IRC log for #htc-linux on 20100430

00:01.14*** join/#htc-linux skodde (~skodde@unaffiliated/skodde)
00:19.22*** join/#htc-linux leviathan (~quassel@c-82-192-226-27.customer.ggaweb.ch)
00:19.42*** join/#htc-linux leviathan_ (~quassel@c-82-192-226-27.customer.ggaweb.ch)
00:27.26*** join/#htc-linux jumoit1 (~samuel@121.77.87.145)
00:36.57stinebdmy god, i could save so much time if i didn't do stupid shit
00:38.34*** join/#htc-linux swc|666 (~infidel20@unaffiliated/swc666/x-4934821)
00:42.56*** join/#htc-linux jumoit (~samuel@121.77.87.145)
00:43.40*** join/#htc-linux fakker (~noobhands@host213-123-190-138.in-addr.btopenworld.com)
00:44.43IceBonestinebd: you just described everybody's life.
00:44.55stinebdyes but i do exceptionally stupid stuff sometimes
00:45.05stinebdlike adding Ts where they don't belong
00:45.49stinebdand then trying to figure out why a buildspec flag isn't doing anything for 3 hours
00:46.08stinebdlike half an hour really, but still
01:04.33*** join/#htc-linux br1ck (~br1ck@xdslck159.osnanet.de)
01:06.46ToAsTcfhstinebd: any luck?
01:06.56stinebdToAsTcfh: still working on aosp
01:07.01ToAsTcfh:/
01:07.08ToAsTcfhaosp takes forever
01:07.19stinebdthere's a lot to do
01:07.29ToAsTcfhwhat u got going on
01:07.43ToAsTcfhwhat u fixin
01:07.44stinebdtrying to fix renderscript so livewallpapers will work for us
01:07.52ToAsTcfhah
01:07.59stinebdgoing to eventually see if launcher2 works as well
01:08.14ToAsTcfhi know some have done it for the hero
01:08.23stinebdit's a bit more interesting on vga devices
01:08.28ToAsTcfhtru
01:08.33ToAsTcfhunderstood
01:09.17ToAsTcfhwe have an issue with the leaked htc .29 kernel and touch not working at all in aosp
01:09.48ToAsTcfhit seems htc went all proprietary with their new .29 kernel for the hero
01:10.21ToAsTcfhwe only have touch screen use in htc built frameworks
01:10.30ToAsTcfhkinda strange
01:16.04*** join/#htc-linux thedicemaster2 (~thedicema@e200071.upc-e.chello.nl)
01:31.24*** join/#htc-linux swc|666 (~infidel20@unaffiliated/swc666/x-4934821)
01:47.20*** join/#htc-linux ravishi (~stormy@c-98-249-6-188.hsd1.va.comcast.net)
01:57.28stinebdnice
01:57.32stinebdhad grass livewallpaper working
01:57.38stinebdthen nexus wallpaper had to ruin it for me
01:58.02stinebdphh: if we could steal some more ram for gpu that would be nice, i think
02:04.53stinebdat least they work in software
02:13.46stinebd12 bytes short on that last gralloc
02:14.01stinebdthat kinda sucks
02:19.43stinebdnope thats the errno :D
02:24.50*** join/#htc-linux Metamucil (proxy@secursurf.org)
03:01.24*** part/#htc-linux kupo (~kupo@pinky.ratman.org)
03:04.44*** join/#htc-linux BHSPitMonkey (~stephen@unaffiliated/bhspitmonkey)
03:08.28*** join/#htc-linux vpeezy (~vpeezy@c-76-111-21-238.hsd1.md.comcast.net)
03:09.50vpeezysup peeps. i have a verizon wireless cdma touch pro 2 with winmo 6.5 and sense 2.1. if i install the project android port can i add themes or cooked roms for droid or no?
03:19.40|Jason8|no
03:19.58|Jason8|it's gotta be from the project
03:20.16vpeezyi coud have asked you that in xda devs lol!
03:36.57*** part/#htc-linux vpeezy (~vpeezy@c-76-111-21-238.hsd1.md.comcast.net)
04:08.13*** join/#htc-linux enjalot_ (~enjalot@c-68-35-253-116.hsd1.fl.comcast.net)
04:10.43*** part/#htc-linux ravishi (~stormy@c-98-249-6-188.hsd1.va.comcast.net)
04:19.29*** join/#htc-linux goxboxlive (~jrs@mail2.hjellnesconsult.no)
04:44.31*** join/#htc-linux Jos (~Jos@Lithium.Jossie90.info)
04:51.02*** join/#htc-linux BHSPitMini (~BHSPitMon@pool-71-170-176-30.dllstx.fios.verizon.net)
05:32.59*** join/#htc-linux Metamucil (proxy@secursurf.org)
05:39.58*** join/#htc-linux jumoit (~samuel@121.77.87.145)
05:50.25*** join/#htc-linux NoMercy (~NoMercy@p4FDE6A4B.dip.t-dialin.net)
05:50.42NoMercygmorning from germany guys
05:55.25|Jason8|yo
06:02.56*** join/#htc-linux NoMercy_ (~NoMercy@p4FDE7A4F.dip.t-dialin.net)
06:34.06*** join/#htc-linux DaveWM (dave@h-96-13-62-164.ip.alltel.net)
06:42.28*** join/#htc-linux balans (~Gebruiker@53546758.cable.casema.nl)
06:50.29*** join/#htc-linux IceBone (~IceBone@89-212-106-54.dynamic.dsl.t-2.net)
06:57.23*** join/#htc-linux Forsaken|Laptop (~Forsaken|@p54B01832.dip0.t-ipconnect.de)
06:57.48*** join/#htc-linux zachy (~george@ip-85-160-10-222.eurotel.cz)
07:01.39*** join/#htc-linux balans1 (~Gebruiker@53546758.cable.casema.nl)
07:11.58*** join/#htc-linux balans (~Gebruiker@53546758.cable.casema.nl)
07:12.39*** part/#htc-linux balans (~Gebruiker@53546758.cable.casema.nl)
07:25.19*** join/#htc-linux kvaster (~kvaster@93.84.112.80)
07:40.43*** join/#htc-linux Markinus (~Miranda@gtng-4db04f9b.pool.mediaWays.net)
07:42.19*** join/#htc-linux kiozen (~kiozen@p549215BE.dip0.t-ipconnect.de)
07:53.37*** join/#htc-linux obrienmd (~obrienmd@173.160.148.166)
08:12.27*** join/#htc-linux rakeem (~rak@n220246173025.netvigator.com)
08:12.54rakeemHi fellas.  Long time I've not been around here.
08:14.09rakeemI'm gonna go and make a quick impulse purchase.  Which device is best supported these days.  Last time I was around here it was the Uni, and googling seems to suggest it's still the case.  Am I wrong?
08:21.08*** join/#htc-linux Wout (~a@s55927d21.adsl.wanadoo.nl)
08:27.08rakeemHello?  Anyone there?
08:49.49*** join/#htc-linux DenPal (~hoper@post1.centr.by)
09:05.58*** part/#htc-linux DenPal (~hoper@post1.centr.by)
09:32.52*** join/#htc-linux Captnoord (~Captnoord@dc5147a47b.adsl.wanadoo.nl)
09:53.44*** join/#htc-linux leviathan (~quassel@etx-public-dock-42-dhcp.ethz.ch)
09:53.54*** join/#htc-linux leviathan_ (~quassel@etx-public-dock-42-dhcp.ethz.ch)
09:56.20*** join/#htc-linux Dinde (kayser@sur-internet.net)
10:03.49*** join/#htc-linux jazka (jaska@193.104.35.159)
10:20.22*** join/#htc-linux Wout (~a@s55927d21.adsl.wanadoo.nl)
10:25.00*** join/#htc-linux GNUtoo (~GNUtoo@host139-39-dynamic.117-80-r.retail.telecomitalia.it)
10:45.25*** join/#htc-linux mickey|office (~Mickey@dialbs-092-079-168-007.static.arcor-ip.net)
10:50.43*** join/#htc-linux jumoit (~samuel@121.77.87.145)
10:52.54*** join/#htc-linux Graduated (~Graduated@HSI-KBW-078-042-078-049.hsi3.kabel-badenwuerttemberg.de)
10:53.18GraduatedMarkinus: hi how are things going
10:53.46GNUtoohi, phh, ping me when you want me to test something
10:56.56*** join/#htc-linux mainuser (~quassel@HSI-KBW-078-042-078-049.hsi3.kabel-badenwuerttemberg.de)
11:08.13*** join/#htc-linux Alex[sp3dev] (~alex_dfr@86.110.163.21)
11:08.40Alex[sp3dev]hello everyone! phh, are you here?
11:14.46*** join/#htc-linux obrienmd (~obrienmd@173.160.148.166)
11:14.59*** join/#htc-linux leviathan (~quassel@etx-public-dock-42-dhcp.ethz.ch)
11:15.10*** join/#htc-linux leviathan_ (~quassel@etx-public-dock-42-dhcp.ethz.ch)
11:22.49phhAlex[sp3dev]: yes
11:23.10phhGNUtoo: there was something I wanted to try yesterday, but as I haven't noted it, it will take some time.
11:23.27phhrakeem: vogue or raphael
11:23.43GNUtoook np
11:23.49GNUtootell me when you updated the test
11:23.51Alex[sp3dev]well, on kovsky the h2w headset gpio is inverted. how do we handle it? i don't want to have dirty hacks like if mach_is_kovs() in the audio driver, i have the idea to add the jack_inverted field to  platform data. what do you think about it?
11:23.51phh[04:02:39] <stinebd> phh: if we could steal some more ram for gpu that would be nice, i think <-------------- it seems that hardware is limited at 8MB
11:24.18phh[00:59:44] <GNUtoo> <phh> stupid uint8/uint32. => new test? <--- fair enough.
11:24.41GNUtoois adicted to test,once I start I can't stop until it works
11:24.49phh[01:17:10] <jonpry> phh: your right that 0x8000... works great. that is the right data is there. but it doesn't change, or the driver is otherwise unhappy with the contents <---- this space is not touched by us. perhaps try to run gps from wince first.
11:25.01GNUtoosame issue sometime with kernel hacking
11:25.42phhAlex[sp3dev]: meh ?
11:25.48phhAlex[sp3dev]: H2W settings are declared in board file
11:26.11Alex[sp3dev]but h2w doesn't support indicating that detection gpio is inverted, does it?
11:26.27phhAlex[sp3dev]: ah you mean 0=1 and 1=0 ?
11:26.31Alex[sp3dev]yes
11:26.40phhoh ok
11:26.42phhwell no
11:26.47phhbut you're sure you have H2W ?
11:26.49phhH2W is HTC
11:27.13Alex[sp3dev]how do i tell for sure then?
11:27.21phhhaha
11:27.22phhno clue.
11:27.32phhAlex[sp3dev]: well, it's quite complicated stuff
11:27.34phhif just inverting logic works
11:27.36phhthen it's H2W.
11:27.53phhAlex[sp3dev]: add a inverted_gpio in the platform data struct
11:27.55Alex[sp3dev]yes, it does. even headset button toggles music playback in android
11:28.05phhok so it definitely is H2W
11:29.47Alex[sp3dev]ok, i'll make the patch some time later. probably a separate patch for other kovsky changes too. ok, gotta go do my homework now. well, it's stupid to make the project and roadmap once you've written your application. and doing it in M$ project and excel makes me sick
11:30.31phhlol
11:48.41phhGNUtoo: guess what... same url :p
11:48.56GNUtoook
11:49.37rakeemphh:  Cheers for the reply, but I'd left the house by that time.  I've just come home with a shiny (well 2nd hand) SonyEricSonyEricSonyEricSon... X1.  That's a Kovsky right?
11:49.56phhrakeem: yup
11:50.05phhrakeem: it works fine too
11:50.16rakeemIs there any weird hardware shit I should know before I nuke the Chinese WM ROM in favour of an English one?
11:50.23phhI think it's just a raphael with WVGA
11:50.38rakeemphh:  Nice to hear.
11:51.51*** join/#htc-linux leviathan (~quassel@etx-public-dock-42-dhcp.ethz.ch)
11:52.00*** join/#htc-linux leviathan_ (~quassel@etx-public-dock-42-dhcp.ethz.ch)
11:52.53rakeemMy 2nd Tornado died in my Pocket (quite weirdly, white screen, turns on by itself) so I've been without a phone for most of the day.  Let me flash an English WM travesty onto it then I'll be back to discuss the nerdier, fun stuff....
11:53.18GNUtoophh, http://pastebin.com/VsE9xPcS + reboot + nothing at all on /dev/smd27
11:53.54phhbetter.
11:54.14phhkind of.
11:55.02rakeemNice to be back on this channel BTW.  Big up to the lads: BabelO, cr2, goxboxlive, dcordes, etc....
11:55.47phhyou've not been here for a long while uh ? :p
11:55.51phhonly dcordes is still activ
11:55.52phhe
11:56.47GNUtoowhat is cr2 becomming?
11:56.58GNUtoolast time I saw him it was at fosdem
11:58.03phhdon't know
11:58.29GNUtoook
12:00.29*** join/#htc-linux infernix (nix@unaffiliated/infernix)
12:03.43*** join/#htc-linux DaveWM (~dave@173-19-136-12.client.mchsi.com)
12:03.56phhGNUtoo: new test
12:06.50GNUtoook
12:08.33GNUtoohttp://pastebin.com/tGkKpHvD + no reboot + nothing on smd27
12:09.14phhclnt_call(376) 3000005b:90380d3d error decoding results.
12:09.15phhmeeeeeeeeeeeeh ?
12:11.50phhoh right.
12:12.26phhupdated
12:12.40phhGNUtoo: now I push the source, test.c at the same time as the binary
12:12.40GNUtoook
12:12.43phhif you want to see the changes
12:15.47GNUtoohttp://pastebin.com/VgL56iB3 + reboot
12:15.59GNUtoo<PROTECTED>
12:16.01GNUtoook
12:16.15GNUtooI'm mostly interested in the source file for when it'll work
12:16.22phhright :p
12:24.38phhGNUtoo: new test
12:24.44GNUtoook
12:25.47GNUtoohttp://pastebin.com/c8jX8yz3 + reboot + nothing on /dev/smd27
12:26.10GNUtoothe reboot appear some seconds after unlocking the screen
12:26.14GNUtoomaybe it's a coincidence?
12:26.32phhhum don't know
12:28.53GNUtoono it seem a coincidence
12:29.09GNUtoopowered on it does the same but you have to wati
12:29.11GNUtoo*wait
12:30.00phhperhaps it crashes when it got a fix ? :p
12:30.41GNUtoono
12:30.53GNUtooit would output the identifier first
12:32.27*** join/#htc-linux Vito_| (~Sysel@94.142.234.1)
12:33.23phhmuh ?
12:33.25phhwhich identifier ?
12:34.00GNUtoothe htc one...let me find it
12:35.47GNUtoo$PSTIS,*61
12:36.07GNUtooanyway it also reboots while screen is off
12:37.11phhsmd27 isn't the only way to give GPS infos unluckily
12:38.06GNUtooah?
12:38.08phhnew test
12:38.10GNUtoopmem?
12:38.11GNUtoook
12:38.14phhno, rpc
12:38.19GNUtoook
12:39.17phhbtw you're using replicant or cm mod ?
12:39.21phhI mean there is a libgps.so ?
12:40.04GNUtoono there isn't
12:40.07GNUtooreplicant
12:40.13GNUtoothere could be something
12:40.17GNUtoobut it's a fake thing
12:40.43phhok fine
12:40.59phhI feared some conflict
12:41.04GNUtooBoardConfig.mk:BOARD_HAVE_GPS_HARDWARE := false
12:41.12GNUtoothat's what I have
12:41.23phhthat's alright
12:42.55GNUtoohttp://pastebin.com/QXktqWwU + reboot + nothing on /dev/smd27
12:43.00GNUtoodo you want me to test on SHR?
12:43.07GNUtoobut you would have to compile with -static
12:45.10phhI don't know how to static build this thing -_-'
12:45.19phhanyway it's not needed
12:45.52GNUtoook
12:49.46phhalways crashes ~ 10s after start hum
12:50.06GNUtooyes
12:50.37GNUtoo10s exactly
12:52.37*** join/#htc-linux Captnoord (~Captnoord@dc5147a47b.adsl.wanadoo.nl)
12:53.27phhGNUtoo: quite same boat as us then.
12:53.32phhjust that our timeout seems way lower.
12:53.41*** join/#htc-linux enjalot_ (~enjalot@c-68-35-253-116.hsd1.fl.comcast.net)
12:53.51GNUtoook
12:54.32GNUtoobut we nearly got it working before(the test where there were some line starting with $GP)
12:55.19phhI know, but I can't reproduce it, and it can't loop /o\
12:56.02GNUtooI can reproduce it but it doesn't stay
12:56.08GNUtooit's reseted at the end
12:56.27phhyou get only one result
12:56.31phhyou have the time to get 10
12:56.40GNUtooah?
12:57.26GNUtooI had several lines
12:57.38phhnew test, I removed the pdsm_get_position loop
12:57.44GNUtoook
12:58.44*** join/#htc-linux hfctsaot (~hfctsaot@173-137-183-133.pools.spcsdns.net)
12:59.20GNUtoohttp://pastebin.com/C9JWuWkq + reboot + nothing on /dev/smd27
12:59.21hfctsaot\o/
12:59.52phhGNUtoo: 10seconds too ?
12:59.55GNUtoono
12:59.56GNUtooless
13:00.12phhhum.
13:00.48*** join/#htc-linux hfctsaot (~hfctsaot@173-137-183-133.pools.spcsdns.net)
13:01.13*** join/#htc-linux hfctsaot (~hfctsaot@173-137-183-133.pools.spcsdns.net)
13:01.27phhnew test
13:01.35phhI changed a magic value for pd_reg
13:01.39GNUtooabout 6s
13:01.40GNUtoook
13:02.19phh(both magic values have been seen so ...)
13:03.16GNUtoo10s
13:03.40GNUtoohttp://pastebin.com/7BzYqz80
13:03.46GNUtooand nothing on /dev/smd27
13:06.23*** join/#htc-linux Wout (~a@s55927d21.adsl.wanadoo.nl)
13:12.06phhGNUtoo: new test
13:12.17GNUtoook
13:13.05GNUtoohttp://pastebin.com/zMymRzZZ + reboot after 10s + nothing on smd27
13:15.08*** join/#htc-linux thedicemaster2 (~thedicema@e200071.upc-e.chello.nl)
13:19.12*** join/#htc-linux leviathan (~quassel@guest-docking-nat-1-001.ethz.ch)
13:20.10*** join/#htc-linux leviathan_ (~quassel@guest-docking-nat-1-001.ethz.ch)
13:21.01phhwhat's the ?!?
13:21.15phhok that's going to be funny.
13:21.23*** join/#htc-linux bais (~quassel@92.246.34.60)
13:21.43baishi
13:22.44phhah no fiou
13:22.51phhI forgot I deleted some useless bytes.
13:22.54GNUtoook
13:28.20*** join/#htc-linux Marcin_88 (Marcin@chello089078146089.chello.pl)
13:28.28phhnew test
13:28.38*** join/#htc-linux Forsaken|Laptop1 (~Forsaken|@p54B003B9.dip0.t-ipconnect.de)
13:29.58phhGNUtoo:
13:30.05phh(don't know if you need highlights :p)
13:30.19GNUtooyes I do
13:31.14GNUtoohttp://pastebin.com/5TesvSBz + nothing on smd27 + 10s reboot
13:32.03phhso nothing changed at all but values -_-'
13:32.11GNUtoook
13:34.17phhnew test
13:35.30GNUtoook
13:37.12GNUtoohttp://pastebin.com/TM7u3LSg etc...
13:37.41GNUtoops : http://pastebin.com/S8zn2A3i
13:37.51phhgood
13:37.52phhI think.
13:37.59phhGNUtoo: now get outside, and wait 10minutes.
13:38.21GNUtoook can I have all the sources?
13:38.29phhhttp://husson.hd.free.fr/test.c
13:38.39GNUtooincluding Android.mk etc...?
13:38.56phhI changed librpc folder, I'll tar it
13:39.14GNUtoook thanks a lot
13:39.19phhhttp://husson.hd.free.fr/librpc.tar.bz2
13:40.24phhcan I have working logs now ? :p
13:40.31phhI mean with some gps satellites fixed at least
13:40.48GNUtoook I'll paste the whole dump
13:41.10phhif the loop happens to change
13:41.12GNUtooHTTP request sent, awaiting response... 404 Not Found
13:41.26phhs/tar.bz2/tbz2/
13:41.45phhlet's test this program on rhod.
13:41.50phhwith some hope it will work.
13:42.02GNUtoohttp://pastebin.com/szkJw9ZJ
13:42.14GNUtooat the end it stops
13:42.19GNUtoobut it's very long....
13:42.42phhmeh it doesn't get any satellite ?!?
13:43.06phhyou have 1meters wide walls ?
13:43.59GNUtooI'll go out
13:44.04GNUtoobut I'll try navit under SHR
13:44.06*** join/#htc-linux AstainHellbring (~AstainHel@unaffiliated/astainhellbring)
13:44.11*** join/#htc-linux johnb81 (~john@85.104.134.179)
13:44.55phhA9 crashed. how surprising.
13:45.46GNUtoobut not now,in 15min or something like that
13:46.10phhok
13:48.15phhinteresting
13:48.17phhthis time A9 didn't crash
13:48.34phhI get nothing on smd27, but dmesg says
13:48.36phh[   32.327606] OOPS - smd_tty_buffer mismatch?!
13:48.42phhit's definitely good :D
13:48.49phhCaptnoord: interested ? :p
13:52.37baisI have a big question
13:52.59baishow to use my linux workstation for gateway of my android by wifi or usb?
13:53.15baisI tried to find something on internet but nothing explain this case
13:53.43baisdream ---> usb or wifi ---> my kubuntu ---> company proxy--> internet
13:54.00baisany ideas?
13:56.37baiszzzzzzz
13:56.53Woutphh: what are you working on at the moment?
13:56.55GNUtoohi mickey|office
13:57.00GNUtoois that a fix: $GPGSV,2,1,05,05,63,260,27,28,50,139,23,10,57,192,,15,21,296,*78 ?
13:57.15phhGNUtoo: that's a satellite info
13:57.16phhI think.
13:57.19GNUtoook
13:57.29phhWout: open gps driver for dream ? :p
13:57.42Woutcool! through that wince driver method?
13:57.57phhWout: dream = htc dream = htc g1
13:57.59phhthere is no wince for them :p
13:58.20Woutah, could that driver be ported to rhod?
13:58.25phhsure
13:58.31phhit's better than previous one
13:58.33phhit doesn't crash A9.
13:58.36Woutnice :)
13:58.38phhit just gives nothing.
13:59.04phhGNUtoo: a fix is GPGGA
13:59.59GNUtoook
14:00.16GNUtoo$GPGGA,140006.8,,,,,0,,,,,,,,*73?
14:00.18phhdon't worry, you'll take easily more than 2 minutes to get fixes
14:00.25phhGNUtoo: time is correct :p
14:00.28phh(first field is time)
14:00.30GNUtoook
14:00.39bais:)
14:01.26phhnow you have a network-less possibility to give SHR the correct time ! :p
14:01.56*** join/#htc-linux gauner1986 (~gauner198@p5B385E13.dip.t-dialin.net)
14:02.03gauner1986hi
14:02.11Woutoi
14:02.39phhGNUtoo: your GPGSV line states you have 5 satellites in line
14:02.50phhso it's really unlikely it get a fix.
14:04.02GNUtoook
14:04.09GNUtooI'll get a bigger cable
14:05.18gauner1986cable? satellites? what?
14:05.23GNUtoousb cable
14:05.56gauner1986for gps?
14:05.56phhgauner1986: I got GPS working on htc dream with an open program, now GNUtoo has to test it, and thus have good view to satellites
14:06.06gauner1986ah ok
14:06.26phhand now I'd like to know why it doesn't work on rhod.
14:06.58phhperhaps our tty driver is broken
14:06.59gauner1986hmmm
14:07.27gauner1986damn... i ate so much at yesterdays barbeque that i'll not be hungry for at least 24 hours
14:07.28gauner1986:D
14:07.56phhno way
14:11.56GNUtoogot a fix
14:12.02GNUtoobut I was grepping
14:12.11GNUtooI'll paste in private
14:12.58GNUtoobut I was grepping so I bet it stopped just after
14:13.26GNUtooI'll paste you in private more logs
14:17.10mickey|officeGNUtoo: woot!
14:17.13mickey|officecongrats
14:17.36GNUtoomickey|office, congrat phh not me
14:17.36*** join/#htc-linux AstainHellbring (~AstainHel@unaffiliated/astainhellbring)
14:17.47GNUtooI only did testing
14:17.53phhmickey|office: still, it stops after one fix :/
14:18.14phhGNUtoo: testing is so annoying that I congrats you too :p
14:18.19mickey|office:)
14:18.20GNUtoook
14:18.25phhI can't make more than 3 tests in a row before getting bored ...
14:19.04phhthat's why, even if now we have this code, it's not going to work on rhod any soon .
14:19.07GNUtoois adicted to kenrel hacking and test, I want always want to test it one more time,because maybe it'll work
14:19.34phhmickey|office: anyway it's still far from finished, we just have magic values currently.
14:19.44phhwith no clue of what they mean.
14:25.36mickey|officehmm, k. it's a start anyways
14:25.43mickey|officeso the at commands are completely meaningless?
14:25.48mickey|officeor is there to be a combination?
14:27.14phhmickey|office: I think AT commands are meaningless
14:27.21phhbut perhaps there are some AGPS stuff involved with them
14:27.32mickey|officeright
14:27.34phhbut android's libgps isn't linked at all with ril afaik
14:28.02mickey|officeeven if we have reliable NMEA, it would be helpful to upload ephimeris data
14:28.12mickey|officeas this dramatically improves cold starts
14:29.24phhyou know how to get ephimeris at least ? :p
14:29.32phhbefore even thinking of uploading them
14:29.51mickey|officeyes
14:30.12mickey|officethere are some gps vendors offering those
14:30.16*** join/#htc-linux DenPal (~hoper@59-243-252-87-dynamic-pool.gprs.mts.by)
14:30.30*** part/#htc-linux DenPal (~hoper@59-243-252-87-dynamic-pool.gprs.mts.by)
14:46.51phhGNUtoo: if you try to do a new fix, it takes ages too ?
14:47.01GNUtooyes
14:47.11GNUtooI'll mesure the time of fix
14:47.43GNUtoo14s
14:47.46GNUtoonot ages
14:47.57GNUtoobetween 10 and 14 s
14:48.37GNUtoofor trying to build a static executable I did mm clena
14:48.40GNUtoo*clean
14:48.48GNUtooand it cleaned everything
14:49.05GNUtooso I'll have to wait for GPS under SHR
14:49.31GNUtoomaybe I should build with openembedded?
14:49.37phhI tried arm-blabla-gcc *.c -static
14:49.40phhbut it doesn't work either :p
14:49.49GNUtoook
14:50.07phhGNUtoo: can't you run it from a chroot ?
14:50.12GNUtoommm maybe
14:50.23GNUtoobut chroot requires same libc
14:50.25GNUtoono?
14:50.28phhno
14:50.32GNUtooah?
14:50.42GNUtoook
14:50.51GNUtooah yes you're right
14:50.59GNUtooit's 2.4 vs 2.6 which differs
14:51.15phhpossible
14:52.58phhGNUtoo: I don't understand... my rhod behaves *exactly* like your dream, if I readd the pdsm_get_position loop, it kills A9 after 10seconds
14:53.01phhI just get nothing on smd27.
14:53.35GNUtooouch mmm
14:54.16phhperhaps I need to deinit what wince inited before
14:55.22phh(oh and when I say it behaves exactly the same way, even the values are the same.)
14:55.50*** join/#htc-linux jumoit1 (~samuel@121.77.87.145)
14:58.21GNUtoohow do I chroot in android
14:58.27GNUtooit's not FSH standard
14:59.03phhchroot /android /system/bin/sh ?
14:59.23GNUtoochroot /android => chroot: can't execute '/bin/sh/': Not a directory
14:59.30GNUtoooops
14:59.46*** join/#htc-linux jumoit (~samuel@121.77.87.145)
14:59.48GNUtooI'll mount the whole android
14:59.57phhthat might help.
15:00.50GNUtoook works
15:06.47GNUtoophh, any idea on how to increase fix numbers?
15:07.11phhsome yes, but I'd like to work on gps for my rhod now :p
15:09.41GNUtooas soon as it builds I'll try your thoughts
15:10.19*** join/#htc-linux dekar (~dekar@e180228124.adsl.alicedsl.de)
20:19.14*** join/#htc-linux apt (ibot@rikers.org)
20:19.14*** topic/#htc-linux is Welcome to the http://htc-linux.org project! Please read the Wiki: http://htc-linux.org/wiki || Logs: http://irclog.netripper.com/?chan=htc-linux || This is no android support channel.
20:20.40chamonixphh: testing zImage.20100401_202405 now, kernel 03.04 tested negative
20:21.22phhchamonix: negative to what ?
20:21.23hfctsaotG2g ill check ToAsTcfh logs when I get home
20:21.28phhdon't you got back your old battery life ?
20:21.57chamonix15 hours
20:22.24chamonixyes I did with the latest but it was supposed to be better so I'm till looking for the commit that made it worse
20:22.46chamonixs/till/still
20:24.22*** join/#htc-linux jonpry (~jonpry@184.78.140.199)
20:24.31chamonixphh: btw I broke my ubuntu as planned, and repaired it while learning a few new things :)
20:24.42phhlol
20:40.32jonpryphh, i was looking at your test.c. it is different from htcraphael-gps, and gpsdriver.  Like after calling act(), they call pa_reg, init(extra_reg), and act() again
20:43.04phhjonpry: I know
20:43.13phhwait what ?
20:43.20phherr no I don't know that one
20:43.35phh(next time I'll read full sentence.)
20:43.57jonpryso you understand? or do i need to explain?
20:44.15phhI understand
20:44.38phhI've to check who is on which clientid
20:44.52jonpryyour doing that on purpose?
20:45.13phhiirc yes
20:45.15phhbut I have to recheck
20:46.17jonpryi would like to import remote_apis_clnt.dll. In theory it just uses a librpc like interface. XDR_Sndint and the like. But when it first loads if accesses memory at 0xE12...., and jumps to whatever address is located there!
20:46.45phhjonpry: I do like htcraphael-gps, init, pd_reg, ext_reg, pd_act
20:46.52phhjonpry: easy one
20:46.59phhmore or less.
20:47.02phhjonpry: is it on a BX R3 ?
20:48.36jonpryBX LR
20:48.47phhjonpry: somewhere the librpc/oncrpc stuff, gives an array of available functions
20:48.54phhah. don't know that one
20:49.37jonpryso ONCRPC is loaded to 0xe12 or something?
20:49.52phhsounds stupid.
20:50.05phhbut perhaps yes
20:50.20jonprythey couldn't just use the linker?
20:50.30phh:p
20:50.37phhthat's why it sounds stupid to me
20:50.44phhi'm not really sure here :/
20:50.57*** join/#htc-linux dcordes (~dccordes@unaffiliated/dcordes)
20:51.43jonpryin theory the dll's extracted from rom are not relocatable. the extractor attempts to infer linkage somehow, but it may not be working
20:52.40jonpryI'm using ROMExactractor.exe, and there is no source. seems to make bad sections in the PE
20:53.48Parad0XUAdcordes: any luck with hd2 kernel panic?
20:53.52jonpryit would be cool if we could find out the memory map of all dll's in rom. then recreate it in the loader so there are no problems
20:54.12dcordesParad0XUA, no. you wanna try something?
20:54.21Parad0XUAdcordes: sure :)
20:59.11*** join/#htc-linux kiozen (~oeichler@rgnb-5d87cc50.pool.mediaWays.net)
20:59.20*** join/#htc-linux [acl] (~abel@cpe-69-203-141-229.si.res.rr.com)
21:06.09ToAsTcfh\o/
21:13.08*** join/#htc-linux Neo31 (~Neo31@unaffiliated/neo31)
21:14.43dcordesNetRipper, ping
21:14.48NetRipperpong
21:15.06dcordesNetRipper, wow that was fast :D did you read about the leo initrd location thing ?
21:15.14dcordeshi btw ^^
21:15.20NetRipperlol i just came on coincidental :)
21:15.39NetRipperi think i did.. it's configured in haret
21:15.55NetRipperthere's a fixed offset
21:16.05NetRipperlike kernel image + 5MB then initrd starts
21:16.29NetRipperkernel image base, + 5MB i mean
21:18.32*** join/#htc-linux M3DLG (~M3DLG@bb-87-81-252-83.ukonline.co.uk)
21:18.43NetRipper#define PHYSOFFSET_INITRD (PHYSOFFSET_KERNEL + 0x500000)
21:18.59dcordesNetRipper, can we add a switch to move it 5mb further for leo?
21:19.05NetRippersure
21:19.10NetRipperwell
21:19.25NetRipperwe can just compile a specific version for leo
21:19.33NetRipperdo you have a kernel > 5mb then?
21:20.18NetRipperit's not configurable per-arch atm
21:22.06NetRipperdcordes, what are these "interesting findings" you talk about on twitter on the codeaurora branch?
21:22.32dcordesseen the markinus branch?
21:22.46NetRipperyes
21:23.00NetRipperthat doesnt answer the question :p
21:23.15dcordeswe need answers
21:23.26dcordeswhen I asked him markinus said he didn't see any panics
21:23.35dcordesin the CA kernel
21:23.38NetRipperok
21:23.58NetRipperi think the panics are coming from drivers not compatible with wince devies
21:23.59dcordesI couldn't try it myself yet because of the screwed bandwidth
21:24.04NetRipperaha
21:24.06NetRipperstill on dialup?
21:24.19dcordesyes dialup speed via hsdpa connection
21:24.23NetRipperah
21:24.42NetRipperyea they wont like it if you download >500mb per month
21:24.42NetRipper:p
21:24.55dcordeswell almost. 1000
21:25.04NetRipperoh better than here then
21:25.12dcordesdepends on how much you pay...
21:25.19NetRipperyea true
21:25.24NetRipperi got the cheapest
21:25.38dcordesParad0XUA, can you pastebin the patch?
21:25.44dcordesParad0XUA, so we can add it in haret in some way?
21:26.13Parad0XUAthe haret patch?
21:26.13dcordesyes
21:26.24dcordesNetRipper, is haret maintainer
21:26.37NetRipperwhat's the haret patch for?
21:26.45dcordesbrb
21:26.58Parad0XUAto run codeaurora kernel on hd2
21:27.09Parad0XUAcause it's almost 6MB
21:27.14NetRipperahh ok the size issue
21:27.17Parad0XUA#define PHYSOFFSET_INITRD (PHYSOFFSET_KERNEL + 0x1000000)
21:27.22Parad0XUAmaybe make it 7 not 10
21:27.37NetRipper0x10 = 16mb
21:27.58Parad0XUA0x50 = ?
21:28.05NetRipper5 mb
21:28.13Parad0XUA0XF0 = ?
21:28.18NetRipper15mb
21:28.24Parad0XUAconfused
21:28.28NetRipperhehe
21:29.54NetRipperi see, 5.1 mb kernel
21:29.56NetRipperthats huge
21:30.09NetRipperguess it has lots of debugging stuff or something
21:30.31Parad0XUAstack debugging is on
21:30.45Parad0XUAactually a lot of debugging flags are on
21:31.05NetRipperyea that would explain
21:31.11Parad0XUANetRipper: explain me this :)
21:31.18Parad0XUAecho used greatest stack depth: 6252 left
21:32.47Parad0XUAirc://chat.freenode.net/#define PHYSOFFSET_INITRD (PHYSOFFSET_KERNEL + 0x780000)
21:32.51Parad0XUAhow about that?
21:33.09NetRipperyea i just set it to 0xA
21:33.10NetRipper10mb
21:33.21NetRipperbut i think im not going to commit that
21:33.34dcordesNetRipper, I was just going to ask if you see problems for other devices
21:33.44Parad0XUAyeah I guess you shouldn't
21:33.51NetRipperi dont know, it might yes.. with limited memory
21:34.00NetRipperso i was thinking of making a variable that can override this offset
21:34.01NetRipperruntime
21:34.15Parad0XUAyeah that'd be the best
21:34.49dcordessounds good
21:35.33dcordesNetRipper, maybe we should start a discussion on haret ml
21:36.00dcordesNetRipper, so kevin etc can review
21:36.05NetRipperhm i got the same panic that i got when i tried booting android 0.8 on nexus kernel
21:36.26NetRipperyea i'll have to check if it is really possible first
21:36.39NetRipperthe var idea
21:36.50NetRipper(there's a reason it's currently a #define)
21:37.05dcordesyes others could also have thoughts on this..
21:37.10dcordesso why does it work for Parad0XUA ?
21:37.24NetRippereh? what works?
21:37.32NetRipperthis patch also works for me if you mean haret
21:37.42Parad0XUAhd2 has enough memory ;)
21:37.51dcordesI mean <NetRipper> hm i got the same panic that i got when i tried booting android 0.8 on nexus kernel
21:38.01dcordesParad0XUA, you also get the panic?
21:38.05NetRipperah yes but this just means the kernel panic might be coming from elsewhere
21:38.09Parad0XUAyup
21:38.17dcordesok
21:38.22NetRipperah you also tried the v0.8 initrd?
21:38.23NetRipper:)
21:38.42Parad0XUAdo you have a good initrd and system image I could use for testing?
21:39.10NetRipperi have an angstrom image and an v0.8 android image, both of which i used on raphael
21:39.17NetRippernot sure if they're 100% ok for leo
21:39.30Parad0XUAeven if they're 75% ok can I have them? :)
21:39.38dcordeshehehe
21:39.55NetRipper[   11.967862] BUG: sleeping function called from invalid context at include/linux/pagemap.h:323
21:39.59NetRipper[   12.024014] in_atomic(): 0, irqs_disabled(): 128, pid: 1, name: init-real
21:40.01dcordesNetRipper, I've also been wondering if the old builds could cause problems
21:40.33NetRipperdcordes, theoretically it could.. which is why i built a new busybox with the same toolchain as i build the kernel with
21:40.43NetRipperalthough it should be backwards compatible
21:41.17dcordesit would be most bullet proof to use some 'official' qsd8 builds for testing I think
21:41.33dcordeslike stuff from nexusone images
21:43.13jonpryphh, know anything about pdsm_atl_l2_proxy_reg?
21:43.35Parad0XUANetRipper so where can I get those initrd and system images?
21:43.52NetRipperParad0XUA, give me a moment
21:43.56NetRipperParad0XUA, only initrd btw
21:44.14NetRipperthe android v0.8 fits on an initrd so to not bother with sdcard yet
21:44.28NetRipperbut i see the CA kernel doesn't have msm_fb, which is required for android
21:44.53dcordeslol?
21:46.08NetRipperwhats funny?
21:46.37dcordesthat ca kernel lacks msm_fb
21:46.50NetRipperwell its not enabled anyway, i havent checked for the existence
21:47.05Parad0XUAcool
21:47.41Parad0XUAwait what? I thought Marknus merged that in today or yesterday
21:47.56Parad0XUAeclair_caramel branch
21:49.27dcordesmsm_fb ?
21:50.27NetRippermsm_fb files are there yes
21:50.31NetRipperit's not enabled i guess
21:50.40NetRipperwhen i boot i dont see it switch anyway ;)
21:50.53NetRipperi screwed up my angstrom usb image i think
21:50.59Parad0XUAohnoes :(
21:51.22Parad0XUAwhich config are you using?
21:51.26NetRippernot just now..
21:51.31NetRipperhtcleo_defconfig
21:52.03Parad0XUAit's not in eclair_caramel :\
21:52.05Parad0XUAweird
21:57.08dcordesNetRipper, can you upload some haret-bigkernel ?
22:00.36NetRipperyes
22:00.37NetRippersec.
22:01.06NetRipperParad0XUA, http://www.netripper.com/leo/initrds/ currently one initrd, the android one
22:01.31NetRipperParad0XUA, you'll need a patched haret to use it (otherwise you'll get a warning message about too large initdr)
22:01.49NetRipperdcordes, Parad0XUA: http://www.netripper.com/leo/haret-pre-0.5.3-20100430_233234.exe
22:02.03NetRipperallows for 10mb kernel + dunno what amount exactly for initrd
22:02.14NetRipperdidnt calculate it
22:02.21Parad0XUANetRipper thanks!
22:03.53NetRippernp i wish i had more time to play
22:03.55NetRipper:)
22:05.02Parad0XUAthat's no game!
22:05.12NetRippersure it is :)
22:05.32NetRipperor do people's life rest on linux for hd2? :p
22:05.42Parad0XUAof course it does!
22:05.49NetRipperhehe
22:06.00Parad0XUAI dunno.. I'm so excited about having android on hd2 for some reason
22:06.12*** join/#htc-linux slavenode (~knoppix@93-34-144-236.ip50.fastwebnet.it)
22:06.24NetRipperwe're having many more weird problems than we had on raphael
22:06.25NetRipperit's annoying :P
22:07.00Parad0XUAhow did you guys get into kernel hacking? is that something you do for living? or a hobby?
22:09.06NetRipperhobby for me
22:09.16NetRipperstarted when i owned the raphael
22:10.18Parad0XUAyeah same here... but I'm thinking of maybe doing this for living :) that'd be fun
22:11.29dcordesParad0XUA, you work in IT field?
22:11.32dcordesor CS
22:11.39NetRipperCS?
22:11.51Parad0XUAcomputer science
22:11.54Parad0XUAweb developer
22:11.56NetRipperah
22:12.18Parad0XUAfrom php straight to hardcore kernel C :)
22:12.28AstainHellbringhey NetRipper!
22:12.34NetRipperhi astain
22:12.50AstainHellbringNetRipper did I give you the passion_c rom?
22:12.55NetRipperyes
22:13.03NetRipperhaven't sen cr2 yet
22:13.20AstainHellbringk
22:13.26AstainHellbringforgetful sometimes
22:13.29NetRipper:)
22:14.04NetRipperis cr2 on holiday or something?
22:14.05NetRipper:p
22:17.40dcordesNetRipper, does the uploaded haret have #define PHYSOFFSET_INITRD (PHYSOFFSET_KERNEL + 0x780000) ?
22:17.52dcordesNetRipper, nobody knows. it's getting scary
22:18.00NetRipper0xA, so 10mb
22:18.07dcordesok
22:20.52Parad0XUANetRipper using your initrd and no luck :(
22:21.28Parad0XUAok back to where we started I guess :)
22:21.46dcordesthe whole kernel panic situation isn't documented anywhere
22:22.35Parad0XUANetRipper so this image actually works on some other device?
22:22.42NetRipperthis initrd works on raphael yes
22:22.58NetRipperhttp://wiki.xda-developers.com/index.php?pagename=RaphaelLinux
22:22.58dcordesmsm7201A
22:23.04NetRipperyes
22:23.21NetRipperdoesnt work on any qsd device yet (sory if you meant that)
22:23.29NetRipperraphael is msm7x00a
22:23.35NetRipper7x01a
22:23.48dcordesNetRipper, do you think we should we mention leo panic on ml?
22:23.48Parad0XUAinteresting... why kernel is in that context when it runs init
22:23.59Parad0XUAwhat's ml?
22:24.02NetRippermailing list
22:24.04NetRipperdcordes, which?
22:24.09dcordesmobile-linux
22:24.17dcordesdon't know the link but it is in the wiki
22:24.33NetRipperis it always the same kernel panic?
22:24.42dcordeshttp://htc-linux.org/wiki/index.php?title=Contact
22:24.43NetRipperi get a different kernel panic on CA tree than on nexus tree
22:24.54dcordesinteresting
22:24.56Parad0XUAnope it's same exact one I think
22:25.00dcordesdidn't boot any CA yet ..
22:25.23Parad0XUAcomming from get_signal_to_deliver
22:25.28NetRipperyea but that's pretty generic
22:25.43Parad0XUAright
22:25.53NetRipperdoabort is handled when a processor exception is caught
22:25.57dcordesthe panic I get on nexusone codebase is attempting to kill init - because some binary in the rootfilesystem dies
22:26.18NetRipperprobably because init dies
22:26.22dcordeswith a segmentation fault
22:26.45dcordeshm yes but sometimes also other things get started
22:26.51Parad0XUAoh btw!
22:27.00Parad0XUAI think today android actually started to boot on my leo :)
22:27.14Parad0XUAwas trying to load the image
22:27.34dcordesNetRipper, but do you think we should post it?
22:27.53dcordesNetRipper, because if so it might be better if you could set it off. you have a better understanding of the failure mechanism
22:28.58Parad0XUAwho's on that list?
22:29.06Parad0XUAany of the google, arm, htc guys?
22:29.20dcordesno with google guys it's interesting story
22:29.30dcordesat the very beginning they gave us hints but guess they changed policy
22:29.47dcordesthey being the guys working on the early msm7k dev boards
22:30.14Parad0XUAdon't they want to help? :\
22:30.21Parad0XUAI'm sure most of our issues is no big deal for them
22:30.22NetRipperwe havent asked them, really
22:30.31dcordesmaybe they don't want to lose their jobs
22:30.52Parad0XUAwell yeah :) only if they're not allowed to help
22:31.00Parad0XUAbut that's the whole point of open source etc
22:31.02dcordesif we discuss everything on the list we can point to the archives
22:31.15dcordesor also refer to it on lkml...
22:31.41Parad0XUAwell people can always create some fake emails on chinese email servers :)
22:31.59NetRipperdcordes, you can post it but i think its rather specific
22:31.59dcordesyes but android is no true foss not community driven no gpl project
22:32.01dcordesetc blabla
22:32.51dcordesParad0XUA, well I think nobody is stopping you them from email brian swetland, san mehat or whoever is working on qsd and hope for a chinese mail
22:33.05Parad0XUA:-D
22:33.08Parad0XUAlol
22:33.14*** join/#htc-linux AlienCat (~Cat@c-136-14-vas-l3.cust.mdfnet.se)
22:33.28Parad0XUAbut has anyone tried emailing them?
22:33.38dcordesnot me (not regarding qsd)#
22:36.14dcordes~seen swetland
22:36.37aptswetland <n=swetland@nat/google/x-f87b65f208eda3f2> was last seen on IRC in channel #htc-linux, 444d 18h 16m 24s ago, saying: 'engineering verification test'.
22:36.54Parad0XUAheh
22:36.56Parad0XUA~seen zavin
22:36.56apti haven't seen 'zavin', Parad0XUA
22:37.26dcordes~seen sanmehat
22:37.27aptsanmehat <n=san@nat/google/x-87831aaa828ad288> was last seen on IRC in channel #htc-linux, 356d 8h 41m 56s ago, saying: 'yeah'.
22:37.41Parad0XUA~seen Parad0xUA
22:37.41aptparad0xua is currently on #htc-linux. Has said a total of 63 messages. Is idling for 0s, last said: '~seen Parad0xUA'.
22:37.42dcordesbirthday
22:37.46*** join/#htc-linux ali12341 (~al@robotfuzz.co.uk)
22:37.53Parad0XUAnice :)
22:39.37dcordesNetRipper, ok I will try to describe the problem on the list. by rather specific, you mean device specific?
22:41.08NetRipperwell so far, yes.. there are no other devices we're working on
22:41.21jonpryanyone know if this was made using HaRET http://pastebin.com/ZeQEzn8V and more importantly how to make one on TP2?
22:41.22NetRipperdevice + kernelspecific probably
22:41.45dcordesdriver shutdown problem
22:41.55IceBoneAnyone else experiencing really poor quality of the contact images in android?
22:41.56dcordesI've heard that similar times
22:41.59dcordesseveral
22:42.44dcordesIceBone, yes I'm just accepting them
22:42.52IceBoneApparently it wasn't always so.
22:42.59*** join/#htc-linux Kensan (~ken@mail.codelabs.ch)
22:43.04IceBoneThey get screwed up in the sync or something.
22:43.05Parad0XUAok guys have a great weekend... I'm taking off
22:43.47dcordesIceBone, the problem is if you don't like it you are fucked. because the community has no voice in the androed project (well only the one marketing people are processing) or you write a patch
22:44.11dcordesParad0XUA, thanks you too
22:44.24IceBonedcordes: if I choose the photo from the high quality photo I have on the phone, crop it, it looks great, until the sync kicks in.
22:45.09dcordesNetRipper, do you have some ramconsole with panic handy?
22:45.43dcordesNetRipper, some dump so we have something to append
22:46.26IceBoneDamn, google maps is nice on wifi + pinch to zoom.
22:48.19*** join/#htc-linux LTxda (~anon@unaffiliated/ltxda)
22:48.48dcordesNetRipper, I'm also off. maybe we can discuss it tomorrow
22:49.07NetRipperno i dont have log handy
22:50.12stinebdToAsTcfh: have you seen this? http://jaysonking.com/bfs-2.6.27-backport/
22:50.42stinebdapplies with fuzz on our kernel
22:51.57ToAsTcfhi couldnt get it to apply
22:52.25ToAsTcfhit had stuff from .30 in it and i didnt wanna backport that stuff too
22:52.42NetRipperim off as well, night :)
22:53.01ToAsTcfhstinebd: fuzz?
22:53.24stinebdToAsTcfh: no rejects, but offsets needed to patch
22:53.44ToAsTcfhah tru which one did u use
22:53.53stinebdall those patches must be applied
22:54.02ToAsTcfhoh i see
22:54.25*** join/#htc-linux enjalot_ (~enjalot@c-68-35-253-116.hsd1.fl.comcast.net)
22:54.29ToAsTcfhyeah i just used the original bfs .31 patch and backported it to .27
22:55.37stinebdi'll try it out later
22:55.44stinebdassuming it builds
22:56.52ToAsTcfhyeah the original patch for me applied clean but i had to work the magic to get it to compile
23:07.11*** join/#htc-linux jumoit (~samuel@121.77.87.145)
23:24.56*** join/#htc-linux [acl] (~abel@cpe-69-203-141-229.si.res.rr.com)
23:26.10*** join/#htc-linux ali12341 (~al@robotfuzz.co.uk)
23:29.24obrienmdhave you all been seeing significant performance gains w/ bfs on .27?
23:29.34obrienmdor "user experience" I guess?
23:32.09stinebdhaven't tried it yet
23:32.17stinebdbuild failed, so i brought our kernel up to 2.6.27.46
23:32.27stinebdit's building now
23:32.34stinebdwe'll see how she does
23:34.33obrienmdcool
23:38.10ToAsTcfhstinebd how did u bring the kernel up
23:38.27ToAsTcfhmerge
23:38.28stinebdthe patch
23:38.53ToAsTcfhoh there is a patch for that lol i didnt know
23:39.10ToAsTcfhu want my patch?
23:39.14ToAsTcfhfor bfs
23:39.37stinebdnah
23:39.40ToAsTcfhok
23:40.14ToAsTcfh<PROTECTED>
23:40.25stinebdthere are a few manual merges to worry about
23:40.45stinebdseems like someone was cherry picking some security commits
23:40.50ToAsTcfhyeah
23:41.01ToAsTcfhform .30
23:41.08ToAsTcfhfrom
23:41.28ToAsTcfhat least compared to gentoo kernels
23:41.53ToAsTcfhi wish it were easier to move from .27 to .29
23:44.27ToAsTcfhi was talking to con about my patch and asking him to review it. he replies with this lol "If it works, it works. Schedulers tend to just explode if something is wrong
23:44.28ToAsTcfh:) I'll take a quick look but I doubt I'll see anything obvious."
23:44.36ToAsTcfhthats one cool cat
23:46.12ToAsTcfhidk but i guess the issues some had with bfs in the past have been fix in builds after 300
23:46.37ToAsTcfhim at 316. its a easy merge after 314
23:56.41obrienmdhow long does build take on your sys stinebd? also, where did last build attempt fail?
23:57.29stinebdobrienmd: no idea how long, last attempt failed with a bad merge in sysctl.c
23:57.35stinebdit's into the net core now
23:58.29stinebdi don't really expect any failures anymore
23:58.50ToAsTcfhdid it compile sched_bfs yet?
23:59.00stinebdbut i'm building a version without bfs before i build one with it
23:59.09ToAsTcfhah tru
23:59.12ToAsTcfhnm
23:59.18stinebdthat way i have something to test with so i know if any problem is in my 2.6.27.46 merge or the bfs merge
23:59.35ToAsTcfhwhere did u find that merg
23:59.42stinebdwhich?
23:59.51ToAsTcfh.46
23:59.57ToAsTcfhgoogle
23:59.59ToAsTcfh?

Generated by irclog2html.pl Modified by Tim Riker to work with infobot.