05:04:09 | andersee | hi all |
05:15:04 | Russ | hello |
05:15:23 | Russ | did a few more things for a udhcp release |
05:15:36 | Russ | added a client, with the option of shared binary |
05:16:43 | andersee | wow! awesome! |
05:17:10 | Russ | it calls three scripts, deconfig, bound, and renew |
05:17:12 | andersee | I'm busy flashing my phone with the latest and greatest. I see the keyboard is working now... :-) |
05:17:23 | Russ | it writes a file with all the options as well as sets them in inv vars |
05:17:28 | Russ | mostly |
05:17:54 | andersee | deconfig? Perhaps better named 'release'? |
05:18:08 | Russ | its called at the very start or when a lease is lost |
05:18:25 | Russ | the script would be ifconfig $interface 0.0.0.0 |
05:20:02 | andersee | Hmm. I can see I need to grab a copy of the latest and greatest and take a look. |
05:20:13 | Russ | I haven't commited it yet |
05:20:19 | andersee | BTW, did you see I got busybox released... |
05:20:33 | Russ | was going to add the LOG() stuff, and docs |
05:20:39 | andersee | I fixed a bunch of little things before the release. |
05:20:40 | Russ | I just keep up on the CVS |
05:20:50 | andersee | sysloging |
05:21:40 | Russ | I never use syslog on an embedded system (yet) |
05:21:54 | andersee | what is LOG() then |
05:22:57 | Russ | syslog... |
05:23:07 | Russ | I realize people want that |
05:23:20 | Russ | if syslog isn't defined, it goes to stdout |
05:23:37 | andersee | ok, I follow you now. |
05:23:57 | andersee | What did you have to do to make the kbd work? |
05:24:25 | Russ | 1266 baud, odd parity, 8 data bits, one stop bit |
05:25:07 | Russ | then I made an input driver, mapped the scancodes |
05:25:24 | Russ | but then I had to write a silly x86ish backend for keyboard.c |
05:25:48 | andersee | So it was just an issue of setting a funky speed then to get the input, then shoving scancodes into the keyboard driver. |
05:25:55 | Russ | right |
05:26:13 | Russ | looked at the signals on tangent's scope |
05:26:28 | andersee | wishes he had a scope |
05:26:37 | Russ | so do I |
05:27:26 | Russ | you really went to town with libc5 |
05:27:50 | andersee | I've finished my latest project stuff at work, so starting Monday I'm back on uClibc work full time |
05:28:35 | Russ | any chance of C++ support? I don't really care for it, but I'm sure there are a lot of people who would love uclibc and qpe |
05:28:42 | andersee | yeah, there is a group of people that I know use libc5. Already reported some warnings (they are gcc 2.73 bugs, not problems)... |
05:29:52 | andersee | Actually, yes. In fact m68k already has it... I've just never bothered with moving that over to the other arches. It really just requires some crt0 mods for proper global constructor handling, and destructor handling |
05:30:25 | Russ | does that include a libustdc+++ |
05:32:35 | andersee | You just have to recompile gcc vs uClibc, and libstdc++ gets built... The header files are all fixed up for c++ support so it is just a recompile issue (once the crt0 c++ support is in for the arch) |
05:33:29 | Russ | not fun |
05:34:27 | andersee | not too bad really -- those wanting c++ can pay the price... I personally don't see much call for it on embedded systems, but... |
05:35:14 | Russ | its the silly qt people |
05:35:24 | andersee | Any idea where the code for turning off the backlight on the phone might need to go? |
05:35:27 | Russ | I really wish wvdial wasn't in c++ |
05:35:48 | Russ | andersee: I'm not sure, I have some stuff in sa1100_frontlight |
05:35:56 | andersee | heh, yup. And the folks who think they need a JVM on their embedded systems... |
05:36:08 | Russ | but I don't like it there, it should be some tty or fb ioctl's I think |
05:37:06 | andersee | I'll take a look about for that stuff |
05:37:16 | andersee | I'm pretty pleased so far. |
05:37:49 | Russ | the intensity, contrast, and backlight power are all aparently setable, but I've never done it |
05:38:03 | andersee | Another area I want to poke at is the phone code... Since Linuxing it, the speakerphone is gone, and dialing is now pulse, not tone |
05:38:29 | andersee | Presumable the inferno code was poking some magic registers somewhere |
05:39:15 | Russ | I have that stuff |
05:39:27 | andersee | oh? |
05:39:34 | Russ | but pcmcia, cleaned up keyboard, and X running with toughscreen is ahead in the queue |
05:39:36 | andersee | Is that stuff documented? |
05:39:41 | Russ | not per se |
05:39:57 | Russ | tim passed me some stuff that I'm not sure I should pass along or not |
05:40:18 | Russ | but he should have schematics when he comes back from vacation, and that will be a big help |
05:44:22 | andersee | Well, I need to take off and get some sleep |
05:44:49 | andersee | I've got a bunch of consulting work I need to be awake enough to do tomorrow, er, today. |
05:45:04 | Russ | ew....consulting? |
05:45:55 | andersee | Yeah. The work is kindof lame, but the pay is great (which is how I got roped into doing it) |
05:46:56 | andersee | My wife wants to buy a minivan or an SUV or something to haul about the kids... So I get to do some consulting... |
05:47:18 | Russ | heh, navigator or something? |
05:48:15 | andersee | I dunno. Just something that is going to be more then I want to spend I'm sure. So I figure it is best to earn it first and pay cash. |
05:49:43 | andersee | g'night -- thanks for your work on the phone. Its looking very nice... |
05:49:49 | Russ | alright, cya |