00:18.46 | A2Sheds | looking good, sample connectors are arriving |
00:26.42 | ManoftheSea | yay, finally |
00:35.25 | A2Sheds | \0/ \\0 0// /0\ \0/...... |
00:37.38 | *** join/#arm-netbook eflatun (~eflatun@46.1.118.222) |
02:41.45 | *** join/#arm-netbook hipboi (~hipboi@221.4.213.95) |
06:02.59 | hipboi | zenitraM: i have a question |
06:03.24 | hipboi | zenitraM: your mele box, before it doesn't work |
06:04.08 | hipboi | zenitraM: how did you powered off the box |
06:05.00 | hipboi | zenitraM: did you powered off by the android system or unplugged the power adapter directly |
08:24.07 | mnemoc | hi, dumb question, how can I increase the brightness of a a10-based tablet from adb shell? :( ... searching in /proc with android's `ls` is not fun :< |
08:31.50 | *** join/#arm-netbook QingPei (~qingpei@114.243.173.190) |
08:35.59 | zenitraM | hipboi: from android |
08:36.20 | zenitraM | mnemoc: use busybox find |
08:36.36 | zenitraM | most of the useful commands are in busybox but they aren't symlinked |
08:36.52 | mnemoc | made an script iterating over ls, no brightness file in proc or sys :< |
08:37.09 | mnemoc | zenitraM: nice for the hint :) |
08:37.23 | mnemoc | zenitraM: didn't know busybox was there by default |
08:37.37 | zenitraM | it is, luckily |
08:38.09 | zenitraM | it even has chroot |
08:38.35 | hipboi | mnemoc: hi |
08:38.43 | mnemoc | hipboi: hi! |
08:38.48 | hipboi | mnemoc: have received your mele box? |
08:39.05 | mnemoc | hipboi: not yet |
08:39.07 | hipboi | someone in uk have got it without any customs |
08:39.11 | mnemoc | left .se 3 days ago |
08:40.23 | zenitraM | airmail does rarely stop in customs at least here in spain |
08:40.47 | mnemoc | received an onda vi10 elite yesterday, so I have an a10 device to play again.... but installed with power control widget, hit brigthness to my world went black :< |
08:41.07 | mnemoc | s/installed with/installed the/ |
08:41.26 | zenitraM | mmm… if nothing else works, try reflashing from livesuit |
08:41.57 | mnemoc | yes, i'll start downloading the firmware |
08:42.37 | zenitraM | oh, here you have |
08:42.38 | hipboi | you can swipe the data partition to get the default setting |
08:42.38 | zenitraM | http://www.slatedroid.com/wiki/index.php/Fix_a_Brightness_Setting_of_0 |
08:43.55 | hipboi | mnemoc: i mean the customs duty |
08:44.11 | hipboi | the guy in uk doesn't pay any customs duty to get the mele box |
08:45.13 | mnemoc | hipboi: yes, very nice indeed, only the DHL guy has been screwed until now |
08:45.31 | mnemoc | DHL guy = zenitraM :p |
08:45.44 | zenitraM | yeah :( i'm the unlucky one i guess |
08:45.52 | hno | oh, my mele is finally on route to delivery. A good chance I get it today. |
08:46.00 | zenitraM | great news |
08:46.03 | mnemoc | \o/ |
08:47.09 | mnemoc | hno: are you sure about https://github.com/amery/linux-allwinner/commit/ee7cda129c8d22db70e57bf200bf365cef6e9c86 ? hipboi reverted that change in the official tree back in december |
08:47.12 | hno | swedish post sitting on the item for 4 days trying to figure out where to send it within same city :/ |
08:47.24 | mnemoc | hahah |
08:47.40 | hno | mnemoc, it should be reverted. |
08:47.46 | zenitraM | i've had packages coming from HK to Spain in 3 days and then having to wait 10 days to be delivered to my city |
08:50.55 | zenitraM | hipboi: when/if you ship the mele back, could you put a lower value for customs, i.e $20? |
08:51.14 | zenitraM | packages under 25€ (~$28) don't have to pay custom duty here |
08:51.35 | hipboi | ok |
08:51.44 | hipboi | i will ship it by swenden post |
08:51.59 | zenitraM | okay, no problem :) |
08:52.30 | mnemoc | zenitraM: thanks a lot about that link, using sqlite3 solved the problem smoothly |
08:54.57 | zenitraM | you're welcome |
08:56.29 | mnemoc | for the records, it was "merely" set to 103... but understood by the driver as 0 |
09:08.15 | *** join/#arm-netbook gordan (~gordan@78-86-195-86.zone2.bethere.co.uk) |
09:16.23 | CIA-91 | rhombus-tech: Peter master * r9494891da404 /allwinner_a10/orders/p.steenbergen__64__j1nx.nl.mdwn: |
09:52.28 | *** join/#arm-netbook baybal (~baybal@gateway/tor-sasl/baybal) |
10:35.56 | CIA-91 | rhombus-tech: Santeri master * ra3b377789114 /allwinner_a10/orders/crysaz.mdwn: |
11:26.59 | *** join/#arm-netbook eflatun (~eflatun@46.1.30.66) |
13:42.41 | *** join/#arm-netbook Guest85176 (~mike@c-71-63-115-202.hsd1.va.comcast.net) |
14:06.03 | *** join/#arm-netbook mikey_w (~mike@c-71-63-115-202.hsd1.va.comcast.net) |
14:37.28 | *** join/#arm-netbook mikey_w (~mike@c-71-63-115-202.hsd1.va.comcast.net) |
14:46.04 | *** join/#arm-netbook mikey_w (~mike@c-71-63-115-202.hsd1.va.comcast.net) |
15:02.23 | *** join/#arm-netbook ibot (~ibot@rikers.org) |
15:02.23 | *** topic/#arm-netbook is EOMA: Embedded Open Modular Architecture - Don't ask to ask. Just ask! - http://elinux.org/Embedded_Open_Modular_Architecture/EOMA-68 - ML arm-netbook@lists.phcomp.co.uk - Logs http://ibot.rikers.org/%23arm-netbook - http://rhombus-tech.net/ |
15:03.23 | CIA-91 | rhombus-tech: master * rb8e68de01a48 /allwinner_a10/orders/evgeniypl.mdwn: |
15:03.36 | CIA-91 | rhombus-tech: master * r95404cad2e54 /allwinner_a10/orders/evgeniypl.mdwn: |
15:05.57 | *** join/#arm-netbook mikey_w (~mike@c-71-63-115-202.hsd1.va.comcast.net) |
15:11.51 | *** join/#arm-netbook eflatun (~eflatun@46.1.30.66) |
15:39.42 | *** join/#arm-netbook QingPei (~qingpei@114.243.173.190) |
15:59.42 | *** join/#arm-netbook mikey_w (~mike@c-71-63-115-202.hsd1.va.comcast.net) |
16:16.41 | *** join/#arm-netbook eflatun (~eflatun@46.1.30.66) |
16:29.47 | *** join/#arm-netbook mikey_w (~mike@c-71-63-115-202.hsd1.va.comcast.net) |
16:37.11 | *** join/#arm-netbook mikey_w (~mike@c-71-63-115-202.hsd1.va.comcast.net) |
16:37.35 | *** join/#arm-netbook mike_ (~mike@c-71-63-115-202.hsd1.va.comcast.net) |
17:24.12 | *** part/#arm-netbook QingPei (~qingpei@114.243.173.190) |
19:10.18 | hno | mele arrived, no customs. |
19:11.03 | hno | well, it was checked by customs but no fee or taxes. |
19:21.28 | mnemoc | \o/ |
19:32.09 | hno | hm.. wonder how I file this in the books... ah, well, i'll sort it out. |
19:34.07 | hno | pulls out a screwdriver. |
19:36.33 | hno | damn, those screws need a tiny screwdriver. Odd proportions. |
19:36.42 | specing | mele? |
19:37.10 | hno | yes |
19:40.29 | hno | and that's a too tiny screwdriver. None of the right size. |
19:40.54 | mnemoc | specing: an inexpensive allwinner-a10 devkit disguised as cheap tv set-top |
19:45.03 | specing | $100? no thanks |
19:46.08 | specing | it looks kinda stupid, too |
20:03.24 | specing | Is AW A10 armv7-a -r -m or what is it? |
20:05.10 | destinal | A10 is a Cortex-A8 (ARMv7) SoC (little endian) |
20:05.48 | mnemoc | it's a 1.0-1.5Hz Cortex A8 with mali400p |
20:06.20 | specing | I know that far |
20:06.30 | destinal | or hmm I guess a8 has either endian mode but armel seems to be the one commonly used for code on it |
20:06.36 | specing | But there are 3 subarches of the ARMv7 acritecture |
20:06.53 | mnemoc | specing: cortex-a* are all armv7a |
20:07.04 | specing | Ah ok |
20:07.12 | mnemoc | cortex-m* are armv7m, and so on |
20:07.56 | mnemoc | destinal: yes |
20:08.35 | mnemoc | specing: for those interested in a devkit $100 (and more the $65 we paid in this special order) is a very good price |
20:09.05 | specing | Is neon better than vfp? |
20:09.53 | specing | choosing between -mfpu=neon and -mfpu=vfp3-d16 |
20:11.13 | mnemoc | gentoo has lots of info in that topic |
20:12.12 | specing | That is what Im configuring :D |
20:12.38 | specing | I just read neon is newer on wikipedia |
20:13.18 | mnemoc | that's what I thought. it's the kind of question gentoo people asks |
20:14.05 | mnemoc | and yes, neon > vfp |
20:14.14 | specing | ofcourse |
20:14.15 | ka6sox | neon is hardware support...nice stuff. |
20:14.32 | specing | mnemoc: Im a ricer :D |
20:14.41 | *** join/#arm-netbook mikey_w (~mike@c-71-63-115-202.hsd1.va.comcast.net) |
20:14.52 | *** join/#arm-netbook mike_ (~mike@c-71-63-115-202.hsd1.va.comcast.net) |
20:15.01 | ka6sox | specing, I"m not...but use OpenEmbedded all the time so semi-ricer :D |
20:15.30 | mnemoc | hahaha |
20:18.45 | specing | is scared to put -mthumb in CFLAGS |
20:22.38 | mnemoc | what about just -mcpu=cortex-a8 ? :p |
20:23.05 | specing | I did that |
20:23.08 | specing | already |
20:23.50 | specing | I've read thumb-2 is almost as fast as regular ARM I-set |
20:25.15 | mnemoc | just forget about it, trust the compilerand move forward :p |
20:25.45 | specing | I've heard there was some galactic breakdown resulting from the use of thumb |
20:27.43 | *** join/#arm-netbook mikey_w (~mike@c-71-63-115-202.hsd1.va.comcast.net) |
20:35.56 | Gumboot | Does -mfpu=neon actually do anything without extra options like -fvectorise-stuff ? |
20:47.47 | specing | Gumboot: Im going to turn it on and see if it burns |
20:49.20 | specing | Ah, its already on |
21:06.15 | *** join/#arm-netbook mike_ (~mike@c-71-63-115-202.hsd1.va.comcast.net) |
21:14.06 | Gumboot | Autovectorisation still has a long way to go before it becomes useful. |
21:16.40 | Gumboot | There are a few idioms you can use in loops to get better use out of it. |
21:17.21 | Gumboot | And I've been pushing, for a couple of months, for a particular modification to be made so that gcc can autovectorise a whole other class of loops it currently misses. |
21:34.15 | *** join/#arm-netbook Guest44957 (~chatzilla@ip70-173-29-15.lv.lv.cox.net) |
22:03.22 | CIA-91 | rhombus-tech: David master * rf5b69b4f3c51 /allwinner_a10/hacking_the_mele_a1000.mdwn: |
23:07.33 | hno | specing, On Cortex-A8 neon is generally a lot faster than vfp, but on A9 the field is different. Neon is not following IEEE math as closely however so it's not a simple question of selecting one or the other. |
23:07.53 | hno | vfp is alsi in hardware. |
23:10.18 | hno | specing, not all applications is thumb-compatible. But we by accident did compile 80% of Fedora in thumbs mode without too much issues, and in most cases one can mix Thumb and ARM-32 code freely. A notable exception is anything related to Qt. |
23:10.53 | hno | and Kernel can not be compiled in thumbs mode. |
23:28.35 | specing | <@ralos> specing: in my tests many things break when using -mthumb2 |
23:28.59 | specing | hno: So these are compile time issues and not runtime ones? |