| 00:07.42 | *** join/#htc-linux surge (surge@pool-98-118-157-221.bflony.fios.verizon.net) |
| 00:09.35 | *** join/#htc-linux furtardo (~mks@nat/yahoo/x-zqkpcbamlfgxjldt) |
| 00:12.58 | *** join/#htc-linux hardwalker (~hardwalke@122-117-115-146.HINET-IP.hinet.net) |
| 00:22.48 | *** join/#htc-linux AstainHellbring (AstainHell@unaffiliated/astainhellbring) |
| 00:25.05 | *** join/#htc-linux D3tul3 (~Oliver@cpe-174-109-223-235.nc.res.rr.com) |
| 00:51.58 | *** join/#htc-linux jonpry (~jon@199.85.236.239) |
| 01:06.23 | *** join/#htc-linux ORi| (~ORi@91.181.137.104) |
| 01:59.25 | *** join/#htc-linux SG1 (~SG@cpe-76-186-50-40.tx.res.rr.com) |
| 02:02.55 | *** join/#htc-linux kslen^^ (~idkfa@static243-203-166.mimer.net) |
| 02:03.57 | *** join/#htc-linux tmzt_ (~tmzt@adsl-99-164-51-188.dsl.akrnoh.sbcglobal.net) |
| 02:04.04 | *** join/#htc-linux toer (~tore@179.81-166-86.customer.lyse.net) |
| 02:04.43 | *** join/#htc-linux KungFunk (~cow@84-55-99-121.customers.ownit.se) |
| 02:09.13 | *** join/#htc-linux arrrghhh (~arrrghhh@unaffiliated/arrrghhh) |
| 02:11.44 | *** part/#htc-linux SG1 (~SG@cpe-76-186-50-40.tx.res.rr.com) |
| 02:34.21 | *** join/#htc-linux Termana (~bradley@122.151.74.100) |
| 02:38.29 | *** join/#htc-linux SG_ (~SG@cpe-76-186-50-40.tx.res.rr.com) |
| 02:46.25 | *** join/#htc-linux programmer8922 (~Evan@D-128-95-134-61.dhcp4.washington.edu) |
| 03:04.23 | *** join/#htc-linux exo-squad (~exo@24.166.164.63) |
| 03:14.47 | *** join/#htc-linux NeoMatrixJR (~chatzilla@173-20-63-176.client.mchsi.com) |
| 03:48.49 | *** join/#htc-linux Rob2222 (~Miranda@p4FFF3F37.dip.t-dialin.net) |
| 03:52.54 | *** part/#htc-linux exo-squad (~exo@24.166.164.63) |
| 04:00.07 | *** join/#htc-linux dhd (5b7639bc@gateway/web/freenode/ip.91.118.57.188) |
| 04:04.02 | dhd | could anybody help me porting meego to the htc desire hd? I've already a meego rootfs and a zImage, copied it to the sdcard and now i'm trying to start it.. but i only fall back to recovery mode, so what's the best way to debug this? |
| 04:07.33 | arrrghhh | with a hammer |
| 04:18.10 | *** join/#htc-linux programmer8922 (~Evan@D-128-95-134-61.dhcp4.washington.edu) |
| 04:40.28 | *** join/#htc-linux bzo (~chatzilla@c-76-126-175-200.hsd1.ca.comcast.net) |
| 04:49.43 | *** join/#htc-linux Andreyxxl[HD2EU] (~Andreyxxl@94.52.236.39) |
| 05:18.45 | jonpry | bzo, have you tried the 39 kernel? |
| 05:19.18 | bzo | nah, barely touched the phone lately |
| 05:20.02 | jonpry | its good stuff. breaths life into the beast |
| 05:20.20 | jonpry | as long as you don't want bluetooth |
| 05:20.20 | bzo | should give it a try, can drop into an exiting system/data right? |
| 05:20.42 | bzo | well, bt was barely usable anyways |
| 05:20.48 | jonpry | i added acoustic yesterday so you need a compatible system.img |
| 05:21.04 | bzo | I'm using hyc's, so that ought to work |
| 05:21.11 | jonpry | indeed |
| 05:21.31 | jonpry | need a large kernel haret and big initrd offset |
| 05:21.45 | jonpry | http://autobuild.wirelesstcp.net/ |
| 05:22.36 | bzo | I'll make like a user and check out arrrghhh's thread on xda, lol |
| 05:23.02 | jonpry | you need a user for that? |
| 05:23.32 | bzo | I mean act like a regular user |
| 05:24.54 | jonpry | you'd have to start running on about how you can't get it to work |
| 05:29.18 | bzo | ok, booting now |
| 05:29.48 | bzo | hey what's with the readable console font |
| 05:30.41 | jonpry | haven't managed to change it |
| 05:31.13 | jonpry | moving to 2x4 binary rsn |
| 05:31.23 | bzo | it wouldn't do for people to actually be able to read what is on the screen |
| 05:31.34 | jonpry | not at all |
| 05:32.24 | jonpry | once it settles down. the graphics are super fast |
| 05:32.34 | jonpry | which rhod do you have? |
| 05:32.40 | bzo | 400 |
| 05:33.00 | jonpry | supposedly cdma devices are working now |
| 05:33.55 | jonpry | networking was just too fast and it would blow up the virtual memory system |
| 05:35.03 | bzo | is that the fix you and hyc were working on, the smd copy stuff? |
| 05:35.09 | jonpry | yeah |
| 05:35.22 | *** join/#htc-linux BHSPitMonkey (~stephen@unaffiliated/bhspitmonkey) |
| 05:35.38 | bzo | everything seems to be working for me, looks good :) |
| 05:36.07 | jonpry | he made it faster. i originally got it to stop trying to stuff kilobytes into some stupid self resizing buffer |
| 05:37.44 | bzo | 39 doesn't do anything with the boot overclock param right? |
| 05:38.18 | jonpry | i think it will if you change it to acpuclock_a11.????? |
| 05:38.38 | jonpry | er acpuclock-arm11 |
| 05:39.53 | bzo | being at the stock 528 on 39 may even be more responsive that 768 on 27 |
| 05:40.26 | jonpry | yeah its pretty fast |
| 05:40.53 | bzo | the thing really noticeable with the faster fb is that you can actually see the screen decelerating on scroll |
| 05:41.07 | *** join/#htc-linux emwe (5040bc35@gateway/web/freenode/ip.80.64.188.53) |
| 05:41.20 | jonpry | i like all the animations |
| 05:41.43 | emwe | jonpry: that missing microp ctrl over the tpa and adc seems to drain some power. wen't down to 11% over night... |
| 05:42.11 | bzo | hi emwe |
| 05:42.18 | emwe | hi bzo |
| 05:43.06 | bzo | I don't think I can go back to 27 now |
| 05:43.26 | emwe | thanks jonpry :P |
| 05:43.58 | jonpry | :) |
| 05:44.15 | jonpry | so 90% drain in 8 hours? |
| 05:44.25 | emwe | alex and me really need a marketing department for .35 |
| 05:44.32 | jonpry | lol |
| 05:44.52 | jonpry | superior goods are always easier to sell |
| 05:45.04 | bzo | seriously though, why split our work on all these different kernel versions? |
| 05:45.08 | emwe | jonpry: yeah, either's acoustic (or some component of it) because it will constantly suspend/resume after playing something with earphones on topa.... or the missing microp ctl over the audio chips |
| 05:45.32 | emwe | bzo: that's what it looks like. |
| 05:45.54 | jonpry | could be h2w. i haven't tried turning it on |
| 05:46.07 | emwe | i think so too |
| 05:46.12 | emwe | when i just play sound via speaker, all fine |
| 05:46.24 | emwe | btu i can't see yet on rhod. no sleep lights.... |
| 05:46.31 | emwe | sure it is though. |
| 05:46.40 | bzo | emwe: is there anything about 35 that you think is better than 39? |
| 05:46.51 | jonpry | if you turn on the tpa gpio it will suck power. but you can hear that |
| 05:49.09 | emwe | bzo: i haven't tried .39. should likely for a comparison. can't judge. what is likely different is the code. alex and me (but really mostly alex) did pull some stuff apart. that per board config and such... so from a code pov it might look cleaner on .35 and we did not copy .27 code over to the new version like some others did.... :P |
| 05:49.19 | emwe | but look how fast it got up |
| 05:50.36 | jonpry | 35 is a lot of new code. 39 has its share, but nothing architectural really |
| 05:50.43 | bzo | emwe: think it would be a lot of work to migrate the clean 35 stuff over to 39? |
| 05:51.35 | emwe | bzo: once it's all done, no, don't think. just need someone to sit down and bisect mmc and rfkill. i am still wanting to write some microp led abstraction |
| 05:51.49 | *** join/#htc-linux mitsutaka (~mitsutaka@61.173.132.169) |
| 05:51.50 | emwe | microp-klt is really a pita to look at. |
| 05:52.28 | bzo | the microp stuff is a mess |
| 05:52.32 | emwe | ofc, it just works and that's likely jons motivation to just take it over to get things done. |
| 05:52.48 | emwe | and i really can understand that motivation. these devices are not getting younger. |
| 05:52.55 | emwe | there's hardly some activity for topa already. |
| 05:52.58 | jonpry | the 27 htc-linux written code is terribly well debugged |
| 05:53.09 | emwe | it's terrible well been butchered. |
| 05:53.26 | emwe | ah sry, "htc-linux written" |
| 05:53.32 | emwe | then yeah. |
| 05:53.45 | jonpry | i won't say its pretty |
| 05:54.47 | jonpry | mostly i just think you can't rewrite the platform stuff and switch kernel versions at the same time |
| 05:55.38 | emwe | yah, that's holding back a bit. but it's just a matter of implemenation now. |
| 05:55.47 | bioterror | http://www.youtube.com/watch?v=27x8ih9NWGg Nokia N9 MeeGo |
| 05:55.54 | bioterror | announced |
| 05:59.00 | jonpry | gotta sleep. cya |
| 05:59.05 | emwe | bye jonpry |
| 06:00.08 | emwe | so looks like i gotta have to try that .39 |
| 06:00.55 | bzo | it does perform noticeably better than 27 |
| 06:01.11 | emwe | .35 too imho |
| 06:02.03 | emwe | as .39 relies on acoustic userland... i think it's time to push some unpolished binaries... |
| 06:02.53 | bzo | yeah, seems like we should just get it over with |
| 06:08.07 | *** join/#htc-linux mitsutaka (~mitsutaka@61.173.132.93) |
| 06:17.01 | *** join/#htc-linux kiozen (~kiozen@p5DDF099F.dip.t-dialin.net) |
| 06:56.02 | *** join/#htc-linux rpierce99 (~rpierce99@c-98-237-190-180.hsd1.wa.comcast.net) |
| 07:08.55 | *** join/#htc-linux dobrin (~dobrin@85.91.150.26) |
| 07:11.05 | *** join/#htc-linux Termana_ (~bradley@122.151.67.54) |
| 07:23.34 | *** join/#htc-linux Andreyxxl[HD2EU] (~Andreyxxl@94.52.236.39) |
| 07:24.10 | *** join/#htc-linux gauner1986 (~Miranda@46.115.18.247) |
| 07:29.45 | *** join/#htc-linux netson-hd10 (~netson-hd@202.159.118.21) |
| 07:30.24 | *** join/#htc-linux gauner19861 (~Miranda@46.115.18.247) |
| 07:54.42 | *** join/#htc-linux HardDisk_WP (~marco@wikipedia/harddisk) |
| 07:54.45 | *** join/#htc-linux LordDeath (~LordDeath@cable-81-173-166-52.netcologne.de) |
| 08:09.57 | *** join/#htc-linux arif-ali (~arif-ali@ip-81-23-53-226.ask4internet.com) |
| 08:18.01 | *** join/#htc-linux balans (~barry@82-170-217-205.ip.telfort.nl) |
| 08:20.34 | *** join/#htc-linux gauner1986 (~Miranda@p5B383FE9.dip.t-dialin.net) |
| 08:30.53 | *** join/#htc-linux arif-ali (~arif-ali@ip-81-23-53-226.ask4internet.com) |
| 09:08.17 | *** join/#htc-linux emwe (5040bc35@gateway/web/freenode/ip.80.64.188.53) |
| 09:09.57 | *** join/#htc-linux Segnale007 (~Segnale00@ppp-201-16.26-151.libero.it) |
| 09:12.32 | *** join/#htc-linux stroughtonsmith (~steven@86-44-83-169-dynamic.b-ras2.bbh.dublin.eircom.net) |
| 09:18.43 | *** join/#htc-linux helicopter88 (~helicopte@host37-31-dynamic.26-79-r.retail.telecomitalia.it) |
| 09:31.56 | *** join/#htc-linux Cass (~Cass@nat/sun/x-nvtdyemozpvkungn) |
| 09:59.47 | *** join/#htc-linux Segnale007 (~Segnale00@ppp-201-16.26-151.libero.it) |
| 10:29.44 | *** join/#htc-linux Ceesheim (~ceesheim@p5B347A5E.dip.t-dialin.net) |
| 10:34.45 | *** join/#htc-linux GNUtoo (~GNUtoo@host148-3-dynamic.244-95-r.retail.telecomitalia.it) |
| 11:03.51 | *** join/#htc-linux D3tul3 (~Oliver@cpe-174-109-223-235.nc.res.rr.com) |
| 11:51.51 | *** join/#htc-linux kiozen (~kiozen@p5DDF099F.dip.t-dialin.net) |
| 12:46.02 | *** join/#htc-linux MethoS- (~clemens@134.102.106.250) |
| 12:48.35 | *** join/#htc-linux helicopter88 (~helicopte@host37-31-dynamic.26-79-r.retail.telecomitalia.it) |
| 13:04.56 | *** join/#htc-linux arif-ali (~arif-ali@ip-81-23-53-226.ask4internet.com) |
| 13:07.20 | *** join/#htc-linux nyl (IceChat09@109.96.18.104) |
| 13:08.06 | *** join/#htc-linux Berger (~noreply@c65wn1.wifi.halden.net) |
| 13:14.38 | *** join/#htc-linux GNUtoo (~GNUtoo@host148-3-dynamic.244-95-r.retail.telecomitalia.it) |
| 13:37.14 | *** join/#htc-linux helicopter88 (~helicopte@host37-31-dynamic.26-79-r.retail.telecomitalia.it) |
| 13:42.08 | *** join/#htc-linux dekar (~dekar@drms-4d01457f.pool.mediaWays.net) |
| 13:52.12 | *** join/#htc-linux jonpry (~jon@199.85.236.239) |
| 14:13.05 | *** join/#htc-linux rzk_ (~rzk@95-25-113-117.broadband.corbina.ru) |
| 14:20.40 | *** join/#htc-linux emwe (~emwe@cable-86-56-10-158.cust.telecolumbus.net) |
| 14:21.29 | *** join/#htc-linux Cass (~Cass@nat/sun/x-dijphtzsaazctbkb) |
| 14:35.08 | *** join/#htc-linux programmer8922 (~Evan@D-128-95-134-61.dhcp4.washington.edu) |
| 14:36.24 | *** join/#htc-linux mastermerlin (~Adium@p4FEE4988.dip.t-dialin.net) |
| 14:36.29 | *** join/#htc-linux mastermerlin1 (~Adium@p4FEE4988.dip.t-dialin.net) |
| 14:37.26 | *** join/#htc-linux rajkosto (~rajkosto@cable-94-189-158-244.dynamic.sbb.rs) |
| 14:37.57 | *** join/#htc-linux |lippa| (~lippa@ppp118-209-176-109.lns20.mel6.internode.on.net) |
| 14:48.13 | *** join/#htc-linux toer (~tore@179.81-166-86.customer.lyse.net) |
| 15:09.10 | *** join/#htc-linux AstainHellbring (~AstainHel@unaffiliated/astainhellbring) |
| 15:17.17 | *** join/#htc-linux dekar (~dekar@drms-4d01457f.pool.mediaWays.net) |
| 15:44.11 | *** join/#htc-linux rpierce99 (~rpierce99@c-98-237-190-180.hsd1.wa.comcast.net) |
| 15:53.44 | *** join/#htc-linux GlemSom (~glemsom@0x5da34bca.cpe.ge-1-1-0-1105.sdnqu1.customer.tele.dk) |
| 15:55.54 | *** join/#htc-linux VALERA (5f43c43d@gateway/web/freenode/ip.95.67.196.61) |
| 16:13.08 | *** join/#htc-linux |Jeroen| (~jeroen@d5152B25B.access.telenet.be) |
| 16:20.52 | *** join/#htc-linux fish1209 (~fish1209@unaffiliated/fish0912) |
| 16:23.24 | *** join/#htc-linux arrrghhh (~arrrghhh@unaffiliated/arrrghhh) |
| 16:27.27 | *** join/#htc-linux |lippa| (~lippa@ppp118-209-72-198.lns20.mel4.internode.on.net) |
| 16:32.23 | *** join/#htc-linux joplayer (~jonathan@AClermont-Ferrand-256-1-59-149.w90-10.abo.wanadoo.fr) |
| 16:35.15 | *** join/#htc-linux Echo31 (~olivier@mir31-4-82-240-194-54.fbx.proxad.net) |
| 16:35.48 | *** join/#htc-linux rpierce99 (~rpierce99@c-98-237-190-180.hsd1.wa.comcast.net) |
| 16:56.00 | *** join/#htc-linux joplayer (~jonathan@AClermont-Ferrand-256-1-75-164.w90-10.abo.wanadoo.fr) |
| 16:56.56 | *** join/#htc-linux rpierce99 (~rpierce99@c-98-237-190-180.hsd1.wa.comcast.net) |
| 17:06.53 | *** join/#htc-linux joplayer (~jonathan@AClermont-Ferrand-256-1-80-109.w90-10.abo.wanadoo.fr) |
| 17:11.30 | *** join/#htc-linux Andreyxxl[HD2EU] (Andreyxxl@94.52.236.39) |
| 17:18.27 | *** join/#htc-linux rpierce99 (~rpierce99@c-98-237-190-180.hsd1.wa.comcast.net) |
| 17:23.26 | *** join/#htc-linux onen|openBmap (~quassel@vbo91-1-89-87-201-85.dsl.sta.abo.bbox.fr) |
| 17:26.21 | *** join/#htc-linux arif-ali (~arif-ali@ip-81-23-53-226.ask4internet.com) |
| 17:29.36 | *** join/#htc-linux technologeek (~jonathan@AClermont-Ferrand-256-1-12-174.w86-209.abo.wanadoo.fr) |
| 17:45.35 | *** join/#htc-linux Alex[sp3dev] (~alexander@ip-95-220-1-205.bb.netbynet.ru) |
| 17:46.37 | *** join/#htc-linux rajkosto (~rajkosto@cable-94-189-158-244.dynamic.sbb.rs) |
| 17:48.30 | arrrghhh | jonpry, i noticed the IO % on .39 under top is much higher than usual. |
| 17:48.48 | arrrghhh | trying to figure out where this slowness is coming from, i don't see any process stealing the show. |
| 17:49.27 | arrrghhh | see the horror |
| 17:49.27 | arrrghhh | http://i.imgur.com/48J1a.png |
| 17:49.51 | Alex[sp3dev] | Windoze on my ircs? NO WAI |
| 17:50.03 | arrrghhh | lol |
| 17:50.13 | arrrghhh | i'm at work, sorry to muck up the place. |
| 17:50.33 | Alex[sp3dev] | idle 59%? you're 12309ed |
| 17:50.59 | arrrghhh | oh i thought that % was IO |
| 17:51.02 | arrrghhh | that's idle %? wth. |
| 17:51.08 | arrrghhh | why is it so freakin slow. |
| 17:51.22 | Alex[sp3dev] | really, this is io |
| 17:51.36 | arrrghhh | ... |
| 17:51.39 | arrrghhh | is confused |
| 17:52.09 | arrrghhh | the top updates are surprisingly fast |
| 17:52.10 | Alex[sp3dev] | maybe you need iotop? |
| 17:52.17 | arrrghhh | but the rest of the actual system on the phone is painful |
| 17:52.31 | arrrghhh | hrm |
| 17:52.36 | arrrghhh | our builds don't seem to have that... |
| 17:52.58 | Alex[sp3dev] | maybe it's easier to install debian in chroot to have all gnu software. |
| 17:53.03 | arrrghhh | heh |
| 17:53.09 | arrrghhh | sounds easy |
| 17:53.12 | arrrghhh | :P |
| 17:54.22 | Alex[sp3dev] | arrrghhh: like that http://img202.imageshack.us/img202/8426/20110621215327731x484sc.png |
| 17:54.35 | arrrghhh | nice |
| 17:54.57 | arrrghhh | heh i don't even have iotop on my server @ home. |
| 17:55.08 | Alex[sp3dev] | this is my laptop |
| 17:55.18 | arrrghhh | just installed it on the server. |
| 17:55.19 | arrrghhh | very cool |
| 17:56.19 | arrrghhh | this would be nice to have in Android |
| 17:56.33 | Alex[sp3dev] | i think it needs python crap |
| 17:56.39 | arrrghhh | bleh |
| 17:57.04 | arrrghhh | it only installed iotop_0.4-1_all.deb on my server, but that doesn't mean i don't already have the poop it needs on there. |
| 18:00.07 | *** join/#htc-linux NeoMatrixJR (~chatzilla@173-20-63-176.client.mchsi.com) |
| 18:01.02 | arrrghhh | between io and sys, the phone never gets to do much. |
| 18:01.27 | *** join/#htc-linux |lippa| (~lippa@ppp118-209-168-170.lns20.mel6.internode.on.net) |
| 18:03.59 | Detule | i seem to be idling at around 90% with connectbot good for another ~4% |
| 18:04.52 | Detule | .39 must not like something about your setup |
| 18:05.04 | arrrghhh | i dunno |
| 18:05.07 | arrrghhh | it randomly screams |
| 18:05.10 | arrrghhh | but mostly drags. |
| 18:05.16 | arrrghhh | this is kinda how .27 is for me too tho |
| 18:05.23 | arrrghhh | although it never screams :P |
| 18:05.29 | *** join/#htc-linux helicopter88 (~helicopte@host37-31-dynamic.26-79-r.retail.telecomitalia.it) |
| 18:05.32 | arrrghhh | and the dragging doesn't seem as dramatic either... very strange. |
| 18:15.36 | *** join/#htc-linux gauner1986 (~Miranda@p5B383FE9.dip.t-dialin.net) |
| 18:17.38 | *** join/#htc-linux swc|666 (~x00@unaffiliated/swc666/x-4934821) |
| 18:19.09 | Detule | oh iotop looks nice.... |
| 18:19.30 | arrrghhh | it does indeed |
| 18:40.01 | jonpry | iotop sounds like a good idea |
| 18:42.02 | jonpry | honestly i have no idea what io time is |
| 18:42.39 | arrrghhh | hrm |
| 18:42.55 | jonpry | but it seems like you are out of memory, and kswapd is hammering the card |
| 18:43.00 | arrrghhh | i have no idea why .39 is so slow, then randomly fast, then slow again. it's mostly slow. |
| 18:43.01 | arrrghhh | hrm |
| 18:43.05 | arrrghhh | i guess that could be possible |
| 18:43.15 | arrrghhh | feel like shoehorning more RAM into my TP2? :P |
| 18:43.58 | jonpry | ramzswap maybe |
| 18:44.33 | arrrghhh | 14mb of RAM free? |
| 18:45.10 | jonpry | system server is using 130.7% of ram |
| 18:45.17 | jonpry | 223mb |
| 18:45.19 | arrrghhh | Oo |
| 18:45.34 | arrrghhh | that's not the case now, i didn't notice that in the screenie |
| 18:45.56 | arrrghhh | hrm maybe it is still the case |
| 18:45.59 | arrrghhh | damn system_server |
| 18:47.42 | *** join/#htc-linux mitsutaka (~mitsutaka@61.173.132.200) |
| 18:58.39 | *** join/#htc-linux NeoMatrixJR (~chatzilla@173-20-63-176.client.mchsi.com) |
| 19:04.23 | *** join/#htc-linux Segnale007 (~Segnale00@ppp-201-16.26-151.libero.it) |
| 19:04.43 | jonpry | i wonder if 27 has similar numbers for system_server |
| 19:05.17 | arrrghhh | jonpry, any ideas on that new haret causing reboots to WinMo? |
| 19:05.28 | arrrghhh | i asked the guy to get ramconsole logs, not sure it'll happen or not. |
| 19:05.41 | arrrghhh | basically dude gets the doublevibe, so i think that means the kernel is getting loaded... |
| 19:05.42 | jonpry | probably put the initrd offset at the end of the file or something |
| 19:05.50 | jonpry | pretty sure its a startup.txt |
| 19:05.52 | arrrghhh | i checked his startup.txt |
| 19:05.57 | *** join/#htc-linux kiozen (~kiozen@ppp-93-104-89-57.dynamic.mnet-online.de) |
| 19:06.15 | arrrghhh | initrd offset is in a good place, and the value is correct. |
| 19:06.18 | jonpry | maybe he's not copying the kernel correctly. had 0 byte kernels or something |
| 19:06.24 | arrrghhh | hrm... |
| 19:06.25 | arrrghhh | ok |
| 19:07.14 | arrrghhh | he claims the same thing happens on old kernels too |
| 19:07.22 | arrrghhh | which, i haven't had a problem booting an old kernel on the new haret... |
| 19:07.32 | jonpry | me neither |
| 19:07.45 | arrrghhh | hrm |
| 19:07.53 | arrrghhh | well i guess we'll see if the dude can get some logs. |
| 19:08.01 | jonpry | if there is no difference between kernels. the kernel is likely not being changed |
| 19:08.08 | arrrghhh | he's literally the only person who has reported issues booting .39 |
| 19:08.21 | arrrghhh | ironically he has a 210 |
| 19:08.27 | arrrghhh | so i told him it can't be the device :P |
| 19:10.16 | jonpry | he can't boot 27 either :p |
| 19:10.58 | arrrghhh | lol |
| 19:11.01 | arrrghhh | on the new haret, no. |
| 19:11.06 | arrrghhh | on old haret... no problems evidently. |
| 19:11.59 | jonpry | best thing i can think of is to dd a card image |
| 19:12.23 | arrrghhh | sounds painful |
| 19:13.06 | jonpry | its either that or ramconsole. not sure of any other way to help |
| 19:13.40 | arrrghhh | indeed |
| 19:13.43 | arrrghhh | i suggested ramconsole |
| 19:13.50 | arrrghhh | probably should've walked him thru it, bleh |
| 19:14.06 | jonpry | if it vibes then there should be some data |
| 19:14.23 | Detule | i guess these new kernels carry this io information themselves -> supposedly enabling "config TASKSTATS" in init/Kconfig writes in /proc/$PID/io |
| 19:15.16 | *** join/#htc-linux gauner1986 (~Miranda@p5B383FE9.dip.t-dialin.net) |
| 19:15.44 | jonpry | you can see from the regular top that system_server is doing a little something which has got kswapd, loop0 and mmcqd going |
| 19:17.15 | arrrghhh | just kill all of those processes. |
| 19:17.59 | jonpry | :) |
| 19:18.15 | *** join/#htc-linux ORi| (~ORi@91.181.29.198) |
| 19:19.04 | *** join/#htc-linux bzo (~chatzilla@c-76-126-175-200.hsd1.ca.comcast.net) |
| 19:19.09 | jonpry | one of these days need to try CONFIG_SCHED_AUTOGROUP |
| 19:20.26 | bzo | jonpry: interesting thing is that 39 doesn't bench any faster on fps2d |
| 19:20.38 | bzo | however, the variance is much lower |
| 19:20.52 | bzo | getting a consistent 30fps instead of jumping way low and high |
| 19:20.57 | jonpry | no benchmarks seem to show any real improvement |
| 19:21.06 | jonpry | except maybe download speed |
| 19:21.27 | bzo | I think the perceived fb perf is much better because it is consistent |
| 19:21.44 | arrrghhh | HRM |
| 19:21.47 | arrrghhh | oops. hrm. |
| 19:21.48 | jonpry | feels much smoother |
| 19:22.05 | bzo | yeah, the 27 one spends like half the time at 15fps |
| 19:22.37 | *** join/#htc-linux cr2_ (~cr2@46.115.19.156) |
| 19:22.41 | bzo | the graph showing the variance is very tight on 39, looks like a regular distribution graph |
| 19:22.51 | arrrghhh | nice |
| 19:23.02 | arrrghhh | bzo, do you get random odd slowdowns at all? |
| 19:23.36 | bzo | maybe - haven't spent a lot of time on 39 yet |
| 19:23.40 | arrrghhh | ok |
| 19:23.55 | arrrghhh | i keep blaming my SD card, but i've tried a few different ones and they all more or less have the same result. |
| 19:23.59 | jonpry | i think your just out of memory |
| 19:24.01 | arrrghhh | none of the cards are particularly new tho |
| 19:24.05 | bzo | but when you really tax it, like with angry birds or something, it seem maybe worse than 27 |
| 19:24.07 | *** join/#htc-linux dekar (~dekar@drms-4d01457f.pool.mediaWays.net) |
| 19:24.10 | Willd | arrrghhh: Have you checked the sd card cache? |
| 19:24.11 | arrrghhh | jonpry, but why wouldn't i be out of memory on .27 as well? |
| 19:24.18 | arrrghhh | Willd, explain |
| 19:24.21 | Willd | Sec |
| 19:24.23 | arrrghhh | k |
| 19:24.39 | arrrghhh | gonna go make some slop for lunch, brb. |
| 19:24.51 | jonpry | i am not sure. maybe there is a memory leak caused by a missing driver or some random failure |
| 19:24.58 | *** join/#htc-linux GNUtoo (~GNUtoo@host148-3-dynamic.244-95-r.retail.telecomitalia.it) |
| 19:25.02 | Willd | arrrghhh: http://forum.xda-developers.com/showthread.php?t=1010807 |
| 19:26.07 | bzo | the nook guys seem really enthused by the sdcard cache setting |
| 19:26.15 | bzo | there is an app on market that will set it as well |
| 19:26.28 | jonpry | hyc has it set to 2MB |
| 19:26.32 | Detule | my guess is arrrghhh is on hyc's build which incorporates readahead |
| 19:27.03 | Willd | Hm |
| 19:27.55 | bzo | nook guys have also done a lot of testing on sdcards |
| 19:28.08 | bzo | I switched to sandisk because it is the only one with ok random write performance |
| 19:28.19 | bzo | it is literally 100x faster |
| 19:28.32 | Willd | bzo: Yikes :) |
| 19:28.40 | cr2_ | ~ping NetRipper |
| 19:28.40 | apt | pong NetRipper |
| 19:28.51 | bzo | that is not an indication that sandisk is fast, only that others are horrible :) |
| 19:28.57 | jonpry | arrrghhh's 35MB of cache seem really low. the dex processes seem to use massive amount of memory but most of it is just malloc'd files. such large really used memory is unusual |
| 19:28.59 | bzo | .01mb/s vs 1-2mb/s |
| 19:29.11 | jonpry | er mmap'd files |
| 19:29.53 | bzo | anyways, the sandisk card seems to reduce a lot of the random slowdowns |
| 19:30.00 | bzo | and perhaps 39 is more sensitive to it right now |
| 19:31.15 | bzo | hmm, quadrant benches .39@768mhz as fast as a nexus one (pre 2.2) |
| 19:31.34 | arrrghhh | hrm |
| 19:31.37 | arrrghhh | i have a sandisk card |
| 19:31.44 | arrrghhh | it's an old 'classless' card tho |
| 19:31.55 | arrrghhh | the only 'classed' card i have is a c4 kingston. |
| 19:32.18 | bzo | arrrghhh: run crystaldiskmark on them, 50mb test 5x |
| 19:32.56 | arrrghhh | hrm |
| 19:32.57 | arrrghhh | ok |
| 19:33.18 | jonpry | bzo is it overclocking with the -arm11? |
| 19:33.26 | bzo | yep |
| 19:33.37 | arrrghhh | just oddness. .27 has random slowdowns, but they're bearable. the slowdowns in .39 are crippling. |
| 19:33.42 | cr2_ | kiozen: ping |
| 19:34.05 | kiozen | cr2_: pong |
| 19:35.34 | Detule | jonpry: hm so is .39 compiled with TASK_IO_ACCOUNTING and TASKSTATS enabled |
| 19:36.46 | cr2_ | kiozen: i've used the magic xor7 decryption, and find out that they use http://de.wikipedia.org/wiki/MicroC/OS-II for parsing NMEA and writing to SD on the logger ;) |
| 19:37.31 | cr2_ | kiozen: this os is not free, but "open source" |
| 19:38.09 | kiozen | and it's for safety relevant uses |
| 19:38.31 | kiozen | thus you can safely build an atomic bomb with it |
| 19:38.36 | cr2_ | it should be easy to hack |
| 19:38.39 | cr2_ | lol |
| 19:39.11 | cr2_ | there are other free rtos around, why did they use this one.. hmm |
| 19:40.14 | cr2_ | kiozen: still better than grmn and iblue747 |
| 19:40.25 | jonpry | Detule, neither of those. accounting doesn't exist and it says taskstats is to export the stats over netlink |
| 19:40.56 | arrrghhh | bzo, still running the later tests but 19.2 read and 10.9 write |
| 19:41.41 | bzo | the first one doesn't matter, the last 2 random write tests are what we want to see |
| 19:41.50 | arrrghhh | oh... :P |
| 19:42.17 | arrrghhh | so seq & 512k don't matter |
| 19:42.20 | arrrghhh | 4k and 4k qd32 do? |
| 19:42.41 | bzo | well, the seq matters if you need high throughput like recording HD video |
| 19:42.41 | *** join/#htc-linux NeoMatrixJR (~chatzilla@173-20-63-176.client.mchsi.com) |
| 19:42.52 | Detule | jonpry: oh i guess i was misled by line 330 in init/Kconfig in your tree |
| 19:43.02 | bzo | but for OS/App use bad random write screws things up |
| 19:43.07 | arrrghhh | i'd just like Android to run without me wanting to throw my phone out a window to improve speed. |
| 19:43.14 | fakker | lol |
| 19:43.30 | *** join/#htc-linux dobrin (~dobrin@85.91.150.26) |
| 19:44.21 | *** join/#htc-linux xawen (~xawen@pool-173-64-109-129.bltmmd.fios.verizon.net) |
| 19:45.06 | arrrghhh | holy hell. |
| 19:45.22 | arrrghhh | these numbers are starting to make me understand the madness lol |
| 19:46.31 | bzo | ? |
| 19:47.18 | *** join/#htc-linux cr2_ (~cr2@46.115.19.156) |
| 19:47.24 | arrrghhh | so far it's a 0.017 random writes |
| 19:47.34 | arrrghhh | lol |
| 19:48.23 | arrrghhh | random read isn't bad, 3.5mb/s |
| 19:48.28 | bzo | you probably have a fake sandisk |
| 19:48.37 | arrrghhh | noes this is my c4 kingston |
| 19:48.38 | arrrghhh | i started with it |
| 19:48.51 | bzo | ah, that's expected then |
| 19:49.15 | arrrghhh | crap, i only have one other card with me |
| 19:49.21 | arrrghhh | and it doesn't have a manufacturer name. impressive. |
| 19:49.38 | arrrghhh | eh i'll still give it a whirl |
| 19:49.49 | arrrghhh | 0.021 was the random write on the second test. atrocious. |
| 19:51.32 | rajkosto | 200 KILOBYTES A SECOND ! |
| 19:51.35 | rajkosto | wait no |
| 19:51.37 | rajkosto | 20 KILOBYTES |
| 19:51.39 | rajkosto | HOLY SHIT BATMAN |
| 19:51.40 | arrrghhh | lol |
| 19:51.51 | arrrghhh | i think i can type faster than that |
| 19:52.47 | arrrghhh | hrm. this classless card got .1 better on read. odd. |
| 19:53.21 | arrrghhh | lol, 0.013 random write. damn knockoff chinese crap. |
| 19:53.36 | *** join/#htc-linux dobrin (~dobrin@85.91.150.26) |
| 19:53.39 | arrrghhh | i know i have a sandisk card, just not here @ work evidently. |
| 19:54.05 | bzo | yeah, .01 random writes are typical of just about all sd cards |
| 19:54.20 | arrrghhh | wow. |
| 19:54.23 | arrrghhh | all but sandisk eh? |
| 19:54.36 | arrrghhh | class doesn't matter, it's the manufacturer? |
| 19:54.38 | NetRipper | cr2_, pong |
| 19:54.40 | bzo | they seem to all be optimized for sequential perf, just like old SSDs |
| 19:54.47 | NetRipper | i saw your patch |
| 19:54.51 | NetRipper | will process it soon |
| 19:54.57 | jonpry | random write = kswapd |
| 19:55.00 | cr2_ | NetRipper: thanks |
| 19:55.05 | bzo | lower class seem to have better random write perf |
| 19:55.15 | bzo | but I've only seen sandisk have good random perf |
| 19:55.18 | arrrghhh | hrm |
| 19:55.36 | jonpry | isn't the class supposed to be the minimum write performance? |
| 19:55.44 | bzo | only sequential |
| 19:55.51 | Echo31 | hello cr2_ |
| 19:55.53 | bzo | which is probably why they ignore random - not marketable |
| 19:56.12 | arrrghhh | yea this card ain't doin too much better. i'll see what the sandisk card scores, and if it is quite a bit better i'll try using Android on it for a few days. |
| 19:56.20 | arrrghhh | perhaps it's time to buy a new card too :/ |
| 19:56.36 | bzo | and really for typical use, like a digital cam or playing/recording music/movies, strong sequential is usually better |
| 19:56.53 | cr2_ | hi Echo31 |
| 19:57.10 | cr2_ | Echo31: any news on athena ? |
| 19:57.19 | arrrghhh | yea |
| 19:57.28 | cr2_ | CONFLICT (add/add): Merge conflict in linux-2.6.35.4/drivers/usb/gadget/isp1583_udc.h |
| 19:57.32 | arrrghhh | no many hooligans are trying to run an OS off of their SD cards. |
| 19:57.35 | cr2_ | Echo31: pulling your latest tree |
| 19:57.36 | arrrghhh | s/no/not/ |
| 19:57.51 | Echo31 | cr2_ : I pushed the epson driver |
| 19:58.28 | *** join/#htc-linux helicopter88 (~helicopte@host232-118-dynamic.2-79-r.retail.telecomitalia.it) |
| 19:58.36 | Echo31 | cr2_ the polling mode work. I have some issue for dma mode |
| 19:59.12 | arrrghhh | yea, that classless/nameless card almost scored identically. random read speed was actually slightly higher. |
| 19:59.46 | arrrghhh | gonna go to the store and open all the SD packaging. test out the cards before i buy lol |
| 19:59.51 | cr2_ | Echo31: what is left to be done ? |
| 20:00.34 | Echo31 | cr2_: I created a new partition on the microdrive of Athena. I can boot on the rootfs. |
| 20:00.44 | bzo | arrrghhh: good luck with the blister packaging :P |
| 20:00.52 | arrrghhh | razor blade FTW |
| 20:00.58 | arrrghhh | i'll borrow one of those while i'm at it. |
| 20:01.05 | jonpry | Detule, i see the accounting option now :), but it still says its exported through netlink |
| 20:01.18 | cr2_ | Echo31: wow, that sounds really good. |
| 20:01.47 | cr2_ | Echo31: how did you repartition it ? |
| 20:02.31 | *** join/#htc-linux nyl (me@109.96.18.104) |
| 20:03.38 | Echo31 | cr2_: for the ata driver, in dma mode, the data from disk send successful to ide fifo. The pxa dma channel shows a bus error |
| 20:04.21 | *** join/#htc-linux noobUK (~UKmrrd32@cpc6-haye18-2-0-cust91.haye.cable.virginmedia.com) |
| 20:04.26 | Detule | jonpry: damn i thought it just write something accessible to /proc/$PID |
| 20:04.34 | *** part/#htc-linux noobUK (~UKmrrd32@cpc6-haye18-2-0-cust91.haye.cable.virginmedia.com) |
| 20:04.54 | *** join/#htc-linux NeoMatrixJR (~chatzilla@173-20-63-176.client.mchsi.com) |
| 20:04.59 | jonpry | i guess its possible |
| 20:05.19 | jonpry | i can enable and push it |
| 20:05.39 | cr2_ | Echo31: weird. i need to try it. |
| 20:05.53 | Echo31 | cr2_: You mount the microdrive as a mass storage with WM5storage on Wince |
| 20:06.31 | Detule | since i am probably the only one that cares how about i pull the kernel and compile it myself |
| 20:06.55 | cr2_ | Echo31: will fdisk from ramdisk work ? |
| 20:07.00 | *** join/#htc-linux dobrin (~dobrin@85.91.150.26) |
| 20:07.51 | Echo31 | cr2_ : after you create a new partition EXT3. You modify the default.txt of haret to read sda2. |
| 20:08.02 | jonpry | Detule, i don't even have a phone atm. |
| 20:08.17 | cr2_ | Echo31: ok |
| 20:08.28 | Detule | ha, no worries.... |
| 20:09.47 | Echo31 | cr2_: On the Wince, with regedit (commander), you unmount the second partition, because wince can overwrite with the fat format. |
| 20:09.54 | *** join/#htc-linux Berger (~noreply@c65wn1.wifi.halden.net) |
| 20:10.24 | cr2_ | Echo31: but it will ask afair |
| 20:10.57 | cr2_ | Echo31: one more stupid question: does gps work ? |
| 20:12.35 | Echo31 | cr2_: does not work. The message from uart is not understanding. Very strange |
| 20:14.18 | cr2_ | ok |
| 20:15.46 | cr2_ | Echo31: about the dma bus error. are you doing byte transfers ? |
| 20:16.46 | cr2_ | i forgot the details, but there was some BE , and 16bit issue |
| 20:17.55 | Echo31 | cr2_. : I use the values of register from wince dump of ide |
| 20:18.15 | Echo31 | cr2_: 16 BE |
| 20:20.03 | cr2_ | Echo31: hm, but does the pxa dma controller support it ? |
| 20:21.49 | Echo31 | cr2_ The pxa dma can use the width (8,16,32) |
| 20:22.27 | cr2_ | BE too ? |
| 20:22.40 | cr2_ | or do you flip the bytes ? |
| 20:23.50 | *** join/#htc-linux NeoMatrixJR (~chatzilla@173-20-63-176.client.mchsi.com) |
| 20:24.00 | Echo31 | cr2_ : the default mode is little endian |
| 20:24.21 | cr2_ | afair BE ? |
| 20:25.12 | cr2_ | but it should not cause a bus error, though |
| 20:26.42 | Echo31 | cr2_: the chipselect register gives 8bit 16 BE or 16 strobe but Wince set on 16 BE |
| 20:27.05 | cr2_ | ok |
| 20:29.02 | Echo31 | cr2_ I used pxa gnu-haret to analyse the behaviour |
| 20:32.29 | cr2_ | ok |
| 20:33.18 | Echo31 | cr2_ : The bus error is often trigged by the a wrong alignment or burst value |
| 20:34.03 | cr2_ | who needs to fix the alignment ? |
| 20:34.10 | cr2_ | there should be a kernel api for that |
| 20:35.26 | Echo31 | cr2_ Normally, it is ok, because I used the values from wince |
| 20:36.04 | cr2_ | ok |
| 20:36.43 | Echo31 | cr2_ But many values from DMA and IDE registers cannot be read |
| 20:37.21 | cr2_ | this is strange |
| 20:37.25 | cr2_ | buggy hardware ? |
| 20:37.41 | cr2_ | there must be a reason ms uses BE |
| 20:38.03 | *** join/#htc-linux Raymonddull (~raymonddu@ip174-69-65-80.ri.ri.cox.net) |
| 20:38.29 | Echo31 | cr2_ a few registers are Write only and others can be locked by the DMA |
| 20:38.51 | cr2_ | wince probably caches WO regs |
| 20:40.38 | Echo31 | cr2_ the polling mode ATA_PIO4 works |
| 20:41.39 | cr2_ | ok |
| 20:41.51 | *** join/#htc-linux Cass (~Cass@188-220-34-222.zone11.bethere.co.uk) |
| 20:43.29 | Echo31 | cr2_: I installed as rootfs the image hawkboard from angstrom repository |
| 20:44.35 | cr2_ | Echo31: can you write it down in wiki ? |
| 20:44.44 | Echo31 | cr2_ : after, you can add the gpe and x11 by opkg. |
| 20:45.06 | Echo31 | cr2_ : If you want. |
| 20:45.30 | cr2_ | i think it's a good idea in general |
| 20:50.11 | Echo31 | cr2_ : The 4-1 cable is mandatory to establish the ethernet connection. Isp1583 does not work. |
| 20:51.01 | cr2_ | ethernet ? |
| 20:51.17 | cr2_ | btw, afair wince does not use epson dma too |
| 20:51.37 | cr2_ | argh, where is my 4-1 cable ... |
| 20:51.51 | Echo31 | cr2_ : I used usb/ethernet adaptator |
| 20:54.18 | Echo31 | cr2_: You can use directly usb wire by usbnet if you have this device. |
| 20:56.47 | Echo31 | cr2_ the epson uses the DMA channel (channel 2 in Wince) |
| 20:57.23 | cr2_ | ok |
| 20:57.35 | cr2_ | can't find the 4-1 cable |
| 21:01.08 | cr2_ | probably i'd also take apart the 1:1 cable |
| 21:02.31 | Echo31 | cr2_: It is important to success the isp1583 work to avoid to use 4-1 cable |
| 21:03.25 | cr2_ | yeah |
| 21:03.51 | cr2_ | can't find the sl5500 serial cable too |
| 21:04.45 | *** join/#htc-linux HardDisk_WP (~marco@velirat.de) |
| 21:04.45 | *** join/#htc-linux HardDisk_WP (~marco@wikipedia/harddisk) |
| 21:12.52 | Echo31 | cr2_ : On Wince, the Gpio1 (isp1583) triggers. Surely, a missed gpio/register is needed for linux. I have the problem with tps65021 regulator. The read command on the tps65021 version register by i2c send a error ! |
| 21:15.09 | cr2_ | omg, found sl5500 cable :) |
| 21:15.25 | cr2_ | Echo31: are you sure it's tps65021 ? |
| 21:15.50 | cr2_ | the video out chip is not what is documented in wiki (but still very similar) |
| 21:20.45 | cr2_ | heh. where have i lost the 4-1 cable??? |
| 21:25.18 | Echo31 | cr2_ The tps65021 ( similar tps65023) is the regulator with client usb power |
| 21:28.29 | cr2_ | does the i2c id match ? |
| 21:28.44 | cr2_ | because the video i2c is does not match the datasheet in wiki |
| 21:29.16 | cr2_ | s/ is / id / |
| 21:29.39 | *** join/#htc-linux MethoS- (~clemens@134.102.106.250) |
| 21:34.17 | Echo31 | cr2_: the other handheld devices use tps65023 with 0x48 id |
| 21:37.16 | *** join/#htc-linux T-Junk (C7xx@f147035.upc-f.chello.nl) |
| 21:38.47 | cr2_ | which id do we have ? |
| 21:39.27 | *** join/#htc-linux NeoMatrixJR (~chatzilla@173-20-63-176.client.mchsi.com) |
| 21:41.00 | *** join/#htc-linux programmer8922 (~Evan@D-173-250-148-24.dhcp4.washington.edu) |
| 21:44.40 | Echo31 | cr2: I set 0x48 id.. i 2c_check_functionality is successed |
| 21:48.57 | cr2_ | Echo31: but it should be traceable in wince |
| 21:56.57 | Echo31 | cr2_: Perhaps. Will you be able to recover these values? |
| 22:08.03 | *** join/#htc-linux FlawlesStyle (~LOL@unaffiliated/flawlesstyle) |
| 22:10.41 | cr2_ | Echo31: yes |
| 22:11.09 | cr2_ | but need to trigger some event causing the tp communication |
| 22:11.21 | cr2_ | hm, without the 4-1 cable ;) |
| 22:12.20 | *** join/#htc-linux Cass (~Cass@188-220-34-222.zone11.bethere.co.uk) |
| 22:13.00 | Echo31 | cr2_: ok |
| 22:13.16 | *** join/#htc-linux MACK1E (~MACK1E@50-88-143-107.res.bhn.net) |
| 22:25.25 | *** join/#htc-linux cr2_ (~cr2@46.115.19.156) |
| 22:34.06 | *** join/#htc-linux tyween (~typhoon@pool-71-163-117-57.washdc.fios.verizon.net) |
| 22:52.50 | *** join/#htc-linux Segnale007 (~Segnale00@ppp-201-16.26-151.libero.it) |
| 23:21.51 | *** join/#htc-linux D3tul3 (~Oliver@cpe-174-109-223-235.nc.res.rr.com) |
| 23:48.29 | *** join/#htc-linux Detule (~Oliver@cpe-174-109-223-235.nc.res.rr.com) |
| 23:57.20 | *** join/#htc-linux mitsutaka (~mitsutaka@61.173.132.169) |
