| 00:10.17 | *** join/#htc-linux jumoit (n=samuel@121.77.87.145) |
| 00:20.53 | *** join/#htc-linux Unholy_ (n=IceChat7@24.227.27.77) |
| 00:25.10 | *** join/#htc-linux leviathan_ (n=quassel@c-82-192-226-27.customer.ggaweb.ch) |
| 00:25.10 | *** join/#htc-linux Ekips_ (i=spike@d51A460C9.access.telenet.be) |
| 00:25.10 | *** join/#htc-linux kupo (i=kupo@pinky.ratman.org) [NETSPLIT VICTIM] |
| 00:25.10 | *** join/#htc-linux polyrhythmic (n=charles@c-67-171-16-39.hsd1.wa.comcast.net) [NETSPLIT VICTIM] |
| 00:25.10 | *** join/#htc-linux bkero (n=bkero@ash.osuosl.org) [NETSPLIT VICTIM] |
| 00:25.10 | *** join/#htc-linux chamonix (i=chamonix@netburst.org) [NETSPLIT VICTIM] |
| 00:25.10 | *** join/#htc-linux NetRipper (n=netrippe@tikkie.net) [NETSPLIT VICTIM] |
| 00:25.10 | *** join/#htc-linux mrmoku|away (n=mrmoku@ppp-93-104-27-92.dynamic.mnet-online.de) |
| 00:25.11 | *** join/#htc-linux Foxdie (i=Foxdie@opencanvas.kicks-ass.net) |
| 00:25.11 | *** join/#htc-linux furtardo (n=mks@203.83.248.32) [NETSPLIT VICTIM] |
| 00:25.13 | *** join/#htc-linux TehCAPS (i=CAPS@86.124.242.20) [NETSPLIT VICTIM] |
| 00:25.13 | *** join/#htc-linux the_sys0p (n=the_sys0@cpe-76-93-251-206.bak.res.rr.com) [NETSPLIT VICTIM] |
| 00:25.13 | *** join/#htc-linux makkonen (n=makkonen@cpe-66-69-229-9.austin.res.rr.com) [NETSPLIT VICTIM] |
| 00:25.13 | *** join/#htc-linux Forsaken|Desk (n=Forsaken@dslb-084-058-244-232.pools.arcor-ip.net) [NETSPLIT VICTIM] |
| 00:30.21 | *** join/#htc-linux stinebd (n=bryan@maduin.southcape.org) [NETSPLIT VICTIM] |
| 00:38.51 | makkonen | for the record... I really want dropbear. being able to leave an ssh port open would feel way better than leaving my phone exposed to telnet all the time (minor reason). SCP/SFTP would also make my weird wifi debugging system way less annoying (major reason). |
| 00:40.36 | stinebd | i built a static version earlier for the hell of it, but haven't used it |
| 00:40.52 | *** join/#htc-linux DarkMasterHalo (n=DarkMast@modemcable187.118-80-70.mc.videotron.ca) |
| 00:41.08 | makkonen | you fool. you're sitting on a GOLD MINE. |
| 00:41.14 | stinebd | meh |
| 00:41.23 | makkonen | good point. |
| 00:41.50 | makkonen | yay, touchscreen calibration (sort of) works. |
| 00:42.39 | *** join/#htc-linux Forsaken|Desk1 (n=Forsaken@dslb-084-058-244-232.pools.arcor-ip.net) |
| 00:55.07 | makkonen | boo, touchscreen calibration app doesn't works. |
| 00:55.23 | stinebd | this is an emotional rollercoaster |
| 00:57.26 | makkonen | I bring the thrills, what can I say. |
| 01:02.41 | babijoee | stinebd> babijoee: latest bundle and updated system.sqsh works great, thanks <-------- thanks :) |
| 01:03.20 | MrPippy | hmm this presentation is interesting , about msm7k security http://www.ics.uci.edu/~dutt/cs295-sq09/mwei_CS%20295%20Presentation.pdf |
| 01:08.40 | *** part/#htc-linux DarkMasterHalo (n=DarkMast@modemcable187.118-80-70.mc.videotron.ca) |
| 01:08.43 | *** join/#htc-linux ghostdogg (n=ghost-do@cpe-76-179-206-39.maine.res.rr.com) |
| 01:13.39 | *** join/#htc-linux DarkMasterHalo (n=DarkMast@modemcable187.118-80-70.mc.videotron.ca) |
| 01:14.09 | *** join/#htc-linux M1DLGpc (n=M1DLGpc@bb-87-81-252-83.ukonline.co.uk) |
| 01:15.05 | stinebd | babijoee: fyi: http://gizmodo.com/5456573/how-to-circumvent-android-21s-word-filter |
| 01:15.08 | stinebd | :D |
| 01:18.09 | makkonen | god bless neil gaiman. |
| 01:24.03 | babijoee | XD |
| 01:24.26 | babijoee | hmmm |
| 01:24.34 | babijoee | might try android on nand now :) |
| 01:29.58 | dcordes | MrPippy, nice finding. wiki is cited there :) |
| 01:30.41 | dcordes | Immense amount of research done |
| 01:30.41 | dcordes | <PROTECTED> |
| 01:30.41 | dcordes | <PROTECTED> |
| 01:30.41 | dcordes | <PROTECTED> |
| 01:48.38 | *** part/#htc-linux MrPippy (n=pip@75.37.167.3) |
| 01:48.48 | *** join/#htc-linux MrPippy (n=pip@75.37.167.3) |
| 01:52.53 | *** join/#htc-linux Unholy_ (n=IceChat7@24.227.27.77) |
| 01:58.55 | *** part/#htc-linux DarkMasterHalo (n=DarkMast@modemcable187.118-80-70.mc.videotron.ca) |
| 02:00.45 | *** join/#htc-linux x29a_ (n=x29a@unaffiliated/x29a) |
| 02:01.44 | *** join/#htc-linux jeremychang_ (n=jeremych@61-30-10-70.static.tfn.net.tw) |
| 02:28.42 | tmzt | NetRipper: what's this about ati? |
| 02:28.52 | tmzt | you find how to do mdp on 8k? |
| 02:36.12 | *** join/#htc-linux BHSPitMonkey (n=stephen@unaffiliated/bhspitmonkey) |
| 02:36.36 | tmzt | maejrep: I told pershoot or whoever how to backport sta |
| 02:36.47 | tmzt | I don't know what the interest in that driver is though |
| 02:36.54 | maejrep | sta? |
| 02:37.00 | tmzt | sta- tiwlan |
| 02:37.02 | tmzt | sta_ |
| 02:37.16 | tmzt | no copy/paste |
| 02:54.58 | stinebd | hey, qualcomm has a pop an hour from here. i should try to get in and pilfer some documents |
| 02:59.24 | babijoee | hahah |
| 03:00.43 | *** join/#htc-linux jooprzol (n=jooprzol@ip68-0-126-178.tu.ok.cox.net) |
| 03:02.05 | MrPippy | hell i live in san diego, everything around here is qualcomm |
| 03:02.17 | stinebd | i live in eastern pa, everything around here is horse and carriage... |
| 03:02.18 | MrPippy | i think htc amss and driver source code would be more useful though |
| 03:20.42 | *** join/#htc-linux Forsaken|Desk (n=Forsaken@dslb-084-058-244-232.pools.arcor-ip.net) |
| 03:33.13 | makkonen | ok, who lives in taiwan? |
| 03:37.18 | tmzt | stinebd: yeah, no Amish carriers? or carriers are also livery related? |
| 03:42.17 | babijoee | stinebd: ping |
| 03:42.42 | stinebd | babijoee: pong |
| 03:43.53 | babijoee | i recieved a message from the bug and error webpage and phh says theres some progress on the camera branch |
| 03:44.19 | babijoee | would it be possible for you to compile it |
| 03:44.54 | babijoee | phh: Some work has been done in htc-msm-2.6.27-camera branch on gitorious. |
| 03:44.54 | babijoee | Now we can talk to the camera I2C device, and initialize the camera part of the modem. It seems that the VFE init is still lacking. (Needs to be confirmed though, it could be userland problem.) |
| 03:45.01 | stinebd | yup |
| 03:45.04 | stinebd | give me a few mins |
| 03:45.06 | babijoee | thanks :) |
| 03:53.03 | tmzt | everyone seen pdk.android.com ? |
| 03:54.22 | stinebd | tmzt: that's basically how i was able to do the fuze keyboard layout |
| 03:54.33 | stinebd | they have some pretty useful notes in there |
| 03:55.01 | tmzt | good |
| 03:58.53 | stinebd | babijoee: i'm gonna warn you, here. i can't boot to test this right now and i'm winging it on the kernel config, so if anything goes wrong we might have to wait for phh |
| 03:59.05 | babijoee | nps |
| 03:59.22 | stinebd | modules are building now, after it's done i'll upload it |
| 03:59.26 | babijoee | thanks |
| 03:59.36 | babijoee | hopefully magic happens |
| 04:01.49 | tmzt | what's with the config? |
| 04:03.41 | stinebd | babijoee: http://maduin.southcape.org/~bryan/android/kernel-2.6.27-00741-gb8dd4a0-camera.zip |
| 04:04.13 | stinebd | turned off a bunch of tuners and v4l1, whatever i needed to get it to build |
| 04:04.25 | stinebd | should work i guess |
| 04:04.30 | stinebd | it passes the compile test |
| 04:06.01 | tmzt | just disable compat probably |
| 04:06.12 | tmzt | or just enable what you need |
| 04:06.44 | stinebd | kconfig autoselects a bunch of tuners unless you explicitly tell it not to |
| 04:07.01 | stinebd | pretty sure those are entirely useless on these systems ;) |
| 04:07.26 | stinebd | and yes v4l1[-compat] doesn't build |
| 04:07.26 | tmzt | probably |
| 04:08.08 | babijoee | not working :( |
| 04:08.34 | tmzt | why are you expecting it to? |
| 04:09.26 | babijoee | :p so i can test |
| 04:09.29 | babijoee | but nvm |
| 04:10.42 | tmzt | what changed? |
| 04:11.10 | babijoee | wont even go into booting' just hangs after vibrate |
| 04:11.26 | babijoee | anyways there was some development for camera |
| 04:11.35 | babijoee | so i was testing out different camera apps and libs |
| 04:13.23 | stinebd | hmm |
| 04:13.27 | tmzt | kernel won't boot? |
| 04:13.33 | stinebd | maybe it's too big |
| 04:14.13 | tmzt | check the haret error log |
| 04:21.12 | stinebd | babijoee: try http://maduin.southcape.org/~bryan/android/kernel-2.6.27-00742-gc60433a-camera.zip |
| 04:24.23 | babijoee | same =/ |
| 04:24.29 | stinebd | oh well |
| 04:24.35 | babijoee | yeah nvm |
| 04:51.47 | *** join/#htc-linux ImCoKeMaN (n=imcokema@pool-71-251-145-207.hrbgpa.fios.verizon.net) |
| 04:56.52 | *** part/#htc-linux swc|666 (n=infidel2@unaffiliated/swc666/x-4934821) |
| 05:08.58 | *** join/#htc-linux droid0011 (n=g1@p4FDCA76F.dip.t-dialin.net) |
| 05:09.01 | *** join/#htc-linux goxboxlive (n=jrs@mail2.hjellnesconsult.no) |
| 05:44.57 | *** join/#htc-linux MrPippy_ (n=pip@adsl-75-37-167-3.dsl.sndg02.sbcglobal.net) |
| 05:45.44 | *** join/#htc-linux MrPippy_ (n=pip@adsl-75-37-167-3.dsl.sndg02.sbcglobal.net) |
| 05:51.47 | *** join/#htc-linux ghostdogg (n=ghostdog@cpe-76-179-206-39.maine.res.rr.com) |
| 06:02.42 | ghostdogg | anyone hanging out? |
| 06:03.17 | ghostdogg | sorry if that was a double comment, couldn't see the previous if it went through |
| 06:06.03 | tmzt | your here |
| 06:06.33 | ghostdogg | :) |
| 06:08.47 | NexVision | http://www.newegg.com/Product/Product.aspx?Item=N82E16820183245 |
| 06:09.24 | ghostdogg | had seen a few comments from pphusson the other day wanting a rhodium owner for testing. have a rhod300 and very willing to help. |
| 06:11.15 | ghostdogg | i've been reading through the chat transcripts and following everything I can find to read. You are all far more advanced than I so my help may be of no use. |
| 06:23.33 | tmzt | hmm |
| 06:23.45 | tmzt | talk to phh |
| 06:25.05 | ghostdogg | will do, thanks |
| 07:00.09 | *** join/#htc-linux marcydarcy (i=5b298af9@gateway/web/freenode/x-rbakrmecosbblwus) |
| 07:19.28 | *** join/#htc-linux zachy (n=george@ip-85-160-28-91.eurotel.cz) |
| 07:27.01 | ghostdogg | *phhusson |
| 07:40.20 | *** join/#htc-linux BabelO (n=fcr@unaffiliated/babelo) |
| 08:11.32 | *** join/#htc-linux leobaillard (n=leobaill@leobaillard.org) |
| 08:16.07 | *** part/#htc-linux Ekips_ (i=spike@d51A460C9.access.telenet.be) |
| 08:16.43 | *** join/#htc-linux Ekips (i=spike@d51A460C9.access.telenet.be) |
| 08:32.49 | *** join/#htc-linux polyrhythmic (n=charles@c-67-171-16-39.hsd1.wa.comcast.net) |
| 08:37.31 | *** join/#htc-linux FR^2 (n=frzwo@2001:41d0:1:ed2f:0:0:0:cafe) |
| 08:46.21 | *** part/#htc-linux FR^2 (n=frzwo@2001:41d0:1:ed2f:0:0:0:cafe) |
| 09:24.31 | *** join/#htc-linux ghostdogg (n=ghostdog@cpe-76-179-206-39.maine.res.rr.com) |
| 09:35.53 | *** join/#htc-linux Billyonthemounta (n=Miranda@stiitcsc126.epfl.ch) |
| 09:37.29 | *** join/#htc-linux kiozen (n=oeichler@p54920FD7.dip0.t-ipconnect.de) |
| 09:44.22 | *** join/#htc-linux M1DLGpc (n=M1DLGpc@bb-87-81-252-83.ukonline.co.uk) |
| 09:50.13 | *** join/#htc-linux yozgatg (i=yozgatg@94-225-177-9.access.telenet.be) |
| 09:53.18 | *** join/#htc-linux marex (n=marex@vasut.kolej.mff.cuni.cz) |
| 10:02.41 | *** join/#htc-linux kvaster_ (n=kvaster@93.84.112.80) |
| 10:29.51 | *** join/#htc-linux zycho (n=zycho@g227201254.adsl.alicedsl.de) |
| 10:47.45 | *** join/#htc-linux Markinus (n=Miranda@gtng-4db04246.pool.mediaWays.net) |
| 10:51.57 | *** join/#htc-linux wei (n=wei@c-67-161-5-64.hsd1.ca.comcast.net) |
| 11:27.39 | *** join/#htc-linux sxe (n=quassel@ip-62-143-102-161.unitymediagroup.de) |
| 11:28.44 | *** join/#htc-linux babijoee (n=joee@110.33.180.85) |
| 11:42.15 | *** join/#htc-linux MethoS- (n=clemens@134.102.106.250) |
| 12:35.40 | *** join/#htc-linux kvaster (n=kvaster@93.84.112.80) |
| 12:36.05 | *** join/#htc-linux tids2k (n=tids2k@122.110.123.151) |
| 12:38.30 | *** join/#htc-linux marex (n=marex@vasut.kolej.mff.cuni.cz) |
| 12:48.57 | *** join/#htc-linux marex (n=marex@vasut.kolej.mff.cuni.cz) |
| 13:13.23 | *** join/#htc-linux marcydarcy (i=5b2989f6@gateway/web/freenode/x-jknwanhmgdavaqnu) |
| 13:14.24 | *** join/#htc-linux jumoit (n=samuel@121.77.87.145) |
| 13:25.20 | *** join/#htc-linux GNUtoo (n=GNUtoo@host144-130-dynamic.54-79-r.retail.telecomitalia.it) |
| 13:33.42 | *** join/#htc-linux Markinus (n=Miranda@gtng-4db051f0.pool.mediaWays.net) |
| 13:43.23 | *** join/#htc-linux GeekLad (n=GeekLad@adsl-178-231-40.jax.bellsouth.net) |
| 13:48.17 | *** join/#htc-linux ghostdogg (n=ghostdog@cpe-76-179-206-39.maine.res.rr.com) |
| 13:53.42 | *** join/#htc-linux leaigor (n=laigor@188.134.16.241) |
| 13:56.45 | *** join/#htc-linux sxe (n=quassel@ip-62-143-102-161.unitymediagroup.de) |
| 14:03.28 | *** join/#htc-linux lalaleche (n=jooprzol@32.170.47.84) |
| 14:11.15 | *** join/#htc-linux tids2k (n=tids2k@122.110.125.103) |
| 14:14.47 | *** join/#htc-linux x29a (n=x29a@unaffiliated/x29a) |
| 14:15.58 | *** join/#htc-linux StarLite (n=nnscript@s55916cb1.adsl.wanadoo.nl) |
| 14:41.09 | *** join/#htc-linux babijoee (n=babijoee@110.33.180.85) |
| 14:42.04 | babijoee | hi people |
| 14:45.23 | MarcLandis | hi babijoee |
| 14:46.20 | babijoee | hi maejrep |
| 14:46.25 | babijoee | hi MarcLandis* |
| 14:46.41 | MarcLandis | still need to test your latest package |
| 14:46.46 | MarcLandis | no time - no time |
| 14:46.51 | babijoee | lol |
| 14:46.59 | *** join/#htc-linux sxe_ (n=quassel@ip-62-143-102-161.unitymediagroup.de) |
| 14:47.08 | babijoee | i'm working on fixing layouts for other devices |
| 14:47.29 | babijoee | but nexus one development is growing and fast |
| 14:47.31 | babijoee | XD |
| 14:47.48 | MarcLandis | but without 3D support it will be too slow |
| 14:47.58 | MarcLandis | or is that magically fixed? |
| 14:48.11 | babijoee | ? |
| 14:48.37 | babijoee | as in hw key layout/ kb layout |
| 14:49.01 | *** join/#htc-linux mickey|cafe (n=M@80.187.217.48) |
| 14:49.12 | leaigor | hello all!!! |
| 14:49.18 | leaigor | Who can help me with haret? |
| 14:50.02 | leaigor | i want add to regs_omap.py some registers |
| 14:50.43 | leaigor | Anybody please help!!! |
| 14:53.09 | babijoee | i think you'll need to speak with cr2 and i can't remmeber the other pesons name |
| 14:53.27 | babijoee | fuZion i think |
| 14:54.03 | *** join/#htc-linux sxe_ (n=quassel@ip-62-143-102-161.unitymediagroup.de) |
| 14:56.28 | *** join/#htc-linux emwe (n=michael@86.56.50.180) |
| 15:00.08 | ghostdogg | hello babijoee - would you happen to know how familiar with linux kernel I'd need to be to help phh with the rhodium panel. he had mentioned wanting a rhod owner for debugging. |
| 15:00.41 | *** join/#htc-linux teh_eb (i=ebcain@cpe-24-210-123-84.insight.res.rr.com) |
| 15:02.28 | *** join/#htc-linux swc|666 (n=infidel2@unaffiliated/swc666/x-4934821) |
| 15:02.41 | babijoee | you'll need to have basic knowledge of linux |
| 15:03.13 | babijoee | as he doesnt have the device i'm not sure |
| 15:03.35 | babijoee | afaik there are still some minor problem for blackstone panels |
| 15:03.38 | babijoee | but i'm not sure |
| 15:03.44 | babijoee | you'll have to ask him directly |
| 15:04.16 | ghostdogg | i have a pretty good handle on linux- worked with some of the devs on the kaiser port so I may be able to help. have a rhod300 |
| 15:05.42 | babijoee | ah k, you should be fine then :p phh usually jumps on in about 2 hours or so |
| 15:06.01 | ghostdogg | excellent, i'll be sure to be around |
| 15:06.14 | ghostdogg | thanks again for all your hard work! |
| 15:06.34 | babijoee | hehe not doing as much as i used to |
| 15:06.43 | ghostdogg | understandable |
| 15:06.47 | babijoee | nexus one development is too exciting |
| 15:06.50 | ghostdogg | i hear you got a nexus |
| 15:07.06 | ghostdogg | i'm sure! thats why I want to get it running on my rhod :) |
| 15:07.26 | babijoee | true |
| 15:07.45 | babijoee | but honestly to really appreciate Android i think you need an actual Android phone |
| 15:08.22 | babijoee | everything works out of box and its just so pleasing :p |
| 15:08.39 | babijoee | ghostdogg: btw i thought u had a diam/tp |
| 15:08.48 | ghostdogg | really? thats unfortunate. i did just get a milestone but I'm not as happy with it as I thought i would be |
| 15:08.59 | babijoee | o right |
| 15:09.08 | babijoee | you got the milestone for cheap, |
| 15:09.11 | ghostdogg | nope, i have a raph and milestone and now rhod from insurance |
| 15:09.18 | babijoee | ooo |
| 15:09.21 | babijoee | very nice |
| 15:09.25 | ghostdogg | super cheap, yes. so cheap i thought it was a scam :) |
| 15:09.45 | babijoee | so how is rhodium running with latest xdandroid? |
| 15:10.54 | *** join/#htc-linux sxe (n=quassel@ip-62-143-102-161.unitymediagroup.de) |
| 15:11.30 | ghostdogg | excellent, the system is super responsive and it returns from sleep without the "burn" screen look |
| 15:13.03 | emwe | hi everybody. |
| 15:14.49 | ghostdogg | hello emwe |
| 15:15.38 | babijoee | only thing bad bout nexus is lack of kb |
| 15:15.46 | babijoee | i wish it had one XD |
| 15:16.43 | emwe | ghostdogg: phh disabled the screen pm for rhod. that's why it now "works" when returning from sleep. |
| 15:16.58 | ghostdogg | i haven't seen anything in the change logs about the rhodium so I haven't played with it much. using my fuze with android exclusively though. |
| 15:17.02 | emwe | ghostdogg: does it instantly resume from sleep or does it take some secs until you see something? |
| 15:17.21 | emwe | (your rhod) |
| 15:17.40 | ghostdogg | i did see that, it typically takes just a second perhaps 2 but much faster than older builds |
| 15:17.49 | ghostdogg | it used to take up to 5 seconds |
| 15:17.58 | emwe | oh, that fast. hm. my topaz takes up to 5 yeah. |
| 15:18.08 | babijoee | emwe: on latest xdandroid? |
| 15:18.22 | ghostdogg | i have no added apps or anything |
| 15:18.23 | emwe | babijoee: 2010-01-25 ... |
| 15:18.27 | ghostdogg | i think that makes a big difference |
| 15:18.34 | babijoee | ok |
| 15:18.53 | emwe | babijoee: i think it's been this way since Markinus added panel pm. |
| 15:19.12 | babijoee | ah k |
| 15:19.21 | emwe | it's so sad when I see my girlfriends tattoo working flawlessly ;) |
| 15:19.32 | ghostdogg | gggrrr |
| 15:19.36 | ghostdogg | i hear you! |
| 15:19.39 | babijoee | get a n1 :) |
| 15:19.49 | emwe | rather the bravo then |
| 15:20.03 | ghostdogg | i honestly would shell out the cash today if it had a keyboard |
| 15:20.06 | babijoee | not out yet |
| 15:20.14 | babijoee | true |
| 15:20.43 | emwe | the thing is... if i have a n1 or bravo, i think i am quickly loosing interesst tinkering on the kernel for topaz :) |
| 15:20.56 | babijoee | for sure |
| 15:21.03 | ghostdogg | I'm eligible for a new upgrade on my phone right now so I plan to wait until AT&T releases their android line |
| 15:21.11 | emwe | have been looking at the tattoo kernel in order to get some changes back to 2.6.27 for the non contigious memory banks. |
| 15:21.23 | babijoee | waaaaaaa so many phones ghostdogg |
| 15:21.24 | emwe | but i am still stuck. getting a bit further on boots, but sooner or later it freezes. |
| 15:21.54 | ghostdogg | just lucky as of late. bought a tilt for my last contract upgrade |
| 15:22.12 | ghostdogg | they replaces 4 of those because they didnt work right. finally gave me a free fuze and let me keep my tilt |
| 15:22.21 | babijoee | emwe how much more mem could we possibly get? |
| 15:22.50 | emwe | babijoee: 77MB extra if pmem-setup sticks to 51MB in total |
| 15:23.15 | ghostdogg | had the tilt insured and they called me to tell me they would replace it with a tilt2 and that was a no brainer |
| 15:23.16 | emwe | but i see no difference if pmem uses 32 or 24MB (apart from gpu and adsp sections) |
| 15:24.25 | emwe | babijoee: the secodn bank is used for pmem only. (51MB) the rest is unused. there seem to be issues when the physical addresses are non-contiguious. as is the case for topaz. that's why i though the tattoo source can help, as it has a similar layout. |
| 15:24.43 | emwe | just different offsets... |
| 15:25.44 | babijoee | yeah its quite a difficult task, MrPippy was playing around for it awhile but in the end was too unstable. |
| 15:26.15 | babijoee | hopefully revealing how the nexus1 can get backs its lost memory may help |
| 15:26.25 | babijoee | but its working on the 2.6.32 kernel |
| 15:26.30 | babijoee | so i dont know |
| 15:26.39 | babijoee | if it will work or not. |
| 15:27.37 | emwe | haven't i read phh saying porting his work to 2.6.32? ;) |
| 15:28.32 | babijoee | yeah but i dont know if haret can support it or something like that |
| 15:28.37 | emwe | i am tempted getting topaz into the tattoo source for testing out if that direction works. (instead of trying to incorporate tattoo and version changes into 2.6.-27) |
| 15:28.41 | babijoee | due to its size? |
| 15:29.15 | emwe | ah. perhaps there's just a small amount of space available for loading the kernel into mem. might be. but i am too un-low-level to judge about that :) |
| 15:30.14 | babijoee | but afaik its just a matter of updating haret for it |
| 15:31.43 | babijoee | ghostdogg: is there any forum for development on milestone? |
| 15:34.00 | ghostdogg | i did find one thats none xda- haven't gotten into it much though. |
| 15:34.27 | ghostdogg | i have the gsm version and all the development has been on the cdma side that i can see |
| 15:34.55 | ghostdogg | i have been reading everything i can find about porting linux kernel to arm so I hope to be useful to everyone in the near future. |
| 15:35.18 | ghostdogg | if you have any great reads i should know about please send them my way |
| 15:37.24 | ghostdogg | I think I will sell the milestone for an insane profit so I can justify buying a new android phone when at&t gets theirs |
| 15:37.34 | ghostdogg | have you heard what the htc release will be for att? |
| 15:38.37 | ghostdogg | babijoee: how is the landscape soft keyboard for n1? i would think with the screen size it wouldnt be horrible |
| 15:41.00 | *** join/#htc-linux zycho_ (n=zycho@g227201254.adsl.alicedsl.de) |
| 15:41.34 | babijoee | its actually decent |
| 15:41.52 | babijoee | still no where as fast as my tp kb though |
| 15:42.11 | babijoee | ghostdogg: i dont live in US so i have no clue |
| 15:42.12 | babijoee | :p |
| 15:42.44 | ghostdogg | do you think you'll get better with it? my wife is incredible with hers but she has long nails :) |
| 15:43.17 | ghostdogg | ohhh, gotcha. i'll just have to engadget that one then |
| 15:44.19 | ghostdogg | babijoee: have you ever used a blackberry storm? |
| 15:47.29 | *** join/#htc-linux nozze (n=weechat@u193-11-162-41.studentnatet.se) |
| 15:53.35 | *** join/#htc-linux yozgatg (i=yozgatg@94-225-177-9.access.telenet.be) |
| 15:55.19 | yozgatg | hello all |
| 15:59.32 | ghostdogg | hello yozgatg |
| 16:01.09 | yozgatg | a bit quiet in here, no? |
| 16:01.35 | ghostdogg | yes, i think people are still trying to start their engines :) |
| 16:01.39 | yozgatg | ps, check this out: Gen.Y DualBOOT |
| 16:01.40 | yozgatg | http://img690.imageshack.us/img690/7862/genydualbootug.jpg |
| 16:02.13 | ghostdogg | already obsessed with that! will be on my raphael as soon as its out! |
| 16:02.36 | MarcLandis | that is cool |
| 16:02.51 | yozgatg | yep i'm going to make a Touch Pro & WVGA versions too |
| 16:02.51 | ghostdogg | anyone know if it will put android on nand? |
| 16:03.13 | ghostdogg | oh your the cook! Your INCREDIBLE! |
| 16:03.21 | yozgatg | hehe |
| 16:03.34 | ghostdogg | i LOVE everything you do! |
| 16:03.36 | yozgatg | no, this is only a bootlauncher, u will have to put android on your nand yourself |
| 16:03.49 | yozgatg | but my PX series will include XDANDROID |
| 16:03.59 | ghostdogg | ok, so it'll work from sd or nand? |
| 16:04.07 | yozgatg | yep |
| 16:04.23 | ghostdogg | sweet :) wvga ie rhodium? |
| 16:04.43 | yozgatg | in development ;) |
| 16:10.48 | ghostdogg | yozgatg: little something to help you with that development! thanks for your hard work! |
| 16:11.17 | ghostdogg | yozgat: Confirmation number: 89K29034BR9307026 |
| 16:11.31 | yozgatg | wth? oh thanks m8 |
| 16:11.47 | ghostdogg | none needed, thank you |
| 16:12.12 | yozgatg | First test with Gen.Y DualBOOT: hmm GUI & programming is in place, but got a little problem reading from registry |
| 16:12.32 | yozgatg | worked fine on desktop however.. |
| 16:13.11 | ghostdogg | babijoee: on the raphael android page you have my listed twice under supporters and used my real name. would it be possible to change it to my nick: ghostdogg? |
| 16:13.45 | ghostdogg | do you develop in windows or linux? |
| 16:14.28 | yozgatg | windows |
| 16:14.41 | ghostdogg | ohh bummer :) |
| 16:16.32 | ghostdogg | your dual boot idea is just awesome man! always had concerns with losing winmo all together for the exact reasons your making the dual boot! |
| 16:18.41 | ghostdogg | the only addition i could see as a benefit for the stripped winmo would be the ability to watch divx/xvid movies with tcpmp. doesnt look like there will be android support for that anytime soon. |
| 16:21.28 | yozgatg | afaik video playback should work in PX-D |
| 16:22.41 | yozgatg | and Gen.Y DualBOOT is officialy functional now! |
| 16:22.50 | yozgatg | just needs some minor GUI corrections |
| 16:24.07 | yozgatg | well i'm off |
| 16:24.19 | yozgatg | see u all later |
| 16:30.38 | *** join/#htc-linux AstainZZZZZZ (n=AstainHe@unaffiliated/astainhellbring) |
| 16:36.44 | *** join/#htc-linux x29a (n=x29a@unaffiliated/x29a) |
| 16:39.43 | *** part/#htc-linux jumoit (n=samuel@121.77.87.145) |
| 16:43.09 | *** join/#htc-linux lalaleche (n=jooprzol@209.12.14.250) |
| 16:55.31 | *** join/#htc-linux Boydell (n=Boydell@69.49.58.28) |
| 16:55.49 | Boydell | morning all |
| 17:03.31 | *** join/#htc-linux dcordes-kais (n=zsirc@ip-109-41-88-166.web.vodafone.de) |
| 17:07.01 | leobaillard | plop Boydell |
| 17:09.46 | *** join/#htc-linux pleemans (n=toi@d54C2A96D.access.telenet.be) |
| 17:17.43 | *** join/#htc-linux nullkuhl (n=N3cR0M4n@196.221.170.72) |
| 17:18.00 | nullkuhl | hello, can i remap the buttons for my android1.6 kaiser system.sqsh ? |
| 17:23.52 | *** part/#htc-linux nullkuhl (n=N3cR0M4n@196.221.170.72) |
| 17:24.01 | *** join/#htc-linux luc_ (n=luc@89-115-128-35.cl.ipv4ilink.net) |
| 17:25.20 | *** join/#htc-linux Boydell (n=Boydell@69.49.58.28) |
| 17:25.59 | *** join/#htc-linux GlemSom (n=glemsom@0x5da34bca.cpe.ge-1-1-0-1105.sdnqu1.customer.tele.dk) |
| 17:28.19 | *** join/#htc-linux surge (n=surge@pool-98-118-154-21.bflony.fios.verizon.net) |
| 17:31.18 | *** join/#htc-linux surgewmwr (n=surge@151.sub-174-215-88.myvzw.com) |
| 17:45.43 | *** join/#htc-linux malgon (n=simon@ABordeaux-257-1-88-249.w90-50.abo.wanadoo.fr) |
| 17:46.45 | *** join/#htc-linux yozgatg (i=yozgatg@94-225-177-9.access.telenet.be) |
| 17:47.34 | *** join/#htc-linux surge (n=surge@84.sub-174-213-67.myvzw.com) |
| 17:47.55 | *** join/#htc-linux Zoolooc (n=fredsiba@p54954551.dip.t-dialin.net) |
| 17:48.13 | *** join/#htc-linux ali1234 (n=al@robotfuzz.co.uk) |
| 17:52.06 | *** join/#htc-linux Squarc (n=Squarc@145.81.0.68) |
| 17:53.07 | phh | [00:12:43] <cr2_> phh: any good idea on gps ? <---- still not. |
| 17:54.49 | *** join/#htc-linux c2d (n=Mike@port-92-200-45-82.dynamic.qsc.de) |
| 17:57.06 | phh | [00:20:37] <cr2_> we know the programming interface only for brf6150 <--- a lot of people (wimo AND linux) will be interested in the commands for FM (nexus one has bcm4329 but no FM software ...) |
| 17:57.07 | *** join/#htc-linux MarcLandis (n=drettsch@port-92-195-10-37.dynamic.qsc.de) |
| 17:57.56 | *** join/#htc-linux surgex (i=surge@pool-98-118-154-21.bflony.fios.verizon.net) |
| 17:58.52 | *** join/#htc-linux MrPippy (n=pip@adsl-75-37-167-3.dsl.sndg02.sbcglobal.net) |
| 18:00.10 | phh | [01:49:47] <makkonen> yay, touchscreen calibration (sort of) works. <------ mmmmmmmmmmuh? |
| 18:00.49 | stinebd | see the next line where he says boo it doesn't work ;) |
| 18:01.12 | phh | [02:37:56] <dcordes> MrPippy, nice finding. wiki is cited there :) <---- that's always what happens when I search something about msm7k -_-' |
| 18:01.26 | *** join/#htc-linux kvaster (n=kvaster@live.bn.by) |
| 18:08.09 | phh | [04:51:52] <babijoee> i recieved a message from the bug and error webpage and phh says theres some progress on the camera branch <----- it doesn't mean it works you know :p |
| 18:09.58 | phh | [07:17:24] <ghostdogg> had seen a few comments from pphusson the other day wanting a rhodium owner for testing. have a rhod300 and very willing to help. <---- not today nor tomorrow, but I'll be free from thursday to sunday for some tries. |
| 18:12.43 | phh | [16:29:18] <emwe> have been looking at the tattoo kernel in order to get some changes back to 2.6.27 for the non contigious memory banks. <------ if the problem is really about being non contiguous, try using the second bank as mtd device and then swap on it. |
| 18:14.05 | yozgatg | Gen.Y DualBOOT now fully functional, both NAND & SDMemory ready (simple registry switch)! (VGA only atm, WVGA soon) |
| 18:14.56 | phh | "nand ready" ? what does that mean ? |
| 18:15.05 | phh | and did you partition the nand as I asked you to ? |
| 18:15.16 | phh | (linux can't access to wimo's partition) |
| 18:15.28 | yozgatg | well nand meaning the internal storage of the Diamond |
| 18:15.38 | yozgatg | so now it can be used both for Diamond & TouchPro's |
| 18:15.39 | phh | ah. |
| 18:15.45 | phh | I don't call that one nand but mmc |
| 18:15.49 | phh | (even if it's still a nand -_-') |
| 18:15.55 | yozgatg | hehe |
| 18:16.29 | yozgatg | how the hell are we going to partition the internal flash nand? |
| 18:16.55 | phh | michyprima did it, I don't know how |
| 18:16.56 | yozgatg | afaik when u flash a rom, the bootloader (SPL) takes care of the partitionng & formatting right? |
| 18:17.16 | yozgatg | gotta eat, brb |
| 18:24.55 | makkonen | phh: if I manually feed in the values 1 through 5 to /sys/class/input/input3/calibration, and click in each of the spots, and then feed a 0 in, it updates the calibration. If I use the app, it only seems to read a value for points 4 and 5 (sometimes just 5). I don't know what's up with that. |
| 18:24.57 | *** join/#htc-linux kiozen (n=oeichler@rgnb-5d87dfbf.pool.mediaWays.net) |
| 18:27.39 | *** join/#htc-linux Forsaken|Desk (n=Forsaken@dslb-084-058-244-232.pools.arcor-ip.net) |
| 18:38.37 | phh | makkonen: ah. you have to put a 0 at the end ? |
| 18:38.38 | phh | mm |
| 18:39.52 | makkonen | yeah. I tweaked the program to do it (and fixed a few other little bits)... but it still doesn't read in correctly. |
| 18:40.00 | phh | bah |
| 18:40.22 | phh | oops. |
| 18:40.30 | phh | ligne 122, add close(calib_fd); |
| 18:41.43 | phh | line* |
| 18:42.00 | makkonen | also the spots to click really like to disappear. I guess that's just the screen redrawing over them. |
| 18:43.10 | phh | yes, we are fighting against android :p |
| 18:43.30 | phh | I'll make it on-boot, before android boots |
| 18:45.03 | yozgatg | ps just want to show my respect to all you folks here getting android working on WM devices |
| 18:45.05 | phh | yozgatg: formatting yes, not partitioning |
| 18:45.21 | phh | yozgatg: our respect to you, playing with a closed source thing is tough :p |
| 18:45.28 | makkonen | If I stick it in rootfs, will all the fb stuff be initialized? |
| 18:45.44 | phh | makkonen: you'll maybe need a mknod |
| 18:46.00 | makkonen | ok |
| 18:46.16 | yozgatg | ps anyone here with a WVGA device? might need some testers for the WVGA version of Gen.Y DualBOOT tomorrow |
| 18:46.22 | phh | but just don't take care of this problem for now, first see the communication with the kernel problem |
| 18:47.02 | phh | yozgatg: so about partitioning, AFAIK, there is some space where you declare the partition table |
| 18:47.48 | phh | it's at the beginning of "output.nb.payload" |
| 18:48.00 | yozgatg | yes i was just about to say |
| 18:48.33 | yozgatg | XIP does do some partitioning |
| 18:48.56 | phh | michyprima gave some link to infos about it |
| 18:49.25 | yozgatg | e.g. just an example, u got FAT Start, FAT Size & FAT End defining the FAT partition |
| 18:49.45 | phh | and there is nothing after FAT right ? |
| 18:49.54 | yozgatg | playing with the partitions is possible, however how on earth can one create an EXT partition? (for linux) |
| 18:50.11 | yozgatg | phh: yep |
| 18:50.31 | phh | don't create an ext partition |
| 18:50.37 | phh | just set the last partition before the end of the nand |
| 18:51.08 | phh | then you can give the address of the available space on the nand to linux |
| 18:51.34 | *** join/#htc-linux luc_ (n=luc@89-115-128-35.cl.ipv4ilink.net) |
| 18:51.41 | yozgatg | ahh, wonderful, and then we can format it at boot (haret) i guess? |
| 18:52.54 | makkonen | what's a good way to transfer files to/from a phone connected via wifi? |
| 18:52.57 | phh | yozgatg: in linux |
| 18:53.08 | phh | makkonen: still no dropbear ? |
| 18:53.22 | makkonen | unless something happened with it this morning, I haven't heard anything. |
| 18:53.27 | phh | ok |
| 18:53.46 | makkonen | although stinebd said he built a static version that might've worked. |
| 18:53.48 | phh | yozgatg: you found how/where to set partition table or I search in my logs for the link michyprima gave me ? |
| 18:54.49 | yozgatg | XIPTools? |
| 19:01.27 | *** join/#htc-linux GregLem (n=greg@lal69-5-88-171-98-250.fbx.proxad.net) |
| 19:01.54 | phh | I don't remember how to do sql -_-' |
| 19:02.02 | phh | (and yes it's needed to search in my irc logs, why ? :p) |
| 19:02.15 | *** join/#htc-linux balans (n=Gebruike@212-123-149-239.ip.telfort.nl) |
| 19:04.06 | yozgatg | just one sec phh, i was building the last build of PX-D, flashing now on my device to see if my dualboot program works as it should |
| 19:04.18 | phh | bah I haven't much time today anyway |
| 19:04.29 | phh | (one homework for tomorrow and two for thursday -_-') |
| 19:04.55 | yozgatg | well i'm not that experienced with XIP actually, but on the forum there are some people who really now how to handle the partitioning stuff |
| 19:05.57 | yozgatg | however i don't think it's that big of a priority atm, since we need hardware opengles acceleration for the real speed bump imho |
| 19:06.32 | phh | no. |
| 19:06.38 | phh | opengles is used only for 3D |
| 19:07.04 | phh | I still haven't figured out why eclair is slow while donut is so smooth |
| 19:07.24 | yozgatg | how bout the animations? |
| 19:07.40 | yozgatg | though actually animations are quite smooth |
| 19:08.07 | makkonen | even the slide between home screens is really stuttery in eclair. |
| 19:08.15 | phh | I really don't know what the problem is... real animations are fine, but as makkonen says, this slide is slow. |
| 19:08.33 | phh | might be because cpufreq settings are bad |
| 19:08.43 | phh | yozgatg: http://forum.xda-developers.com/showpost.php?p=2915353&postcount=6 |
| 19:08.57 | yozgatg | hmm, can we force cpu to run at 528mhz, just for testing? |
| 19:09.11 | phh | use the cpu "overclocker" for normal android devices it works as well |
| 19:09.43 | stinebd | i'll take a look at cpufreq tunables a little later and see what i can get from it |
| 19:09.58 | phh | might be idle sleep too |
| 19:10.05 | stinebd | it was like that before idle sleep |
| 19:10.06 | phh | err no, not working. |
| 19:10.48 | stinebd | cpufreq is a pretty good guess i think |
| 19:11.04 | phh | makkonen: so did you try adding the close() i told you to ? |
| 19:11.13 | phh | stinebd: it doesn't sound stupid at least :p |
| 19:11.26 | phh | but iirc using the same kernel, donut is way faster than eclair |
| 19:11.37 | stinebd | donut may also be more efficient |
| 19:11.48 | stinebd | cpufreq governors can be kinda touchy |
| 19:12.07 | makkonen | phh: yeah. I think I stuck the closing "0\n" in the wrong spot. |
| 19:12.25 | stinebd | even on desktops, h.264 at a certain size could make the cpu hover just below the ondemand governor's raise threshold |
| 19:12.57 | makkonen | ...but there's also a problem where it reads the x max/x min and y max/y min incorrectly, and even with correct values seems to not calibrate correctly. testing now. |
| 19:13.11 | makkonen | (could just be confused because it's not getting all the settings in one go) |
| 19:13.25 | makkonen | (I'll be quiet now until I have something useful to say) |
| 19:14.30 | phh | makkonen: here is how I do it: http://husson.hd.free.fr/calibrate.c.2 |
| 19:14.52 | *** join/#htc-linux onen|openBmap (n=quassel@vbo91-1-89-87-201-85.dsl.club-internet.fr) |
| 19:16.39 | yozgatg | ps Gen.Y DualBOOT tested, working perfectly, will release a CAB tomorrow for all VGA devices, hopefully makes testing/running android more appealing to the public |
| 19:17.23 | yozgatg | phh: will look into the partitioning stuff tomorrow, have to go now |
| 19:17.25 | phh | makkonen: so I added the close (that might be really problematic.) and the write 0 |
| 19:17.27 | yozgatg | take care all |
| 19:17.27 | phh | yozgatg: ok thanks |
| 19:19.12 | phh | stinebd: rendering the menu should take the whole cpu :/ |
| 19:33.48 | *** join/#htc-linux luc__ (n=luc@89-115-128-35.cl.ipv4ilink.net) |
| 19:35.22 | *** join/#htc-linux bzo (n=chatzill@netblock-68-183-234-14.dslextreme.com) |
| 19:37.16 | makkonen | phh: ok. app is now working as it's supposed to... now the only problem is at the end, it reports x max, x min, y max, and y min to be ridiculously low, which means the whole responsive area of the touchscreen is compressed into the bottom right hand corner of the screen. This is probably because I reverted the 2048x2048 to 240x320 because the values were even more messed up with 2048x2048. anyway... progress! |
| 19:38.11 | phh | ok, tell me when you tried back with 2048x2048 |
| 19:38.17 | makkonen | (need to reboot and see what it reports for those max/min values when I just echo something to calibration_points, too) |
| 19:40.44 | phh | I'm just thinking. |
| 19:41.06 | phh | we are actively working on FM, BT, GPS, flashing, usb host will be started soon too |
| 19:41.11 | phh | is there anything we aren't trying to do ? :p |
| 19:41.35 | makkonen | (I also have no idea if calibration persists through reboots) |
| 19:41.47 | phh | makkonen: there is no way it stays |
| 19:41.55 | makkonen | yeah, I figured. |
| 19:41.55 | phh | if it says, we'll have huge problems. |
| 19:42.21 | makkonen | because it'd be writing something somewhere it really shouldn't? |
| 19:42.29 | phh | chamonix: we will also need an option to be able to (re)enable automatic backlight |
| 19:42.32 | phh | makkonen: yes |
| 19:42.46 | phh | makkonen: the only persistant memory is nand, and there is no way you want it to save it there on his own |
| 19:42.50 | MrPippy | we should have it save to /sdcard, i don't want to have to calibrate on every boot |
| 19:43.03 | phh | MrPippy: yes that will be done later :p |
| 19:43.10 | stinebd | don't reboot :) |
| 19:43.30 | Boydell | sometimes, you dont CHOOSE to reboot :P |
| 19:44.00 | phh | yeah, sometime you want to test this stupid camera :p |
| 19:44.29 | Boydell | yah, i wanna try goggle googles :( |
| 19:44.34 | makkonen | well, right now I just have a line in eclair.user.conf that sends the right values to calibration_points. Just as easy to look for a ts-calibration file in the root of the sd card that does the same. and then write a few lines into the init script somewhere. and then have the calibration program... grab the values. All simple things. |
| 19:44.59 | phh | Boydell: mmm I'll try virtual relity first I guess |
| 19:45.11 | phh | but without GPS it will be tough I think |
| 19:45.19 | Boydell | lol |
| 19:46.54 | phh | oh btw everyone. I'd like you to use the bug tracker to have some traces of what is done ( http://pjottrr.no-ip.org:81/redmine/projects/show/androidphh ) |
| 19:49.08 | Boydell | curious question, we arnt using all the memory possible right now right? |
| 19:49.49 | *** join/#htc-linux pH5 (n=ph5@e178210235.adsl.alicedsl.de) |
| 19:49.55 | phh | no |
| 19:50.07 | phh | well on diam gsm yes :'( |
| 19:50.14 | Boydell | lol |
| 19:50.27 | Boydell | raph gsm? |
| 19:50.28 | phh | Boydell: there is something like 70MB left unused |
| 19:50.36 | Boydell | on bank2? |
| 19:50.38 | phh | yes |
| 19:50.43 | Boydell | wow |
| 19:51.09 | Boydell | yah I remember MrPiipy was playing around with bank2, but wasnt able to get anything stable.. |
| 19:51.24 | phh | I suggest trying mtd phys ram driver and then swapon |
| 19:52.26 | bzo | speaking of memory, is compcache working? |
| 19:52.27 | Boydell | out of my scope. I dont know anything about hardware programmnig really.. |
| 19:52.31 | phh | bzo: yup |
| 19:52.46 | MrPippy | i'm not really trying bank2 until i can figure out why its unstable just with bank1 |
| 19:53.04 | bzo | on the cdma diamond build, lsmod only shows xvmalloc |
| 19:53.15 | phh | bzo: xvmalloc and no ramzswap ? |
| 19:53.23 | MrPippy | yeah i stopped building ramzswap into my modules |
| 19:53.33 | bzo | phh: yes, insmod ramzswap gives error |
| 19:54.04 | bzo | aha, makes sense |
| 19:54.35 | MrPippy | we already only have 102MB of RAM, ramzswap takes 25%, and the system seems worse with it |
| 19:54.54 | bzo | wonder why xvmalloc is loading though, isn't it only used by ramzswap? |
| 19:55.12 | phh | bzo: yes |
| 19:55.19 | phh | MrPippy: it doesn't take 25% of it ... |
| 19:55.31 | phh | read /proc/ramzswap to know how much it takes |
| 19:55.48 | phh | when using the whole swap (25MB), it takes for me less than 10MB |
| 19:56.15 | *** join/#htc-linux cyberdesigner (n=cyberdes@port-92-206-189-164.dynamic.qsc.de) |
| 19:56.23 | bzo | doesn't it only take what you configure it with cc_disksize? |
| 19:56.31 | phh | no |
| 19:56.37 | phh | this setting isn't taken in account |
| 19:56.42 | phh | I have to make it work :p |
| 19:57.12 | bzo | I see, that is why the nonsense parameter in eclair.user.conf of 100 does not cause problem? |
| 19:57.33 | phh | it's not nonsense :p |
| 19:57.38 | phh | I used to run it with 70 |
| 19:57.44 | phh | (but hardcoded in the initscript) |
| 19:57.47 | phh | (not in the conf file) |
| 19:58.35 | bzo | I think some dream folks might be using a number higher than 70 |
| 19:59.02 | *** join/#htc-linux |Jason8| (i=JayAte@server1.tiltshellz.org) |
| 19:59.16 | bzo | MrPippy: is the stability problem only on cdma, or for gsm too? |
| 20:01.04 | MrPippy | i think it might be worse on cdma, but i don't have any gsm to test |
| 20:02.21 | bzo | that is my impression also. I don't seem to read as much about crashes from the GSM users |
| 20:02.56 | makkonen | MrPippy: you've stopped using compcache entirely? |
| 20:03.43 | MrPippy | yeah i wasn't really seeing the benefits |
| 20:03.58 | makkonen | cool. I'll give it a try. |
| 20:05.00 | bzo | when I was running on Vogue, compcache felt slower to me as well, unless you were very memory constrained (hero builds) |
| 20:06.12 | makkonen | well, we are very memory constrained. |
| 20:06.49 | bzo | yes, but with the SenseUI, you are VERY memory constrained |
| 20:07.15 | makkonen | I only tried it for one boot on the vogue. it was so agonizingly slow I immediately dumped it. |
| 20:07.49 | MrPippy | yeah vogue has a little bit less ram than us (~100 MB) |
| 20:08.53 | phh | bzo: home is getting killed at every single app launch |
| 20:09.09 | bzo | I think right now, vogue is greater/equal in speed, despite slower cpu and slightly less memory |
| 20:09.32 | makkonen | no question it's faster than cdma touch pro. |
| 20:10.51 | MrPippy | i still have a vogue, i should try it out |
| 20:11.06 | makkonen | love that phone. |
| 20:11.31 | makkonen | good form factor, good dpad... |
| 20:11.46 | bzo | phh: I guess that makes sense with the huge amount of gc logcat entries |
| 20:11.56 | phh | bzo: :) |
| 20:12.02 | phh | that's one of the reason. |
| 20:12.08 | makkonen | phh: ok... with 2048x2048 it goes nuts in a totally different way. |
| 20:12.19 | makkonen | [ 236.698852] touch_calibration: x max 133, x min 872. |
| 20:12.19 | makkonen | [ 236.698944] touch_calibration: y max 73, y min 1653. |
| 20:12.32 | makkonen | whereas correct is: |
| 20:12.35 | makkonen | [ 249.672363] touch_calibration: x max 130, x min 851. |
| 20:12.35 | makkonen | [ 249.672454] touch_calibration: y max 125, y min 930. |
| 20:12.49 | bzo | vogue is great, I wouldn't have switched to the diamond if it had a higher res screen |
| 20:13.21 | bzo | that and the lack of hardware buttons for android... |
| 20:14.20 | MrPippy | i thought the vogue was too wide to hold comfortably for a while, diamond is better that way |
| 20:14.28 | makkonen | I really found the diamond to be on the whole practically a step down from the vogue. screen difference was amazing, but the diamond (under winmo) was so restricted in storage memory, and the dpad was so terrible. |
| 20:14.45 | bzo | on vogue, with the weird keymapping, when you switch between winmo and android, it feels like someone reversed your accelerator and brake on your car :) |
| 20:14.53 | makkonen | hah |
| 20:15.36 | bzo | yes, diamond keypad is horrible. I still can't press left most of the time |
| 20:16.17 | MrPippy | yeah for me the touchscreen is better enough to make up for it |
| 20:18.38 | bzo | was the kernel forked from dzo's vogue kernel or another source? |
| 20:18.38 | MrPippy | i'll try disabling cpufreq to see if it makes a difference, but...whenever i run top, the cpu is always at 100% anyway, its always doing gc or servicing interrupts for mmc |
| 20:19.00 | makkonen | speaking of awful capacitive buttons... in android, the scrolly-zoomy thing is really awful. everything else about the navi pad is better than winmo, but that thing is way too sensitive. Is there no built in keycode (or whatever you call it) in android for zoom? it doesn't seem like having that go up and down is a very useful feature. |
| 20:19.59 | MrPippy | yeah i wonder if theres a way to slow down the accel on the scroll ring |
| 20:20.49 | bzo | if the scroll circle actualy worked, it would be great. But for me it keeps stopping and starting |
| 20:21.02 | MrPippy | i don't think theres any hw way to zoom in android |
| 20:21.12 | phh | bzo: make a patch |
| 20:22.47 | bzo | phh: maybe some day :) just started poking around the source - learning curve is a little steep |
| 20:22.52 | makkonen | phh: so I'm just guessing at values for CALIBRATION_SCREEN_WIDTH and _HEIGHT to make the calibration come out right... that doesn't seem to be the best way to go about it. |
| 20:24.08 | phh | makkonen: pfff |
| 20:24.19 | makkonen | :-) |
| 20:28.05 | bzo | anyway, I was asking about kernel origin thinking about memory stability problem. I think the Vogue kernel is still on 2.6.25 |
| 20:28.32 | bzo | so backports of newer kernels should not be necessary to make one bank work perfectly |
| 20:31.26 | *** join/#htc-linux Squarc (n=Squarc@82-217-32-29.cable.quicknet.nl) |
| 20:37.01 | *** join/#htc-linux kvaster (n=kvaster@live.bn.by) |
| 20:38.09 | stinebd | every day with the twidroid updates |
| 20:38.36 | makkonen | that's what you get for interacting with twitter. |
| 20:38.45 | stinebd | i haven't even set it up yet |
| 20:38.51 | stinebd | really should |
| 20:39.05 | stinebd | would make the constant market notifications worth something |
| 20:40.12 | makkonen | hmm. things seem slower without ramzswap. probably some sort of confirmation bias. |
| 20:41.51 | stinebd | heh |
| 20:42.04 | stinebd | phh: setting ondemand up_threshold to 40 makes app drawer noticeably better |
| 20:42.15 | stinebd | i cringe at the battery life effects though |
| 20:42.15 | bzo | makkonen: do you see much difference in frequency of gc? |
| 20:42.16 | phh | hum I'll try that thanks |
| 20:42.34 | phh | stinebd: the A11 will still be down most of the time |
| 20:42.37 | phh | don't worry about that |
| 20:42.46 | makkonen | bzo: haven't looked. this is based on just 2 reboots after stopping building them. |
| 20:42.51 | *** join/#htc-linux Captnoord (n=Captnoor@dc5147a47b.adsl.wanadoo.nl) |
| 20:42.53 | phh | I'll have to wait for chamonix to test if dynamic freq actually changes anythin |
| 20:42.54 | phh | g |
| 20:43.10 | stinebd | i'd say 40 is the best value i've found so far |
| 20:43.34 | stinebd | phh: you mean in terms of saving power? |
| 20:43.44 | phh | yes |
| 20:43.47 | stinebd | ok |
| 20:44.07 | phh | I think setting threadshold higher migt actually help :p |
| 20:44.14 | makkonen | ...why am I still building compat-wireless. what a waste of time. |
| 20:44.28 | Captnoord | wines...... as he's on old school pc now.... "damn slow windows" thats something different that "damn small linux" |
| 20:45.23 | stinebd | phh: default threshold is 80 |
| 20:45.30 | stinebd | so 40 is quite a bit lower |
| 20:45.31 | phh | ah. |
| 20:45.37 | phh | what about threshold down ? |
| 20:45.54 | stinebd | there is none |
| 20:46.01 | stinebd | i think that was a recent addition, after .27 |
| 20:46.20 | MrPippy | wow the vogue defconfig doesn't use cpufreq at all |
| 20:46.23 | stinebd | might help to set sampling_rate to 100000 (minimum) |
| 20:46.36 | stinebd | right now it's 200000 |
| 20:47.15 | *** join/#htc-linux slight (n=23hdci27@217.130.58.175) |
| 20:47.25 | MrPippy | heroc uses max 528 MHz and min 245.76, and a "msm7k" governor that is a tweaked conservative |
| 20:47.39 | stinebd | tweaked how? |
| 20:47.50 | stinebd | algorithm? |
| 20:49.04 | MrPippy | yeah some minor additions |
| 20:50.03 | Captnoord | I think its related to the delta cpu power drawn |
| 20:50.06 | Captnoord | tweaked |
| 20:50.10 | stinebd | wow |
| 20:50.24 | stinebd | this is a lot smoother |
| 20:50.29 | Captnoord | oO |
| 20:54.22 | makkonen | phh: so 2048x1024 seems to be the magic numbers to make it calibrate nearly, but still not completely correctly. and those are some weird numbers to make it come out right. |
| 20:55.14 | Captnoord | makkonen: ts calibration? |
| 20:55.29 | makkonen | yeah |
| 20:55.32 | Captnoord | cool |
| 20:56.29 | makkonen | phh: should I be messing with the calibration_screen stuff in the calibration program? could that be throwing it off? |
| 20:56.56 | phh | makkonen: you have to define the position of the calibration points somewhere |
| 20:57.03 | phh | it is supposed to be done through calibration_screen |
| 20:57.12 | phh | I haven't checked if the code actually works for that |
| 20:57.33 | makkonen | well, it gets passed to calibration_screen. I don't know if that causes anything to happen elsewhere. |
| 21:05.01 | *** join/#htc-linux [1]Captnoord (n=Captnoor@dc5147a47b.adsl.wanadoo.nl) |
| 21:06.35 | *** join/#htc-linux malgon_ (n=simon@62.201.142.3) |
| 21:06.43 | *** part/#htc-linux [1]Captnoord (n=Captnoor@dc5147a47b.adsl.wanadoo.nl) |
| 21:06.52 | *** join/#htc-linux Captnoord (n=Captnoor@dc5147a47b.adsl.wanadoo.nl) |
| 21:08.10 | emwe | phh: thx for the earlier reply regarding mtd. though i'd rather have bank2 working, but perhaps worth a try. though I have not the slightest idea how to accomplish the mtd thingy. ;) |
| 21:08.27 | phh | mtd will make bank2 "working" |
| 21:08.31 | phh | the ugly though |
| 21:08.36 | emwe | MrPippy: just read up the logs... gsm with bank1 is just fine here on topaz. |
| 21:08.46 | *** join/#htc-linux herpez|away (n=herpez@bl15-53-4.dsl.telepac.pt) |
| 21:09.14 | emwe | phh: so it will just act as a swap device, right? |
| 21:09.23 | phh | yes |
| 21:10.29 | emwe | phh: i have just read up about mtd when you mentioned it. what's required in order to make it work? publish the device somehow with the physram location adresses? |
| 21:10.48 | emwe | and then mount the published device as swap device? |
| 21:10.56 | phh | emwe: search how some users use the memory of their graphic card (for x86) as swap |
| 21:10.57 | phh | it's the same. |
| 21:11.13 | Captnoord | makkonen: from flash or by using the hero calibration code |
| 21:11.22 | Captnoord | reading from flash* |
| 21:11.37 | emwe | phh: thx. will read up about it. (when time permits :/) |
| 21:12.14 | makkonen | captnoord: using the tattoo calibration code. |
| 21:12.20 | Captnoord | k |
| 21:12.26 | makkonen | (well, it came from the hero source, but it's for the tattoo) |
| 21:12.38 | Captnoord | hehe |
| 21:12.44 | Captnoord | tomato tometo |
| 21:13.12 | Captnoord | dude |
| 21:13.16 | Captnoord | and if its not 1024 |
| 21:13.17 | Captnoord | but |
| 21:13.23 | Captnoord | 1064 |
| 21:13.24 | Captnoord | ? |
| 21:13.45 | Captnoord | dono why I says that |
| 21:14.01 | Captnoord | just because its in the touch.dll doesn't mean its meaning full |
| 21:14.36 | makkonen | I'm... not sure what you're saying. |
| 21:14.49 | *** join/#htc-linux kam187 (n=kam187@81-179-8-102.dsl.pipex.com) |
| 21:14.50 | Captnoord | k |
| 21:14.55 | Captnoord | thank god |
| 21:15.03 | makkonen | :-) |
| 21:19.03 | Captnoord | cool |
| 21:19.24 | Captnoord | in the touch.dll there is a similar function to parse the ts calibration stuff |
| 21:19.42 | Captnoord | kinda the same as we use |
| 21:24.58 | *** join/#htc-linux sxe_ (n=quassel@ip-62-143-102-161.unitymediagroup.de) |
| 21:27.44 | *** join/#htc-linux cr2_ (n=cr2@ip-109-84-89-207.web.vodafone.de) |
| 21:29.36 | *** join/#htc-linux swc|666 (n=infidel2@unaffiliated/swc666/x-4934821) |
| 21:36.50 | Captnoord | ts power down / powerup |
| 21:36.51 | Captnoord | is |
| 21:37.11 | Captnoord | TSSC_CTL = TSSC_CTL & ~0x1; to powerdown |
| 21:37.15 | Captnoord | TSSC_CTL = TSSC_CTL | 0x1; to powerup |
| 21:40.35 | *** join/#htc-linux Boydell_ (n=Boydell@69.49.58.28) |
| 21:45.30 | MrPippy | i think theres a patch in the nexusone tree that adds wakelock/earlysuspend to tssc_manager |
| 21:46.04 | Captnoord | lets have a look |
| 21:46.09 | Captnoord | thanks |
| 21:49.32 | MrPippy | has anyone tried out a higher CONFIG_DEFAULT_MMAP_MIN_ADDR? heroc defconfig has it at 32768, so does the tattoo defconfig |
| 21:53.09 | emwe | MrPippy: yap. doesn't seem to help. |
| 21:53.44 | emwe | MrPippy: copied in some stuff which "looked" helpful from the tattoo tree. (as it's the onyl device i see which has dual-bank setup with non-contiguious banks) |
| 21:54.07 | *** join/#htc-linux leaigor (n=laigor@188.134.16.241) |
| 21:54.09 | MrPippy | yeah |
| 21:54.28 | emwe | MrPippy: am not getting further though. i used your inital attempt with pfn_valid() and then another one regarding memmap_valid_within() and some other small things wich looked interssting. |
| 21:56.45 | Captnoord | not in the |
| 21:56.45 | Captnoord | http://github.com/cyanogen/cm-kernel/blob/379c6bc09b589cb545c8b17276e8f38671d6db88/drivers/input/touchscreen/msm_ts.c |
| 21:57.15 | Captnoord | ponders... tinyurl... |
| 21:57.29 | Captnoord | #ifdef CONFIG_HAS_EARLYSUSPEND |
| 21:57.29 | Captnoord | static void synaptics_ts_early_suspend(struct early_suspend *h); |
| 21:57.29 | Captnoord | static void synaptics_ts_late_resume(struct early_suspend *h); |
| 21:57.29 | Captnoord | #endif |
| 21:57.30 | Captnoord | yuip |
| 21:57.36 | Captnoord | in the i2c ts driver |
| 21:57.50 | MrPippy | synaptics? isn't that the capacitive driver? |
| 21:57.57 | bzo | MrPippy: perhaps, it could help with the cdma stability on the one bank. I just set mine to 32768 via /proc/sys/vm |
| 21:58.01 | Captnoord | MrPippy: yup |
| 21:58.17 | Captnoord | but the system to suspend a driver is generic |
| 21:58.27 | MrPippy | emwe: yeah i wonder if any tattoos actually shipped that use the 2 bank config |
| 21:58.31 | Captnoord | suspend hardware* |
| 21:58.36 | MrPippy | that code could be left in there because they tried it and found it wasn't stable |
| 21:58.57 | Captnoord | hmm I would second guess that |
| 21:59.06 | Captnoord | code that they release seem to be paranoirdly checked |
| 21:59.19 | emwe | MrPippy: do not destroy my last hope! would be worth a try adb-ing to my girlfriends tattoo, eh? ;) |
| 21:59.25 | dcordes | hello |
| 21:59.42 | phh | Captnoord: stoping TS when in sleep would be nice. |
| 21:59.45 | MrPippy | you have access to a tattoo? definitely, get a dmesg if you can |
| 21:59.52 | phh | I don't think it's the current highest power consumption yet though |
| 21:59.59 | Captnoord | phh true |
| 22:00.05 | Captnoord | but its always nice |
| 22:00.09 | phh | but as much things as possible has to be fixed. |
| 22:00.11 | emwe | MrPippy: can i get it only via dmesg or is it somewhere else on fs? /var/log? |
| 22:00.16 | Captnoord | hehe |
| 22:00.17 | phh | Captnoord: we can do the same for microp |
| 22:00.23 | Captnoord | yup |
| 22:00.24 | phh | Captnoord: again, read mahimahi's code |
| 22:00.33 | Captnoord | ? |
| 22:00.40 | Captnoord | who's code? |
| 22:00.49 | Captnoord | point me to it |
| 22:00.51 | Captnoord | or |
| 22:00.58 | Captnoord | you mean the guy who cioded the microp |
| 22:01.20 | emwe | phh: is setting up phram via kernel args supposed to work? |
| 22:01.29 | phh | emwe: don't know |
| 22:01.37 | phh | Captnoord: mahimahi = nexus one's board :p |
| 22:01.43 | MrPippy | yeah just do dmesg, i don't think its anywhere in /proc |
| 22:01.50 | Captnoord | hehe |
| 22:01.53 | Captnoord | phh i'm reading that |
| 22:02.21 | emwe | phh: what is setting up /dev/mtd0 called msm_nand? |
| 22:02.37 | phh | emwe: don't use that one |
| 22:02.46 | phh | *don't* write to it. |
| 22:02.51 | emwe | phh: no, was just curious how that one is set up :) |
| 22:03.02 | phh | just loading msm_nand module :p |
| 22:03.17 | emwe | phram=bank2,0x20000000,32Mi << didn't work |
| 22:04.01 | phh | you built it ? |
| 22:04.33 | MrPippy | emwe: the kernel config from a running tattoo would be good too, its the only msm linux tree i've seen that could do sparsemem, but the config in-tree doesn't show it enabled |
| 22:05.36 | emwe | it has the HAS_HOLES_MEMORYMODEL and memmap_valid_within which i've seen in 2.6.30 only. tattoo was 2.6.29, right? |
| 22:06.27 | emwe | phh: ok, did i fool myself into believing enabled CONFIG_MTD_PHRAM=y does not suffice? |
| 22:06.45 | MrPippy | yeah the tattoo source we have is 2.6.29 |
| 22:06.54 | phh | emwe: don't know |
| 22:06.56 | emwe | MrPippy: so i thought that was/is the backported fix for non-contig banks |
| 22:07.07 | phh | I guess it should |
| 22:07.27 | emwe | phh: ok, retrying again and looking up symbols in the Image then... just to make sure it's in ;) |
| 22:08.30 | phh | emwe: read dmesg to see if ths parameter is ignored or not |
| 22:08.38 | emwe | it was not on last boot |
| 22:09.01 | emwe | but cat /proc/mtd shows no "bank2" |
| 22:09.37 | phh | I don't know how this works yet |
| 22:09.59 | MrPippy | has anyone tried tickless kernel? |
| 22:10.03 | emwe | me neither. will retry. thanks for the assistance and hints into the right direction. |
| 22:10.12 | dcordes | cr2_, ping |
| 22:10.14 | phh | MrPippy: only stinebd afaik |
| 22:10.36 | emwe | MrPippy: oh as you mention it. tattoo has CONFIG_NOHZ, iirc |
| 22:11.10 | MrPippy | yeah the msm7k governor uses a function that only works under tickless, and heroc uses no_hz |
| 22:11.57 | MrPippy | i just built a kernel with it that paniced on boot, early console looked fine but fb never came up |
| 22:12.16 | phh | MrPippy: needs to set generic_clocksource bac |
| 22:12.16 | phh | k |
| 22:12.47 | MrPippy | i put generic_time back |
| 22:13.43 | phh | ah. |
| 22:13.52 | phh | ask stinebd then |
| 22:24.05 | *** join/#htc-linux rafyvitto (n=rafyvitt@76-195-44-215.lightspeed.wlfrct.sbcglobal.net) |
| 22:24.17 | stinebd | tickless .27 works fine for me, all i had to do was update the config |
| 22:25.51 | MrPippy | ok maybe its the cpufreq then |
| 22:26.14 | stinebd | i can try it i guess |
| 22:27.11 | stinebd | you got it from heroc kernel? |
| 22:27.23 | *** join/#htc-linux mickey|sofa (n=M@e180139189.adsl.alicedsl.de) |
| 22:28.47 | MrPippy | yeah the governor |
| 22:28.53 | stinebd | ok |
| 22:31.15 | phh | MrPippy: for me this governor is stupid, but it's still worth the try. |
| 22:31.55 | phh | well. stupid if you have idle even with screen on working, but maybe they don't. |
| 22:32.37 | MrPippy | they use power collapse for idle sleep and power collapse suspend for sleep mode |
| 22:33.23 | MrPippy | the heroc defconfig also sets cpu_freq_min_ticks to 1, and the sampling latency multiplier to 500, both WAY lower than what we were using |
| 22:33.38 | phh | MrPippy: if they have vsync on everytime, idle doesn't work |
| 22:34.27 | phh | and afaik hero has still no vsync off :) |
| 22:34.35 | phh | but maybe 2.6.32 branch has it |
| 22:35.16 | MrPippy | hmm mahimahi uses min_ticks 2 and same multiplier |
| 22:37.05 | Captnoord | MrPippy have ya seen the sdcc highspeed patch on nexus branch? |
| 22:37.12 | tmzt | how is tickless different from no_hz? |
| 22:37.23 | stinebd | same thing |
| 22:38.13 | stinebd | tickless refers to the kernel timer running at "0hz" |
| 22:38.27 | MrPippy | which patch is that? i brought over most (maybe all?) of the nexus sdcc patches |
| 22:39.37 | Captnoord | http://github.com/cyanogen/cm-kernel/commit/6585a85fe21edc70136859464a7c6d2652f17dab |
| 22:40.46 | MrPippy | looks like thats only sdio |
| 22:41.09 | MrPippy | our sd clock is running pretty slow (32 MHz), but i haven't tried increasing it |
| 22:41.13 | Captnoord | the comment is correct..... |
| 22:41.21 | Captnoord | that isn't slow |
| 22:41.22 | Captnoord | 32mhz |
| 22:41.25 | Captnoord | thats default speed |
| 22:41.37 | Captnoord | I tought normal speed at current spec's was 35 mhz |
| 22:41.49 | phh | MrPippy: I think you should reenable multiple voltages in mmc btw |
| 22:41.54 | phh | (or I should, ok.) |
| 22:42.01 | Captnoord | lolz |
| 22:42.03 | phh | constant voltage is bad. |
| 22:42.21 | Captnoord | and also consuming power |
| 22:42.30 | Captnoord | even if its very little |
| 22:42.30 | Captnoord | :P |
| 22:42.47 | phh | I'm not sure it's very little actually |
| 22:42.50 | MrPippy | i'm not sure what the spec says, but the other phones are running at 48-50 MHz |
| 22:43.38 | Captnoord | 35mhz was the default speed of 3 year old cards |
| 22:43.43 | Captnoord | so new ones yes |
| 22:44.20 | Captnoord | at least its interesting to try |
| 22:44.37 | Captnoord | if the current clock is considert slow |
| 22:45.01 | Captnoord | it may increase responciveness |
| 22:46.10 | phh | I don't think so |
| 22:46.28 | Captnoord | i'm just guessing... |
| 22:47.01 | Captnoord | downloaded all the delta's |
| 22:47.02 | Captnoord | whooo |
| 22:47.17 | phh | it will help only bandwidth imgo |
| 22:47.19 | phh | imho* |
| 22:47.48 | Captnoord | it will only inprove the already improved startup time |
| 22:48.03 | *** join/#htc-linux xfight81 (n=quassel@host228-143-dynamic.8-87-r.retail.telecomitalia.it) |
| 22:48.17 | Captnoord | s/inprove/improve |
| 22:49.14 | phh | for startup time, I think we would need perflock |
| 22:49.22 | phh | to lock the cpu at 528MHz at startup |
| 22:50.40 | cr2_ | phh: dex call |
| 22:50.59 | emwe | wth.. drivers/input/touchscreen/msm_ts.c:26: fatal error: mach/msm_iomap.h: No such file or directory ?? |
| 22:51.05 | makkonen | calibration is working. whee. stupid error with the order of points on calibration_screen being different from in calibration_points. |
| 22:51.08 | phh | cr2_: sure... any subtitle maybe ? :p |
| 22:51.23 | phh | makkonen: oops. |
| 22:51.24 | stinebd | makkonen: for real this time? ;) |
| 22:51.49 | MrPippy | yeah heroc uses perflock for boot and screen-on min freq is 480 mhz |
| 22:51.55 | makkonen | for serious |
| 22:56.20 | *** join/#htc-linux apt (i=ibot@rikers.org) |
| 22:56.20 | *** topic/#htc-linux is Welcome to htc-linux | Project homepage and wiki http://htc-linux.org | Livelogs: http://irclog.netripper.com/?chan=htc-linux | Logs: http://apt.rikers.org/%23htc-linux | htc-linux is not android |
| 22:56.21 | phh | cr2_: why do you tell me "dex call" ? |
| 22:56.44 | cr2_ | we talked about arm9 slow speed |
| 22:57.05 | cr2_ | RTFW :) |
| 22:57.29 | phh | :) |
| 22:58.02 | phh | cr2_: have you added the values of the "reboot_reason" thing ? (the smem byte to set before pulling the power off gpio) |
| 22:58.37 | stinebd | hmm |
| 22:59.02 | cr2_ | phh: it's wince-only i think |
| 22:59.10 | cr2_ | the 0x81000 area |
| 22:59.15 | cr2_ | ? |
| 22:59.21 | *** join/#htc-linux Markinus (n=Miranda@gtng-4db051f0.pool.mediaWays.net) |
| 22:59.21 | phh | I don't remember :/ |
| 22:59.28 | cr2_ | phh: i can't access wiki |
| 22:59.46 | cr2_ | which is the "slow" call with parameter ? |
| 23:00.22 | phh | cr2_: "reboot reason" might be 0xfc044 |
| 23:00.31 | cr2_ | hm |
| 23:01.03 | phh | low speed dex is 0x1d |
| 23:02.05 | phh | the only setting I know that goes from 0 to 7 is the VDD of A11 |
| 23:02.10 | phh | don't know if it's related to the slow call |
| 23:02.26 | cr2_ | =2 |
| 23:02.38 | phh | always 2 ? |
| 23:02.45 | cr2_ | fow slow |
| 23:02.54 | cr2_ | =7 for 528 |
| 23:03.09 | cr2_ | but it may adjust arm9 only. you need to check |
| 23:03.26 | Captnoord | lol |
| 23:03.35 | phh | 0..7 setting make me think it might be just A11 VDD setting |
| 23:03.37 | Captnoord | if I want to implement that suspend stuff to the ts |
| 23:03.46 | Captnoord | it needs some major overhoaling |
| 23:03.47 | Captnoord | lolz |
| 23:03.50 | phh | cr2_: wince calls it ARM9 slow down ? |
| 23:03.54 | phh | Captnoord: overhoaling ? |
| 23:03.57 | cr2_ | yes |
| 23:04.03 | phh | ok |
| 23:04.05 | Captnoord | overhauling* |
| 23:04.25 | Captnoord | overhaulin'* |
| 23:04.32 | Captnoord | http://dhd.discovery.com/tv/overhaulin/overhaulin.html |
| 23:04.32 | phh | Captnoord: I'd suggest using tssc-manager |
| 23:06.25 | *** join/#htc-linux Tinyboom_ (n=nahh@ti0121a340-dhcp0200.bb.online.no) |
| 23:07.05 | tmzt | http://cia.vc/stats/project/linux-2.6/.message/176aaf0 |
| 23:09.45 | Captnoord | phh yea |
| 23:09.52 | Captnoord | it seems to be the logic choice |
| 23:10.07 | phh | Captnoord: maybe see if 2.6.32 testing has some newer code |
| 23:10.39 | phh | ah. they have just a diffrent driver. |
| 23:10.44 | Captnoord | I just re-downloaded the gitourius repo |
| 23:10.49 | Captnoord | url? |
| 23:11.04 | phh | git://codeaurora.org/quic/le/kernel/msm.git |
| 23:11.10 | phh | branch eclair |
| 23:11.20 | Captnoord | k |
| 23:11.26 | phh | there is msm_ts.c and msm_touch.c -_-' |
| 23:12.20 | phh | tssc_manager is cleaner. |
| 23:12.56 | phh | and has calibration. |
| 23:13.17 | makkonen | is the main branch going to switch over to it? |
| 23:13.37 | makkonen | (I don't know how well msm_ts works on the gsm devices.) |
| 23:13.50 | Captnoord | msm_ts is just like our current one |
| 23:13.52 | Captnoord | just cleaner |
| 23:14.07 | makkonen | oh. I thought msm_ts was the current one. |
| 23:14.08 | Captnoord | I like the one on the nexus branch |
| 23:14.14 | phh | makkonen: it's yet another one :p |
| 23:14.15 | Captnoord | nah |
| 23:14.19 | Captnoord | yup |
| 23:14.25 | phh | we have 3 or 4 different msm touchscreen drivers -_-' |
| 23:14.32 | Captnoord | its very funny how many people write multiple diff drivers |
| 23:14.34 | makkonen | whatever the current one on the main branch is, it works terribly on the raph800. jumpy and weird. |
| 23:14.37 | Captnoord | for exactly the same device |
| 23:14.48 | phh | Captnoord: I wonder if there is a different driver for vogue too :p |
| 23:14.58 | makkonen | hence tssc-manager. which seems much nicer. |
| 23:14.58 | *** part/#htc-linux c2d (n=Mike@port-92-200-45-82.dynamic.qsc.de) |
| 23:15.02 | phh | makkonen: we will switch to tssc_manager if it works and blackstone's pad is coded |
| 23:15.09 | makkonen | though still imperfect. |
| 23:16.02 | Captnoord | perfection can only be reached by imperfections |
| 23:16.06 | makkonen | good. so the calibration can get rolled into the rootfs on the main distribution. I hate rolling my own rootfs. |
| 23:16.20 | phh | makkonen: just get black's pad working :p |
| 23:16.33 | Captnoord | I need to sleep |
| 23:16.35 | Captnoord | tired |
| 23:16.36 | makkonen | oh. think it won't work for that? |
| 23:16.38 | Captnoord | talk to ya all later |
| 23:16.50 | tmzt | http://feeds.tuaw.com/click.phdo?i=a63c399b208c17cec7b356facb15182c |
| 23:16.55 | tmzt | it has bluetooth? |
| 23:16.57 | phh | makkonen: black's buttons uses touchscreen |
| 23:17.01 | phh | so you'll have to rewrite that part |
| 23:17.05 | makkonen | ah. ugh. |
| 23:17.06 | phh | shouldn't be too hard though |
| 23:17.49 | phh | tmzt: they say they use a dongle |
| 23:18.38 | tmzt | oh |
| 23:19.04 | phh | no fun uh ?:p |
| 23:22.43 | MrPippy | hmm vogue and kaiser don't use the msm touchscreen, they have an i2c one |
| 23:23.30 | phh | MrPippy: yeah, not funny :( |
| 23:23.42 | makkonen | phh: ok. back to 2048x2048 on the calibration, still works fine. (except the targets are no longer square :-p ). Everything seems good. (have to have the same width, height, and all calibration points in the kernel as in the calibration app). Just have to figure out now when the interfaces are enumerated, so I can see if I can add it to the... init script? |
| 23:24.07 | phh | which are interfaces ? |
| 23:24.15 | phh | you mean TS device ? |
| 23:24.23 | makkonen | /sys/class/input/input3/calibration_points |
| 23:24.44 | phh | what do you mean with enumerated ? (ok just explain the whole sentence :p) |
| 23:24.49 | makkonen | haha, sorry. |
| 23:25.12 | makkonen | I mean, when do those bits show up so I can read from and write to them from the command line. |
| 23:25.37 | makkonen | are they there while the init script is running off of rootfs, or do they not appear until something in android creates them? |
| 23:25.58 | phh | should be there when calling the /init script |
| 23:26.06 | makkonen | good. then it should be easy. |
| 23:26.23 | phh | maybe not the /dev/graphics/fb0 device though, you might have to create it by hand |
| 23:26.25 | makkonen | except for possibly having to add a mknod for the fb |
| 23:26.31 | makkonen | right. |
| 23:27.06 | phh | can you read the calibration parameters from the kernel now ? |
| 23:27.45 | makkonen | which parameters? |
| 23:27.56 | makkonen | after a successful calibration, calibration_points shows the good values. |
| 23:28.04 | phh | ok |
| 23:28.12 | phh | so you can write it back after restart ? |
| 23:28.19 | makkonen | so I'll just 'cat /sys/class/input/input3/calibration_points > /sdcard/ts-calibration' |
| 23:28.25 | phh | fine :) |
| 23:28.36 | makkonen | then check for that on boot, and if it exists, dump it back in. |
| 23:28.48 | makkonen | otherwise run the calibration app. easy peasy. |
| 23:29.00 | phh | yup |
| 23:29.06 | phh | we might conflict with the console though |
| 23:29.14 | phh | (not a big deal to fix, it's just an ioctl to add) |
| 23:29.21 | makkonen | ok |
| 23:33.29 | emwe | gnight everyone. giving mtd phram another shot the next days. not worked out yet. no sign of life. not even debug output.. whonder if it is really "in" ... anyways. gnight. |
| 23:36.43 | *** join/#htc-linux mufumbo (n=mufumbo@71-6-82-200.static-ip.telepacific.net) |
| 23:52.49 | stinebd | MrPippy: i got msm7k governor going with nohz |
| 23:53.04 | stinebd | it's at the bootanim now, we'll see how it performs |
| 23:53.26 | stinebd | $ ~/adb shell cat /sys/devices/system/cpu/cpu0/cpufreq/scaling_governor == msm7k |
| 23:53.28 | stinebd | :) |
| 23:53.38 | MrPippy | cool |
| 23:53.58 | MrPippy | i'm trying out the performance governor, just want to see what its like |
| 23:54.34 | stinebd | oh noes, i left cpufreq debug on |
| 23:54.42 | stinebd | lots of output |
| 23:57.08 | stinebd | looks good |
| 23:57.38 | stinebd | we get the conservative tunables too so at least you can tweak to compensate for eclair slowness |
| 23:59.27 | *** join/#htc-linux jumoit (n=samuel@121.77.87.145) |
