| 00:00:14 | Russ | who's the maintainer? |
| 00:00:28 | GoRK | rmk himself. |
| 00:01:00 | GoRK | but rmk's notes from thursday say he just got the tux to boot up so "now he has an sa1100 platform to play with"... so did he write that without testing most of it? |
| 00:01:13 | Russ | dunno |
| 00:01:31 | Russ | he's on irc from now and then |
| 00:02:27 | GoRK | yeah i was asking some folks about double storage on arm and they asked rmk and he came over to chat |
| 00:02:46 | GoRK | back when i discovered the uclibc's libm arm problems |
| 00:12:04 | GoRK | Russ: do you have any ideas on this? i dont understand how the MCP works... supposedly it's just sort of a 'magic' interface that abstracts talking to the ucb1x00's sib |
| 00:12:38 | Russ | dunno |
| 00:12:54 | GoRK | me neither. it is highly boggling |
| 00:13:50 | Russ | I just haven't looked at it |
| 00:33:57 | jwalden | Hello anyone... wuz playing with buildroot today, and it died trying to get t |
| 00:34:54 | jwalden | ... trying to get touchscreen.patch - wget did not seem to fetch this - anyone seen it? |
| 00:36:28 | jwalden | Also trying to find the tuxphone source and still could not find it... |
| 00:37:03 | Russ | tuxscreen.net |
| 00:37:47 | GoRK | it's all in cvs, jwalden |
| 00:37:55 | GoRK | on sourceforge's tuxscreen project |
| 00:38:07 | GoRK | you should be using buildroot-tux from cvs also |
| 00:40:06 | jwalden | Gork: - I got the CVS version of buildroot-tux 4 hours ago from sourceforge on cvs, and selected microwindows as my graphic evironment in the makefile, so I'm not sure.. but I think I did what I was supposed to? |
| 00:40:15 | scanline | morning Sammy |
| 00:40:15 | Sammy | hi all |
| 00:41:10 | Sammy | hello scanline |
| 00:42:15 | jwalden | Gork - buildroot-tuxscreen did a wget for all the files in CVS etc... then I did a make world, and the build stopped with an error message on the touchscreen.patch part of the build |
| 00:45:56 | jwalden | Gork: ie. +patch -d /jwalden/buildroot-tux/linux -p Error: /bin/sh /jwalden/buildroot-tux/sources/linux-2.4.7/touchscreen.patch: No such file or directory |
| 00:53:30 | as_ | Nite all. |
| 01:14:05 | Hiraghm | well |
| 01:14:19 | | salut, Hiraghm |
| 01:14:19 | Hiraghm | Hi |
| 01:14:26 | Hiraghm | Hi ibot |
| 01:14:49 | Hiraghm | anybody here? |
| 01:17:17 | Hiraghm | waves near MicroChip32 |
| 01:17:31 | MicroChip32 | ducks |
| 01:18:52 | Hiraghm | won't get his DSL connection til next wednesday |
| 01:18:55 | Hiraghm | feels sad |
| 01:19:48 | thekernel | Don't feel bad... |
| 01:20:21 | thekernel | I had almost 2 weeks without it and I didn't have any notice. |
| 01:20:28 | Hiraghm | Cox cut my email and website before I could get the last of my email and data out. |
| 01:20:36 | jwalden | Scanline.. Thanks for you patience - I was thinking stupid - sorry about that. Should use my eyes before my hands |
| 01:20:36 | thekernel | Ouch |
| 01:20:42 | Hiraghm | the dirty poopy heads |
| 01:21:01 | scanline | jwalden: huh? |
| 01:21:15 | thekernel | jwalden: Hey, I did some stuff on that board with the dual ethernet |
| 01:21:17 | Hiraghm | pokes scanline real slow |
| 01:21:24 | scanline | ducks |
| 01:21:42 | Hiraghm | anybody know enough about routers to advise me? |
| 01:21:43 | MicroChip32 | grabs the back of Hiraghm's undies and pulls them over Hiraghm's head! |
| 01:21:51 | thekernel | ouch... |
| 01:22:02 | thekernel | Hiraghm: I know I like my linksys router... |
| 01:22:02 | Hiraghm | wasn't wearing undies |
| 01:22:09 | jwalden | thekernel: cool! Did Cosmo whack you or what? |
| 01:22:11 | scanline | debugs C code whilst listening to cheesy trance music |
| 01:22:12 | thekernel | backs away slowly |
| 01:22:15 | MicroChip32 | ack, i dont wanna know what i grabbed then |
| 01:22:16 | Hiraghm | how many ports? |
| 01:22:31 | thekernel | Cosmo? |
| 01:22:34 | Hiraghm | The back of my chair I think |
| 01:22:37 | thekernel | or Cosmic? |
| 01:22:50 | jwalden | thekernel: you know the CosmicPenguin |
| 01:22:54 | MicroChip32 | Hiraghm, no wasnt that, it wasnt hard |
| 01:23:11 | thekernel | jwalden: I'm putting together a small root filesystem using uClibc and BusyBox |
| 01:23:30 | Hiraghm | neither's my chair |
| 01:23:34 | thekernel | I'm also using syslinux as the boot loader |
| 01:23:38 | Hiraghm | has BZ been in? |
| 01:23:45 | Hiraghm | I got my braindead tuxscreen today |
| 01:23:46 | | BZFlag was last seen on #bzflag 3 hours, 6 minutes and 6 seconds ago, saying: sup? [Tue Nov 13 22:17:40 2001] |
| 01:23:46 | thekernel | seen BZFlag |
| 01:23:57 | MicroChip32 | brainded ? |
| 01:24:01 | jwalden | thekernel: nice! |
| 01:24:10 | Hiraghm | yeah, BZ blobbed it |
| 01:24:22 | Hiraghm | ibot's smart |
| 01:24:25 | MicroChip32 | ah |
| 01:24:25 | jwalden | thekernel: you were right about autoconf... |
| 01:24:34 | MicroChip32 | so now you gotta load the rest of the stuff eh Hiraghm |
| 01:24:39 | Hiraghm | Micro, have you blobbed yours yet? |
| 01:24:42 | Hiraghm | nods |
| 01:24:48 | thekernel | Though, I havn't finished enough to test it. The syslinux boot requires 1MB off the front of the 8MB DOC as an MSDOS partition. |
| 01:24:50 | MicroChip32 | Hiraghm: nope, too busy with other things |
| 01:24:57 | thekernel | It really does seem easier than lilo though. |
| 01:25:03 | Hiraghm | yeah. The first one is still running inferno. Wish I knew how to run it.. 8c) |
| 01:25:19 | thekernel | On what part was I right? |
| 01:25:29 | MicroChip32 | like writing a device driver, a program (maybe two), two electronics projects, and preparing for a vacation |
| 01:25:31 | jwalden | thekernel - oh forgot to mention we are going to 16MB for the DCO |
| 01:25:54 | thekernel | Great, so what your saying is I have space for a proper libc |
| 01:26:03 | jwalden | thekernel: it only took a few hours... |
| 01:26:38 | jwalden | thekernel: yes... you can thank the recession and the sinking cost of silicon for that |
| 01:28:44 | thekernel | Hey, on ftp.viewml.com is /pub/filesystem.tar.gz a filesystem you would expect to use on your devices? |
| 01:30:28 | jwalden | thekernel: I think that was for the touchscreen thingy and for demo purposes only - |
| 01:30:51 | jwalden | thekernel: that is to say the dual touchscreen thingy |
| 01:30:53 | thekernel | I'm sure I could use it as a base to install the busybox stuff into. |
| 01:31:09 | jwalden | thekernel: that'd be cool... |
| 01:31:12 | thekernel | It's always easier to take from a working tree than generate a new one. |
| 01:31:32 | thekernel | Where did you guys get those touchscreens? |
| 01:32:01 | Hiraghm | what's a vacation? |
| 01:32:18 | jwalden | thekernel: well phillips - the touchscreen is ELO - the LCD just got discontinued, and it was a phillips part |
| 01:32:28 | thekernel | vacation: that thing I'll be going on in a month when I take my '74 Bug on a road trip. |
| 01:32:45 | jwalden | national semi is picking up where Phillips left off, however at a greater cost... |
| 01:32:47 | Hiraghm | oh |
| 01:33:22 | jwalden | 2 |
| 01:33:23 | Hiraghm | anybody know if Tyan Thunderbird motherboards are flakey or not? |
| 01:33:30 | thekernel | For those in professional jobs vacations also include paid time off. |
| 01:33:55 | Hiraghm | waitaminit.. you get PAID.. even though you're not *working*?? |
| 01:33:57 | Tangent | Hiraghm : I've never heard anything but strong praise for Tyan motherboards |
| 01:33:59 | Hiraghm | that's so cool! |
| 01:34:07 | thekernel | Well, national semi always makes things costly. Our business partnership is a fine example |
| 01:34:14 | Hiraghm | Tangent, you must not have visited my local PCClub story 8c |
| 01:34:17 | Hiraghm | 8c( |
| 01:34:26 | Tangent | Hiraghm : I guess not |
| 01:34:44 | thekernel | Hiraghm: Well, I do get paid to work on stuff for jwalden... But I'm not at work now. |
| 01:35:29 | Hiraghm | I went in the other day to price the thunderbird (I want to set up a dual athlon system) and the salesdroid said they wouldn't carry them because the boards were flakey |
| 01:35:35 | thekernel | has 2 weeks of paid time off... |
| 01:35:37 | Hiraghm | then pointed me at a dual pentium board that would only go up to 1 ghz |
| 01:35:41 | jwalden | thekernel: you should not be allowed to have vacations if you live with your parents... |
| 01:35:42 | GPSFan | + |
| 01:36:05 | thekernel | Hey, they just havn't kicked me out yet. It's wonderful |
| 01:36:05 | Hiraghm | jwalden, oh, so *that's* why I never got a vacation. |
| 01:36:10 | jwalden | theke |
| 01:36:43 | thekernel | Then again I only graduated from highschool a year and a half ago |
| 01:36:48 | Hiraghm | Oh |
| 01:36:53 | jwalden | Higtaghm: well my parents kicked me out when I was 12, and sent me to work at the local factory.. |
| 01:37:02 | Hiraghm | ack! |
| 01:38:00 | jwalden | thekernel: Well - I never "offically" graduated from High School... ;-) they just got sick of me and gave me a diploma to get rid of me |
| 01:38:11 | thekernel | My parent's let me live at home, then again I provide email services on my Linux box. I plan on offloading that to either my RS/6000 or DEC Alpha (Whatever one will work) |
| 01:38:11 | Hiraghm | that was my tactic too 8c) |
| 01:38:24 | Hiraghm | oooh.. |
| 01:38:29 | thekernel | jwalden: I really wish I could have gotten away with that. |
| 01:39:10 | jwalden | thekernel: well - I guess you did not try hard enough - you probably has nice parents who like you too huh? |
| 01:39:26 | thekernel | If anyone has some large 72pin FP RAM simms they would like to contribute to the "thekernel vintage risc systems" it would be much appreciated. |
| 01:39:35 | Hiraghm | has anybody here ever gotten the amulet of yendor? |
| 01:39:52 | thekernel | I once got a mallet of rubber |
| 01:40:07 | jwalden | thekernel: My tuxscreen link is barfing at me as in the excorsist (a movie that was shot b4 you were born) I need help!! |
| 01:40:18 | Hiraghm | is trying out an isometric nethack |
| 01:40:20 | thekernel | jwalden: the nice parents thing is one I really have going for me. |
| 01:40:36 | jwalden | thekernel: wanna trade? |
| 01:40:39 | Hiraghm | they didn't really shoot the movie (though they should have) |
| 01:40:43 | thekernel | tux screens? |
| 01:40:51 | jwalden | no parents |
| 01:41:15 | thekernel | Oh |
| 01:41:17 | thekernel | No |
| 01:41:26 | jwalden | thanks! |
| 01:41:54 | jwalden | if you help me with my build - I'll give you a FREE review of "Lord of the Rings" |
| 01:42:33 | thekernel | Hey, my mom is a massage therapist. When I damage myself I go to her and have her fix what I did. Last weekend was a fine example... I got myself pounded quite nicely into a wall and she did her stuff and now my face is fully symmetrical again. |
| 01:43:28 | jwalden | Hey - my mom is a Math Prof, and she can't (won't) do diddly for me... |
| 01:46:36 | Niddix_away | jwalden: So what your saying is Algebra isn't helping you in any way. |
| 01:47:29 | jwalden | Niddix: Actually only infinite albelian group theory helps - the rest is just useless |
| 01:47:32 | Niddix_away | jwalden: I knew my math teachers were lying ;) |
| 01:49:02 | jwalden | -o vmlinux |
| 01:49:02 | jwalden | init/main.o: In function `start_kernel': |
| 01:49:02 | jwalden | init/main.o(.text.init+0x68c): undefined reference to `cpu_CPU_NAME_check_bugs' |
| 01:49:02 | jwalden | init/main.o(.text.init+0x68c): relocation truncated to fit: R_ARM_PC24 cpu_CPU_NAME_check_bugs |
| 01:49:02 | jwalden | arch/arm/kernel/kernel.o: In function `__dabt_svc': |
| 01:49:02 | jwalden | arch/arm/kernel/kernel.o(.text+0x150): undefined reference to `cpu_CPU_NAME_data_abort' |
| 01:49:04 | jwalden | arch/arm/kernel/kernel.o(.text+0x150): relocation truncated to fit: R_ARM_PC24 cpu_CPU_NAME_data_abort |
| 01:49:06 | jwalden | arch/arm/kernel/kernel.o: In function `__dabt_usr': |
| 01:49:21 | Russ | are you using the handhelds.org toolchain? |
| 01:49:29 | jwalden | yup |
| 01:50:01 | Russ | I think that one has problems |
| 01:50:16 | Russ | try the embdebian one, or erikm's toolchain |
| 01:50:55 | jwalden | thanks - the one I'm using is the "skiff" toolchain - what does skiff stand for? |
| 01:51:15 | Russ | its the name of a platform |
| 01:51:19 | Russ | compaq made them |
| 01:51:49 | thekernel | is going away |
| 01:51:56 | jwalden | oh! - so that is the one that you a referring to as the "handhelds.org" toolchain |
| 01:51:58 | Niddix_away | Once again I the entire world has delcared war on me. |
| 01:52:02 | thekernel | jwalden: are you ever on during the day? |
| 01:52:06 | jwalden | yes |
| 01:52:36 | jwalden | thekernel: tomorrow AM about 10:00 PST |
| 01:52:41 | thekernel | make sure your around sometimes so I can ask you questions |
| 01:53:03 | jwalden | thekernel: will I be able to ask you questions in return? |
| 01:53:09 | thekernel | I'm going to go away now and talk to a person that called me |
| 01:53:17 | thekernel | No, I'll do all the asking :) |
| 01:53:17 | jwalden | thekernel: I should give you my cell number |
| 01:53:44 | thekernel | jwalden: email it to shaunk@censoft.com |
| 01:54:00 | jwalden | thekernel: will do - have a festive evening! |
| 01:54:00 | thekernel | That way I won't loose it. |
| 01:54:10 | thekernel | c-ya tomorrow |
| 01:54:28 | Russ | who told you to use the handhelds.org toolchain? |
| 01:55:29 | jwalden | Russ: I should not reveal his identity - I do have much respect for him, and he is a member of this project - but if pressed I'll happily point him out |
| 01:56:05 | Russ | who was it? |
| 01:56:18 | jwalden | Ready... The CosmicPenguin |
| 01:56:34 | Russ | ah |
| 01:56:46 | jwalden | Russ: pls refine "ah" |
| 02:04:11 | | i think there is an external program bbkeys for this purpose, (never used it though) |
| 02:04:11 | jwalden | Russ: still there? |
| 02:05:10 | Russ | yah |
| 02:05:26 | | russ: wish i knew |
| 02:05:26 | Russ | ibot: what are you smoking? |
| 02:05:53 | jwalden | Russ: so you choose not to refine "ah"? |
| 02:06:02 | Russ | oh |
| 02:06:15 | Russ | I don't think the hh.org toolchaing works right |
| 02:09:10 | jwalden | Russ: would you mind pointing me in the direction of the toolchain that you are using? |
| 02:10:10 | Russ | http://www.lart.tudelft.nl/lartware/compile-tools/ |
| 02:10:18 | Russ | the 2.95.2 one |
| 02:10:29 | Russ | arm-linux-cross.tar.bz2 |
| 02:10:43 | Russ | er, the 2.95.3 one |
| 02:10:59 | Russ | cross-2.95.3.tar.bz2 |
| 02:12:15 | jwalden | Thanks Russ! |
| 02:40:59 | MicroChip32 | anyone with rh 6.2 and kernel 2.2.19 able to find usb-skeleton.c on their system ? |
| 02:45:53 | Sammy | it have on rh kernel 2.2.17 |
| 02:46:18 | Sammy | also 6.2 too |
| 02:47:25 | MicroChip32 | can you dcc it to me (if dcc will get thru my firewall) |
| 02:48:36 | MicroChip32 | for some reason i dont find it on my system |
| 02:48:36 | Sammy | send it now... |
| 02:49:20 | MicroChip32 | if youre trying to dcc it aint getting here, did you get the email addr i sent ? |
| 02:55:51 | jacques | waits impatiently for his `screens. |
| 02:56:44 | MicroChip32 | Sammy: did you email that file? |
| 02:57:17 | Sammy | yes , i can't DCC to you , so I send the mail already ... |
| 02:57:38 | cory | hm/say |
| 02:57:41 | cory | weird |
| 03:00:27 | MicroChip32 | ok thanks Sammy |
| 03:00:47 | MicroChip32 | watches his email box, wondering if someone is holding it up filling it with anthraz |
| 03:00:52 | MicroChip32 | s/z/x/ |
| 03:01:54 | MicroChip32 | kicks his isp, hey send me my email already |
| 03:04:19 | MicroChip32 | Sammy: got it, thanks |
| 03:04:37 | MicroChip32 | gives his isp another kick, just cuz it deserves it |
| 03:04:50 | Sammy | : ) |
| 05:53:11 | GoRK | what time is it in the uk now? is it 7am or 8am? |
| 06:00:29 | GoRK | andersee: hey |
| 06:01:12 | | hello, GoRK |
| 06:01:12 | GoRK | BZFlag: hello |
| 06:01:19 | andersee | howdy |
| 06:01:20 | | Sorry, GoRK, I'll keep my mouth shut. |
| 06:01:20 | GoRK | ibot shut up |
| 06:01:49 | andersee | GoRK: progress with the shanip2? |
| 06:01:59 | GoRK | ok question: i got recording working in the new kernel ucb1x00 drivers, but there is sort of a political problem |
| 06:02:07 | andersee | oh? |
| 06:02:10 | andersee | whazzup? |
| 06:02:14 | BZFlag | biab |
| 06:02:46 | jacquesZzZz | political? |
| 06:02:48 | GoRK | well... the new ucb1x00 drivers (and they *do* work much better than the old ones) have stripped their nice robust interface down to absolutely nothing |
| 06:03:13 | GoRK | all in the name of the kernel being 'simply an arbiter of resources' and all that crap |
| 06:03:19 | andersee | Ok. I'm with you so far... |
| 06:03:22 | GoRK | anyway |
| 06:04:20 | GoRK | given: since the ucb1200 sib clock has to be synced with the sa1100 mcp, it is technically not possible to record and playback at different sample rates. |
| 06:04:52 | andersee | Ok |
| 06:05:08 | andersee | So you are thinking of re-fusing them? |
| 06:05:08 | GoRK | taken to the bare minimum capabilities of the hardware, it is then technically not possible to record or playback independantly |
| 06:05:27 | GoRK | so... recording works only during playback |
| 06:05:49 | andersee | Thats pretty goofy |
| 06:05:57 | GoRK | the old driver had some workarounds to make them behave a little bit indepenantly that rmk ripped out |
| 06:06:04 | scanline | why not have the OSS interface simulate a half-duplex sound card, and let the record code do its own synchronization? |
| 06:06:29 | GoRK | because "sample rate conversions dont belong in the kernel blahblahblah" |
| 06:06:47 | andersee | I'm with scanline. I think his idea is good. |
| 06:06:56 | scanline | why would you need sample rate conversion? |
| 06:07:04 | scanline | Well, i mean for the recording... |
| 06:07:16 | GoRK | i have seen several mails where rmk or ac rips someone a new one for making a kernel driver that makes hardware behave as expected and not expose every little flaw out to userspace |
| 06:07:21 | scanline | i know that the ucb1200 has the strange sample rates that it would be nice to have resampling for |
| 06:08:03 | scanline | heh... I haven't ventured into kernelspace yet. so there's a lot of political issues there i hear :) |
| 06:08:12 | andersee | GoRK: So the idea is to teach every user space app about broken drivers? |
| 06:08:17 | GoRK | that is why the new udb1200 touchscreen driver does not have the calibration code anymore |
| 06:08:27 | scanline | yipes |
| 06:08:39 | GoRK | andersee: that is exactly what ac says.. "Fix the tools." |
| 06:09:14 | scanline | I've been hoping they will make a standard touchscreen interface, so the same calibrators and apps work with different hardware. Same idea as the framebuffer device, but for input |
| 06:09:14 | GoRK | andersee: afaik, i dont know a single app that supports the 46.8KHz sample rate necessary to make the ucb1200 output quality audio! |
| 06:09:51 | andersee | GoRK: So the idea is to code for the least common demominator... And then adjust all the drivers to match (i.e. provide little functionality) |
| 06:10:03 | scanline | heh. So rmk wants to put resampling code in every app rather than make the drivers behave normal... blech |
| 06:10:24 | andersee | scanline: I suppose a libresample could be written... |
| 06:10:25 | GoRK | andersee: that's what i would think, but all that stuff is gone from the new driver.. it was all already written too |
| 06:10:39 | scanline | andersee: Yeah, but how many apps would use it? |
| 06:11:12 | GoRK | andersee: and then every audio app would have to link to it? ... i mean we could use artsd or esd or something but that's going to be sh*t for voip due to the extra delays associated with moving the audio data over sockets |
| 06:11:47 | scanline | it would be easier if the OSS used a shared lib for audio, rather than have each app open /dev/dsp directly |
| 06:12:07 | GoRK | if all that were the case, then why is there stuff like "software raid", "lvm" or even routing stuff in the kernel?!?!?! |
| 06:12:35 | scanline | hehe |
| 06:12:38 | GoRK | scanline: most modern cards allow multiple writers (if not readers) to /dev/dsp though |
| 06:12:47 | andersee | looks forware to ALSA taking over the world in 2.5 |
| 06:13:17 | scanline | GoRK: Right... Oh yeah, I decided to upgrade my computer :) I'm probably going to get a good soundcard for it too |
| 06:13:32 | GoRK | heck even the old ucb1200 drivers supported multiple readers/writers |
| 06:13:49 | GoRK | scanline: sblive 5.1... audigy work is coming along at a rocket pace as well |
| 06:13:51 | scanline | yeah, i tried playing 2 mp3s just for fun on the Tux :) |
| 06:14:17 | GoRK | alsa is too fractured though.. they have at least 3 different incompatible api's out there :( |
| 06:14:32 | scanline | GoRK: I've already ordered Athlon barebones system, memory, hard disk, and KVM switch. Just need to get a monitor, and maybe sound and video |
| 06:14:52 | andersee | I suspect we will see just the winning API showing up in 2.5 |
| 06:15:19 | scanline | andersee: They're including a sound API in the kernel itself, or will it still be a libalsa? |
| 06:15:53 | scanline | wait, that was a dumb question... |
| 06:15:57 | andersee | There will surely be an API for both layers.... |
| 06:15:59 | andersee | :) |
| 06:16:07 | GoRK | i believe the winning api will be OSS without tying itself to the OSS codebase. the emu10k1 project has proven that that approach works and the OSS api is not missing the new/good features |
| 06:16:40 | GoRK | the only OSS-free code that emu10k1 ties itself to is code to support /dev/sequencer for midi (it is impossible to support it without tying to oss-free) |
| 06:17:21 | andersee | GoRK: unless you use timidity and such... |
| 06:17:31 | GoRK | no... /dev/midi and /dev/dsp are all free of oss ties, yet they use the same api and support features such as dsp patches, ac3 passthrough, multichannel output, etc etc |
| 06:18:15 | GoRK | there are only a few midi apps left that use /dev/sequencer anyway... most use /dev/midi |
| 06:18:17 | andersee | So just /dev/sequencer |
| 06:18:41 | andersee | I guess the world can stand to lose it then |
| 06:19:07 | GoRK | it wouldnt be impossible to support it without oss-free, but all the drivers would have to implement it independantly then |
| 06:20:41 | GoRK | anyway the issue is that to move to the new kernel we have to jump through hoops for audio support and all of our touchscreen drivers have to be rewritten to perform some sort of userspace calibration |
| 06:21:27 | BZFlag | back. |
| 06:21:34 | andersee | GoRK: ick. |
| 06:21:51 | scanline | Does the iPaq use the ucb1200? |
| 06:21:56 | andersee | GoRK: Any other projects already done the touchscreen? |
| 06:22:07 | andersee | BZFlag: howdy |
| 06:22:34 | andersee | looks for ice cream |
| 06:22:54 | GoRK | andersee: well the calibration code can be taken directly from the old kernel driver. it is simple and works. it just has to be ported to userspace and put into touchscreen drivers. fbtuxcal will even still continue to work if all the input drivers use the same coordinate translation code |
| 06:23:36 | GoRK | honestly, it is really more the audio that im worried about |
| 06:24:17 | GoRK | the other alternative may be that the audio is not actually working correctly yet on the tux and these bizzare problems are simply the result of more bugs that i have not discovered :) |
| 06:25:30 | GoRK | i havent been able to contact any other developers working with post-2.4.9 ucb1x00 drivers.... |
| 06:26:33 | GoRK | finally, the third option is that we maintain our own kernel patch to put all the "forbidden" code back into the newer kernels |
| 06:28:05 | andersee | Hmm. |
| 06:28:19 | andersee | I'm going to defer that one to tim. |
| 06:28:45 | andersee | I suspect that long term we are better off tracking what ac, amk and friends do... |
| 06:28:50 | GoRK | but btw, i still get hard locks on the touchscreen device |
| 06:29:12 | andersee | sigh |
| 06:29:17 | BZFlag | reads... |
| 06:29:18 | andersee | thats not good |
| 06:29:36 | GoRK | anyone know if printk() works after lock_kernel() has been called? |
| 06:29:51 | GoRK | ie inside a realtime thread? |
| 06:30:32 | GoRK | the ts device is implemented with much trickery :) |
| 06:31:06 | BZFlag | we need to try to get the new code working. If audio works ar one frequency then we change the user code. |
| 06:31:26 | GoRK | BZFlag: well audio works. i have working audio |
| 06:31:28 | BZFlag | if it does not work at all, we do whatever it takes to fix the driver. ;-) |
| 06:31:40 | GoRK | im onto ts device now :) |
| 06:31:44 | BZFlag | GoRK: it's just a sampling rate issue? |
| 06:32:13 | GoRK | BZFlag: no it is more an issue of the hardware is capable of some of the basic audio things like recording without playing back only through software assistance it seems |
| 06:32:43 | GoRK | BZFlag: play/record rates are locked together by the rate the sa1100's mcp audio codec is going |
| 06:33:32 | GoRK | the manifestation in userspace is that in order to record, software has to know that and open the interface accordingly, it seems |
| 06:34:25 | BZFlag | so it can do full duplex as long as the rates are the same? |
| 06:35:03 | GoRK | yes but it can't seem to record without playing back either (even if the playback is of no sound |
| 06:35:27 | BZFlag | well that much should be handled in the driver. |
| 06:35:42 | BZFlag | asking an app to playback nothing just to get record is silly. |
| 06:35:45 | GoRK | rmk seems to have taken it out |
| 06:36:02 | GoRK | BZFlag: but technically that's what the hardware requires :) |
| 06:36:22 | GoRK | people who think that way should just go work on HURD |
| 06:36:28 | BZFlag | true, but there are broken pci bridges that the kernel fixes too. |
| 06:36:59 | BZFlag | I suppose we should just require hacked pci apps for all broken bridges? I think not. |
| 06:37:51 | GoRK | yeah i understand all that; the old driver (albeit a bit broken and lacking support for the telecom codec) was marvelously featureful for audio - multiple writers/readers, any sample rates you wanted, etc |
| 06:38:56 | GoRK | i kind of find it ironic that something such as a software modem that requires such explicit timings and could never be located in linux userspace seems to belong there according to the powers that be |
| 06:40:18 | BZFlag | heh |
| 06:41:43 | GoRK | anyway i can forge on with the ts device and perhaps just start a discussion on arm-kernel or something :) |
| 06:43:40 | BZFlag | I'll ponder it some more. |
| 06:44:17 | BZFlag | if we could show that it would be the only way to do a soft modem, that might make it easier to push through. ;-) |
| 06:44:58 | GoRK | well if you have a 2ghz machine you can do a 56k modem in userspace :) |
| 06:46:01 | GoRK | porting any existing softmodem code into kerel space also not only poses a coding challenge, but the creator of almost all the working code has passed away leaving it without a license, which poses a problem :) |
| 06:48:26 | andersee | A least you can be sure he won't sue.. ;-) |
| 06:48:43 | GoRK | sgi could. it was code originally writen for indy |
| 06:52:46 | GoRK | woo i just think i found another bug that should fix telecom right up to the functionality of the regular audio |
| 06:52:50 | GoRK | answering machine here i come |
| 06:59:02 | scanline | nifty |
| 07:24:26 | GoRK | ok ts device is almost working |
| 07:25:34 | andersee | cool |
| 07:26:07 | andersee | BTW, today I made the first thread work with uClibc. :) |
| 07:26:33 | scanline | claps |
| 07:26:40 | GoRK | sweet! picogui clients can now be multithreaded... that will come in mighty handy when we want to make an ip phone app :) |
| 07:26:53 | GoRK | linked against uClibc of course |
| 07:27:09 | andersee | I wrote up the most important bits of a pthread lib. Now in CVS. |
| 07:27:09 | scanline | IMHO the multithreaded libpgui is a bit of a hack... |
| 07:27:22 | andersee | Lots of things missing still |
| 07:27:48 | GoRK | haha got it |
| 07:27:54 | andersee | but it was enough for me to make a multithreaded disk copier app that works. |
| 07:27:56 | GoRK | Ts event: p=86 x=629 y=204 |
| 07:27:59 | GoRK | p is pressure |
| 07:28:04 | GoRK | this driver is pretty darn raw |
| 07:29:06 | GoRK | seems safe to use ts with audio and no skips! |
| 07:29:09 | GoRK | wh00t |
| 07:29:12 | scanline | nice |
| 07:29:24 | scanline | Hmm... i'd forgotten that the ucb1200 supported pressure |
| 07:29:47 | andersee | Hmm. Maybe we can make it into a midi instrument... ;-) |
| 07:29:49 | GoRK | yeah the old driver doesnt return it... it just uses it to detect touches |
| 07:30:28 | GoRK | yes or control brush size with pressure |
| 07:32:54 | andersee | Oh yeah. Seems readdir64 segfaults. Guess I should fix that next (or my app won't work) |
| 07:32:57 | scanline | Shouldn't be too hard to add some GUI infrastructure for handling the pressure :) |
| 07:37:28 | GoRK | hmmm doing extra ts tests now |
| 07:39:28 | GoRK | works fine yup |
| 07:41:03 | GoRK | testing brightness/contrast |
| 07:42:22 | GoRK | hard freeze! yay |
| 08:20:18 | BZFlag | hmmm... how do I find out what nameservers the root nameservers list for a domain? (other than whois) |
| 08:20:31 | GoRK | whoot ok brightness/contrast going! |
| 08:21:26 | GoRK | BZFlag: dig @a.gtld-servers.net <domain> ns |
| 08:22:16 | BZFlag | ahh. very good. thanx. got the answer, just not the right one. |
| 08:23:25 | GoRK | cool |
| 08:23:35 | GoRK | yet uncool |
| 08:24:42 | ctr | GoRK: got any kernel and root images with the new stuff working? |
| 08:25:17 | GoRK | ctr: heh not nearly :) i am still churning and want to run everything by russ first as well |
| 08:25:38 | ctr | k |
| 08:26:11 | GoRK | as far as a root image goes, it'll probably be pretty limited. gui's wont be able to work without input driver rewrites, not to mention calibration |
| 08:26:23 | BZFlag | is that like "Yes, I have seen them. And yet, no, I have not." ? |
| 08:26:44 | GoRK | BZFlag: if you like. heh |
| 08:27:11 | GoRK | BZFlag: it's kind of like... i *might* have this new kernel working or i *might* be lying :) |
| 08:27:46 | GoRK | im going to go eat some wheaties to celebrate though either way |
| 08:28:22 | GoRK | oh, hey, does anyone here right now know much about mtd? |
| 08:28:26 | GoRK | looks towards andersee |
| 08:28:33 | ctr | nope |
| 08:28:40 | ctr | i suck at all this stuff you are doing |
| 08:29:02 | GoRK | spacecoaster got the corruption problem fixed with the serial flash but it doesnt currently have enough support to run jffs2.. he was hoping someone could help him w/ it |
| 08:29:27 | GoRK | all he needs to know is what to make the functions do with what bytes exactly.. |
| 08:36:47 | andersee | gork? |
| 08:36:52 | GoRK | yo |
| 08:37:59 | andersee | I'm not the major guru on jffs2 internals. I just read the existing code when I rewrite mkfs.jffs2. I didn't actually try to understand the underlying structures... Just made sure I wrote the same sort of stuff... |
| 08:38:11 | GoRK | ahh ok |
| 08:38:15 | andersee | :-) |
| 08:39:41 | GoRK | well jffs2 wont work on the serial flash for some unknown reason but it's just a matter of figuring out how an mtd driver needs to present read/write/erase chunks to the system... it's not obvious since the serial flash can read/write single bytes without needing block erases |
| 08:40:17 | BZFlag | jffs2 expects to be able to write to bytes, not blocks. serial flash is only accessed at the block level. |
| 08:40:37 | andersee | Set the block size = 1 byte? |
| 08:40:38 | BZFlag | getting jffs2 on the serial flash will require and block read/mod/write layer. |
| 08:40:57 | BZFlag | serial flash is fixed size at 256 (well, 264) bytes. |
| 08:41:10 | GoRK | i understood that the serial flash could write without requiring erase or something though |
| 08:41:45 | GoRK | hmm so perhaps jffs2 is a terribly awful choice for it anyway :) |
| 08:41:57 | BZFlag | GoRK: perhaps. not sure. but you cannot just write to a byte. you will have to read a block, make the mods, and rewrite the whole thing for any partial block writes. |
| 08:42:26 | BZFlag | yes. I'd say if you want jffs2 on it, use a blk2mtd driver and then the block device. |
| 08:42:30 | GoRK | ouch well that burns |
| 08:43:03 | GoRK | hmm ok |
| 08:43:22 | BZFlag | course I think we could write the kernel up there. ;-) |
| 08:43:45 | BZFlag | now just teach blob to read/write from there. ;-) |
| 08:43:51 | GoRK | oh yeah one other question... the new kernel here has clock changing support. is there something that is preventing that from working on this platform? |
| 08:44:04 | BZFlag | just testing and debugging. |
| 08:44:31 | GoRK | so for all intents and purposes it might already work fine? |
| 08:44:52 | BZFlag | might, but I suspect there might be memory timing bugs |
| 08:45:16 | GoRK | is there a special userspace tool to change the frequency? |
| 08:45:25 | GoRK | or do you talk to something in /proc? |
| 08:46:54 | BZFlag | andersee: you around? can you add a ns1.rikers.org (166.70.15.250) to codepoet.org ? |
| 08:47:05 | BZFlag | dotster does not seem to have a brain after all. |
| 08:48:17 | andersee | k |
| 08:49:02 | BZFlag | dotster will happily add any already existant nameserver to a domain, but appears to not have a clue on how to create a new entry. |
| 08:49:28 | BZFlag | and I don't find this out till I've got all my domains moved over. =( |
| 08:49:38 | BZFlag | time to hunt for a new registry. |
| 08:49:52 | andersee | Just a A record or what? |
| 08:49:56 | GoRK | BZFlag: i like bulkregister.com |
| 08:50:36 | GoRK | BZFlag: although the name might imply differently, the only problems with them that i have had is the inability to change the same field on a large number of domains en masse |
| 08:51:28 | BZFlag | GoRK: dotster has a usable web interface for managing NS and contact info across many domains. |
| 08:51:46 | BZFlag | they just can't create/edit name server whois records. |
| 08:52:05 | BZFlag | andersee: just add it to a domain. any domain. so it gets a whois entry. |
| 08:52:25 | GoRK | bulkregister has a pretty decent web interface... i was talking about something that said "change all these domains ns records to the following:" etc |
| 08:52:48 | BZFlag | hmm... I suppose now is my chance to build a microsoft.a.looser.company.without.a.brain.not.used.at.rikers.org ;-) |
| 08:52:51 | GoRK | but it adds dns servers automagically |
| 08:53:19 | BZFlag | GoRK: yeah, the dotster will let you mod 1 or all or just a list all at once. |
| 08:53:41 | BZFlag | so I can select all and change contact info for example, or nameservers. |
| 08:54:15 | BZFlag | but the nameserver change form does not have a spot for IP address. it _assumes_ there is already a nameserver record in existance. |
| 08:54:36 | andersee | K. I added "ns1.rikers.org. IN A 166.70.15.250" to db.codepoet.org |
| 08:54:52 | BZFlag | andersee: hmmm... I'm not being clear.... |
| 08:55:03 | andersee | figured. :) |
| 08:55:05 | GoRK | i dont think that is sufficient to get a ns entry. you need to set it as a secondary NS for some domain |
| 08:55:11 | andersee | I don't either |
| 08:55:17 | BZFlag | andersee: I need you to go to your registrar and add a nameserver to the whois info. |
| 08:55:36 | GoRK | BZFlag: i have some unused domains i can set that server as NS on, i think |
| 08:55:49 | GoRK | BZFlag: how about misterspongybrain.com ? |
| 08:55:51 | BZFlag | GoRK: exactly. and dotster will only add ones that already exist on some other domain. |
| 08:56:33 | BZFlag | GoRK: fine by me, I just already am unpublished primary for andersee's domains so they won't be broken. ;-) |
| 08:57:10 | GoRK | ah ok either way. :) |
| 08:57:14 | BZFlag | andersee: clear now? |
| 08:57:40 | GoRK | BZFlag: you could have NS for godiamsofuxinleet.net ... that one is not used for anything |
| 08:57:43 | BZFlag | GoRK: is that an unmapped domain? if so, it would be nice if you added it. ;-) |
| 08:57:49 | GoRK | BZFlag: or beefamato.com |
| 08:57:59 | GoRK | BZFlag: i have several unmapped domains |
| 08:58:21 | BZFlag | GoRK: any will do. Just needs to be a domain at a registrar with a brain. |
| 08:58:37 | BZFlag | will be searching for a new registrar, that's for sure. |
| 08:59:02 | GoRK | ooh beverageranch.com is another that i have |
| 08:59:06 | GoRK | and letstalkbusiness.com |
| 08:59:10 | andersee | Unable to add nameserver: Command failed: unable to verify existence of nameserver ns1.rikers.org |
| 08:59:26 | andersee | geez |
| 08:59:54 | BZFlag | andersee: hmm.... Perhaps they need DNS propigated first? I don't know. did it ask for the IP address as well? |
| 09:00:03 | andersee | nope |
| 09:00:08 | andersee | Just the name. |
| 09:00:12 | GoRK | BZFlag: what is the ip for ns1.rikers.org? |
| 09:00:16 | BZFlag | ah. that's broken too. |
| 09:00:17 | GoRK | BZFlag: i can do it |
| 09:00:19 | andersee | I suspect it hit rikers.org for it |
| 09:00:38 | andersee | Yup. Pretty stupid. |
| 09:00:43 | BZFlag | GoRK: 166.70.15.250 |
| 09:01:12 | GoRK | ok you got a nameserver now and it supposedly hosts letstalkbusiness.com |
| 09:01:16 | BZFlag | nameservers for root domains are looked up via nsi registry, NOT via dns lookups. that's why it is... |
| 09:01:31 | BZFlag | valid to have a ns1.example.com for example.com |
| 09:01:53 | BZFlag | so the nameserver registration should include the IP address. dotster does not seem to know this. |
| 09:02:57 | BZFlag | GoRK: cool! when root servers get updated tomorrow I'll add it to my domains. thanx! |
| 09:03:09 | GoRK | BZFlag: no problem |
| 09:03:21 | GoRK | ssh gork@tuxtiles.com |
| 09:03:21 | BZFlag | you are: larry@LTBN.COM ? |
| 09:03:24 | GoRK | whoops sorry |
| 09:03:33 | GoRK | eh? that's not me |
| 09:03:57 | BZFlag | that's the contact on letstalkbusiness.com |
| 09:04:34 | GoRK | lol i guess i have the domain name letstalkbusinesses.com then |
| 09:05:04 | BZFlag | johnl@blurbco.com ? |
| 09:05:10 | GoRK | yeah that's more like it |
| 09:05:23 | BZFlag | so that's the one you added it to? |
| 09:05:32 | GoRK | yeah |
| 09:05:53 | BZFlag | what does BULKREGISTER.COM charge per domain (for < 10) and for transfers? |
| 09:07:18 | BZFlag | GoRK ? |
| 09:07:36 | GoRK | per domain is $12/yr flat. after you register 50 domains and continue to register more than 50 per month the price goes down i think to 10 or 8 |
| 09:08:08 | BZFlag | ah. that's decent. dotster is 14.95/yr |
| 09:08:09 | GoRK | transfers are $12 also and include one year. they have an automated process to transfer from NSI only i think |
| 09:08:26 | GoRK | include one year added onto the end of your current expiration that is |
| 09:08:55 | BZFlag | define "automated" ? ie: I assume they can do transfers from other domains, just perhaps not look them all up for you. |
| 09:10:07 | GoRK | automated means if i have the authority to do the transfer in good faith, i type the domain name in and the admin contact listed in the nsi registry gets an email. They click on the confirm link in the email and the domain transfers. all the info stays the same and new handles are created as needed. i can then go in and change it after that |
| 09:10:28 | GoRK | whenever the transfer goes, my cc gets a $12 charge |
| 09:11:14 | GoRK | i have gotten plenty of people to transfer simply because they find it inconvenient to make changes to their domain records through nsi's absurd systems |
| 09:11:49 | GoRK | i have never transferred from anything except NSI... i would email support@bulkregister.net if i had to |
| 09:12:23 | GoRK | the website says "we will automate the process to accept name transfers from other registrars in the future, so please stay tuned." |
| 09:12:32 | GoRK | but that was like 2 years ago |
| 09:14:08 | BZFlag | I guess I should have taken them up on the "free domain registration" offer I got a while back. =( |
| 09:20:16 | GoRK | wh00t! welcome back brightly lit screen of yore! |
| 09:21:03 | GoRK | wonders why this can't happen on boot again... russ said something about how it couldnt |
| 09:26:23 | GoRK | gets a diff for russ ready |
| 09:40:39 | BZFlag | sweet. nite. |
| 10:03:58 | scanline | is away: ZzZzZzZz |
| 10:04:05 | GoRK | later |
| 10:09:21 | andersee | night |
| 10:18:29 | GoRK | oh i didnt leave |
| 10:18:34 | GoRK | i was telling scanline later |
| 10:18:40 | GoRK | im posting this kernel patch |
| 10:18:44 | andersee | cool |
| 10:18:46 | GoRK | this kernel is ready |
| 10:18:55 | andersee | Excellent! |
| 10:19:07 | GoRK | cept for mount -a and new touchscreen drivers, we can update everything to use it |
| 10:19:15 | andersee | I just finished fixing LFS support so readdir64 doesn't segfault anymore |
| 10:19:29 | GoRK | err new touchscreen drivers for the gui systems that is |
| 10:20:18 | GoRK | i am shooting to release a new picogui image this weekend with all the new goodies :) |
| 10:20:48 | GoRK | new kernel + lpp + tuxphone + pgtuxphone |
| 10:21:12 | GoRK | mount -a is pretty trivial to work around for the time being |
| 10:21:21 | GoRK | least for a test image |
| 10:21:31 | andersee | cool. |
| 10:21:46 | andersee | Definately something to dig into |
| 10:22:22 | GoRK | and my stuff for shanip2 prototyping should be in by the end of the week, so maybe boards can get ordered early next week too |
| 10:22:34 | andersee | I've never aactually tried picogui, so I'm looking forware to giving it a go... |
| 10:22:40 | andersee | GoRK: that would be sweet |
| 10:23:24 | andersee | this is cool. I can build and use my app for the gov't cpntract again with uClibc. |
| 10:23:48 | andersee | I had to use glibc for the last week... Finally got some time to fix things. |
| 10:24:07 | GoRK | heh |
| 10:24:13 | GoRK | glibc.... the horror! |
| 10:24:32 | andersee | I figure if I have to use my own stuff, it keeps me honest. :) |
| 10:24:39 | GoRK | ok time to make sure this patch applies, compiles up in a monolithic kernel and goes |
| 10:25:07 | GoRK | ooooh yeah like the govt. is all about that honesty stuff |
| 10:25:15 | andersee | heh |
| 10:25:16 | GoRK | "hey can we use honesty as a weapon?!" |
| 10:25:29 | andersee | Well, I'm soing work for the IRS, so they like honest. :) |
| 10:25:47 | GoRK | make sure to code in the anti gork audit feature :) |
| 10:26:02 | andersee | I'll see what I can do. |
| 10:26:56 | GoRK | but seriously... one of my offices is in a building with the irs. they bring in metal detectors on tax day... i have keys to the super secret enterences though so i can bypass them... i dont want to have to try to get a laptop through all that every day :) |
| 10:27:17 | andersee | heh |
| 10:27:43 | andersee | Can you imagine, I'm working for their criminal investigation division... |
| 10:28:31 | andersee | But this group is mostly worried about money laundering and such |
| 10:28:32 | GoRK | haha well if i harass the irs cop that likes to grill me before he lets me on the elevator any more they're going to open a file on me! |
| 10:28:53 | andersee | heh |
| 10:29:20 | andersee | You must learn to only hassle those that can't hassle you back.... |
| 10:29:24 | GoRK | i tried to get him to give me a high five last year... tried for like 10 minutes. he got really gruff |
| 10:29:42 | andersee | heh. |
| 10:29:57 | GoRK | ooh yeah like palace guards in the uk |
| 10:30:04 | andersee | Sounds about par |
| 10:32:42 | ctr | :) |
| 10:32:57 | GoRK | compiles this pos |
| 10:48:58 | GoRK | works :) cept serial_flash wont compile except as a module |
| 10:49:04 | GoRK | spacecoaster can take a gander at that later |
| 10:51:41 | GoRK | kernel patch and binary are up at www.blurbco.com/~gork/tuxscreen :) and im going to bed!!! |
| 10:53:00 | ctr | what about a root image! |
| 10:53:01 | ctr | :) |
| 10:53:22 | GoRK | ctr: i could tar up and mail you my debian nfs directory |
| 10:53:43 | GoRK | ctr: just comment out the /dev/mtdblock2 line from /etc/fstab and it'll boot with the image you are using |
| 10:54:05 | ctr | nah |
| 10:54:10 | ctr | dont worry about ti for the moment |
| 10:54:46 | GoRK | if you have done the audio mod, /dev/dsp and /dev/dsp1 are back to their standard major/minors, and ucb1200-ts is gone in favor of /dev/ucb1x00-ts at major 10, minor 14 |
| 10:55:02 | GoRK | and it wont work with any drivers anymore1 |
| 10:57:44 | GoRK | oh and the screen sets brightness/contrast on open of that -ts device so the easiest way to get the screen to where you can read it is to cat /dev/ucb1x00-ts and press ctrl-c |
| 10:57:49 | GoRK | goodnight |
| 12:36:37 | cory | who's alive? |
| 12:57:01 | MacAtack | nobody its too early |
| 13:01:52 | cory | hehe |
| 13:28:36 | jacquesZzZz | yeah |
| 13:34:34 | Echelon | hi |
| 13:52:41 | CosmicPenguin | morning all |
| 13:53:57 | SpaceCoaster | hi |
| 13:54:47 | CosmicPenguin | SpaceCoaster: It seems like a while since I've talked to you... how are things? |
| 13:55:48 | SpaceCoaster | I was thinking the same thing, things are well, I'm making gradual progress on the various bits and pieces which is good, what about you? |
| 13:58:23 | SpaceCoaster | CosmicPenguin: Are you fully bluetoothed these days? |
| 14:03:13 | prpplague | morning all |
| 14:36:43 | CosmicPenguin | SpaceCoaster: Sorry, got pulled in to a meeting I didn't want to be in.... |
| 14:36:52 | CosmicPenguin | Yep, I'm going crazy with the bluetooth.... |
| 14:38:47 | SpaceCoaster | CosmicPenguin: is that good or bad and if it's bad is that good bad or bad bad? |
| 14:42:45 | thekernel | Is that kind of like the difference between mostly dead and all dead? |
| 14:42:55 | CosmicPenguin | SpaceCoaster: Well, actually it works, so thats good.... |
| 14:42:59 | CosmicPenguin | SpaceCoaster: http://linuxdevices.com/news/NS4820277296.html |
| 14:43:36 | pattieja | prpplague: morning |
| 14:43:38 | pattieja | morning everyone |
| 14:43:40 | thekernel | CosmicPenguin: are there any USB bluetooth devices to work on :) |
| 14:43:42 | pattieja | prpplague: hacked the Compaq iPAQ IA-1 last night (till almost 2am, zzzzz) |
| 14:44:20 | SpaceCoaster | CosmicPenguin: do you see it as being useful? |
| 14:45:30 | thekernel | I wonder if bluetooth could be applied to wireless keyboards and mice... |
| 14:46:52 | CosmicPenguin | thekernel: Why USB? |
| 14:47:44 | thekernel | You seem to have done some work with that too |
| 14:48:30 | thekernel | Anyway, I'm coming in now. See you in a few |
| 14:48:30 | collord | CosmicPenguin: are specs out for that natsemi/censoft gateway thing? |
| 14:51:12 | collord | CosmicPenguin: err, it seems more "conceptual" sorry |
| 14:51:48 | CosmicPenguin | collord: Sadly it is very conceptual... Its a cool idea though, and *I* think it has a future (but then again, I am emotionally involved) |
| 14:53:50 | collord | i've always loved "personal server"... and 802.11b AP functionality is near and dear too |
| 14:53:54 | prpplague | pattieja: make any progress? |
| 14:54:05 | CosmicPenguin | collord: I would have loved to add 802.11A..... :) |
| 14:55:20 | prpplague | collord: i've not done any research yet, but i'd like run 802.11b AP functionality on a linux box, is that easy to implement? |
| 14:55:31 | collord | CosmicPenguin: is intersil going to be as linux-friendly/hackable with A as they are B? |
| 14:56:51 | collord | prpplague: for Prism2 chips there are drivers... http://www.epitest.fi/Prism2/ |
| 14:57:22 | CosmicPenguin | collord: I haven't heard.... There were a few people showing 802.11a at Comdex, but not many.... |
| 14:57:48 | collord | prpplague: everyone I know of has had good luck especially with the newer versions of this driver. but we haven't e.g. tried to associate 50 clients or anything yet |
| 14:59:23 | collord | prpplauge: you'll want a card with an external antenna and there are a few. zcomax.com is tested |
| 15:02:23 | prpplague | collord: http://www.abcsinc.com/small-linux/dsc00044.jpg |
| 15:02:58 | Niddix | Whats the prefered cross-compile toolchain to use for the tux? |
| 15:03:27 | collord | prpplague: pricey? |
| 15:04:24 | prpplague | collord: retail with 16meg ram,16meg Disk-On-Module and dual pcmcia $465 |
| 15:05:06 | prpplague | collord: no a good choice as a retail item for an access point, but it will be competative as a wireless p-t-p router |
| 15:05:46 | collord | prpplague: ack! there's a group called musenki that is supposedly building for this purpose boxen with 2 PCMCIA and 2 10base for under $250 |
| 15:07:03 | prpplague | collord: ya, the thing is we already produce this unit as a thin client, this way we don't have to stock a whole different component |
| 15:07:40 | prpplague | collord: thanks for the zcomax link |
| 15:08:48 | collord | prpplague: here's a list... http://www.personaltelco.net/index.cgi/Prism2Card |
| 15:10:35 | prpplague | collord: is there a url on the misenki box? |
| 15:15:48 | collord | prpplague: hrm... not yet an official one i don't think. actually that was like $300 not $250 as i remember |
| 15:17:40 | prpplague | collord: if we stripped some of the features from the tcsx-2 we could get it down to that price, but that would compromise the functionality in out areas of use for us |
| 15:18:01 | prpplague | s/out/other |
| 15:18:21 | prpplague | anyway, sorry guys about the OT conversation.... |
| 15:21:06 | CosmicPenguin | prpplague: No worries... this is a hacking forum first and foremost.... |
| 15:22:43 | collord | prpplague: there's no getting around that "small" hardware is still expensive unless in huge quantities... old half-broken laptops serve my network... |
| 15:23:35 | prpplague | collord: ya i know what you mean, thats what i do a lot of my x86 prototyping on |
| 15:24:01 | Niddix | I've got several old 486 Compat Contura's with dead LCD panels.. They make nice "small" linux boxes. |
| 15:24:45 | prpplague | collord: our 5 year plan is to have a flexiable X-windows workstations for under $200 retail |
| 15:24:49 | Niddix | s/Compat/Compact |
| 15:25:24 | prpplague | Niddix: those are pretty nice to do lcd development on |
| 15:25:30 | collord | prpplague: LCD-based, included? |
| 15:26:19 | prpplague | collord: no svga based |
| 15:26:33 | Niddix | prpplague: In truth the LCD's are fine. Its just the housing around them has broken. |
| 15:26:45 | prpplague | Niddix: ahh |
| 15:27:15 | RedBaron | howdy all |
| 15:27:20 | prpplague | howdy |
| 15:27:32 | Niddix | prpplague: But they are useless as user laptops. |
| 15:28:08 | prpplague | RedBaron: found a source of $$ yet? |
| 15:28:39 | RedBaron | prpplague: Not yet...but I might soon |
| 15:29:44 | prpplague | RedBaron: crackin' crypto pdf files? |
| 15:30:43 | RedBaron | prpplague: *chuckle* Eh? |
| 15:33:36 | prpplague | RedBaron: just kidding, syslog: crc error in /proc/prpplague/humor |
| 15:34:22 | |\ | oops |
| 15:35:36 | Niddix | Can anyone give me a link to a decent sniffer? |
| 15:36:20 | prpplague | Niddix: i've got several guys here in the office that swear by etherRape |
| 15:37:38 | RedBaron | prpplague: *chuckle* No prob...just didnt follow yah. |
| 15:38:47 | RedBaron | Oh...thats EtherApe .... I had read it as 'Ether' 'Rape' ... hehe |
| 15:39:24 | Niddix | RedBaron: No he wrote etherRape.. not etherApe |
| 15:40:03 | RedBaron | Niddix: *nod* he did...but I cant find any etherrape projects...I can how ever find an etherApe project that is a sniffer. |
| 15:40:05 | jacquesZzZz | ethereal |
| 15:40:24 | prpplague | sorry its etherape, but everyone around here calls it ether-rape... |
| 15:40:33 | Niddix | ;) |
| 15:40:53 | RedBaron | *chuckle* |
| 15:41:11 | Niddix | Looks interesting.. I'll give it a shot. Trying to figure out why udchpc isn't able to contact my Novell DHCP server. |
| 15:47:34 | pattieja | prpplague: yep. It's running Linux now. :) |
| 15:47:48 | prpplague | pattieja: cool |
| 15:47:56 | pattieja | prpplague: it was really simple too |
| 15:48:04 | prpplague | pattieja: whatcha think? ltsp? |
| 15:48:24 | pattieja | prpplague: you could drop the image you have on your tcsx-1 in place with 0 modifications |
| 15:49:02 | prpplague | pattieja: what a nic? |
| 15:49:05 | pattieja | actually, I don't see any reason why you wouldn't get it to work by simply dd'ing an image onto the iPAQ IA-1 from a tcsx-1 image |
| 15:49:12 | pattieja | prpplague: doesn't have nics |
| 15:49:18 | pattieja | 4 USB ports |
| 15:49:24 | pattieja | don't you have yours yet? |
| 15:49:35 | pattieja | I thought you ordered them awhile ago? |
| 15:49:42 | prpplague | pattieja: thats what i thought, i've have to try the dlink usb-nic |
| 15:50:00 | Russ|werk | Niddix: what? |
| 15:50:04 | Russ|werk | something about udhcpc? |
| 15:50:13 | Russ|werk | btw, ethereal is the best sniffer |
| 15:50:26 | prpplague | pattieja: probably in inventory, but i've barely had time to take bathroom break much less leave the building.... |
| 15:50:28 | pattieja | prpplague: I've got both the Linksys USB100TX and a 3Com USB adapter working |
| 15:50:36 | pattieja | :) |
| 15:51:01 | pattieja | As far as using them for LTSP, they'll make great thin clients |
| 15:51:06 | pattieja | the audio is supported |
| 15:51:21 | pattieja | but I don't think the builtin modem is supported (but who cares?) |
| 15:53:27 | CosmicPenguin | Russ: We were using udhcpc at Comdex earlier this week, it worked great... thanks! |
| 15:54:35 | Niddix | Russ|werk: udhcpc can't contact my Netware DHCP server. |
| 16:11:28 | Russ|werk | what does ethereal show? |
| 16:12:20 | kernelwork | Hey MicroChip32 |
| 16:12:25 | MicroChip32 | straw |
| 16:13:02 | kernelwork | looks confusedly at MicroChip32 |
| 16:13:33 | | OK, jacquesZzZz. |
| 16:13:33 | jacquesZzZz | ibot: message for bzflag there is a typo on the wiki - 23 lbs 2 oz for a 4-pack should be 34 lbs 2 oz |
| 16:14:39 | MicroChip32 | kernelwork: hey -> hay -> straw |
| 16:15:12 | kernelwork | That really isn't funny. That sounds almost as bad as something I would say. |
| 16:15:18 | MicroChip32 | =) |
| 16:15:38 | kernelwork | Note: almost |
| 16:15:59 | MicroChip32 | yep, i can do worse tho, and have ;) |
| 16:16:38 | kernelwork | I don't doubt that |
| 16:17:54 | MicroChip32 | almost killed a girl in college simply by making one of those "its all in the timing" comments |
| 16:18:15 | MicroChip32 | she nearly died laughing ... literally |
| 16:30:09 | collord | prpplague: all of these way cheaper than celestix though |
| 16:30:09 | kernelwork | sharp drops nearby? |
| 16:33:28 | MicroChip32 | kernelwork: that to me? |
| 16:33:51 | kernelwork | yes, though poorly timed |
| 16:33:51 | sammy_wms | night all... |
| 16:33:55 | kernelwork | night? |
| 16:34:11 | MicroChip32 | kernelwork: hehe, no, was in the college commons ... she laughed herself blue, literally |
| 16:34:13 | kernelwork | hehe, and I though I sayed up late... |
| 16:36:05 | kernelwork | The great thing about laughing until it hurts is the fact that laughing releases those endorphins that act as a natural pain killer |
| 16:36:48 | MicroChip32 | hehehe yea, but if she'd kept laughing more than a few more seconds she wouldnt have had to worry bout pain no more, thats how close she was |
| 16:37:14 | kernelwork | Wow, it makes me think of some people laughing themselves silly at 2:00 AM |
| 16:37:31 | Russ|werk | Niddix? |
| 16:38:41 | kernelwork | I have some basic one-liners I use ocassionaly, and sometimes I have good timing and get a good laugh. |
| 16:38:48 | kernelwork | Most of the time I get wierd looks. |
| 16:39:03 | kernelwork | Then again, my parents were the first ones to give me wierd looks... |
| 16:39:13 | MicroChip32 | kernelwork: mine are usually off-the-cuff, or at least the ones that are any good are |
| 16:39:29 | MicroChip32 | kernelwork: parents have a way of doing that dont they :) |
| 16:39:51 | Niddix | Russ|werk? |
| 16:39:54 | kernelwork | Mine are mostly like that, but there are some that just apply so many places I find myself using them again. |
| 16:40:41 | MicroChip32 | true |
| 16:41:08 | kernelwork | That wierd looks from parents one is an old backup... |
| 16:41:31 | MicroChip32 | heheh gotta blame someone ;) |
| 16:41:32 | kernelwork | I usually use it around people that don't know me yet... |
| 16:42:07 | kernelwork | I also find myself cycling through phrases every couple of weeks. |
| 16:43:50 | kernelwork | As soon as I realize I keep saying the same thing over and over I stop then another phrase takes over until I realize I keep saying it. |
| 16:43:57 | kernelwork | I feel so recursive |
| 16:44:04 | MicroChip32 | heh, i dont get out in the public enough to cycle anything. shite i only get out of the house a few times a month |
| 16:44:12 | MicroChip32 | thats one bad thing about working at home |
| 16:44:16 | Niddix | kernelwork: Repeating yourself redundantly? |
| 16:44:34 | kernelwork | all over again too :) |
| 16:44:42 | MicroChip32 | and repeatedly |
| 16:45:01 | Niddix | Why is there all this appletalk crud on my Lan. |
| 16:45:25 | Russ|werk | Niddix: just watch the dhcp ports |
| 16:45:37 | kernelwork | *lame joke withheld for lack of decent context* |
| 16:45:55 | Russ|werk | Niddix: filter "port 67 or port 68" |
| 16:46:53 | Niddix | Russ|werk: Sorry the appletalk stuff distracted me. |
| 16:47:34 | Russ|werk | can you dcc em the dump? |
| 16:47:46 | Russ|werk | er, me |
| 16:47:47 | Niddix | Russ|werk: Ok I've captured udhcpc and pump |
| 16:47:55 | Russ|werk | I can open it in ethereal, you might want to make sure its koshure though |
| 16:48:09 | Niddix | Russ|werk: Pump actually gets an address. |
| 16:48:55 | Russ|werk | crap, I need to leave |
| 16:49:03 | Niddix | Russ|werk: Gimmi a sec.. I'm running ethereal on a different machine.. (My laptop is off the lan. Our firewall blocks IRC.) |
| 16:49:13 | Russ|werk | just email them to spam@asu.edu |
| 16:49:27 | Niddix | Ok. |
| 16:50:48 | Niddix | Russ|werk: How can I save the packet? |
| 16:51:18 | Niddix | Russ|werk: nevermind.. print to file. |
| 17:01:32 | GoRK | hello |
| 17:01:40 | GoRK | did russ have to take off? |
| 17:02:24 | GoRK | Russ: hey. did you get my mail/patch? |
| 17:02:42 | Russ|werk | ya, thanks |
| 17:02:54 | Russ|werk | I haven't had a chance to rummage through it yet |
| 17:03:09 | Russ|werk | Niddix: you forgot the attachments |
| 17:03:30 | GoRK | ok just wanted to make sure. it turns out i was having email server problems last night |
| 17:03:41 | Niddix | Doh |
| 17:03:53 | Russ|werk | anyway, now I do have to go |
| 17:05:06 | GoRK | have fun |
| 17:07:09 | GoRK | SpaceCoaster: you there? |
| 17:14:09 | SpaceCoaster | GoRK: hi |
| 17:14:41 | GoRK | SpaceCoaster: i cant seem to get the latest serial_flash.c to compile into the kernel.. it will only compile as a module... some linker errors |
| 17:14:55 | prpplague | doh |
| 17:15:20 | SpaceCoaster | GoRK: never tried it compiled in, just a moment |
| 17:15:27 | GoRK | SpaceCoaster: my latest patch against russ's 2.4.13-ac5-rmk2-tux2 is on my website.. i tried to put all the drivers where they belong also... |
| 17:15:47 | GoRK | so serial_flash.c actually is in /drivers/mtd/devices |
| 17:15:55 | GoRK | with its own config option |
| 17:16:46 | GoRK | SpaceCoaster: that patch also fixes up ucb1x00 support for all the other stuff also |
| 17:17:14 | SpaceCoaster | wow, you've been busy |
| 17:23:34 | SpaceCoaster | GoRK: was it complaining about serial_flash_init? |
| 17:23:44 | GoRK | yeah |
| 17:23:47 | GoRK | but |
| 17:23:56 | GoRK | i took those entries out of that one file |
| 17:24:05 | GoRK | how vague is that? |
| 17:26:48 | GoRK | SpaceCoaster: also bzflag suggested that the blk2mtd driver may get us with jffs2.. |
| 17:30:08 | Tangent | hears blkmtd mentioned |
| 17:30:16 | Tangent | Huh.. what do you want to do with that? |
| 17:30:41 | Tangent | My old house mate wrote that, because I bugged him to death :) |
| 17:32:13 | GoRK | haha well the atmel serial flash in the tux can only deal with blocks it seems |
| 17:33:02 | GoRK | jffs2 and the like i think expect to be able to read and write bytes and erase blocks |
| 17:33:51 | Tangent | GoRK : Nod.. mtdblk may help out there... |
| 17:34:31 | Tangent | GoRK : spse is the guy to bug about that.. he's not online too often tho... except for when he's not bumming around my house (around 6 days a week, he's sitting right next to me) |
| 17:34:47 | Tangent | GoRK : Sadly, he's not here right now... |
| 17:34:58 | Tangent | GoRK : still sounds like a promising way ahead :) |
| 17:36:27 | Russ | GoRK: what did you change with the backlight? |
| 17:36:35 | prpplague | pattieja: you there? |
| 17:36:56 | GoRK | Russ: you have to enable the mcp before you can write registers... so you dont step on the audio's feet |
| 17:37:54 | Russ | ah, well, the new backlight thing will be tricky |
| 17:38:04 | Russ | if its compiled as a module, then its easy |
| 17:38:12 | GoRK | Russ: actually im not sure if the ioctl's will work due to that fact or not... i didnt write a program to test them... if not then all you'd have to do is call the mcp_enable and mcp_disable things when you read the ioctls |
| 17:38:20 | Russ | but if its in the kernel, I want it to init after the ucb1x00-core |
| 17:38:21 | GoRK | where was the new backlight supposed to go? |
| 17:38:36 | Russ | it hooks onto sa1100fb |
| 17:38:40 | GoRK | it should init after ucb1x00-core. it depends on it |
| 17:38:45 | GoRK | hmm yes that will be difficult |
| 17:39:44 | GoRK | is that an ioctl test app? |
| 17:40:09 | Russ | no, that is my attempt at a standardized way to do backlights |
| 17:40:15 | GoRK | oh |
| 17:40:21 | Russ | (along with slight modification to sa1100fb.[ch]) |
| 17:40:35 | GoRK | so are you looking to try to do things like ipaq frontlight? |
| 17:40:41 | GoRK | isnt theirs on a sa1100 gpio? |
| 17:41:07 | Russ | I want them to be able to make a similar file called ipaq_frontlight.c |
| 17:41:19 | Russ | that compiles, loads, and interfaces with userspace in the same way |
| 17:41:28 | Russ | class |
| 17:41:52 | GoRK | are the ipaq folks going to stand for it? :) |
| 17:42:08 | pattieja | is away: lunchtime |
| 17:54:53 | CosmicPenguin | if RMK likes the idea, they will have to stand for it |
| 17:59:50 | GoRK | true |
| 17:59:57 | GoRK | btw, |
| 18:00:10 | GoRK | CosmicPenguin: the new ucb1x00-ts device requires an input driver overhaul |
| 18:00:38 | GoRK | no more kernel-space calibration or even touch detection... you have to do it all in userspace! |
| 18:00:39 | CosmicPenguin | GoRK: ?? |
| 18:00:53 | CosmicPenguin | GoRK: Ohh.... I see.... |
| 18:01:22 | GoRK | luckily you can most likely jack the code from the old kernel driver :) that's what i plan on doing anyway |
| 18:01:23 | CosmicPenguin | GoRK: I really don't like that idea at all...... |
| 18:01:53 | GoRK | CosmicPenguin: yep know what you mean. i think it is ok for a touchscreen driver to process coordinates, but whatever |
| 18:02:06 | GoRK | <CosmicPenguin> if RMK likes the idea, they will have to stand for it |
| 18:02:32 | GoRK | :) |
| 18:02:37 | CosmicPenguin | GoRK: I know, I know.... But it is a stupid idea born of a stupid philosophy..... |
| 18:02:54 | GoRK | yes i had this talk in here yesterday |
| 18:03:09 | CosmicPenguin | GoRK: But, even if RMK does like it, this is linux, right? |
| 18:03:15 | GoRK | it's not really a stupid idea but it's not suited to linux. the HURD does things that way |
| 18:03:37 | GoRK | daemons for every piece of hardware |
| 18:04:32 | CosmicPenguin | GoRK: Well, great.... there's another 5k of code that we have to add that we could have done previously in 3 lines in the kernel..... |
| 18:05:00 | CosmicPenguin | GoRK: Plus, now everyone needs to gigure out what the hell the UCB1200 is saying and interpet it.... |
| 18:05:25 | GoRK | the new driver returns x, y, pressure |
| 18:05:32 | GoRK | err pressure, x, y actually |
| 18:05:59 | GoRK | but you have to account for anomolous points, jitter, etc. all in userland now |
| 18:06:06 | CosmicPenguin | GoRK: Ohh, I thought that it didn't do touch detection.... |
| 18:06:40 | GoRK | well it doesnt... a point with a low pressure reading is probably an accidental touch or an inconsistent reading |
| 18:06:55 | CosmicPenguin | GoRK: Surprise, surprise, they're standardizing to X11 again.... |
| 18:07:12 | GoRK | you still have to figure out if the data is a real touch or not |
| 18:07:45 | CosmicPenguin | GoRK: Bah.... if I talk about it any more, I will just get more pissed..... lets just adapt and move on.... |
| 18:08:21 | GoRK | heh |
| 18:09:22 | GoRK | to be fair, audio playback works much better, but you should hear the loops we gotta jump through now to playback standard rates/formats not to mention what we gotta do to record... :) |
| 18:23:44 | pattieja | is back (gone 00:41:36) |
| 19:24:30 | CosmicPenguin | Oh, great... there goes the neighborhood..... |
| 19:24:31 | jwalden | Hello All ;-) |
| 19:24:53 | jwalden | Hi Cosmo - well - equal opportunity and all... |
| 19:25:25 | CosmicPenguin | jwalden: No, really.... welcome... |
| 19:25:49 | CosmicPenguin | jwalden: Shaun is on here too.... /msg kernelwork for a good time.... :) |
| 19:26:03 | jwalden | CosmicPenquin: Thanks! I love this box - I have a few guys build out the shanip boards for me |
| 19:26:34 | jwalden | CosmicPenguin: And also got the 5v PCMCIA working on Sunday... |
| 19:27:37 | jwalden | CosmicPenguin: Also had a little friend that I used to work with and a large telco help me out with a little WLAN project on Sun Nite for this box... |
| 19:28:00 | jwalden | And I'm not sure how to properly divulge the info... |
| 19:35:46 | MicroChip32 | jwalden: just spit it out, or better yet, make a really cool wiki page and post it =) |
| 19:48:12 | CosmicPenguin | jwalden: Wanna make a shanip for me? I hate doing hw stuff.... |
| 19:50:16 | CosmicPenguin | MicroChip32: You ever get that USB stuff worked out? |
| 19:51:28 | kernelwork | returns from lunch. |
| 19:51:32 | MicroChip32 | CosmicPenguin: not really. still have no clue to how to communicate with the usb device, wether from an app, or from the driver i need to write |
| 19:53:14 | CosmicPenguin | Want me to msg you the usb-skeleton file? I am surprised you don't have it though.... |
| 19:55:39 | MicroChip32 | CosmicPenguin: i got one from someone, but glancing thru it nothing made alot of sense since im not (yet) familiar with drivers (working on that part) |
| 19:56:06 | CosmicPenguin | MicroChip32: Ahh, trial by fire... |
| 19:56:16 | GoRK | MicroChip32: you can use usb-robot-slave and libusb to rapidly prototype a driver in perl or python or some other language.. then code c from that |
| 19:56:49 | MicroChip32 | hmmmmmm those are perl modules? or what ? |
| 19:57:33 | GoRK | MicroChip32: no they are programs. you can talk to them from a scripting interface. im using perl to do some hacking on the ifeel logitech mice to get the force feedback stuffs working. |
| 19:57:51 | GoRK | MicroChip32: they have fully automated reverse enginerring support also :) |
| 19:58:01 | MicroChip32 | GoRK: actually i just wanna know how to comm with the usb port, wether from an app or a driver. and i cant find anything on it, at all |
| 19:58:15 | MicroChip32 | GoRK: cool |
| 19:58:19 | GoRK | directly? or do you need to write a device driver? |
| 19:58:44 | MicroChip32 | GoRK: i need to write a driver, but before i can do that i need to know how to talk to the usb interface |
| 19:58:56 | GoRK | libusb lets you talk directly to usb devices from userspace (ie. userspace drivers) |
| 19:59:33 | Tangent | MicroChip32 : What you trying to do with USB? |
| 19:59:45 | Tangent | MicroChip32 : On the Tux? |
| 19:59:50 | MicroChip32 | Tangent: http://voipblaster.creative.com/ |
| 20:00:30 | GoRK | MicroChip32: the voipblaster uses a proprietary codec that you can't decompress. there is already some support for it but it's fairly useless |
| 20:00:45 | GoRK | the license fees are exhorbitant |
| 20:01:20 | MicroChip32 | GoRK: not trying to reproduce the codec. just port the drivers to linux so can use it on linux too |
| 20:01:36 | BZFlag | usb on early rev sa1100's is supposed to be very broken. stay away. |
| 20:02:19 | GoRK | MicroChip32: you can't. not legally anyway and finding the codec sources is going to be a hassle to do illegially too :) the only way to talk to it is to use the g.723.1 codec. |
| 20:02:31 | BZFlag | get a mono ipaq for a cheap usb client. don't know of a cheap usb host. |
| 20:03:09 | GoRK | MicroChip32: the only thing you could really do in linux legally with the voipblaster is to do simple IVR by moving g.723.1 frames around without decoding them. |
| 20:03:58 | MicroChip32 | GoRK: the pc shouldnt have to encode/decode anything, all that is, as far as i understand it, done entirely on the blaster |
| 20:04:06 | scanline | is back (gone 10:00:08) |
| 20:04:19 | Russ | GoRK: can you give good justification for the ucb1x00-audio hunk other than it works? |
| 20:04:20 | MicroChip32 | the only thing the driver does is send config commands and the raw audio stream |
| 20:05:50 | GoRK | MicroChip32: the audio data has to go IN and OUT of the voip blashter in g.723.1 period. you can't legally encode or decode g.723.1 on the host cpu without a license |
| 20:06:27 | GoRK | Russ: yes. they were setting audio mcp registers on the mcp and ucb1200 for the telecom codec. i fixed that |
| 20:06:46 | MicroChip32 | GoRK: that cant be, else how can i talk over the blaster using a std phone plugged into the blaster, the phone decoding the 723?? |
| 20:07:09 | Russ | GoRK: is that shannon specific, or ucb1x00 specific |
| 20:07:43 | GoRK | MicroChip32: argh. read the asterisk-pbx list archives for the whole schmiel. the data going in and out of the unit is g.723.1. there are no apps on linux that support g.723.1 because it requires a license. the phone line audio has nothing to do with it |
| 20:08:06 | GoRK | Russ: it can go upstream actually. it's a bug in the driver i think |
| 20:08:18 | GoRK | ok gotta go.. bbiab |
| 20:08:38 | MicroChip32 | GoRK: ok yes, the pc-side data would be 723... but the pc has no need to [en/de]code that at all, just pass it along to the net |
| 20:10:23 | GoRK | MicroChip32: pass it to where? the remote app has to support it |
| 20:11:07 | MicroChip32 | GoRK: at the other end of the internet connection? naturally, presumably another blaster then. unless its a pc->phone connection |
| 20:11:19 | MacAtack | yawns... Good morning |
| 20:11:21 | GoRK | you can get fxo and fxs interfaces for $99 that aren't proprietary. that's not much more than the voipblaster anyway |
| 20:11:53 | CosmicPenguin | Hey! Did somebody add a menuconfig script to the buildroot? Or is that busybox? |
| 20:13:13 | MicroChip32 | GoRK: im not trying to decode the audio or anyting at all in linux, only trying to make the application interface just like in windows and aparently in netbsd |
| 20:15:02 | MicroChip32 | sticks a foot in MacAtack's yawn |
| 20:15:55 | MacAtack | Gork: what MicroChip32 and I are trying to accomplish is hack up a Linux device driver for the VoIP Blaster. Then we can utilize the Open H.323 project and VoIP Blaster to make h.323 calls... not specificly relating to the tux but hell if there is a usb port on the thing why not |
| 20:16:11 | scanline | GoRK: I saw your comment about releasing a new jffs2 image this weekend... I suppose i should fix up pgtuxphone a bit before then? |
| 20:17:47 | AlHazWork | Anybody understand the color codes on axial ceramic fuses? |
| 20:18:32 | RedBaron | AlHazWork: Oh...I used to remember it. |
| 20:18:32 | AlHazWork | I've got a little blue fuse here, size of a resistor, with color bands green-silver-silver-white-orange |
| 20:18:37 | AlHazWork | or the other way around |
| 20:18:53 | MicroChip32 | AlHazWork: i would think theyd be coded essentially the same as resistors |
| 20:19:14 | AlHazWork | MicroChip32: I never memorized those either :) |
| 20:19:19 | MicroChip32 | hehehe |
| 20:19:46 | MicroChip32 | 0-x black brown red orange yellow green blue purple silver white |
| 20:19:53 | MicroChip32 | grey white actually |
| 20:20:16 | GoRK | macattack: you cant use openh323. h323 requires the ability to do codec translation which you cant do without decoding g.732.1 |
| 20:20:41 | MacAtack | Big Buicks Run On Yellow Gas But Voltzwagons Go Without |
| 20:21:00 | MicroChip32 | hahahaha MacAtack |
| 20:21:18 | MacAtack | Black Brown Red Orange Yellow Green Blue Violet White (s/Go/go/ |
| 20:21:31 | GoRK | you can transport g.723.1 over h.323 but unless the recieving end can handle g.723.1 it's not gonna work very well |
| 20:21:54 | MicroChip32 | GoRK: thats our point, the other end WILL support 723. |
| 20:23:02 | MicroChip32 | we just want to use the blaster in linux the same way its used in windows. the windows interface wont let you talk with non 723 endpoints either |
| 20:23:27 | MacAtack | MicroChip32: well on windblows you can use MS-GSM Software Codec |
| 20:23:28 | GoRK | ah i didnt know that about it then |
| 20:23:48 | GoRK | kind of a useless product then, no wonder they dump the hardware |
| 20:24:11 | MicroChip32 | GoRK: real cheap world wide phone calls, is enough for me ;) |
| 20:24:46 | MacAtack | GoRK: works for my application... i bought 4 of them so far and will prolly pick up a few hundered once we get a working driver (then give them to people, and make them use my VoIP termination service :) |
| 20:25:10 | MacAtack | MicroChip32: or FREE if both ends have these wonderfull things |
| 20:25:21 | MicroChip32 | MacAtack: yep yep |
| 20:25:40 | MacAtack | GoRK: but yes they would have been much more usefull if they had PCM |
| 20:26:20 | GoRK | so get the $99 quicknet product that does all the other codecs and stuff (including g.723.1) for $99 instead |
| 20:26:44 | MicroChip32 | snickers waiting for MacAtack's response to that <G> |
| 20:27:33 | MacAtack | blah quicknet sucks... i've got 4 of there cards (Phone and LineJacks) and they've been trying to repair 2 of them for 3 months (3 seperate trips, everytime they've come back they never worked) |
| 20:28:51 | MacAtack | plus i'm a primary developer for the Open H.323 project by equivaliance which is owned by Quicknet and I still don't get any better service |
| 20:29:31 | MacAtack | But i'm not 31337 enough to code my own h.323 stack so I can never get completely away from quicknet |
| 20:30:31 | BZFlag | to work |
| 20:30:40 | RedBaron | Does anyone know if anyone has gotten a larger SODIMM to work with linux and the tux? |
| 20:31:19 | MacAtack | Now someday I ~may~ pick up a PhoneCard from quicknet for laptop VoIP, but that would be the only thing i ever buy from quicknet again |
| 20:31:30 | CosmicPenguin | Everyone see the new yopy? http://www.yopy.com/ |
| 20:31:52 | RedBaron | CosmicPenguin: Yah I did...still waiting for this vaporware project to become reality. :/ |
| 20:32:41 | CosmicPenguin | RedBaron: I have actually held a Yopy in my hand..... the older developer version... way uncool.... |
| 20:32:59 | RedBaron | CosmicPenguin: uncool? What was bad about it? |
| 20:33:08 | AlHazWork | yeah, the old yopy was a brick |
| 20:33:32 | CosmicPenguin | RedBaron: It was exactly that - uncool... slow, annoying, hard to hold... you name it... |
| 20:33:52 | RedBaron | CosmicPenguin: I knew they had some developer versions...that you could get for a large price and a lot less memory than the commercial version was supposed to have. |
| 20:34:07 | RedBaron | CosmicPenguin: Hrm...that sucks. |
| 20:35:50 | Russ | GoRK: I don't think serial_flash likes you calling the cleanup function from where you are calling it and still declaring it static __init/__exit |
| 20:36:21 | SteevD | RedBaron: there is initial experimental support for the 32M SODIMM - but only giving 32M accessible RAM. This worked for me. |
| 20:37:59 | kram | has returned. |
| 20:44:24 | RedBaron | SteevD: Yah...the 8M onboard wasnt accessable right? |
| 20:45:07 | scanline | is away: class |
| 20:46:07 | Tangent | RedBaron : That's right |
| 20:48:26 | SteevD | RedBaron: I know that SpaceCoaster has been trying hard to get all 40 going but so far not succeeded |
| 20:51:48 | RedBaron | SteevD: Ah. okay. |
| 20:53:12 | nanoprobe | Are there any apps like tenux or wiol ported for use on tuxscreen? Would be really cool to connect to a AXE10 switch from a phone :) |
| 20:55:49 | Niddix | Russ: Did you get the attachments the 2nd time I sent it? |
| 20:56:10 | Russ | yah |
| 20:57:25 | Niddix | See anything obvious in the successful lease via pump? |
| 21:01:29 | Russ | haven't had much of a chance yet |
| 21:02:10 | Niddix | No problem. I'm still working on getting my cross-compiler setup. |
| 21:02:28 | Russ | actually |
| 21:02:35 | Russ | I don't see either of them getting a reply |
| 21:04:13 | Russ | it looks like the replies just weren't included |
| 21:04:56 | Niddix | Russ: Hrmm.. Maybe they got filtered. |
| 21:05:21 | Niddix | Russ: I've got to run anyway.. no worries. I'll recapture them tomorrow and send them to you. |
| 21:07:11 | Russ | hmm...it may be that novell needs padding |
| 21:16:06 | as_ | Russ, something offtopic, but do you have a good place to get chip data sheets from? |
| 21:18:21 | as_ | I'm at a serios loose end. |
| 21:18:52 | AlHazWork | http://www.embeddedlinks.com/chipdir/ <-- this can be a good start |
| 21:19:08 | as_ | I'll look tnx. |
| 21:20:01 | as_ | All my usual sites have come up blank. As has the manufacturers homepage. I only find publicity stuff about it. |
| 21:20:27 | AlHazWork | what are you looking for? |
| 21:21:10 | as_ | Its a NEC MPEG chipset. |
| 21:21:17 | Russ | try NEC then |
| 21:21:32 | as_ | Ive tried the webpage. |
| 21:21:47 | AlHazWork | ah |
| 21:22:38 | AlHazWork | they prolly licensed a bunch of other peoples tech. and can't release specs w/o an NDA |
| 21:22:50 | as_ | And unusually IVe found nothing, no family docs, nothing. Just a few scraps of information. |
| 21:22:57 | as_ | Thats possible. |
| 21:23:16 | Russ | email them |
| 21:25:07 | as_ | Re. |
| 21:25:11 | as_ | <as_> I think I will have too. But since I dont work for a large electronics company I dont hold much hope in it. |
| 21:26:38 | collord | there are routines for building G.723.1 in the Intel IPP libraries for SA1100... anyone played with? |
| 21:27:00 | Russ | ok, time to start merging linux-2.4.13-ac8-rmk1-tux1 |
| 21:27:06 | Russ | any one want to throw anything on? |
| 21:27:41 | as_ | I'm still waiting on my tux. Not BZs fault. |
| 21:29:56 | Tangent | throws on a shirt |
| 21:30:11 | Tangent | kinda cold in here |
| 21:30:29 | as_ | Well. It wont dry very fast on a tux will it? |
| 21:30:33 | kernelwork | usually likes to keep his shirt on |
| 21:30:39 | RedBaron | Tangent: Hmm...getting rather civilized there. :P |
| 21:30:49 | Tangent | It'll be trousers next ;) |
| 21:31:12 | as_ | I think we will probably leave at that point. |
| 21:31:21 | Tangent | best to |
| 21:31:22 | kernelwork | Please, I only ate a few hours ago |
| 21:31:50 | RedBaron | I'm eating right now! |
| 21:31:54 | Russ | Tangent: I meant into the next kernel release |
| 21:32:14 | kernelwork | Why is it people keep trying to release me? |
| 21:32:24 | kernelwork | or my counterpart thekernel :) |
| 21:32:28 | Tangent | Russ : As long as it supports the second serial port, I'm happy |
| 21:32:52 | Russ | you have a small patch? |
| 21:33:10 | Tangent | Two lines |
| 21:33:19 | Tangent | in shannon.c |
| 21:33:38 | as_ | This is an interesting develeopment, the last unit had a housekeeping micrcontroller, this one has a FPGA instead. A Xilinx spartan. I wonder if design were cutting corners. |
| 21:34:43 | Russ | care to paste? |
| 21:34:53 | AlHazWork | hmm, if i read this thing backwards, and it's orange-white-silver-silver-green, that might indicate a 3.9ohm resistor . . . |
| 21:34:54 | Tangent | sa1100_register_uart(1, 1); |
| 21:34:55 | Tangent | Ser1SDCR0 |= SDCR0_SUS; |
| 21:35:49 | as_ | Alhaz, do you know someone with a working one you could test? |
| 21:36:18 | AlHazWork | as_: not one that's right near by |
| 21:36:48 | as_ | Ahh, well. The voices in my head still say short it. |
| 21:36:56 | collord | hrm, i guess that the IPP libraries being binary-only are a good reason for no... thought they weren't |
| 21:37:03 | GoRK | Russ: i asked spacecoaster about the serial_flash thing... it's his code :) he was only testing it as modules, afaik |
| 21:37:12 | GoRK | Russ: did you have a fix for it? |
| 21:38:26 | AlHazWork | actually, should be .39 ohms . . . . this doesn't make sense |
| 21:38:46 | AlHazWork | goes and pesters an EE |
| 21:41:18 | Russ | he already commited a fix, about 50 minutes ago |
| 21:41:38 | Russ | I'll download whats in cvs and use that (wheaties->newwheaties.c) |
| 21:54:05 | GoRK | cool |
| 21:55:42 | GoRK | newwheaties? |
| 21:56:41 | GoRK | Russ: at some point wheaties.c and serial_flash.c will share code from spi.c so you might wantt o move it out of mtd into some common location... /arch/arm comes to mind and give it another config option that wheaties and serial_flash depend on |
| 21:57:39 | GoRK | nothing else uses spi at this point on tux, but it could definately on other arch's.. |
| 21:57:49 | GoRK | anyway i gotta go do some consulting junk. bbiaw |
| 21:58:32 | scanline | is back (gone 01:13:25) |
| 21:59:15 | Russ | maybe drivers/misc |
| 22:22:49 | prpplague | check this out from /. - http://www.senseboard.com/products.htm |
| 22:24:08 | Tangent | prpplague : Looks cool to me :) |
| 22:25:46 | prpplague | heads home, had all the cisco @$%@#$% he can stand for today |
| 22:27:06 | prpplague | night all |
| 22:28:13 | Russ | Linux version 2.4.13-ac8-rmk1-tux1 (russ@russ) (gcc version 2.95.2 19991024 (release)) #40 Wed N |
| 22:28:13 | Russ | ov 14 16:15:01 MST 2001 |
| 22:32:53 | Tangent | spse : There may be a use for mtdblk on the Tux to talk to the serial flash |
| 22:33:01 | jacques | hi |
| 22:33:10 | Tangent | jacques : Evening |
| 22:33:36 | spse | what serial flash? |
| 22:33:58 | Tangent | GoRK : You around still? |
| 22:34:32 | jacques | my accelerometers came!!! |
| 22:34:36 | jacques | whoo!! |
| 22:34:55 | Tangent | spse : I don't really know the low down details yet... just that it uses 256byte blocks, and there was some talk of maybe needing to use mtdblk on it to get jffs2 working |
| 22:35:19 | Tangent | jacques : Cool :) |
| 22:36:03 | spse | what type of device is it? |
| 22:37:16 | Tangent | spse : Again, not certain... atmel serial flash, but not sure which one |
| 22:37:19 | jacques | is it possible to use the dsp on the tux as single or multiple a/d converters? |
| 22:38:41 | spse | prob easier to write a straight chip driver, pretty easy if you have the programming docs |
| 22:39:18 | jacques | ?? |
| 22:42:47 | BZFlag | andersee: I think I actually got dotster to create a name server record. |
| 22:43:17 | AlHazWork | BZFlag: I'm sure they'll realize their error eventally . . . |
| 22:43:19 | andersee | oh? |
| 22:43:26 | BZFlag | I have to go under the domain for the nameserver I want to add (ie rikers.org) and then create a ns1 at that point it will accept and address. |
| 22:43:29 | andersee | What did you do? |
| 22:43:39 | BZFlag | there is no way to change or delete one via the web. |
| 22:43:45 | andersee | silly |
| 22:44:40 | BZFlag | and since the name server RIKERS.ORG is in use by my brother-in-law's inactive domain on another registrar, they say NSI will not accept a delete or a modify request. |
| 22:45:17 | andersee | Now that is _stupid_ |
| 22:45:47 | BZFlag | agreed, but it may in fact be NSI's fault, not dotster. |
| 22:45:54 | scanline | is away: food |
| 22:46:30 | andersee | BZFlag: seen http://uclibc.org/uClibc-apps.html |
| 22:46:32 | andersee | ? |
| 22:51:14 | MacAtack | MicroChip32: nice of you to join the party again |
| 22:51:43 | MicroChip32 | yea yea, tell that to my friggin isp |
| 22:51:45 | signal11 | bzflag how many tux screens left? |
| 22:51:57 | jacques | is it possible to use the dsp on the tux as single or multiple a/d converters? |
| 22:54:20 | CosmicPenguin | andersee: Thats awesome.... I've got a few more... |
| 22:56:04 | jacques | cool, pcmcia-cs |
| 22:59:53 | andersee | CosmicPenguin: k. gimme the list (or even better a patch) |
| 23:04:02 | MicroChip32 | CosmicPenguin: earlier was it you that went quiet, or did i get disconnected and didnt know it ? |
| 23:09:46 | CosmicPenguin | MicroChip32: You got so disconnected.... |
| 23:09:47 | BZFlag | signal11: over 700 |
| 23:10:16 | MicroChip32 | CosmicPenguin: heheh *kicks his isp* i just realized that a few minutes ago *lol* wondered why everything was quiet |
| 23:11:02 | MicroChip32 | CosmicPenguin: but anyway, im making [slow] progress, thanks to your info =) |
| 23:11:23 | CosmicPenguin | Good.... |
| 23:12:58 | MicroChip32 | CosmicPenguin: at least now i know what im looking at and for =) |
| 23:23:47 | MacAtack | weeeeee that was fun |
| 23:25:17 | MicroChip32 | i wanna know when theyre gonna start sellin tickets for those rides |
| 23:25:22 | Tangent | Roll up, Roll up... just 10c a ride.. ride the IRC rollercoaster.. just 10C a go |
| 23:26:20 | MicroChip32 | whoops there he goes again, off on a Tangent |
| 23:26:24 | MicroChip32 | ;) |
| 23:27:05 | Tangent | groans |
| 23:27:57 | jacques | this is probably the hardest conversion I have ever seen |
| 23:28:33 | CosmicPenguin | Its pretty cool to watche everyone leave and then come flying back |
| 23:28:41 | MicroChip32 | isnt even going to touch jacques' comment |
| 23:28:45 | scanline | is away: lab |
| 23:31:10 | thekernel | CosmicPenguin: is jeffm still there? |
| 23:31:40 | MicroChip32 | thekernel: got that map stuff all finished yet ? |
| 23:31:42 | CosmicPenguin | thekernel: yeppers |
| 23:31:49 | thekernel | What's his extension? |
| 23:31:55 | thekernel | MicroChip32: no... |
| 23:32:03 | thekernel | *hangs head in shame* |
| 23:32:05 | MicroChip32 | thekernel: hehe |
| 23:35:57 | Tangent | BZFlag : Do you know the part number for the Atmel serial eeprom in the tux? |
| 23:36:58 | jacques | Tangent: what does it do? |
| 23:37:29 | Tangent | jacques : Well, not very much at the moment.. We'll be able to store stuff in there soon.. but who knows what stuff |
| 23:37:47 | CosmicPenguin | scanline: Have you updated picogui to use the new kernel stuff yet? |
| 23:37:49 | Tangent | jacques : I only overheard part of a conversation earlier.. I don't really know the details |
| 23:38:36 | spse | Tangent: is it just a chip on its own or is it part of a subsytem in the tux? |
| 23:38:55 | Tangent | spse : Yeah.. one of those two |
| 23:39:06 | Tangent | spse : It's one of the two SPI devices in the Tux |
| 23:42:08 | spse | well ive found a data sheet for one |
| 23:42:31 | jacques | what's the other SPI device? |
| 23:42:50 | Tangent | jacques : Serial protocol for 2/3 wire bus |
| 23:43:11 | Tangent | jacques : Basically two shift registers hooked back to back.. one on each end of the serial link... |
| 23:43:36 | Tangent | jacques : When the clock ticks (controlled by the master device), the two bytes swap from one device to the other |
| 23:44:28 | BZFlag | Tangent: nope I dont think so. checking the wiki... |
| 23:44:55 | jacques | Tangent: cool, explan of SPI, but I was askng what the other SPI device was - you mentioned there were 2 - is the other the proc? |
| 23:45:31 | Tangent | jacques : Ok.. techincally the CPU is one too.. it's the master device.. the USB1200 DSP is the other one |
| 23:45:49 | Tangent | jacques : Or wheaties.. or maybe they're the same thing... not 100% sure |
| 23:46:12 | BZFlag | Tangent: AT45DB041A-RC on http://TuxScreen.net/wiki/view/Main/Hardware |
| 23:46:20 | Tangent | BZFlag : Many thnaks |
| 23:46:26 | jacques | Tangent: thanks :-) |
| 23:46:38 | BZFlag | amazing what info the wiki has. ;-) |
| 23:47:18 | Tangent | BZFlag : Yeah.. That's the thing about open source stuff... the docs are all there if you can be bothered to look ;) |
| 23:47:38 | jacques | ive been reading everything on the wiki in preparation of getting my `screens (maybe tomorrow?) |
| 23:48:23 | Tangent | puts a link to the datasheet in the wiki |
| 23:49:20 | jacques | this one? http://www.atmel.com/atmel/acrobat/doc1432.pdf |
| 23:49:26 | Tangent | Yep |
| 23:49:47 | CosmicPenguin | All - I will see you tommorrow... |
| 23:51:17 | jacques | so what's it for? |
| 23:51:31 | jacques | why is it needed - there is already 4MB flash on the board |
| 23:52:32 | Tangent | jacques : It's for us to play with :) |
| 23:53:10 | spse | atmel do stuff to interface with sound chips like voice memos etc - maybe that was what they planned to use it for |
| 23:54:07 | jacques | it's only 512kB |
| 23:54:35 | andersee | spse: atmel chips (such as the at75C330) include dual onboard DSPs... |
| 23:57:35 | jacques | how do I physically connect to a LCC package IC? |
| 23:57:53 | jacques | is there a socket they fit in or are they for surface mount only? |
