| 00:20.29 | *** join/#htc-linux WizMaui (n=WizMaui@62.112.90.231) |
| 00:42.37 | *** join/#htc-linux scatterp- (n=zsirc@spc2-ashf1-0-0-cust798.asfd.broadband.ntl.com) |
| 01:31.25 | lkcl | ok i got a build of blueangel, i even got a screen driver |
| 01:31.46 | lkcl | and i got a kernel crash in tsc2200_ts_init on the platform_device_register o well |
| 01:58.23 | lkcl | blueangel is another device, i believe, with mmc 'read-only' either not used, or possibly reversed |
| 01:58.38 | lkcl | or something |
| 02:00.17 | lkcl | wha-hey! blueangel boot! |
| 02:00.30 | lkcl | well... synthesising hotplug events heeeave |
| 02:11.56 | scatterp- | oh |
| 02:12.02 | scatterp- | i found something i think |
| 02:12.16 | scatterp- | lkcl you know anything about the special edition 2002 rom ? |
| 02:12.19 | lkcl | nope |
| 02:12.43 | lkcl | i stay away from wince except for essential reverse-engineering |
| 02:12.48 | scatterp- | seems its compatable with all htc phones but i wanna check before i fry my phone |
| 02:13.12 | scatterp- | im hopeing to run a playstation emulator on mine untill linux is a little more useable |
| 02:13.28 | lkcl | other than to offer you sympathy for consuming valuable linux-porting time i really don't know what else to say :) |
| 02:13.39 | scatterp- | heh |
| 02:14.04 | lkcl | ok commit time because i like to be a pain in the ass |
| 02:14.14 | scatterp- | i have issues on my phone with the ts driver dont know if that helps ? |
| 02:15.30 | lkcl | the blueangel? |
| 02:15.38 | scatterp- | i have a universal |
| 02:15.47 | scatterp- | but i think it uses the same driver |
| 02:16.55 | lkcl | no the universal has a tsc2306 |
| 02:17.07 | scatterp- | oh ok |
| 02:17.09 | lkcl | the ba (and himalaya) have a tsc2200 |
| 02:17.20 | lkcl | tsc2306 same as on sable |
| 02:29.35 | *** join/#htc-linux WizMaui_ (n=WizMaui@62.112.90.231) |
| 03:21.54 | *** join/#htc-linux WizMaui (n=WizMaui@62.112.90.231) |
| 03:43.49 | *** join/#htc-linux WizMaui_ (n=WizMaui@62.112.90.231) |
| 06:37.47 | *** join/#htc-linux skodde (n=skodde@unaffiliated/skodde) |
| 06:41.31 | *** join/#htc-linux WizMaui (n=WizMaui@62.112.90.231) |
| 07:26.16 | *** join/#htc-linux drap_ (n=drap@91.64.94.166) |
| 07:28.20 | *** join/#htc-linux rob_w (n=rob_w@p549B994B.dip0.t-ipconnect.de) |
| 08:36.24 | *** join/#htc-linux TripleDES (n=sergio@84.77.155.74) |
| 08:46.41 | *** join/#htc-linux Fossi (n=cmewes@195.177.48.115) |
| 09:37.34 | *** join/#htc-linux babeloued (n=Fabrice@faim1.itesoft.com) |
| 10:23.03 | *** join/#htc-linux goxboxlive (n=goxboxli@gw.hjelcowi.no) |
| 10:50.25 | *** join/#htc-linux TripleDE1 (n=sergio@89.129.135.220) |
| 10:55.01 | *** part/#htc-linux rmoravcik (n=rmoravci@pc-3s0zt5w2e4y0vzmhnrzq3a21zqajzfw.users.student.utc.sk) |
| 11:03.53 | goxboxlive | cr2: I saw the umts-tools web link yesterday. I have added it to OE lovaly, and trying to build it now. Do you knoe if python will work with Opie? |
| 11:17.41 | *** join/#htc-linux Fossi (n=cmewes@195.177.48.115) |
| 11:30.55 | *** join/#htc-linux TripleDES (n=sergio@89.129.143.241) |
| 11:31.27 | *** join/#htc-linux psokolovsky (n=psokolov@232.usernat.ip.net.ua) |
| 11:31.50 | psokolovsky | Hi! |
| 11:32.29 | psokolovsky | Anybody knows what happened yesterday with that Spyro's massive commit? |
| 11:37.25 | psokolovsky | cr2, any hints ^? |
| 11:45.18 | goxboxlive | hi psok. Has something ahppend in the hh.org cvs ? |
| 11:49.11 | psokolovsky | goxboxlive, yep, some large amount of random code was committed this night |
| 11:49.28 | psokolovsky | anyone made a build today? does it still boot? |
| 11:50.01 | psokolovsky | and either way, I would like to know how that happened, w/o any notification to kernel-discuss or something |
| 11:51.51 | goxboxlive | ok, i can try to build one now, but i cant test it until i am home from work. |
| 11:52.12 | psokolovsky | if that's not too hard for you |
| 11:57.52 | *** join/#htc-linux rob_w (n=rob_w@p549B994B.dip0.t-ipconnect.de) |
| 12:01.25 | *** join/#htc-linux psokolovsky_ (n=psokolov@ip.85.202.124.214.dyn.sub-9.broadband.voliacable.com) |
| 12:12.04 | *** join/#htc-linux psokolovsky__ (n=psokolov@ip.85.202.124.214.dyn.sub-9.broadband.voliacable.com) |
| 12:29.57 | cr2 | psokolovsky_: i did not test universal, but made many changes for blueangel, lkcl has added many ba files too. i still can compile it :) |
| 12:30.26 | cr2 | goxboxlive: the umts-tool are perl scripts ? |
| 12:31.48 | psokolovsky__ | cr2, well, we with Spyro currently negotiate creating separate branch for hist changes still. If it compiles and even boots, that's nice, but what we will do once we hit first issues? Whose those issues - ours, 2.6.19's or those changes'? I don't want to decide, I won't to give users straight 2.6.19 with just our ports changes and sign off of it. |
| 12:36.15 | goxboxlive | psokolovsky__ : The universal source tree compiles. Havent tested it. |
| 12:36.26 | psokolovsky__ | ok |
| 12:41.30 | cr2 | i have a list of modified files, but many of them are certainly legitimate. |
| 12:42.08 | psokolovsky__ | cr2, it's just random import for git |
| 12:42.18 | psokolovsky__ | noone knows what it adds and what it breaks |
| 12:42.51 | psokolovsky__ | I'm switching that stuff to branch now, sorry for any possible breakage ;-(( |
| 12:43.26 | psokolovsky__ | in particular, I would like to refrain from commits in the near 1 hour |
| 12:43.50 | psokolovsky__ | I'll of course port any port changes made during that commit... |
| 12:44.50 | babeloued | hi |
| 12:45.22 | cr2 | ok. |
| 12:46.57 | babeloued | cr2 do you test all change ? |
| 12:48.35 | psokolovsky__ | cr2, btw, what about suspend/resume state with .19 on HTC devices? |
| 12:53.01 | babeloued | cr2 : i put tsc2200_ts.c in input/touchscreen and tsc2200_keys.c in input/keyboard |
| 13:04.51 | *** join/#htc-linux Fossi (n=cmewes@195.177.48.115) |
| 14:10.03 | *** join/#htc-linux Fossi (n=cmewes@195.177.48.115) |
| 14:22.26 | psokolovsky | cr2, revert completed. boots ok for h4000. I ported all your changes, of course. But please check it (compile/run). I apologize for any bits which could have been lost beforehand. |
| 14:25.52 | cr2 | ok, shit happens. never mind. |
| 14:26.47 | goxboxlive | cr2: it's not any plan to support ak4641 in to Asoc |
| 14:27.16 | cr2 | i think no, because they don#t have such hardware. |
| 14:27.23 | goxboxlive | Why is Asoc bether than the one we are using today? |
| 14:28.33 | cr2 | i thought it offered better control options, but now i'm not convinced about the overall architecture at all. |
| 14:29.35 | cr2 | the phone-related sound is under full control of the BB, ak4641 is only a bolt-on thing, so we need a clever userspace switching daemon anyway. |
| 14:30.44 | cr2 | or it seems so now. there are 2+2 audio channels supported by MSM without any clear purpose (and they are not activated in any known operation) |
| 14:32.01 | psokolovsky | goxboxlive, cr2: asoc is not better ;-). it's just solution for some codecs, which didn't exist before. it doesn't cover all codecs, obviously. |
| 14:34.50 | goxboxlive | i c |
| 14:35.16 | goxboxlive | well time to go home, i am fed up tdoay. HHmm have to be taht it's monday. bbl |
| 14:35.19 | *** part/#htc-linux goxboxlive (n=goxboxli@gw.hjelcowi.no) |
| 14:47.55 | *** join/#htc-linux hlbot802 (n=adm@iclem.net) |
| 14:53.03 | babeloued | cr2 why tsc2200 is removed from drivers/soc ? |
| 14:57.32 | *** join/#htc-linux lkcl_ (i=cgiircus@gateway/web/cgi-irc/chriscole.info/x-d947cb89d9d87530) |
| 15:01.26 | *** join/#htc-linux lkcl__ (i=cgiircus@gateway/web/cgi-irc/chriscole.info/x-b4b0ff11b764e135) |
| 15:02.13 | lkcl_work | allo darlins |
| 15:03.27 | babeloued | hi lkcl |
| 15:03.58 | lkcl_work | hi babel |
| 15:04.01 | psokolovsky | babeloued, it's not removed, relax. read commit messages (some people bother to make them sensible) and update you tree. |
| 15:04.14 | lkcl_work | did you notice my cvs commits of the htcblueangel stuff yesterday? |
| 15:04.20 | lkcl_work | allo psoko. |
| 15:04.30 | lkcl_work | did i do any damage, yesterday, psoko? |
| 15:04.45 | lkcl_work | also - i wanted to talk to you about committing pierrox's stuff |
| 15:05.03 | cr2 | babeloued: i have not looked at the current tree. |
| 15:05.51 | lkcl_work | allo cr2 |
| 15:05.52 | cr2 | lkcl_work: hi. did you get my e-mail ? |
| 15:05.52 | lkcl_work | whatwhatwhat? wossgoinon? |
| 15:05.59 | psokolovsky | lkcl_work, Hi! Nope, but I had to do some reverts in repo ;-(. I swear I did everything to preserve your changes, but please check them and fix if anything wrong, I beforehand apologize if that's the case. |
| 15:06.01 | babeloued | cr2 : psokolovsky answer to me, it s ok for BA all compile and boot :) |
| 15:06.17 | lkcl_work | cr2: not yet - hey i installed an imap proxy on my server so i might actually be able to get it. 1sec |
| 15:06.33 | psokolovsky | lkcl_work, as for eten stuff, cr2 would know better, I guess ;-) |
| 15:06.47 | *** join/#htc-linux rmoravcik (n=rmoravci@158.193.198.3) |
| 15:07.16 | lkcl_work | cr2: brilliant! |
| 15:07.24 | cr2 | cvs update: blueangel_ds1wm.c is no longer in the repository |
| 15:07.35 | cr2 | cvs update: blueangel_tsc2200.c is no longer in the repository |
| 15:07.52 | cr2 | lkcl_work: so your changes are reverted now ? |
| 15:08.00 | lkcl_work | psoko: oh oh well oops no problem, i have three cvs checkouts at home so i can update one |
| 15:08.07 | babeloued | lkcl_work : can we move tsc2200_keys and tsc2200_ts to driver / input tree ? |
| 15:08.33 | lkcl_work | cr2: oops :) |
| 15:09.22 | cr2 | babeloued: of course they could be moved. |
| 15:09.46 | psokolovsky | cr2, ok, so I forget to cvs add those files, fixing now... |
| 15:10.27 | lkcl_work | babel: i only put them there because it looked like a good idea - yes drivers/input is a much better place duh. |
| 15:10.46 | lkcl_work | do you also get a kernel oops due to enable_irq being called more than once? |
| 15:11.22 | lkcl_work | i get a log message 'Unbalanced IRQ' when the enable_irq() function is called in _both_ the tsc2200_keys.c _and_ tsc2200_ts.c |
| 15:12.13 | lkcl_work | oh. where's blueangel_sound module? ha ha |
| 15:12.45 | lkcl_work | i wondered why sound wasn't working when i tried it - i just remembered you have to have the gpio initialisation helper module enabled |
| 15:13.02 | psokolovsky | babeloued, lkcl_work, cr2: Ok, so now files are actually resurrected. sorry for hassle. |
| 15:14.39 | cr2 | psokolovsky: blueangel is a bit messed up now anyway. |
| 15:14.54 | lkcl_work | hey, if i wasn't using cgi-irc at work i wouldn't even have known, psoko |
| 15:15.31 | psokolovsky | cr2, it should be at the state it was after all your/lkcl's commits ;-) |
| 15:15.48 | cr2 | yes :) |
| 15:16.20 | psokolovsky | cr2, btw, blueangel_ds1wm.c is not very bright idea, it should be redone on asic3_base level, like hamcop_base do, etc. |
| 15:16.31 | cr2 | i have made some cosmetic changes. |
| 15:17.10 | cr2 | these drivers are not my work, actually. |
| 15:18.03 | cr2 | but there were some longstanding bugs corrected in the process. |
| 15:18.20 | lkcl_work | oh - i notice that nobody's commented on my mad sound-routing email on hh.org |
| 15:18.39 | lkcl_work | i think i'd better write it up as a wiki page for people to be able to comment on / edit, instead... |
| 15:18.52 | cr2 | still i know about some more, and 2 drivers need a complete rewrite. |
| 15:19.35 | cr2 | lkcl_work: audio routing is really important only the phones, and nobody works on that. |
| 15:19.53 | psokolovsky | cr2, ok, just don't put too much effort in ds1wm handling with such adhoc drivers, I'd say - it would need to be redone anyway. |
| 15:20.10 | lkcl_work | yehh... |
| 15:20.19 | psokolovsky | lkcl_work, yep, you really should cc: ozx folks on that ;-) |
| 15:20.36 | psokolovsky | ezx |
| 15:20.38 | cr2 | psokolovsky: agreed. i just don't want to see the blueangel stalled in the stone age. |
| 15:20.55 | lkcl_work | psoko: o yeh good point |
| 15:20.58 | psokolovsky | sure! |
| 15:21.15 | cr2 | psokolovsky: ezx seems to be dead after all the developers moved to moko. |
| 15:21.24 | lkcl_work | cr2: ain't gonna happen. nor the himalaya, neither! i love those phones! well, except for the asic5 of course. |
| 15:21.28 | psokolovsky | ;-) |
| 15:21.44 | cr2 | and openmoko is a joke (on the hardware side) compared to the universal. |
| 15:21.48 | babeloued | lkcl_work : i've irq warning too, but it works under OPIE, i ve testeduniversal opie image :) |
| 15:23.01 | cr2 | asic5 needs some "wi" work. there are 2 pcmcia areas and a serial port there. |
| 15:23.20 | cr2 | only himalaya and alpine have asic5. |
| 15:23.55 | cr2 | i can put an approximate register layout into wiki. |
| 15:24.47 | cr2 | i have some notes also on the asic-i-don#t-know-which in wallaby. but nobody works on this beast. |
| 15:26.41 | cr2 | heh. and if i had "wi" for linux-arm, i could do something for openezx too ;-) |
| 15:26.58 | babeloued | cr2 : for ds1wm i only take pH5 code from magician and made some change |
| 15:29.31 | cr2 | babeloued: it does not really matter, if the driver will work :) try to solve the unknown problems with it: owm clock enable/disable. |
| 15:30.33 | *** join/#htc-linux rmoravcik (n=rmoravci@158.193.198.3) |
| 15:37.54 | lkcl_work | cr2: openmoko is the free software applications suite being developed by the open source developers hired by FIC. the neo1973 is the first phone to _use_ the openmoko apps suite |
| 15:41.17 | rob_w | lkcl_work, i am keen to see how they did those screenshots at openmoko .. woder what widgets those are |
| 15:41.39 | babeloued | cr2 , lkcl_work: we need to do something for MMC readonly on BA. I read a suggestion on kernel mailing list about this |
| 15:42.03 | cr2 | lkcl_work: i know, but will they really build the suite keeping in mind the UMTS, videophone, bluetooth headset, wifi in mind ? |
| 15:42.43 | cr2 | rob_w: afaik it's a new widget set designed for fullscreen operation. |
| 15:43.02 | cr2 | like wince <3.0 :) |
| 15:43.49 | rob_w | cr2, and i wonder how this then performs |
| 15:45.59 | cr2 | lkcl_work: just think about the system design problems we have with the universal: 2 keyboards, 2 cams, 3 speakers, 3 mics, bt, wifi, umts. current openmoko hardware has nothing of it. |
| 15:46.25 | cr2 | rob_w: it should be fast. |
| 15:47.02 | rob_w | hehe it better should ;-) |
| 15:50.03 | cr2 | it will take some effort to merge the etenG500, looking at linux-2.6.18.2-phe-g500.diff |
| 15:50.57 | lkcl_work | cr2: the first phone is going after the gsm market, not 3g. |
| 15:51.01 | lkcl_work | oops about the g500... |
| 15:51.42 | rob_w | i found the black eten nice in its design .. |
| 15:52.19 | cr2 | not oops, but there are many changes to the mainline code. i'd better consider posting it to the k-d. |
| 15:52.36 | lkcl_work | babeloued: yep. currently i've got a #ifndef CONFIG_MACH_HW6900 || CONFIG_MACH_BLUEANGEL around the code that checks the read-write status, in mmc_block.c |
| 15:52.40 | cr2 | it may break other s3c24xx ports |
| 15:52.56 | lkcl_work | ok, then we need to kick pierrox's ass :) |
| 15:55.01 | cr2 | i think he should prepare a split patch for 2.6.19, and put the g500-tests to a mach-s3c24x/g500 subdirectory. |
| 15:56.15 | rmoravcik | cr2: where is linux-2.6.18.2-phe-g500.diff patch? |
| 15:56.20 | *** join/#htc-linux asylumed (n=insanity@196.211.116.2) |
| 15:57.30 | lkcl_work | rmoravcik: handhelds.org wiki, g500... 1sec... |
| 15:58.10 | rmoravcik | lkcl_work: thanks, i have it |
| 15:58.11 | lkcl_work | http://www.handhelds.org/moin/moin.cgi/EtenG500 |
| 15:58.25 | lkcl_work | http://www.handhelds.org/moin/moin.cgi/EtenG500Downloads |
| 15:58.32 | lkcl_work | allo asylumed |
| 15:58.53 | lkcl_work | asylumed: i committed some of your work, yesterday night, and it got reverted already ha ha |
| 15:59.14 | lkcl_work | oops he's gone already o dear what did i say wrong? :) |
| 16:02.05 | rmoravcik | lkcl_work: it look like a rest of my rx3715 kernel patch |
| 16:03.18 | lkcl_work | yep - the only thing not good about it is that it adds a drivers/test subdirectory |
| 16:03.29 | lkcl_work | other than that it's pretty damn good, imo |
| 16:04.03 | rmoravcik | lkcl_work: there is older version of udc driver |
| 16:07.26 | *** join/#htc-linux lkcl_work (i=cgiircus@gateway/web/cgi-irc/chriscole.info/x-28dad6fc8d56bec8) |
| 16:07.52 | lkcl_work | oops got disconnected. damn cgi / irc gateway i'm freeloading off of :) |
| 16:09.10 | lkcl_work | so - sorry: there's an existing udc driver, it's older than the one written by pierrox? |
| 16:10.06 | rmoravcik | lkcl_work: yes, guys from h1940 has it in they git, i have it too |
| 16:10.23 | rmoravcik | lkcl_work: i working on rx3715 (s3c2440 based) |
| 16:12.16 | rmoravcik | lkcl_work: backlight is good, keyboard driver was updated and renamed to s3c_buttons, but pierrox can use gpio_keys from hh.org cvs |
| 16:21.52 | Kevin2 | Hi |
| 16:35.01 | *** join/#htc-linux rmoravci1 (n=rmoravci@158.193.198.3) |
| 16:53.58 | *** join/#htc-linux pH5 (n=ph5@e178200111.adsl.alicedsl.de) |
| 16:59.38 | *** join/#htc-linux rmoravcik (n=rmoravci@158.193.198.3) |
| 17:06.15 | *** part/#htc-linux babeloued (n=Fabrice@faim1.itesoft.com) |
| 17:19.55 | *** join/#htc-linux goxboxlive (n=goxboxli@9.80-202-160.nextgentel.com) |
| 17:31.23 | cr2 | lkcl: what a losers. i share your opinion :) |
| 17:31.28 | cr2 | There is an "open" GPS |
| 17:31.28 | cr2 | hw/fw solution out there, but it is open only because the fw side of it |
| 17:31.28 | cr2 | went belly-up and the hw side decided to keep the silicon in their |
| 17:31.28 | cr2 | catalog. |
| 17:31.40 | cr2 | <PROTECTED> |
| 17:31.40 | cr2 | support it. PhDs have less time on their hands than GUI developers, |
| 17:31.41 | cr2 | evidently. |
| 17:33.53 | cr2 | lol. |
| 17:34.20 | cr2 | as if i was a GUI developer ;-) |
| 17:36.57 | *** join/#htc-linux rob_w (n=bob@p85.212.166.135.tisdip.tiscali.de) |
| 17:37.37 | *** join/#htc-linux booba (n=booba@lns-bzn-47f-81-56-252-135.adsl.proxad.net) |
| 17:49.27 | *** join/#htc-linux rob_w (n=bob@p85.212.166.135.tisdip.tiscali.de) |
| 17:52.03 | lkcl | who are losers, cr2? |
| 17:52.17 | lkcl | what did i do now, ha ha |
| 17:54.36 | cr2 | Here is your answer from a senior engineer at Global Locate: |
| 17:55.24 | cr2 | lkcl: i was reading the openmoko archived mailing list. |
| 17:59.35 | cr2 | i translate it in a such way: "You GUI developers are dumb, have never heard of signal processing and should shut up. use our library and pay $$$" |
| 17:59.49 | cr2 | i think i have heard it already somewhere :) |
| 18:00.46 | *** join/#htc-linux BabelOued (n=Fabrice@lun34-2-82-238-28-28.fbx.proxad.net) |
| 18:05.15 | cr2 | it's a pity there is no e-mail there. i would have sent him this link http://en.wikipedia.org/wiki/Johannes_Kepler |
| 18:07.53 | *** join/#htc-linux rmoravcik (n=rmoravci@pc-3s0zt5w2e4y0vzmhnrzq3a21zqajzfw.users.student.utc.sk) |
| 18:12.29 | *** join/#htc-linux FossiAfk (n=Fossi@e176109186.adsl.alicedsl.de) |
| 18:41.24 | cr2 | http://wiki.xda-developers.com/index.php?pagename=UniversalBluetooth |
| 18:41.31 | cr2 | updated, but not complete. |
| 18:43.54 | *** join/#htc-linux asylumed (n=insanity@196.211.28.91) |
| 18:57.02 | Kevin2 | cr2: I've extended "wi" to support memory polling during interrupts and trace events. |
| 18:57.22 | Kevin2 | I have a new binary in svn -- haret.exe -- rev 138 |
| 19:11.26 | goxboxlive | psokolovsky: I tried out apm --suspend with .19. It doesn suspend, but we dont have any wake up key working now, so i am not able to try resume. |
| 19:12.07 | goxboxlive | s/doesn/does/ |
| 19:12.08 | *** join/#htc-linux asylumed (n=insanity@196.211.28.91) |
| 19:17.53 | *** join/#htc-linux asylumed (n=insanity@196.211.116.2) |
| 19:20.33 | goxboxlive | cr2: help me out here. What else could getkey be used with beside opie. Is it a a application that could be usefull for other stuff? |
| 19:30.59 | *** join/#htc-linux drap__ (n=drap@91.64.94.166) |
| 19:31.40 | *** join/#htc-linux drap_ (n=drap@91.64.94.166) |
| 19:36.27 | cr2 | goxboxlive: gpio0 is setup as a wakeup source. the bug must be somewhere else. |
| 19:37.05 | cr2 | what about "read" in bash ? can it be used as a replacement ? |
| 19:37.48 | goxboxlive | maybe, i can try to edit opie running on my machine an see if it will. |
| 19:39.30 | cr2 | but it may block |
| 19:39.57 | goxboxlive | ok, but how should i use the read? Just replace getkey? |
| 19:40.11 | cr2 | <PROTECTED> |
| 19:40.29 | cr2 | but it's in real gnu bash. |
| 19:40.39 | cr2 | read -t 5 |
| 19:41.36 | cr2 | hmm. how i can have a different bt address in linux than in wince ? |
| 19:41.50 | goxboxlive | I will check that my self now |
| 19:41.52 | cr2 | http://www.digitalmunition.com/setbd-bluez.c |
| 19:42.10 | rob_w | do we have any working angstrom images, yet ? |
| 19:42.14 | cr2 | ogf 0x3f, ocf 0x000d |
| 19:42.16 | Kevin2 | cr2: Did you see my IM about memory polling from wi? |
| 19:42.27 | cr2 | ogf 0x3f, ocf 0x0022 |
| 19:42.38 | cr2 | but it's ericsson specific. |
| 19:42.48 | cr2 | Kevin2: yes, what is the syntax ? |
| 19:43.05 | Kevin2 | addirqwatch P2V(0x0a000000) 0x20ff 16 0 |
| 19:43.20 | Kevin2 | would watch egpio on apache and ignore wifi irq. |
| 19:43.32 | BabelOued | rob_w: yes, goxboxlive have a lot :) |
| 19:44.10 | cr2 | Kevin2: does P2V do ioremap() ? |
| 19:44.52 | cr2 | rob_w: you should use the image for universal, and whine at goxboxlive to implement ${MACHINE} there :) |
| 19:46.06 | Kevin2 | cr2: It just provides the virtual address. (It's a little simpler than trying to remember the wince virtual mappings.) |
| 19:46.30 | cr2 | Kevin2: agreed. |
| 19:46.52 | cr2 | heh. they could have used hcitool to send this command. |
| 19:47.44 | cr2 | BabelOued: you should trace HWUART on ba for the BT init data. |
| 19:48.24 | goxboxlive | cr2: read is not in the opie-image, have to install or add it to the image. |
| 19:49.14 | BabelOued | cr2: ok, i ve to use your script to do this ? |
| 19:51.21 | cr2 | BabelOued: just use "wi" on the hwuart address. |
| 19:51.56 | cr2 | goxboxlive: read is a bash builtin command, like 'cd' |
| 19:52.18 | goxboxlive | ok |
| 19:52.36 | goxboxlive | not in busybox |
| 19:54.45 | goxboxlive | cr2: About my bt adress. xx:yy:zz:nn:mm:oo. The xx:yy are diffrent for me. |
| 19:54.56 | cr2 | yes. the same here. |
| 19:55.08 | goxboxlive | ok |
| 19:55.11 | cr2 | i wonder how it happens. |
| 19:55.34 | goxboxlive | is your starting with 08.00 ? |
| 19:55.40 | cr2 | yes. |
| 19:55.49 | goxboxlive | :-) strange |
| 19:55.53 | cr2 | i have found that here |
| 19:55.55 | cr2 | http://www.handhelds.org/pipermail/familiar/312/31213.html |
| 19:55.59 | cr2 | but no replies. |
| 19:56.19 | cr2 | maybe wince is changing it |
| 19:56.34 | cr2 | by some 0x3f command. |
| 19:56.40 | cr2 | i need to complete my list. |
| 19:57.08 | cr2 | this is only the bts script summary |
| 19:57.09 | goxboxlive | ok |
| 19:57.10 | cr2 | http://wiki.xda-developers.com/index.php?pagename=UniversalBluetooth |
| 19:57.34 | BabelOued | cr2:I do not have a peripheral bluetooth at home |
| 19:58.02 | cr2 | BabelOued: it does not matter. |
| 19:58.14 | cr2 | you will trace the chip init sequence. |
| 19:59.14 | cr2 | goxboxlive: in the post above, that addr is also 08.00 |
| 19:59.22 | cr2 | what do you have in wince ? |
| 20:00.15 | goxboxlive | 00:12 |
| 20:00.48 | cr2 | 00:12 here too :) |
| 20:00.53 | goxboxlive | :-) |
| 20:01.00 | cr2 | it seems wince is playing dirty. |
| 20:01.15 | cr2 | is 08:00 the standard TI setting ? |
| 20:01.23 | goxboxlive | seems like it is |
| 20:01.48 | cr2 | Kevin2: what are the first 2 bytes of your BT addr ? |
| 20:02.13 | cr2 | i'll check ba |
| 20:02.28 | Kevin2 | cr2: How do I check? |
| 20:02.46 | goxboxlive | hciconfig -a |
| 20:02.58 | cr2 | 'hcitool scan' on the notebook |
| 20:03.05 | cr2 | oh yes. in linux. |
| 20:03.28 | Kevin2 | hcitool scan on my desktop or on the pda? |
| 20:03.36 | goxboxlive | desktop |
| 20:03.39 | cr2 | scan on desktop |
| 20:03.47 | cr2 | hciconfig on pda |
| 20:06.01 | Kevin2 | Scanning ... |
| 20:06.01 | Kevin2 | <PROTECTED> |
| 20:06.11 | Kevin2 | That's from under windows. |
| 20:06.25 | goxboxlive | same as ours |
| 20:07.23 | cr2 | and in linux ? |
| 20:08.30 | cr2 | ba is 00:09 |
| 20:09.05 | cr2 | in wince2003 |
| 20:10.16 | goxboxlive | cr2: where lkcl able to test his camera on sable? |
| 20:11.34 | Kevin2 | My familiar image is bailing with .19 kernel.. I can't test under linux on the pda. |
| 20:11.51 | cr2 | ok. |
| 20:12.21 | cr2 | not surprising the bdaddr failed on hx4700. it's for ericsson, not ti. |
| 20:13.01 | cr2 | not true :) |
| 20:13.05 | cr2 | #define OCF_TI_WRITE_BD_ADDR 0x0006 |
| 20:19.08 | *** join/#htc-linux rob_w (n=bob@p85.212.166.135.tisdip.tiscali.de) |
| 20:28.11 | psokolovsky | goxboxlive, you don't have wake up button? ;-D |
| 20:28.18 | cr2 | goxboxlive: can you check the phone mic ? |
| 20:28.22 | psokolovsky | goxboxlive, thanks for that ticket - it's live and assigned to me know. |
| 20:30.23 | *** join/#htc-linux asylumed (n=insanity@196.211.116.2) |
| 20:36.45 | goxboxlive | psokolovsky: We had a wakup button, but after some changes with the keyboard driver we have lost it. I guess cr2 could explain some more in details. |
| 20:37.20 | goxboxlive | cr2: CHeck phone mic how? Running linux and using the phone or debuging with wince? |
| 20:37.34 | psokolovsky | goxboxlive, cr2, "Power" button should be connected to GPIO0, and you guys should really use gpio_keys for it. |
| 20:40.20 | cr2 | psokolovsky: it is. |
| 20:40.40 | cr2 | goxboxlive: call in linux, and send |
| 20:40.55 | cr2 | at@audioset=3,0,0,2 |
| 20:40.58 | cr2 | or |
| 20:41.04 | psokolovsky | cr2, then, it should work. if it doesn't, resume doesn't work for univiersal, you should find and converting any offending drivers... |
| 20:41.05 | cr2 | at@audioset=7,0,0,2 |
| 20:41.26 | cr2 | psokolovsky: yes, it was working in .18 |
| 20:41.52 | psokolovsky | cr2, so fix it for .19 ;-) |
| 20:45.17 | goxboxlive | cr2: while doing a call, or before making the call? |
| 20:45.42 | cr2 | goxboxlive: try before. |
| 20:46.00 | goxboxlive | ok |
| 20:50.05 | BabelOued | cr2: to trace bt on bt it is this : |
| 20:50.06 | BabelOued | <PROTECTED> |
| 20:50.49 | cr2 | what is the btuart address is dmesg ? |
| 20:50.57 | cr2 | hwuart, sorry. |
| 20:51.44 | BabelOued | cr2: oh, i forgot to look at this :) since it does not works |
| 20:52.12 | BabelOued | pxa2xx-uart.3: ttyS3 at MMIO 0x41600000 (irq = 0) is a HWUART |
| 20:52.16 | BabelOued | better :) |
| 20:53.49 | cr2 | set trace 0xaf600000 |
| 20:54.05 | BabelOued | thx |
| 20:54.44 | BabelOued | how do you do this ? it is an offset ? |
| 20:54.50 | lkcl | allo darlins |
| 20:55.19 | goxboxlive | cr2: this one made my speaker that usually works dead: at@audioset=3,0,0,2 |
| 20:55.42 | goxboxlive | The other one gave me "4" as output and i guess that means "failed" |
| 20:56.32 | BabelOued | cr2: tracemask 0xff ? or more ? |
| 20:57.37 | goxboxlive | cr2: This one brings it back to the lcd spekaer : at@audioset=7,0,0,2 |
| 20:59.25 | cr2 | goxboxlive: ok |
| 20:59.38 | cr2 | but does it work ? |
| 20:59.51 | cr2 | BabelOued: i need to look into the source :) |
| 21:00.20 | cr2 | BabelOued: 0xff is ok. |
| 21:00.52 | goxboxlive | The only thing that works is the lcd side speaker. And that one works with at@audioset=7,0,0,2. The mic is dead, both on the universal and when trying the headset. |
| 21:01.25 | goxboxlive | So, still there is no mic working when making a call. |
| 21:01.44 | goxboxlive | Other stuff i should try while i have the phone running? |
| 21:02.07 | cr2 | audioset 0, 3 |
| 21:02.16 | cr2 | instead of 7. |
| 21:02.29 | cr2 | 2 is bluetooth (i think) |
| 21:03.03 | cr2 | 1,4,5,6 are unknown. and not observed. |
| 21:03.18 | BabelOued | cr2: http://linuxbox2.free.fr/blueangel/trace |
| 21:04.04 | cr2 | BabelOued: is it the whole trace ? |
| 21:04.28 | BabelOued | cr2: yes to bluetooth led power on |
| 21:06.46 | cr2 | ok. |
| 21:07.01 | BabelOued | hx4700 is hwuart too ? |
| 21:07.14 | *** join/#htc-linux rmoravcik (n=rmoravci@pc-3s0zt5w2e4y0vzmhnrzq3a21zqajzfw.users.student.utc.sk) |
| 21:08.29 | goxboxlive | cr2: well the others gave me nothing beside killing the lcd speaker. Still no mic working. It's dead. |
| 21:09.45 | cr2 | ok, so 7 enables the lcd speaker ? |
| 21:09.55 | goxboxlive | yes |
| 21:10.05 | cr2 | ok. interesting. |
| 21:12.08 | goxboxlive | i dint try 1,4,5,6 thow |
| 21:12.15 | cr2 | i have not seen them in operation. |
| 21:12.31 | cr2 | btw, if you use the headphone, can you record the call ? |
| 21:13.56 | goxboxlive | cr2: i didnt try it out. But as i said, the headphones are dead when making a phonecall. The only speaker that works is the lcd speaker. |
| 21:15.19 | cr2 | ok. so some gpio/i2c manipulation is still needed. |
| 21:15.57 | cr2 | Kevin2: can i trace a 16bit register with 'wi' ? |
| 21:22.42 | *** join/#htc-linux rejo1 (n=rejon@adsl-75-5-124-98.dsl.pltn13.sbcglobal.net) |
| 21:22.42 | *** part/#htc-linux rejo1 (n=rejon@adsl-75-5-124-98.dsl.pltn13.sbcglobal.net) |
| 21:25.16 | goxboxlive | cr2: this one also make sound in the lcd speaker : AT@AUDIOSET=2,0,0,2 |
| 21:25.51 | cr2 | 2 and 7 are paired. 4 and 5 , 1 and 6, 3 and 8. |
| 21:26.22 | cr2 | i think 7 is called when ak4641 power is down, otherwise 2. |
| 21:27.20 | goxboxlive | When i do the audioset command, i can hear it clicks in the other phone. |
| 21:27.20 | goxboxlive | ok |
| 21:27.43 | cr2 | what if you do 'od -x /dev/dsp' while running the phone call ? |
| 21:29.35 | goxboxlive | is flush the screen with 0050060 837b 7b81 7f81 827e 7f80 7f7f 8081 807f |
| 21:31.02 | cr2 | cat /dev/dsp > junk.wav ? |
| 21:31.15 | cr2 | 7f does not look very good. |
| 21:31.37 | *** join/#htc-linux rmoravcik (n=rmoravci@pc-3s0zt5w2e4y0vzmhnrzq3a21zqajzfw.users.student.utc.sk) |
| 21:31.38 | goxboxlive | o, but the numbers where diffrent all the way |
| 21:36.48 | goxboxlive | cr2: well it created something, but the screen is flushed junk. The file where 90k |
| 21:37.20 | goxboxlive | When i tried it in opie it said missing demux plugin. I can try it on my laptop |
| 21:40.28 | cr2 | ok. |
| 21:41.30 | goxboxlive | well, it didnt play on my laptop either. Well have to go. |
| 21:45.38 | *** join/#htc-linux florian__ (n=fuchs@87.193.63.169) |
| 21:53.51 | Kevin2 | cr2: Trace a 16bit register using the pxa memory watch feature, or using my new memory polling stuff? |
| 21:53.58 | Kevin2 | In either case, yes. |
| 21:55.07 | cr2 | memory watch. |
| 21:56.03 | Kevin2 | You watch a single byte address with the PXA stuff - it will cause an exception when anything writes (or reads) to that byte. This includes 4-byte, 2-byte, and 1-byte accesses. |
| 21:56.36 | cr2 | 10000204 (phys) |
| 21:57.11 | cr2 | i don't see anything at 202, so some reason. |
| 21:57.26 | cr2 | what mask should i use ?? |
| 21:58.19 | cr2 | i can see the data with "wreg16 0x10000204 60", but i'd prefer "wi". |
| 21:59.58 | lkcl | btw where's the blueangel sound init module?? |
| 22:00.13 | Kevin2 | cr2: Is something writing to it at another virtual address? |
| 22:00.20 | Kevin2 | Is the hardware modifying it itself? |
| 22:01.27 | cr2 | Kevin2: it always catches changes in 200 |
| 22:01.48 | Kevin2 | What is at that address? |
| 22:02.13 | Kevin2 | What did you use for "set traceXX" |
| 22:02.19 | cr2 | lkcl: i have pressed the wrong button and downloaded the ephemeris from the evil company :) |
| 22:02.25 | lkcl | mwahahaha |
| 22:02.40 | cr2 | set trace 0x92000202 |
| 22:02.57 | lkcl | i hope you used tor and a proxy |
| 22:03.14 | cr2 | lkcl: no. dhcp :) |
| 22:03.21 | lkcl | ha ha |
| 22:03.34 | Kevin2 | cr2: That will catch reads/writes to x02 -- if you want to catch reads/writes to x04 - use "set trace 0x92000204". |
| 22:04.02 | cr2 | Kevin2: but which mask should i use then ? |
| 22:04.30 | Kevin2 | To catch both? set trace 0x92000200 ; set tracemask 0x9200020f |
| 22:04.37 | Kevin2 | oops |
| 22:04.44 | Kevin2 | set trace 0x92000200 ; set tracemask 0xf |
| 22:05.00 | cr2 | f ? |
| 22:05.18 | Kevin2 | Ignore the bits in the last half-byte. |
| 22:05.47 | Kevin2 | so, trace 0x92000200 - 0x9200020F |
| 22:05.58 | cr2 | it's too much. |
| 22:06.30 | Kevin2 | You're trying to ignore a register at the same time? If so, which address? |
| 22:06.51 | cr2 | i'd like to see 202 (send) and 204 (recv) |
| 22:06.56 | cr2 | both 16bit. |
| 22:07.33 | Kevin2 | set trace 0x92000202 ; set trace2 92000204 ; set tracemask -1 |
| 22:07.52 | cr2 | 200 have some fifo bits and other undocumented stuff. |
| 22:07.59 | cr2 | ok, makes sense :) |
| 22:10.17 | Kevin2 | cr2: I'm not sure what you're doing - but be aware that the "wi" stuff doesn't actually read the registers, so it is okay to trace sensitive hardware addresses. It only catches arm instructions that are writing/reading to those addresses. |
| 22:12.32 | lkcl | has anyone written a pxa ssp asoc driver yet? like snd-htcuniversal_audio.c but ported to asoc? |
| 22:13.25 | Kevin2 | lkcl: Do you still have the hermes phone (a samsung based processor)? |
| 22:13.32 | lkcl | Kevin2: yep. |
| 22:13.57 | BabelOued | lkcl: pH5 have starded one i think |
| 22:14.00 | lkcl | i've been filling in the wiki page |
| 22:14.03 | Kevin2 | Can you see if new mainline haret runs on it? The one from svn at - svn://www.shadowmite.com/home/svn/ppc6700/haret |
| 22:14.33 | lkcl | have you been pulling in stuff from that haret-sc244x.exe version? |
| 22:15.02 | lkcl | does it compile with arm-wince-gcc? |
| 22:15.36 | lkcl | BabelOued: ok, ta. |
| 22:15.46 | Kevin2 | lkcl: Yes, it compiles with gcc. No, I haven't found anything really in the sc244x version to pull in. |
| 22:16.00 | lkcl | i wanted to copy something, for thingy. blueangel. |
| 22:16.10 | lkcl | Kevin2: the runtime cpu detection |
| 22:16.27 | lkcl | and the very basic gpio watching for GPIOA to GPIOH |
| 22:17.51 | lkcl | it's not as sophisticated as wgpio but it does the job. |
| 22:21.29 | Kevin2 | lkcl: The only think I can see different is the gpio stuff - I don't see anything special going on with cpu detection. |
| 22:21.45 | lkcl | ok. cool. |
| 22:22.25 | Kevin2 | Can you give mainline a shot? If it runs okay, I can pull in the gpio stuff. |
| 22:23.01 | lkcl | okie. |
| 22:23.22 | lkcl | got a precompiled version? |
| 22:24.09 | BabelOued | lkcl: http://pastebin.ca/267520 from pH5 magician code |
| 22:24.29 | Kevin2 | Yes - at svn://www.shadowmite.com/home/svn/ppc6700/haret |
| 22:24.51 | lkcl | ta Kevin2. oh - it's svn:// oops |
| 22:28.25 | florian | re |
| 22:28.34 | lkcl | ok which version? haret.exe? gnuharet-noop.exe? burble? |
| 22:28.42 | Kevin2 | haret.exe |
| 22:36.13 | *** join/#htc-linux psokolovsky (n=psokolov@ip.85.202.124.214.dyn.sub-9.broadband.voliacable.com) |
| 22:39.42 | lkcl | ok. |
| 22:39.49 | lkcl | ta BabelOued |
| 22:40.19 | cr2 | hmm. sable does it too: |
| 22:40.22 | cr2 | 005.806157-005.806217 write: '\x08\x08\x01\x1a\xff\x02\x04\x00' |
| 22:40.27 | cr2 | 0x8 0x8 |
| 22:43.41 | BabelOued | lkcl: ok i ve found how to shutdown linux on BA :) |
| 22:43.54 | BabelOued | load snd-htcblueangel_audio :) |
| 22:49.00 | lkcl | hooray! :) |
| 22:49.06 | lkcl | ha ha |
| 22:52.26 | lkcl | damn wince wifi shit. |
| 22:54.22 | cr2 | Kevin2: what does '6' here mean ? 005.856842-005.856853 write: '<01>6<ff><04><00><10><0e><00>' |
| 22:54.52 | Kevin2 | Ascii 6 -- to disable all ascii, modify the python script itself (there is a variable in the first few lines). |
| 22:58.18 | cr2 | ok. thanks. i was doing a lookup with awk 'END{for (i=32;i<128;i++)printf("0x%x %c\n",i,i)}' < /dev/null |
| 23:18.31 | cr2 | Kevin2: universal and sable BT commands are consistent with each other. universal uses 1 more command. |
| 23:18.34 | cr2 | http://wiki.xda-developers.com/index.php?pagename=UniversalBluetooth |
| 23:18.46 | cr2 | you can can it on apache. |
| 23:19.00 | cr2 | s/can can/can check/ |
| 23:20.57 | *** join/#htc-linux cjb (n=cjb@wireless-19-49.media.mit.edu) |
| 23:22.56 | cjb | Hi! Are there any HTC phones with complete (cell, wifi, suspend) Linux support yet? Sorry to ask a newbie question. |
| 23:24.51 | cr2 | lkcl: you have a wrong job, you should have been a jornalist :) |
| 23:25.03 | lkcl | ha ha, who me? |
| 23:25.04 | Kevin2 | cr2: Did you want me to run a test on apache? If so, what commands do you want me to run? |
| 23:25.10 | lkcl | what makes you say that, cr2? |
| 23:25.32 | cr2 | cjb: hi. there is no device with "complete" support. |
| 23:26.01 | cr2 | lkcl: by the ability to manupulate the facts :) |
| 23:26.15 | lkcl | whatwhatwhat? what did i do! i only forget things, that's all! |
| 23:26.32 | lkcl | or not read them properly |
| 23:26.34 | cr2 | Kevin2: trace the BT serial port. start with it switched off, and switch it on. |
| 23:27.02 | cr2 | lkcl: read the e-mail |
| 23:27.05 | lkcl | okie |
| 23:28.12 | cr2 | Kevin2: yes, BTUART. |
| 23:29.54 | lkcl | ok i see what you mean |
| 23:29.56 | lkcl | oops! |
| 23:30.43 | cr2 | but i have enjoyed your fantasy :D |
| 23:31.06 | cr2 | you are too good for a journalist. |
| 23:35.15 | Kevin2 | cr2: There is a lot of data there - what should I look for? |
| 23:38.26 | cr2 | Kevin2: run scanserial.py and look for |
| 23:38.29 | cr2 | 005.806212-005.806217 write: '<01><1a><ff><02><04><00>' |
| 23:38.29 | cr2 | 005.818296-005.818303 write: '<01><1b><ff><02><04><00>' |
| 23:38.29 | cr2 | 005.830826-005.830866 write: '<01><1a><ff><02><05><00>' |
| 23:38.29 | cr2 | 005.843261-005.843269 write: '<01><1b><ff><02><05><00>' |
| 23:38.52 | cr2 | python scanserial.py < typescript | grep write: | less |
| 23:40.09 | lkcl | cjb: we're working on it, dude! it's been a case of 1 step forward, 2 steps back, for a long time. |
| 23:40.56 | Kevin2 | 002.380491-002.380497 write: '<01><1a><ff><02><04><00>' |
| 23:40.56 | Kevin2 | 002.623184-002.623195 write: '<01><1b><ff><02><04><00>' |
| 23:40.56 | Kevin2 | 002.767029-002.767040 write: '<01><1a><ff><02><05><00>' |
| 23:40.56 | Kevin2 | 002.903674-002.903685 write: '<01><1b><ff><02><05><00>' |
| 23:41.06 | cjb | lkcl: I can imagine. :) What are the main hurdles left? |
| 23:41.08 | cr2 | yes. that's the BT init. |
| 23:41.32 | cr2 | the write: '<01> is the BT command |
| 23:41.40 | Kevin2 | That's the second thing I have going to the phone. The first is: |
| 23:41.42 | Kevin2 | 002.380435 09309f3c: e583e000 str lr, [r3] # r14=00000008 r3=ae200000 |
| 23:41.42 | Kevin2 | 002.380476 09309f3c: e583e000 str lr, [r3] # r14=00000008 r3=ae200000 |
| 23:41.52 | Kevin2 | s/going to the phone/going to the bt/ |
| 23:41.53 | cr2 | the next 2 bytes are ogf&ocf |
| 23:42.04 | cr2 | Kevin2: yes. 0x8 0x8 |
| 23:42.25 | lkcl | cjb: well, it depends on the device. |
| 23:42.36 | Kevin2 | cr2: What is ogf and ocf? |
| 23:42.56 | lkcl | _all_ of the devices need a decent applications suite: opie and gpe are simply not designed for phones: they're PDA suites. |
| 23:43.02 | cr2 | you need to do something like that perl -e '{$val=0xff1b; printf "0x%x 0x%x\n",$val>>10,$val&0x3ff}' |
| 23:43.20 | cr2 | 0xff1b means <1b><ff> |
| 23:43.22 | lkcl | but openmoko.org is about to change all that, in approximately january |
| 23:43.43 | Kevin2 | What does the perl thing do? |
| 23:44.03 | lkcl | _other_ than that - there are hardware progress pages for each of the devices, but it goes (approximately) like this (and cr2 will correct me i am sure, ha ha, mr journalist) |
| 23:44.29 | cr2 | hmm. you need to split the 16bits in 6 and 10 |
| 23:44.49 | cr2 | i'll look for a link. i've posted it yesterday. |
| 23:45.06 | cr2 | http://www.geocities.com/blue_prasso/bluetooth/hci_commands.htm |
| 23:45.27 | cr2 | http://www.geocities.com/blue_prasso/bluetooth/hci_list.htm |
| 23:45.32 | lkcl | sable: suspend/resume, wifi, joystick all need 'polishing'. gsm needs a userspace 'wrapper' to multiplex data to the eriksson K500/K700 gsm radio-rom module; gps needs decoding (it's a new proprietary chipset from globallocate.com) |
| 23:46.32 | cjb | ooh, it looks like the Sable is getting a lot of attention at the moment. |
| 23:46.58 | cr2 | cjb: the most advanced port is universal. |
| 23:46.59 | lkcl | universal: neeearrly there: everything works except the audio routing is still a bit 'iffy' as it's quite literally one of the most complex and comprehensive devices that linux runs on. what was the list, cr2? 3 speakers, 3 mics, 2 cameras, GSM/UTMS.. |
| 23:47.07 | cjb | cr2: Thanks. |
| 23:47.24 | lkcl | so, for example, it's possible to hear audio but only in the headphone - at the moment. |
| 23:47.52 | cr2 | yes, the mic switching needs more research. |
| 23:48.46 | lkcl | blueangel: 2.6.12 kernel had _everything_ working but sound (argh!) and the lcd 'resume'. in 2.6.19, it's the most backwards of the ports except for himalaya. someone worked out the lcd/resume sequence just last month. and the sound infrastructure has been so completely reworked (asoc) that _nobody_ is using it ha ha |
| 23:49.21 | cjb | Gotcha. Many thanks for the summary, by the way! |
| 23:49.41 | cr2 | blueangel does not have a working wifi |
| 23:50.00 | cr2 | and the command to switch it to 921600 is also not yet found. |
| 23:50.09 | lkcl | himalaya: 2.6.12 kernel was the most 'complete' at the time except for sound, and last year cr2 and i identified that it has the same sound chip as the blueangel. so it is now in the same boat as the blueangel - except that the porting from xanadux cvs to handhelds cvs is _completely_ fscked up. |
| 23:50.26 | lkcl | oh yes forgot about that lot. |
| 23:51.02 | cjb | Huh. So, for someone wanting to help out, which of the sable/universal would you recommend as having the most mindshare around it at the moment? |
| 23:51.08 | cr2 | camera on blueangel and himalaya is on ATI 2D accel video chip +capture. |
| 23:51.24 | cr2 | it is not really RE'd. |
| 23:51.27 | lkcl | cr2: it might be because there are 5 GPIO pins required, and two of them will be 'permanently' on, whilst the other three are only switched on/off by wince when you powerup/down. just like the sable. |
| 23:51.44 | lkcl | i only found the GPIOs by switching them off, manually, and getting disconnected on wifi ha ha |
| 23:51.49 | cjb | :) |
| 23:52.05 | cr2 | cjb: it depends on what you really want to have. |
| 23:52.16 | cr2 | universal is a micro-notebook, not a phone. |
| 23:52.25 | lkcl | cjb: the hw6915 is the simplest of the two devices, but the universal is much more... how-to-say yeh, cr2 got it. |
| 23:52.25 | Kevin2 | cr2: Okay, I can see the bt command framing. |
| 23:52.26 | cr2 | but it has 640x480 LCD. |
| 23:52.40 | cjb | cr2: GSM, Wifi and suspend would be ideal. I'll take an uncomfortable form factor; beggars can't be choosers. :) |
| 23:53.00 | cr2 | while hw6915 has 240x240. but it has GPS. |
| 23:53.07 | cr2 | builtin. |
| 23:53.31 | cr2 | but we need to extract the PRN pseudoranges from its binary datastream. |
| 23:53.37 | lkcl | it's 240x240 but is' a quite _big_ 240x240. |
| 23:53.41 | cjb | Could you explain PRN? |
| 23:53.45 | cr2 | quite a challenge for a real hacker :) |
| 23:54.01 | cr2 | satellite number, to make things more simple. |
| 23:54.10 | cjb | Ah, got it. |
| 23:54.28 | lkcl | the protocol is proprietary, and globallocate.com are complete xxxxing wxxxxxrs |
| 23:54.28 | cjb | I'm not so interested in GPS, but hax0r1ng satellites is always fun. |
| 23:54.44 | cr2 | usually you (may) know where are the 28 sats and a given time. |
| 23:54.48 | lkcl | ok. their top sales/marketing guy is a dick. the technical team are brilliant. |
| 23:54.49 | *** join/#htc-linux lamikr|lap (n=chatzill@aragorn.kortex.jyu.fi) |
| 23:55.03 | cr2 | it's called ephemeris. |
| 23:55.12 | lkcl | and the product itself is amazing. |
| 23:55.30 | cr2 | and you can measure the time offsets from your clock to the sat clocks. |
| 23:55.57 | cr2 | multiply by a speed of light = you get "pseudoranges" - distances to the satellites. |
| 23:55.59 | lkcl | cjb: what's your level of programming experience / learning-speed / technical ability? |
| 23:56.54 | cjb | lkcl: My background's in software, as a systems programmer -- hardware is limited to what I've picked up in the last few months working for the One Laptop per Child gang. |
| 23:57.01 | lkcl | oh cool. |
| 23:57.07 | cjb | I think drinking from the fire hose would be a fair way to put it. :) |
| 23:57.18 | lkcl | _great_ project, btw. |
| 23:57.36 | cr2 | well, we have mostly done with the hardware on the universal. |
| 23:57.37 | cjb | Thanks! We're having fun, great to have real laptops around at last. |
| 23:57.59 | lkcl | i nearly got a great processor design funded which would have been perfect for the OLPC. |
| 23:58.33 | cjb | That would have made us happy, I think. :) We surveyed 320 different CPUs. |
| 23:58.43 | lkcl | but when it came down to making the funding decision, the company directors started fighting over the $20m cash that was to be earmarked for the project. |
| 23:58.52 | cjb | (and ended up with one with 16K of cache. Score!) |
| 23:59.05 | lkcl | so far they have each spent about $1m in legal fees over ooooooo 16k cache is a bitch |
| 23:59.10 | cjb | haha. |
| 23:59.28 | lkcl | you are _so_ not going to be able to run prelink in anything less than 16 hours, with only 16k cache |
| 23:59.35 | lkcl | try it. seriously. |
| 23:59.48 | cjb | Yeah, tell me about it. I'm the performance optimisations guy. |