of whatever |
| 23:44.16 | Turl | slapin_: https://en.wikibooks.org/wiki/LaTeX/Tables |
| 23:46.17 | Marex | Turl: yep |
| 23:46.44 | Marex | slapin_: you can get calxeda chip with gemac (10g) |
| 23:48.09 | mnemoc | can <math /> do latex tables? |
| 23:51.03 | Turl | mnemoc: try on a sandbox |
| 23:51.04 | Turl | :P |
| 23:53.15 | mnemoc | it doesn't accept \begin |
| 23:54.03 | slapin_ | Marex: that's cool, but I still don't use 100Mbit to its full potential... |
| 23:54.27 | mnemoc | inline svg just to do a nice table of bits on a registry doesn't sounds like a crazy idea to me :< |
| 23:54.29 | slapin_ | Marex: any cheap boxes with that chip? |
| 23:54.39 | *** join/#arm-netbook specing (~specing@unaffiliated/specing) |
| 23:55.39 | slapin_ | mnemoc: I'm lazy person. Too lazy for that. I better learn Tex some more and script the damn thing. |
| 23:55.51 | slapin_ | s/Tex/TeX |
| 23:55.53 | slapin_ | s/Tex/TeX/ |
| 23:56.02 | slapin_ | aww shit |
| 23:56.18 | slapin_ | ibot: you're uncool :( |
| 23:56.45 | mnemoc | slapin_: but doing it in tex won't help improving the documentation in the wiki |
| 23:58.30 | Marex | slapin_: I think you can get a "cheap" server blade with that |
| 23:59.48 | slapin_ | mnemoc: well, it will make me learn some fun bits and will make me like what I do, so it serves its purpose. As for wiki - doesn't it have TeX markup plugin? Regardless, I'll blow my mind filling-out wiki page at this stage with what I have. TeX does have some cool automagic thingies... |
