irclog2html for blob on 2002.03.02

17:33:09prpplaguefinalfangtasydog: i don't believe we've met, you work for an embedded arm company or just a back room hack?
22:16:38prpplaguehowdy there erikm
22:16:56erikmhell prpplague
22:17:00erikmhello, I mean :)
22:17:04prpplaguewe were worried you broke your leg or something since you didn't show up yesterday
22:17:17prpplagueerikm: the first one was probably more accurate!lol
22:18:22erikmprpplague: no, I'm just taking quite some days off. I have about fifty holidays days left before I finish my current job, and I'm gonna use them liberally :)
22:18:52prpplagueerikm: lol, go for it
22:19:29prpplaguehas spent all day digging through sa1100fb.c
22:20:04prpplaguelots of duplicated effort there
22:21:00erikmrmk already cleaned it up
22:21:04erikmit was much worse
22:21:47erikmdo you volunteer to clean it even further?
22:22:15prpplagueerikm: i've never submited a patch to rmk before, he'd probably tear me to shreds!
22:22:35prpplagueerikm: but i'm going to at least try and clean the stuff i'm working on up
22:22:51erikmyou'd better discuss it on the linux-arm-kernel list first
22:23:20prpplagueerikm: ok i will
22:24:44prpplagueerikm: basically the lcd power on the board i'm working on is controled through the SMBus protocol, and instead of using the existing function calls provided in I2C, they apparently duplicated the code just for the lcd power up function
22:25:07erikmLCD power or backlight?
22:25:13prpplagueerikm: lcd power
22:26:19erikmso it's more a cleanup of the machine specific code, right?
22:27:01prpplagueerikm: ya nothing affect thing overall sa1100fb code just machine specific for three or four boards
22:29:00prpplaguethis lcd stuff is pretty interesting, its going to take me a couple weeks to get the hang of adding a new display from scratch
22:30:23prpplagueerikm: did i ask you if your were going to attend the UKUUG conference in bristol?
22:30:23erikmno, that usually takes you an hour or so. the hardest part is to read the LCD data sheets, which are usually written in japanese or korean english
22:30:33erikmyou asked, yes
22:30:41erikmand I said I'm not sure yet
22:30:43prpplagueok didn't remember if i did
22:31:00erikmbut you're in the US, right?
22:31:13prpplagueerikm: i've got 6 lcd panels and 3 arm board, my goal is to get them all working together
22:31:36prpplagueerikm: ya i'm in the US, but my company pays 1/2 for any conference once a year
22:32:22erikmprpplague, meet rmk
22:32:43prpplaguermk: pleasure to meet you
22:32:56rmkand you 8)
22:33:59erikmprpplague: adding an LCD is not difficult. it currently looks like the LCDs are board specific, but that's only partly true
22:34:31erikmprpplague: you could basically share a single LCD structure with many boards
22:35:17prpplagueerikm: yes from what i can tell the major of boards a very standardized, however i purchase some "off brand" lcd
22:35:31prpplagueto learn about problems that you can encounter
22:36:03prpplaguei find i learn more when things don't go well.....
22:37:17rmkwe all do 8)
22:37:41rmkI learn most when solving problems, not when things are moving along smoothly.
22:38:45prpplaguermk: we were just discussing the omnimeter lcd code
22:39:42prpplaguermk: i'm hoping to complete a cleanup of all the omnimeter/elf code soon
22:41:01rmkit doesn't have any pcmcia interfaces, does it?
22:41:28prpplaguermk: it uses a cirrus logic dual pcmcia interface
22:41:54prpplaguermk: inhand decided not to use the sa pcmcia interface
22:42:13prpplaguermk: which makes the pcmcia for the omnimeter/elf nasty
22:42:32rmkumm, what happened to Sally Glass (who registered the machine type)?
22:43:25prpplaguermk: sally and simon glass are with a company in NZ and have completed a port, but have not release the code, they are waiting on someone to fork up the $$
22:44:04rmkas long as they're not distributing binaries...
22:44:27prpplaguermk: correct, to my knowledge they are not distributing anything
22:45:00rmkwhich kernel are you working against?
22:46:21rmkthat's fine.  I'd also like to see 2.5 patches as well so it doesn't get left behind.
22:48:21prpplaguermk: i'll work towards 2.5 as well, my main goal is to remove as many of tentacles extending from the omnimeter fixup code and try to use existing functions and structures
22:50:26rmkok.  if there are too many "tentacles" appearing, I might be tempted to pull some more of the 2.5 changes back into 2.4 (eg, allowing the lcd power and backlight power code to live in the per-machine files)
22:52:37prpplaguermk: i think i'll be able to fix most of the power and backlight functions on the omnimeter/elf could by using the I2C-SMBus functions
22:53:00prpplaguermk: this should keep the need for addition headers and speciality functions
22:53:26prpplaguermk: down
22:55:30prpplaguethe 2.4.17 sa1100fb code is definetly at lot more orginized and readable compared to the 2.2 series
22:56:52prpplagueerikm: if you do decide to attend the UKUUG, let me know, i'll buy a round beer
22:57:03erikmhehe :)
22:57:57rmkbtw, I've been invited to some random event in Scotland in October...
22:58:39prpplaguermk: anything interesting?
22:59:34rmkits in the early stages of being planned atm.
22:59:51rmkthey've managed to get Stephen Tweedie and myself to think about doing a talk there.
23:00:31prpplaguemaybe i should hold off on my trip over to UKUUG
23:01:41prpplaguermk: the company i work for pays 1/2 the costs of one conference a year, i'm trying to find a nice one in the UK
23:02:24erikmjust submitted a paper proposal to OLS
23:03:17prpplagueya i get the feeling that there will be a nice turnout at OLS this year
23:03:51rmkI'm not sure if the october one is even going ahead its that early.
23:04:44prpplaguermk: ohh well, maybe i can talk them into two
23:09:58rmkadds a link off the developer page to the 2.5 kernel BK changeset info
23:13:32prpplaguermk: you getting used to using BK ?
23:16:21erikmprpplague: bk is very nice for kernel work
23:18:44rmkyes, but currently only using it as a means to get stuff into Linus, and for some of my own kernel work.
23:19:32rmkeventually I do see the whole arm kernel going in under BK (and made available as BK, patches and CVS)
23:23:20prpplagueerikm / rmk : don't suppose anyone from sharp has sent you guys a dev board with the new sharp arm core?
23:29:11prpplaguei called about one, but they are being very selective on who gets dev boards
23:30:35prpplaguei think BZFlag mentioned that he'd seen one pass through his offices
23:31:10erikmin that case I'm sure he'll port blob to it :)
23:31:43rmkoh a suggestion to do before sending anything to the patch system or machine database.
23:32:12rmk1. go to my web site, and find the patch system.  go to the add a new patch link.
23:32:19prpplaguermk: read the notes and faq?
23:32:24rmkthat as well, but...
23:32:37rmk2. go to the 'add a new patch' or whatever I've called it.
23:32:53rmkit should ask you for an email address and a password, or ask you to register.
23:33:12rmkif you register, you'll automatically be sent a password via email.
23:33:35rmkwhen you do register, give the name and the mail address as it appears in your email system.
23:33:56rmk(ie, if you're mail client calls you 'Bloggs, Joe' enter it as that)
23:34:31prpplaguermk: got it
23:34:32rmkif you send something via mail to either _before_ registering, you need to bother me to get a password.
23:35:07rmk(since it automatically creates an account with the web access disabled, and I haven't implemented the "forgotten password" button still)
23:35:56prpplaguesaves window buffer to rmk-warnings.txt
23:36:28erikmprpplague: warning: rmk's passwords are usually generated by something like this: dd if=/dev/random bs=6 count=1 | uuencode - | head -2 | tail -1
23:36:33erikmruns for rmk
23:36:45rmkcurrently my "change the password" is a case of typing the right mysql stuff.
23:37:07rmkerikm: its actually a little more crude.  its "hit random keys on the keyboard and see what we come out with"
23:37:25erikmrmk: do you have monkeys for that?
23:37:48rmkerikm: if I did, I'd rig them up to the "forgotten password" button.
23:37:56erikmrmk: hehe.
23:38:17erikmrmk: and what would happen if they actually came up with a shakespeare sonnet? ;)
23:38:46rmkthen it'd hit the news 8)
23:39:23erikmthe sharp board looks nice
23:39:57erikmand the board really looks like an assabet+neponset
23:41:52prpplagueerikm: from talk with the techs at sharp, its sounds like interfacing is a snap with the processor
23:42:38erikmlooks like a fun board to port linux to
23:43:19erikmtakes a shower
23:43:32rmkthis pcmcia chip isn't a CLPS6700 is it?
23:46:10prpplaguermk: on the sharp board?
23:46:44rmkyep (I've not looked at that site in great depth yet)
23:48:08prpplaguermk: i'm not seeing any info on the specfic pcmcia controler, let me check the tech ref
23:51:31prpplaguermk: pcmcia appears to be in the processor package and there are no indications of type, however the registers appear to be similar to the SA series
23:54:21rmkhmm, ok.
23:55:13prpplaguermk: their docs are very sketchy right now
23:55:51prpplaguermk: our sharp rep is suppose to be letting me know next week about getting an eval board
23:57:14prpplaguermk: well it was a pleaseure to finaly (virtualy) met you, i'm headed home to study lcd interfaces and drink beer
23:58:54prpplaguermk: i'm sure you hear it often, but thanks for the hard work in getting and keeping arm-linux going

Generated by by Jeff Waugh - find it at! Modified by Tim Riker to work with infobot logs, split per channel and by date, etc.