| 00:00.17 | Dr{Who} | like 5$ each yuk.. |
| 00:00.54 | Dr{Who} | i had fun tuning my c code to get the results back from the device using the extended temperature calculations.. the docs were not very clear on how to do that. |
| 00:01.16 | Dr{Who} | and if you dont do that you get very innacurate readings.. |
| 00:04.16 | RobotGuy | prpplague: I found one of my I2C devices will run off 3.3V - Honeywell compass. I'll play with that one first. The others are all 5V. :( |
| 00:05.09 | RobotGuy | prpplague: Somehow, I am going to have to deal with level translations at some point. |
| 00:07.10 | ce_geek | Dr{Who}: Thanks for the link, much appreciated. |
| 00:07.23 | RobotGuy | I got a call to report for Jury Duty. :( |
| 00:07.33 | *** join/#edev prpplague (n=dave123_@ppp-70-244-174-210.dsl.rcsntx.swbell.net) |
| 00:08.01 | Dr{Who} | np. |
| 00:08.04 | prpplague | got to get a new router and reformat this PoS laptop |
| 00:08.26 | RobotGuy | prpplague: You have been saying that for awhile. :) |
| 00:08.41 | Dr{Who} | after you do a DS1WReadBuf (buf, 7); |
| 00:09.05 | Dr{Who} | here is my math.. note its floting point emulation as i dont have float on a mega128 |
| 00:09.10 | Dr{Who} | dev->temp = ((16 - buf[6])*10) + ((buf[0] >> 1)*100); |
| 00:09.30 | Dr{Who} | iTempTemp=dev->temp/100; |
| 00:09.45 | Dr{Who} | sprintf(szTempString,"%i.%i",iTempTemp,(int)(dev->temp - (iTempTemp*100))); |
| 00:09.53 | prpplague | RobotGuy indeed |
| 00:10.02 | prpplague | RobotGuy low priority |
| 00:10.06 | adyer | RobotGuy: TI PCA9306 looks pretty simple to use for i2c level translation |
| 00:10.08 | Dr{Who} | you can stitch the rest together teh basic idaa is you get the main value the decimal or precision is the tricky part |
| 00:10.18 | Dr{Who} | thats the second calc.. |
| 00:11.32 | RobotGuy | What is low priority? |
| 00:11.32 | prpplague | i really need to edit a bunch of the wiki pages |
| 00:11.38 | prpplague | RobotGuy router |
| 00:11.58 | RobotGuy | prpplague: Oh. I would have thought it would have higher priority now. |
| 00:12.23 | RobotGuy | prpplague: We are not doing well enough on the Wiki?? |
| 00:12.36 | prpplague | checks ebay for cheap kids laptops |
| 00:13.20 | *** join/#edev Appleman1234 (n=Appleman@CPE-124-186-229-83.qld.bigpond.net.au) |
| 00:13.24 | prpplague | RobotGuy there are some pages that are kid of mixed up |
| 00:13.29 | prpplague | kind |
| 00:14.11 | RobotGuy | prpplague: I'm more interested in how -I- am doing and if I need to do better or not. |
| 00:14.36 | prpplague | RobotGuy based on where you started, doing pretty darn good |
| 00:14.44 | RobotGuy | adyer: I need something in a DIP. |
| 00:14.52 | RobotGuy | prpplague: Thankyou. |
| 00:15.19 | prpplague | http://cgi.ebay.com/Children-Kid-Laptop-Computer-Educational-Learning-Toy-B_W0QQitemZ270225041564QQihZ017QQcategoryZ11734QQssPageNameZWDVWQQrdZ1QQcmdZViewItem |
| 00:15.26 | prpplague | hmm that ones looks good hammering |
| 00:15.30 | RobotGuy | I am guessing I would want a level translater per 5V device, correct? |
| 00:16.07 | adyer | RobotGuy: you should only need one adapter for the whole 5V bus segment |
| 00:16.09 | prpplague | another chinesse knock off - http://cgi.ebay.com/LAPTOP-COMPUTER-LEARNING-SYSTEM-NIB_W0QQitemZ250234537283QQihZ015QQcategoryZ145939QQssPageNameZWDVWQQrdZ1QQcmdZViewItem |
| 00:17.28 | prpplague | hmm |
| 00:17.35 | prpplague | this one is cute - http://cgi.ebay.com/Comquest-Plus-Child-Laptop-Learning-Activity-Computer_W0QQitemZ280215771303QQihZ018QQcategoryZ145932QQssPageNameZWDVWQQrdZ1QQcmdZViewItem |
| 00:17.51 | RobotGuy | adyer: Ah, OK, that's good! :) |
| 00:18.04 | ce_geek | Dr{Who}: you're right, that code does look great. I'll give it a try when I get to the hammer stuff. Thanks! |
| 00:18.25 | adyer | RobotGuy: but good luck on the dip package... |
| 00:18.30 | RobotGuy | prpplague: Hmmmm, I wonder if I could put Linux on that. |
| 00:19.09 | prpplague | RobotGuy not without the hammer |
| 00:19.11 | ce_geek | RobotGuy: You need a level shifter thing in a DIP? |
| 00:19.37 | RobotGuy | ce_geek: Yes, or a module I can use on a solderless breadboard or solder on a proto board. |
| 00:19.50 | ce_geek | RobotGuy: What kind of levels do you need to shift? RS232 to 5V? |
| 00:19.59 | RobotGuy | ce_geek: I2C 3.3V to 5V level shifter. |
| 00:20.03 | prpplague | http://www.toysrus.com/product/index.jsp?productId=2449554 |
| 00:20.11 | prpplague | that one could have linux |
| 00:20.25 | prpplague | but its got CoB's and it would be a major effect to get it on it |
| 00:20.28 | ce_geek | RobotGuy: I've used the MAX232 chip for RS232. Not sure about 3.3V I2C |
| 00:20.47 | ce_geek | RobotGuy: Are you familiar with that device? Might be worth looking at the datasheet if not, maybe it can work. |
| 00:21.00 | adyer | ce_geek: i2c needs bidirectional level shifting. parts exist, it's just that he wants DIP package |
| 00:21.14 | ce_geek | adyer: Ah. |
| 00:21.20 | RobotGuy | ce_geek: I have quite a few MAX232's here. :) |
| 00:21.42 | prpplague | ho ho hum |
| 00:21.55 | prpplague | ce_geek you get your paper turned in? |
| 00:21.58 | RobotGuy | searches Spark Fun |
| 00:22.08 | ce_geek | RobotGuy: Ah okay :). I used one years ago, couldn't remember if it could do other things too. |
| 00:22.14 | ce_geek | prpplague: Yep! All gone. |
| 00:22.18 | prpplague | ce_geek lovel |
| 00:22.21 | prpplague | lovely |
| 00:22.31 | ce_geek | prpplague: Indeed. Defense 1 week from Thursday. |
| 00:22.49 | ce_geek | prpplague: Later this week and next week I'll prepare a presentation. |
| 00:23.06 | prpplague | ce_geek hammer time eh? |
| 00:23.07 | ce_geek | prpplague: Not quite as stressful as writing the thesis itself. The presentation material will all come from the thesis. |
| 00:23.12 | ce_geek | prpplague: yup! |
| 00:23.25 | ce_geek | prpplague: Getting re-acquainted this evening. |
| 00:23.35 | ce_geek | prpplague: gotta get boa to play nice |
| 00:23.42 | ce_geek | prpplague: Then start looking into hardware programming. |
| 00:23.46 | RobotGuy | Sienna sent me an SRF-08, SP03 (Speech Synth), and a Devantec Compass to play with. All of them are 5V I2c, unfortunately. |
| 00:24.05 | prpplague | RobotGuy this is the one i really one to hack - http://www.toysrus.com/product/index.jsp?productId=2647159 |
| 00:24.22 | ce_geek | HAHA |
| 00:24.57 | adyer | RobotGuy: try this one http://www.standardics.nxp.com/products/p82b/datasheet/p82b96.pdf I have no idea if you can find it, but it will do 3.3 to 5 i2c in DIP |
| 00:25.17 | RobotGuy | Could I maybe use a voltage divider to match the 5V to the 3.3V I2C bus?? |
| 00:25.53 | RobotGuy | prpplague: You just want it because it is Star Wars. :) It would be cool if it could be used to console to Hammer. |
| 00:26.32 | adyer | RobotGuy: you'd have to go through the combinations of parts and do the calculations |
| 00:26.33 | RobotGuy | adyer: I can't open that file. |
| 00:27.23 | adyer | RobotGuy: look on the nxp web site for p82b96 |
| 00:27.51 | adyer | RobotGuy: or maybe the url had a line break in it |
| 00:27.57 | *** join/#edev prpplague (n=dave123_@ppp-70-244-174-210.dsl.rcsntx.swbell.net) |
| 00:28.32 | RobotGuy | I wish NXP gave out samples, but you have to have an account with them. :( |
| 00:29.08 | prpplague | hmm looks like its the neighbors wireless |
| 00:29.24 | ce_geek | prpplague: Try different channels n' stuff? |
| 00:29.33 | prpplague | looks like they have one of those pre-approved 802.11n ap |
| 00:29.39 | adyer | RobotGuy: in stock at digikey |
| 00:31.59 | MonMotha | these days, I've found it's often easier to just buy the darn chips from digi-key than attempt to sample them from the manufacturer |
| 00:32.05 | MonMotha | unless you happen to know a sales rep |
| 00:32.08 | adyer | RobotGuy: lol SMT PCA9306, $0.75, DIP P82B96 $3.08, qty. 1 |
| 00:32.21 | prpplague | ce_geek yea, looking at my wispy output |
| 00:32.30 | RobotGuy | adyer: Cool! I used to order from Digi-Key all the time. |
| 00:32.31 | MonMotha | for the cheap stuff, it's not worth messing with, especially since many of them are charging s&h now |
| 00:32.39 | prpplague | ce_geek looks like lots of 2.4ghz cordless phones too |
| 00:32.41 | RobotGuy | adyer: :P I' |
| 00:32.44 | MonMotha | for the big stuff, they are hesitant to sample without some sort of intended commercial application |
| 00:32.49 | MonMotha | prpplague: my solution is 802.11a |
| 00:32.53 | RobotGuy | adyer: I'll pay for the DIP. |
| 00:32.59 | ce_geek | prpplague: What are you measuring this stuff with? |
| 00:33.30 | RobotGuy | MonMotha: Any company could do that. Look how free Microchip is with samples. Maxim and ADI are pretty good also. |
| 00:36.29 | MonMotha | RobotGuy: the issue is getting stuff for personal projects. They (understandably) are hesitant to give out samples of expensive stuff when there's absolutely zero chance they'll get any significant sales off of it |
| 00:36.53 | MonMotha | it's different if you're asking for them as a company. There's a chance that you may come back a year later and want a million of htem |
| 00:37.33 | *** join/#edev prpplague^2 (n=dave123_@ppp-70-242-113-58.dsl.rcsntx.swbell.net) |
| 00:37.39 | prpplague^2 | oh tricky tricky |
| 00:37.54 | Dr{Who} | robot.. that reminds me i need some max232's baddddly.. |
| 00:38.56 | Dr{Who} | i need to level shift some 0-12v serial to values more appropriate to the hammer. |
| 00:40.08 | MonMotha | a couple resistors will do for that |
| 00:40.23 | MonMotha | chuck a 3.3V zener (or 5V, whatever's appropriate) on there if you're paranoid |
| 00:40.24 | adyer | Dr{Who}: usually it's -12/+12 for serial |
| 00:40.29 | Dr{Who} | yes it is |
| 00:40.34 | Dr{Who} | but this is not : c ) |
| 00:41.13 | ce_geek | Anyone use optical isolation for the hammer much? |
| 00:41.43 | *** join/#edev prpplague^2 (n=dave123_@ppp-70-242-113-58.dsl.rcsntx.swbell.net) |
| 00:43.07 | prpplague^2 | jeeze |
| 00:43.28 | Dr{Who} | glues prpplague down |
| 00:43.35 | prpplague^2 | wheres landley when you need him |
| 00:43.40 | prpplague^2 | Dr{Who} thanks |
| 00:43.43 | prpplague^2 | http://shantous.en.china.cn/op/SellInfoDetail/selinfid/1066684335/index.htm |
| 00:43.51 | prpplague^2 | just got a reply for those guys |
| 00:44.12 | prpplague^2 | we can get that partially populated for just $6.68USD |
| 00:44.30 | RobotGuy | MonMotha: I understand that, but some companies are much easier to get stuff from than others. The three I mentioned are my favorites, although I have gotten stuff from FreeScale, ST Micro, and others also. |
| 00:45.03 | prpplague^2 | http://shantous.en.china.cn/op/SellInfoDetail/selinfid/1066684338/index.htm |
| 00:45.07 | prpplague^2 | same one but in a desktop |
| 00:48.26 | RobotGuy | prplague: Those links don't load for me |
| 00:48.38 | prplague | yep, found it |
| 00:48.42 | prplague | perfect one |
| 00:48.45 | prplague | just for ds2 |
| 00:48.48 | prplague | http://chenghaitoys.en.alibaba.com/product/200103767/50062347/English_Learning_Laptop_Toy/Computer_Learning_Toy.html |
| 00:49.27 | RobotGuy | I just want something I can console to Hammer with and carry with me easily. |
| 00:49.37 | adyer | later all |
| 00:49.40 | *** part/#edev adyer (n=adyer@adsl-69-223-13-78.dsl.chcgil.ameritech.net) |
| 00:49.44 | prplague | RobotGuy what about this one - http://chenghaitoys.en.alibaba.com/product/200103767/52073246/English_Learning_Laptop_Toy/Children_Laptop.html |
| 00:49.48 | prplague | RobotGuy does that load? |
| 00:50.22 | prplague | RobotGuy probably can find a zipit v1 pretty cheap |
| 00:50.25 | RobotGuy | prplague: That last one did. |
| 00:51.00 | prplague | we can get that partially populated for just $6.68USD |
| 00:53.13 | prplague | oh oh i like this one - http://chenghaitoys.en.alibaba.com/product/200103772/50786728/Spanish_And_English_Learning_Laptop_Toy/Bilingual_Language_Toy_Laptop/showimg.html |
| 00:54.27 | prplague | RobotGuy found you one |
| 00:56.16 | prpplague | ho ho hum |
| 00:56.38 | RobotGuy | prpplague: Oh? |
| 00:57.08 | prpplague | jokes with RobotGuy |
| 00:57.12 | prpplague | RobotGuy http://chenghaitoys.en.alibaba.com/product/200103767/52072963/English_Learning_Laptop_Toy/English_Laptop_Learner.html |
| 00:57.40 | prpplague | RobotGuy you'll be the envy of all the robotics geek girls |
| 00:58.01 | RobotGuy | prpplague: How am I gonna even get one of those other ones (not the pink one) though? |
| 00:58.27 | prpplague | RobotGuy or would you prefer the darth vader one? |
| 00:58.32 | RobotGuy | prpplague: Looks like it would be easier to just attach a display to hammer directly with some sort of keyboard. |
| 00:59.18 | prpplague | RobotGuy yea most of the others can be had on ebay in onies |
| 00:59.26 | prpplague | RobotGuy the zipit is pretty cheap |
| 00:59.29 | RobotGuy | prpplague: Darth would be OK. I just need something simple with a keyboard to type on and a display I can console to hammer with. Doesn't have to be terribly fancy. Pink and red need not apply. :) |
| 00:59.32 | *** join/#edev Shingoshi (n=shingosh@c-67-169-195-234.hsd1.or.comcast.net) |
| 01:00.44 | prpplague | seems like someone had the zipit v1 for $49.99 |
| 01:00.46 | RobotGuy | Heck, with some of the newer phones and such, I could almost do it that way - especially when I get bluetooth working on the console port. |
| 01:02.17 | prpplague | shoot i was i could remember who had those |
| 01:02.57 | RobotGuy | prpplague: Zipit Z2 looks very cool. |
| 01:03.18 | prpplague | RobotGuy pricy |
| 01:03.42 | prpplague | hehe you can ge a pixter-mm for like $15 |
| 01:04.19 | RobotGuy | All I need is a standard serial port, or even a usb port. I have been figuring I would spend somewhere in the neighborhood of around $200.00 to get something decent. Funny thing is for $499.00 at Best Buy I could get a real lap top. |
| 01:04.41 | prpplague | RobotGuy $299 you could get an eeepc |
| 01:05.31 | RobotGuy | prpplague: How would I get one though? |
| 01:05.39 | prpplague | hmm looks like the $299 ones are gone - http://www.geeks.com/details.asp?invtid=90OA02AF0004111U105Q-DT&cat=NBB |
| 01:06.05 | prpplague | order it |
| 01:06.58 | prpplague | RobotGuy i've got one of these - http://www.halted.com/commerce/ccp20981-fujitsu-pencentra-130-pen-tablet-computer-pencentra-130-80797.htm |
| 01:07.08 | prpplague | RobotGuy they are pretty handy, just not very powerful |
| 01:08.04 | Dr{Who} | robot.. http://www.nutech.com/hardorder2.asp?ID=2 |
| 01:08.09 | prpplague | basically a tablet version of the vr3 |
| 01:08.22 | Dr{Who} | but no display.. |
| 01:08.24 | RobotGuy | prpplague: I don't need real powerful though. Of course, I would want the Eee PC 8G. :) |
| 01:09.18 | prpplague | anyway this is one i'm currently using with a hammer - http://www.toysrus.com/product/index.jsp?productId=2340839 |
| 01:10.00 | Dr{Who} | neet : c ) |
| 01:10.56 | RobotGuy | prpplague: I like http://www.toysrus.com/product/index.jsp?productId=2373385 |
| 01:14.18 | RobotGuy | prpplague: I really like the Eee PC 8G, but need to find out how much it is. |
| 01:15.39 | Dr{Who} | so does that toy actualy have an os? and a serial port? |
| 01:21.53 | *** join/#edev prpplague (n=dave123_@ppp-70-242-113-58.dsl.rcsntx.swbell.net) |
| 01:22.34 | *** join/#edev prpplague (n=dave123_@ppp-70-242-113-58.dsl.rcsntx.swbell.net) |
| 01:40.09 | *** join/#edev CP|home (n=nobody@www.crouse-house.com) |
| 01:40.35 | CP|home | damn - where is the plague? |
| 01:41.56 | GPSFan | CP|home: he's plagued by router problems. ;>) |
| 01:44.46 | *** join/#edev agaffney_ (n=agaffney@gentoo/developer/pdpc.active.agaffney) |
| 02:11.16 | RobotGuy | This is crazy. It should not stop before booting the kernel. :( |
| 02:20.05 | *** join/#edev Appleman1234 (n=Appleman@CPE-124-186-229-83.qld.bigpond.net.au) |
| 02:31.42 | *** join/#edev Appleman1234 (n=Appleman@CPE-124-186-229-83.qld.bigpond.net.au) |
| 02:49.11 | RobotGuy | This kernel configuration stuff seems way too finicky. |
| 02:50.22 | CP|home | huh - thats not something you hear everyday |
| 02:58.28 | Mirell | Heh |
| 02:59.09 | RobotGuy | What, that the kernel configuration is too finicky? |
| 02:59.43 | RobotGuy | Why is the IR and Bluetooth stuff in the Networking section? |
| 02:59.53 | CP|home | yeah - usually the kernel configuration is looked upon as pretty solid |
| 02:59.57 | CP|home | heck, we steal it for other project |
| 02:59.57 | CP|home | s |
| 03:00.00 | RobotGuy | Shouldn't they be in the Device Drivers section? |
| 03:01.47 | RobotGuy | It's totally not intuitive. |
| 03:02.35 | RobotGuy | C'Mon now, some stuff is allowed to be set for modules but won't work that way and vice versa. |
| 03:03.03 | RobotGuy | Certain combinations of stuff have to be built a certain way to work - as modules or in kernel. |
| 03:03.28 | Mirell | RobotGuy: For instance, why is video stuff in character devices. |
| 03:03.31 | Mirell | Legacy crap. |
| 03:03.39 | RobotGuy | Mirell: Exactly my point. |
| 03:04.05 | RobotGuy | Well, actually video stuff could be considered character devices, but what else would they be? |
| 03:04.30 | Mirell | It's just not intuitive from a newcomer's point of view, even a techie. |
| 03:04.38 | RobotGuy | I'd put Video stuff with frame buffer and such. Seems like they could be connected. |
| 03:04.48 | CP|home | But they aren't |
| 03:04.50 | CP|home | not even close |
| 03:05.20 | RobotGuy | Well, I don't think Video stuff belongs in character devices either. |
| 03:05.42 | Mirell | *shrug* |
| 03:05.45 | RobotGuy | Just like IR and Bluetooth do NOT belong in Networking. |
| 03:06.09 | RobotGuy | They should be in Device Drivers like most other stuff. |
| 03:07.19 | RobotGuy | Sometimes I have to go all over the place to get something setup right - networking related stuff is in one place, character device in another, etc, and they may be connected. |
| 03:08.43 | RobotGuy | I'd rather see the Device Type (i.e. Bluetooth, IR, I2C, etc) be the major catagory instead of Networking, Character Devices, etc. |
| 03:08.49 | RobotGuy | sighs |
| 03:09.02 | CP|home | Submit a patch then |
| 03:11.52 | Mirell | I'm sure it's been brought up on LKML, but I'm too tired to search. |
| 03:13.10 | RobotGuy | The likely hood a patch like that would be accepted is pretty slim - this has been this way for a long time. |
| 03:13.50 | RobotGuy | I'm not even sure I want to get into writing kernel modules for my I2C devices. |
| 03:13.59 | CP|home | we keep telling you that you don't have to.. :) |
| 03:14.24 | RobotGuy | I don't have the patience to discuss that right now. |
| 03:14.52 | RobotGuy | I was so frustrated earlier, I started having chest pains. |
| 03:15.12 | Mirell | >.< |
| 03:15.46 | RobotGuy | CP|home: Besides, it's not a matter of whether I have to or not. |
| 03:16.51 | CP|home | It depends on how frustrated you want to be |
| 03:18.42 | RobotGuy | Is the API really that bad? |
| 03:23.39 | RobotGuy | Maybe just accessing the I2C devices bare-metal from Python is really the best way to go. |
| 03:24.10 | RobotGuy | I'm not sure I can do everything that way though. |
| 03:27.42 | RobotGuy | Where does ARM Cortex M3 fit in the ARM hierarchy? |
| 03:52.44 | *** part/#edev RobotGuy (n=robotguy@dsl093-038-072.pdx1.dsl.speakeasy.net) |
| 03:58.09 | *** join/#edev DrZimmerman (n=theo@tbaumgartner.broker.freenet6.net) |
| 04:00.44 | *** join/#edev linac (n=lin@59.61.124.100) |
| 05:07.12 | *** join/#edev RobotGuy (n=robotguy@dsl093-038-072.pdx1.dsl.speakeasy.net) |
| 05:09.13 | Dr{Who} | rehi. |
| 05:09.14 | *** join/#edev steliosk (n=Stelios@85.72.79.254) |
| 05:09.20 | Dr{Who} | robot i have a link that may help? |
| 05:09.57 | Dr{Who} | looks like a some simple code for i2c based upon the via epia board with its own onboard i2c chip you can connect to. |
| 05:10.21 | Dr{Who} | so i presume if the hammer driver is simmilar your talking to the i2c "api" or driver would be the same.. |
| 05:13.54 | Dr{Who} | robot? |
| 05:14.19 | RobotGuy | Yes? |
| 05:14.29 | RobotGuy | Oh, cool |
| 05:14.32 | Dr{Who} | take a look at this |
| 05:14.33 | Dr{Who} | http://codesink.org/eeprog.html |
| 05:14.50 | RobotGuy | Thanks! I will take a look at that. |
| 05:15.04 | Dr{Who} | it looks very simple.. and if it works. it will let you program an eprom and verify you can talk i2c |
| 05:15.27 | Dr{Who} | i have been trying to find other samples for you but not much luck. |
| 05:15.28 | RobotGuy | Yes, indeed, and I do have some I2C EEPROMs here. |
| 05:15.37 | Dr{Who} | i figured you did. |
| 05:15.40 | RobotGuy | I appreciate the help. |
| 05:16.27 | Dr{Who} | well i figure i will be needing to do this soon too so its all part of the project : c ) |
| 05:16.35 | RobotGuy | With so many people trying to help me at the same time, I can get frustrated very easily trying to sort it all out. I was having bad chest pains earlier today, and had to exit for awhile. |
| 05:16.57 | RobotGuy | Yes, indeed, and you know everything I learn will go on the Wiki once I have success. |
| 05:17.12 | Dr{Who} | ya i hear ya.. i get that way too sometimes.. get to hyped up on a big problem i need to get over |
| 05:17.19 | Dr{Who} | sometimes its just good to back off for a day.. |
| 05:18.00 | Dr{Who} | i just spend the last 4 hours trying to figure out why my kernel would not switch from single processor to SMP |
| 05:18.01 | RobotGuy | I know. It's hard for me to let go of a problem sometimes though. I still don't get why my Hammer can't get past uncompressing the kernel. No idea how to fix it. |
| 05:18.12 | Dr{Who} | something like that is SIMPLE.. but it didnt work.. |
| 05:18.27 | RobotGuy | It's the simple things that confound me a lot of times. |
| 05:18.34 | Dr{Who} | some bug in 2.4.32 kernel if you build with NO smp then turn it on it wont build right |
| 05:18.44 | RobotGuy | Hmmm, weird. |
| 05:18.50 | Dr{Who} | so i had to make mrproper then re import my config and it was fine.. |
| 05:19.05 | Dr{Who} | but i spent 4 hours figuring it was just something stupid i did. |
| 05:19.41 | RobotGuy | Now I have to find my Honeywell compass module, because it's the only 3.3V I2C device I have. Everything else is 5V. I need to get some of those NXP I2C level shifter chips, but that can't happen until next month. |
| 05:20.04 | Dr{Who} | ya i think it left a dangeling include or define that would not be reset when you change the kernel config option.. |
| 05:20.06 | RobotGuy | I'm already in the hole this month. :( |
| 05:20.34 | RobotGuy | I hate when it turns out to be something so simple it is so easily missed. |
| 05:20.51 | Dr{Who} | yep.. frustrating.. |
| 05:21.52 | RobotGuy | That happens to me a lot unfortunately. |
| 05:22.09 | Dr{Who} | im guessing from this that if the i2c drivers are working correctly on the hammer you will see /dev/i2c-xx |
| 05:22.18 | Dr{Who} | and you can connect directly to that device file. |
| 05:22.24 | RobotGuy | I am always having to re-find things because I can forget what happened or what I just did right afterwards. :( |
| 05:22.42 | Dr{Who} | wiki is our friend... : c ) |
| 05:22.49 | RobotGuy | Yes, I have the I2C device files already added to my device_table.txt |
| 05:22.52 | Dr{Who} | i update my own daily.. |
| 05:23.08 | Dr{Who} | or i would not know what i had for breakfast..... |
| 05:23.15 | RobotGuy | Dr{Who}: I am starting my own Wiki also at www.thedynaplex.info and forums and a Blog. |
| 05:23.23 | Dr{Who} | cool. |
| 05:23.54 | Dr{Who} | i have one for work i started a few years back.. and i make all my employees wiki all the time.. |
| 05:24.15 | RobotGuy | Dr{Who}: It's a good thing! I am setting mine up using TikiWiki. |
| 05:25.21 | Dr{Who} | cool.. i have been staying with metawiki. |
| 05:25.38 | Dr{Who} | hehe wowl |
| 05:25.40 | Dr{Who} | There are 1,845 total pages in the database. This includes "talk" pages, pages about NutWiki, minimal "stub" pages, redirects, and others that probably don't qualify as content pages. Excluding those, there are 1,152 pages that are probably legitimate content pages. |
| 05:25.40 | Dr{Who} | 188 files have been uploaded. |
| 05:25.40 | Dr{Who} | There have been a total of 37,005 page views, and 12,644 page edits since NutWiki was setup. That comes to 6.85 average edits per page, and 2.93 views per edit. |
| 05:28.22 | Dr{Who} | my guys are nutz.. the main page has this.. |
| 05:28.25 | Dr{Who} | "I was gonna let him do it until he started rubbin em on me" -- Kevin about Nick's balls |
| 05:29.28 | Dr{Who} | referring to an office toy.. a sock with sand in it that well looks like http://www.bumpernuts.com/ |
| 05:32.16 | Dr{Who} | and they put this pic on the main page too.. kids.. http://www.dribbleglass.com/misinspired/quality.htm |
| 05:37.29 | Dr{Who} | ok well i have to head down to my colo facility and replace a server i have been fuxing with all day. |
| 05:59.30 | *** join/#edev DrZimmerman (n=theo@tbaumgartner.broker.freenet6.net) |
| 05:59.34 | *** join/#edev drath_ (i=vmaster@p5B07D932.dip.t-dialin.net) |
| 06:03.09 | RobotGuy | Dr{Afk}: That is some Wiki! |
| 06:16.38 | *** join/#edev harpal (n=harpal@124.125.79.150) |
| 06:28.06 | *** join/#edev harpal_ (n=harpal@124.125.79.150) |
| 06:33.05 | *** join/#edev Appleman1234 (n=Appleman@131.181.47.71) |
| 06:33.14 | *** join/#edev KazaLite (n=kazaLite@203.215.180.205) |
| 07:06.47 | *** join/#edev RobotGuy (n=robotguy@dsl093-038-072.pdx1.dsl.speakeasy.net) |
| 07:09.47 | *** join/#edev harpal (n=harpal@124.125.79.150) |
| 07:23.31 | *** join/#edev [AD]Turbo (n=turbo@82.57.128.131) |
| 07:24.25 | [AD]Turbo | yo |
| 07:34.50 | *** join/#edev magnetron (n=magnetro@unaffiliated/magnetron) |
| 07:38.59 | *** join/#edev roxfan2 (n=dunno@27.97-201-80.adsl-dyn.isp.belgacom.be) |
| 07:40.40 | *** join/#edev KazaLite (n=kazaLite@203.215.180.205) |
| 07:44.21 | RobotGuy | oy |
| 07:44.45 | KazaLite | anyone knowing sign extension in mips? |
| 07:45.04 | KazaLite | hi RobotGuy |
| 07:48.31 | RobotGuy | Hhey KazaLite! |
| 07:48.45 | RobotGuy | I know zero about MIPS |
| 07:49.20 | KazaLite | hehehe np:D |
| 07:49.58 | SpeedEvil | Use a large servo to extend a sign. |
| 07:51.13 | RobotGuy | Who woke SpeedEvil up? ;) |
| 07:51.53 | SpeedEvil | Beyonce. |
| 07:52.00 | SpeedEvil | Alas, not in person. |
| 07:54.19 | RobotGuy | wonders if flashing a hammer is illegal.. |
| 07:58.14 | RobotGuy | It seems crazy to me that a kernel would not even try to boot after being uncompressed. |
| 07:58.45 | SpeedEvil | It does - presumably. |
| 07:58.54 | SpeedEvil | It's just crashing before it does anything. |
| 07:59.56 | RobotGuy | SpeedEvil: Wouldn't it give some sort of error if it did crash?? |
| 08:00.04 | RobotGuy | Some indication of what is wrong? |
| 08:00.07 | SpeedEvil | Depends how it crashes. |
| 08:00.19 | RobotGuy | Silently if it is indeed crashing. |
| 08:00.50 | SpeedEvil | I it's crashing in the place it'd normally print, or the kernel is corrupt enough, it won't even try to report the crash. |
| 08:01.10 | SpeedEvil | Think of what'd happen if you run a random collectiono of bytes |
| 08:01.22 | SpeedEvil | It won't report any errors |
| 08:02.49 | RobotGuy | How am I to know what needs fixing? |
| 08:04.25 | SpeedEvil | JTAG debug'll do it. |
| 08:04.30 | SpeedEvil | In theory. |
| 08:04.30 | *** join/#edev lyakh (n=lyakh@p57BD2B6F.dip0.t-ipconnect.de) |
| 08:05.10 | SpeedEvil | this is your hammer? |
| 08:05.13 | RobotGuy | I've never done any JTAG debugging yet. |
| 08:05.29 | RobotGuy | Yes, the one I broke pin one off of (RXD0). |
| 08:06.12 | SpeedEvil | Reflash it with a known good original version? |
| 08:08.52 | RobotGuy | SpeedEvil: Umm, I neglected to save the last known working config. :( |
| 08:09.56 | *** join/#edev DrZimmerman (n=theo@tbaumgartner.broker.freenet6.net) |
| 08:13.29 | *** join/#edev harpal_ (n=harpal@121.246.75.165) |
| 08:14.23 | RobotGuy | SpeedEvil: I have my Bluetooth module wired to /dev/ttySAC0 now, but it is not currently in its socket. |
| 08:16.13 | SpeedEvil | reflash it with the original config - see if it still boots. |
| 08:16.17 | SpeedEvil | Go from there. |
| 08:16.31 | RobotGuy | SpeedEvil: That's way too easy. ;) |
| 08:16.46 | RobotGuy | I have to have a jffs2 boot setup. |
| 08:18.07 | RobotGuy | I am booting from flash, not using an initrd. |
| 08:20.17 | SpeedEvil | It should be easier to setup the second time. |
| 08:21.17 | RobotGuy | Second time? |
| 08:22.50 | RobotGuy | Hmmmmm, I think I could fit a pico-ITX on W.A.L.T.E.R. |
| 08:30.53 | RobotGuy | I am sure glad I don't have to enter all the build commands by hand - I created some scripts that handle it all. |
| 09:55.54 | *** join/#edev steliosk (n=Stelios@athedsl-282332.home.otenet.gr) |
| 10:21.11 | *** join/#edev hw_ (n=hw@p578b3905.dip0.t-ipconnect.de) |
| 10:42.24 | *** join/#edev howtoo (n=howtoo@195.159.62.66) |
| 10:47.28 | *** join/#edev bsdfox_ (n=h36sa@dialup-4.181.121.122.Dial1.SanJose1.Level3.net) |
| 11:02.52 | *** join/#edev RobotGuy (n=robotguy@dsl093-038-072.pdx1.dsl.speakeasy.net) |
| 11:13.01 | *** join/#edev Appleman1234 (n=Appleman@CPE-124-186-229-83.qld.bigpond.net.au) |
| 11:18.50 | *** join/#edev bsdfox__ (n=h36sa@dialup-4.181.121.155.Dial1.SanJose1.Level3.net) |
| 11:30.58 | *** join/#edev harpal_ (n=harpal@124.125.79.150) |
| 11:40.13 | *** join/#edev dijenerate (n=dijenera@65.48.147.188) |
| 11:42.52 | *** join/#edev bsdfox_ (n=h36sa@c-71-197-73-68.hsd1.ca.comcast.net) |
| 13:04.25 | *** join/#edev GPSFan (n=kenm@12.10.255.246) |
| 13:04.25 | *** mode/#edev [+o GPSFan] by ChanServ |
| 13:14.08 | *** join/#edev prpplague (n=dave@mail.americanmicrosystems.com) |
| 13:48.20 | *** join/#edev chouimat|work (n=dieu@209.217.106.98) |
| 14:12.18 | *** join/#edev prpplague^2 (n=dave@mail.americanmicrosystems.com) |
| 14:13.45 | *** join/#edev CosmicPenguin (n=nobody@163.181.251.103) |
| 14:13.45 | *** mode/#edev [+o CosmicPenguin] by ChanServ |
| 14:21.36 | CosmicPenguin | prpplague^2: http://beertown.org/pr/pdf/2007Top50Release.pdf |
| 14:21.46 | prpplague^2 | looks |
| 14:23.05 | prpplague^2 | nice |
| 14:23.51 | CosmicPenguin | I had no idea new belgium was that big |
| 14:24.05 | CosmicPenguin | I assume they are coast to coast at this point |
| 14:24.25 | CosmicPenguin | anybody on the east coast that drinks beer - can you buy fat tire in the store? |
| 14:24.57 | chouimat|work | never heard of it |
| 14:25.22 | CosmicPenguin | canada doesn't count |
| 14:26.03 | prpplague | CosmicPenguin: yea they are pretty much coast to coast |
| 14:26.14 | prpplague | CosmicPenguin: you can even get it at the airport in san juan |
| 14:26.30 | CosmicPenguin | awesome |
| 14:26.50 | prpplague | CosmicPenguin: fat tire is their most well known |
| 14:26.54 | CosmicPenguin | yeah |
| 14:26.57 | prpplague | CosmicPenguin: their other brews are harder to find |
| 14:27.00 | CosmicPenguin | I don't know how wide spread their other stuff is |
| 14:27.05 | CosmicPenguin | sunshine wheat |
| 14:27.17 | CosmicPenguin | their new IPA is okay - not great - called Mighty Arrow |
| 14:27.24 | prpplague | CosmicPenguin: yea, sunshine wheat, and the lager |
| 14:27.33 | prpplague | CosmicPenguin: those two are available here in most places |
| 14:27.34 | CosmicPenguin | Blue Paddle |
| 14:27.38 | prpplague | CosmicPenguin: yea thats it |
| 14:27.42 | CosmicPenguin | And I like Mothership Wit |
| 14:27.48 | prpplague | CosmicPenguin: but the rest of their stuff is very very hard to find |
| 14:27.54 | CosmicPenguin | its a wheat beer, so you have to be in the mood, but its pretty good |
| 14:28.00 | prpplague | ho ho hum |
| 14:28.14 | prpplague | my 2440 kernel is miscalculating clocks |
| 14:29.09 | prpplague | thats what i get for using a cutting edge kernel :( |
| 14:29.19 | CosmicPenguin | what, 2.6.11? |
| 14:29.20 | CosmicPenguin | runs |
| 14:29.54 | chouimat|work | CosmicPenguin: he uses 2.6.12.6-arm1 now :) |
| 14:30.31 | prpplague | CosmicPenguin: hehe |
| 14:30.36 | prpplague | CosmicPenguin: 2.6.25-rc8 |
| 14:30.40 | CosmicPenguin | wow - leet |
| 14:34.14 | prpplague | CosmicPenguin: hehe |
| 14:36.16 | *** join/#edev CheBuzz (n=CheBuzz@213-66-158.netrun.cytanet.com.cy) |
| 14:38.06 | CheBuzz | I have a couple of old JetDirect 615n cards that died on me and I am looking to make something cool with the (working) parts. Anybody ever attempt anything similar? |
| 14:39.56 | prpplague | ChanServ: not that i'm aware of, you are welcome to start a page on the elinux.org wiki about them, first thing you need to do is catalog and identify all the chips on the board |
| 14:42.10 | CheBuzz | Yep. Already have done that. |
| 14:43.14 | prpplague | got it posted somewhere? |
| 14:43.24 | CheBuzz | Just in a text document |
| 14:43.29 | CheBuzz | So... no |
| 14:43.58 | prpplague | CheBuzz: people can't really help unless they have something to look at |
| 14:44.06 | prpplague | CheBuzz: i'd start a page on the wiki |
| 14:44.26 | Mirell | Ho Hum. |
| 14:44.34 | CheBuzz | Sounds like a plan. I'll tidy up my text document and then transfer it to the wiki. |
| 14:44.54 | prpplague | Mirell: amen |
| 14:45.03 | Mirell | Another day at work. |
| 14:45.35 | prpplague | Mirell: i'm willing to bet, a nudie bar owner doesn't say "ho hum, another day at work" |
| 14:46.20 | Mirell | prpplague: Depends. You see naked girls all day, it begins to get boring. |
| 14:46.38 | prpplague | _never_ |
| 14:46.46 | Mirell | Haha |
| 14:48.51 | prpplague | Mirell: "Dave's Nudie Bar and Embedded Consulting" |
| 14:49.16 | prpplague | Mirell: "JTAG at every table!" |
| 14:49.27 | Mirell | Insert tab A into slot B. |
| 14:50.03 | Mirell | Recompile her kernel all night long |
| 14:50.05 | prpplague | CosmicPenguin: spent all day yesterday tracking down a nand flash problem |
| 14:50.30 | prpplague | CosmicPenguin: turns out the EE had a pull up resistor on the nWP instead of a pull down |
| 14:50.46 | prpplague | CosmicPenguin: so every time i'd reboot the system, the nand flash basically got wiped |
| 14:51.20 | *** join/#edev ccole (n=cole@cpe-71-72-50-240.neo.res.rr.com) |
| 14:58.50 | CosmicPenguin | doh |
| 14:58.54 | CosmicPenguin | that's gotta hurt |
| 15:00.13 | chouimat|work | prpplague: is that kind of mistake the type a newbie will make? |
| 15:00.23 | gandhijee | hey any of you guys ever use any of the gsm/cdma modems from multitech? |
| 15:00.49 | prpplague | gandhijee: i've looked but not used, we have a couple telit models |
| 15:00.55 | prpplague | chouimat|work: *cough* |
| 15:01.01 | chouimat|work | prpplague: :) |
| 15:01.17 | gandhijee | =/ |
| 15:08.47 | *** join/#edev adyer (n=adyer@adsl-69-223-13-78.dsl.chcgil.ameritech.net) |
| 15:10.22 | prpplague | GPSFan: ping |
| 15:11.18 | *** part/#edev CheBuzz (n=CheBuzz@213-66-158.netrun.cytanet.com.cy) |
| 15:12.01 | prpplague | darn, i think i'm getting sdram errors |
| 15:15.29 | prpplague | CPU: Testing write buffer coherency: ok |
| 15:15.29 | prpplague | Unable to handle kernel paging request at virtual address e3530000 |
| 15:15.29 | prpplague | Unhandled fault: alignment exception (0x003) at 0xe2860e49 |
| 15:20.01 | *** join/#edev landley (n=landley@adsl-68-93-84-18.dsl.hstntx.swbell.net) |
| 15:20.10 | *** mode/#edev [+o landley] by ChanServ |
| 15:21.10 | prpplague | landley: greetings |
| 15:21.30 | prpplague | landley: boss said he'd have a few nail boards for you to take with you to CELF |
| 15:22.38 | landley | Morning. |
| 15:22.41 | landley | Coolness! |
| 15:22.54 | landley | wonders how best to take advantage of this... |
| 15:23.08 | landley | I have one, and I was planning to demonstrate it. |
| 15:23.26 | landley | The obvious thing to do with others is to give them to tutorial attendees, but if I don't have enough for everybody... |
| 15:23.33 | landley | I suppose I could have a drawing or something... |
| 15:23.39 | landley | decides to worry about it later. |
| 15:34.24 | *** join/#edev linac (n=lin@121.41.43.208) |
| 15:35.44 | CosmicPenguin | Tim Bird is good at giving stuff away |
| 15:38.21 | prpplague | landley: bjdooks is still jerking me around on the tct hammer patch :( |
| 15:39.29 | *** join/#edev ce_geek (n=ce_geek@roger-blanchette-4.umelst.maine.edu) |
| 15:39.51 | prpplague | CosmicPenguin: how would you handle this? |
| 15:40.11 | landley | Tim Bird is excellent at giving stuff away, when that stuff is funded by Sony. :) |
| 15:40.18 | landley | (Or IBM, now that I think of it.) |
| 15:40.27 | landley | I guess PS3's count as both. :) |
| 15:40.37 | landley | prpplague: did you cc me on the resubmit? |
| 15:40.58 | landley | still hasn't resubscribed to the arm linux list. |
| 15:41.04 | landley | And their web archive remains TOTALLY UNUSABLE. |
| 15:41.15 | landley | (Why pretend to have one when it's _that_?) |
| 15:41.16 | prpplague | landley: i've not resubmitted it yet, i was hoping bjdooks was going to do something about it by now |
| 15:41.41 | landley | By "jerking you around", do you mean he hasn't replied at all? |
| 15:42.49 | landley | Or do you mean something else? |
| 15:42.53 | prpplague | landley: no reply on the list, on irc he first started saying he was busy setting up his own git, then on friday he says he can't decide if wants to go his on git tree or just pass the patches up to LAK, and now today he states he's too busy to deal with it |
| 15:43.11 | prpplague | s/go/do |
| 15:43.40 | prpplague | landley: never you mind he's signed off on plenty of his own company patches |
| 15:44.40 | landley | Could you privately email me the patch that you sent to the list, so I can make sure it matches the version I've got? |
| 15:44.51 | landley | (I tweaked the one you sent me, but I think you applied those tweaks...) |
| 15:45.36 | prpplague | landley: the one i posted is slightly different to comply with the LAK "standards" |
| 15:45.48 | prpplague | landley: i'll send forward you a copy |
| 15:45.51 | landley | Thanks. |
| 15:49.41 | prpplague | CosmicPenguin: how would handle this? |
| 15:49.44 | landley | I just got a recruiting call from Microsoft! |
| 15:49.49 | landley | turbo-boggles. |
| 15:50.11 | CosmicPenguin | prpplague: executive summary me - I wasn't paying attention |
| 15:50.23 | prpplague | <PROTECTED> |
| 15:50.28 | prpplague | <prpplague> landley: never you mind he's signed off on plenty of his own company patches |
| 15:50.57 | landley | I'd resubmit after a week, and I'd cc Andrew Morton, me, and the kernel list, and the kernel-arm list. |
| 15:51.17 | landley | Say this is a slightly updated version of the patch you submitted to kernel-arm last week. |
| 15:51.24 | landley | I'll ack: it publicly. |
| 15:51.29 | landley | After re-testing here. |
| 15:51.38 | landley | Point the rest of the people on this IRC at the post, so they can ack it. |
| 15:51.48 | landley | (Which is part of the reason to post to linux-kernel, since the archives of that are less insane.) |
| 15:52.18 | landley | The cc' to andrew morton is so it can go in the -mm tree. Doesn't have to make it into the final kernel through there, that's just for wider testing. |
| 15:52.35 | prpplague | rmk isn't going to be happy about that |
| 15:52.40 | landley | Get as many signed-off-by (or acked-by) lines as you can from people it worked for. |
| 15:52.43 | landley | prpplague: why? |
| 15:52.53 | landley | You did the right thing, you submitted to the arm list and waited a week. |
| 15:53.13 | landley | You're cc:ing the arm list on here, I just can't reply to stuff on the arm list because their web archive is incomprehensible. |
| 15:53.16 | prpplague | landley: becuase he won;t sign off any s3c24xx stuff without bjdooks signing off |
| 15:53.31 | landley | prpplague: ok. |
| 15:53.42 | landley | That doesnt' mean you have to wait for bjdooks to come out of his coma. Other people can test it. |
| 15:53.52 | landley | It becomes bjdooks responsibility to say what's _wrong_ with it. |
| 15:54.06 | landley | That's the four layer maintainer thing: each level owes yo a response. |
| 15:54.11 | prpplague | landley: this is why i rarerly submit patches, i just have the time to nag people to death about it |
| 15:54.20 | prpplague | landley: indeed |
| 15:54.20 | landley | cc: me. I can nag. |
| 15:54.37 | landley | If you submitted and got zero response a week later, resubmit, and cc: interested parties. |
| 15:55.03 | landley | How long has it been, by the way? |
| 15:56.16 | prpplague | landley: http://lists.arm.linux.org.uk/lurker/message/20080328.192949.4de17fec.en.html |
| 15:56.21 | prpplague | landley: 11 days or so |
| 15:56.38 | landley | Ok, say this is the same patch you submitted back in March. :) |
| 15:56.55 | landley | Actually, lemme look at it here first... |
| 15:58.23 | prpplague | CosmicPenguin: thoughts? |
| 15:59.24 | CosmicPenguin | I agree with landley - send it over his head |
| 15:59.56 | gandhijee | hey prpplague the telit cards you used, what type of bus did they sit on? |
| 16:00.06 | prpplague | gandhijee: rs-232 |
| 16:00.16 | prpplague | gandhijee: or i should say uart ttl |
| 16:00.35 | gandhijee | did you have it ride on top of a winsystems GPS-card? |
| 16:01.35 | prpplague | gandhijee: no the one we got had GPS built in |
| 16:01.43 | prpplague | gandhijee: they are available on sparkfun |
| 16:01.49 | prpplague | CosmicPenguin: okie dokie |
| 16:01.59 | gandhijee | ugh, i might have to tell the customer to get what your using then |
| 16:02.48 | gandhijee | i got this winsystems PC/104 card w/ GPS that takes these lil multitech cards, i can't get it to bring up the second rs-232 port on them damn things though |
| 16:02.54 | gandhijee | you gotta link for them? |
| 16:10.35 | prpplague | http://www.sparkfun.com/commerce/product_info.php?products_id=8246 |
| 16:11.25 | landley | prpplague: patch looks good. Testing it here now. |
| 16:16.29 | prpplague | hmm, i am at a loss to explain this kernel panic :( |
| 16:19.35 | prpplague | this has the feel of an sdram issue |
| 16:29.32 | SpeedEvil | Is this new hardware? |
| 16:30.35 | prpplague | yea |
| 16:36.46 | prpplague | interesting, if i drop the sdram size down to 16mb it doesn't kp, but still doesn't boot |
| 16:51.44 | SpeedEvil | TSOP SDRAM? |
| 16:51.59 | SpeedEvil | And got a logic analyser? |
| 16:52.25 | prpplague | SpeedEvil: bga |
| 16:52.33 | prpplague | SpeedEvil: yea looking into it now |
| 16:52.39 | SpeedEvil | Good luck. |
| 16:53.32 | prpplague | thanks |
| 16:53.45 | prpplague | sdram is always the biggest issue on board bringups imho |
| 17:09.26 | GPSFan | prpplague: pong |
| 17:10.09 | prpplague | GPSFan: hey, i don't suppose you'd have time in the next week or so to do a wiki page on making a web server from the hammer? |
| 17:13.17 | landley | prpplague: in terms of software, or in terms of hardware? :) |
| 17:14.02 | prpplague | landley: for the bringups? |
| 17:14.11 | GPSFan | prpplague: up to my ears atm. The closest I've done it using the webpal as a server. extremely basic, but it's been serving my weather page for almost 4 years. it uses busyboxes httpd. I might be able todo a few "ls -la" of various dir's that are critical and put up a skeleton. |
| 17:14.23 | landley | In terms of making a web server from a hammer module. |
| 17:14.39 | landley | The software side of making a web server out of it doesn't seem to hard. |
| 17:15.06 | landley | The hardware side... I could probably do it with a nail plugged into a wall outlet USB charger with an ethernet dongle in the outbound USB port. |
| 17:15.15 | GPSFan | of course it will need net access, and depending on how the isp is set up some sort of firewall |
| 17:15.26 | prpplague | landley: yea, its a little of both |
| 17:15.35 | prpplague | landley: i can do it, but i'm swampped too |
| 17:16.02 | landley | needs to finish OLS paper by the 16th, CELF paper this weekend, and do two project releases. |
| 17:16.22 | landley | Oh, and catch up on the QEMU weekly news thing I started. (Got 3 days done this morning before work.) |
| 17:16.38 | landley | Alas, those three days were january 15-17. |
| 17:23.30 | *** join/#edev ico2 (n=ico2@modem-2126.chameleon.dialup.pol.co.uk) |
| 17:55.15 | *** join/#edev magnetron (n=magnetro@unaffiliated/magnetron) |
| 17:55.44 | prpplague | something just isn't right and i'm not sure if it is the kernel or the sdram |
| 17:56.03 | prpplague | i'd think that if it was the sdram, the kernel wouldn't even start booting |
| 17:59.19 | *** join/#edev ico2 (n=ico2@modem-429.cougar.dialup.pol.co.uk) |
| 18:08.17 | SpeedEvil | can be randomish |
| 18:08.30 | SpeedEvil | for example if refresh isn't on |
| 18:08.37 | prpplague | SpeedEvil: yea |
| 18:08.46 | prpplague | SpeedEvil: still diging into ir |
| 18:08.47 | prpplague | it |
| 18:10.13 | prpplague | i'm suspecting something isn't exactly right with the sdram init |
| 18:14.40 | prpplague | ho ho hum |
| 18:14.43 | prpplague | goes for coffee |
| 18:17.23 | landley | prpplague: perhaps tighten the sdram timing just a bit? |
| 18:17.41 | landley | If the sdram timing is close but not quite, then it could sort of work but not enough to get all that far. |
| 18:18.06 | prpplague | landley: yea, once i rule out a kernel issue, i'll start tweaking the sdram timming |
| 18:18.09 | landley | Also, the cache might disguise a little of the sdram's issues. Try switching it off for a test boot? |
| 18:18.40 | prpplague | landley: getting ready to test the kernel on a similar board that we already have running |
| 18:18.44 | landley | How does one rule out having a kernel issue? |
| 18:18.48 | landley | Ah. |
| 18:19.01 | landley | has encountered kernel issues in shipped versions of Ubuntu. |
| 18:19.23 | prpplague | landley: a ton of s3c24xx stuff changed between .22 and .25 |
| 18:19.32 | prpplague | landley: so i would not be surprised with a kernel issue |
| 18:19.41 | landley | .25 worked for me on nail. |
| 18:19.46 | landley | Well, -rc7. |
| 18:19.56 | landley | And by "worked" I mean I could boot to a shell prompt and play around with it for 15 minutes. |
| 18:20.07 | prpplague | landley: yes this board is 2440 based whereas the nail is 2410 based |
| 18:20.16 | landley | ah. |
| 18:20.22 | prpplague | landley: i've already found several bugs that relate only to 2440 |
| 18:20.22 | landley | Lost in the "xx" above. |
| 18:20.44 | landley | Not that I likely would have caught it anyway. :) |
| 18:20.55 | prpplague | hehe, np, its alot to keep up with |
| 18:21.22 | prpplague | landley: thats one of my problems, the boss seems to think that the 2440 is just a faster version of the 2410 |
| 18:21.49 | prpplague | landley: so if i have a 2410 working it should be a snap for 2440 to work........NOT |
| 18:23.07 | landley | Er... ok. |
| 18:23.17 | landley | Is 2440 still armv4l-soft? |
| 18:24.02 | landley | Or perhaps armv4lt-soft, I think 2410 supports thumb. |
| 18:24.07 | landley | I'm not using it if so... |
| 18:25.03 | prpplague | landley: yea the 2410 have the same arm920t core, but there are alot of peripheral differences |
| 18:25.38 | prpplague | landley: the 2440 has support for either 2.5v or 1.8 operations as well as a whole host of special functions |
| 18:26.40 | landley | Anything particularly of interest for userspace software programming? |
| 18:28.12 | prpplague | landley: for the 2440? |
| 18:29.15 | wmat | prpplague: ping |
| 18:29.30 | wmat | prpplague: you been on elinux.org server directly lately? |
| 18:30.02 | prpplague | wmat: no not for atleast a month |
| 18:30.10 | prpplague | wmat: i just got word that the page is down |
| 18:30.14 | wmat | prpplague: ok |
| 18:30.17 | prpplague | wmat: it was up this morning |
| 18:30.20 | wmat | prpplague: yes, i'm investigating now |
| 18:30.42 | prpplague | wmat: ok thanks, keep me updated |
| 18:31.19 | prpplague | ~nslookup www.elinux.org |
| 18:31.59 | wmat | no ssh access |
| 18:32.22 | prpplague | someone forget to pay the bill? |
| 18:33.19 | wmat | heh |
| 18:38.57 | *** join/#edev CosmicPenguin (n=nobody@163.181.251.103) |
| 18:38.57 | *** mode/#edev [+o CosmicPenguin] by ChanServ |
| 18:47.22 | *** join/#edev RobotGuy (n=robotguy@dsl093-038-072.pdx1.dsl.speakeasy.net) |
| 18:49.27 | ds2 | hmmmmm |
| 18:55.47 | prpplague | wmat: anything? |
| 18:57.49 | wmat | prpplague: on the phone with godaddy right now |
| 18:57.59 | prpplague | wmat: okie dokie |
| 18:58.04 | prpplague | wmat: dns issue? |
| 18:58.22 | wmat | prpplague: perhaps |
| 18:58.43 | prpplague | wmat: i hope the data is ok |
| 18:58.55 | wmat | prpplague: yeah, me too |
| 19:03.36 | wmat | prpplague: they've opened a priority trouble ticket; should be back up shortly |
| 19:03.56 | prpplague | wmat: they know what the problem is? |
| 19:04.05 | wmat | prpplague: not exactly yet ;) |
| 19:09.40 | prpplague | hmm, looks to be acpi related |
| 19:13.17 | prpplague | wonder why its trying to start acpi |
| 19:14.07 | ds2 | for SMP/Hyperthreading? |
| 19:15.25 | prpplague | ds2: single core processor |
| 19:15.37 | ds2 | single core but you can still have HT |
| 19:15.47 | prpplague | ds2: on arm? |
| 19:16.02 | ds2 | oh nevermind... thought you meant the elinux box |
| 19:16.14 | ds2 | odd to do acpi on anything but x86 |
| 19:21.21 | RobotGuy | I keep getting those dang kernel panics. No idea what is causing them. :( No idea how to fix them. |
| 19:21.54 | ds2 | RobotGuy: got dump somewhere? |
| 19:22.16 | RobotGuy | http://www.pastebin.ca/977316 |
| 19:23.20 | ds2 | since i am neither claravoyant or omnicient, you will need to do at least one thing first |
| 19:23.40 | ds2 | see the hex humbers on line 107/108 of your dump? |
| 19:24.11 | ds2 | resolve those into symbol names with your System.map or reproduce this will the symbols in place if you don't want to do it by hand |
| 19:24.43 | prpplague | RobotGuy: your jffs2 partition is corrupted |
| 19:24.58 | ds2 | someone else here is apparently claravoyant ;) |
| 19:25.09 | RobotGuy | prpplague: How can it be? I just rebuilt the image. |
| 19:25.55 | prpplague | RobotGuy: something probably overwrote the jffs2 partition area |
| 19:26.05 | prpplague | ds2: no, just experienced |
| 19:26.26 | ds2 | prpplague: :) |
| 19:27.33 | ds2 | it would be good practice for people to know how to submit proper bug reports with symbols resolved |
| 19:27.33 | RobotGuy | <PROTECTED> |
| 19:27.34 | prpplague | RobotGuy: so application or function is overwriting your jffs2 partition |
| 19:27.40 | chouimat|work | ds2: this sound like a debuging 101 assigment |
| 19:27.47 | RobotGuy | prpplague: Kernel doesn't even boot, so can't be an application running. |
| 19:28.13 | prpplague | RobotGuy: so you just flashed this jffs2 image? |
| 19:28.22 | RobotGuy | prpplague: Yes. |
| 19:28.24 | ds2 | actually,kernel did boot. you were right on the edge of starting userspace |
| 19:28.50 | RobotGuy | prpplague: I have to reflash everything when I change the kernel because the kernel modules are in the jffs2 image. |
| 19:28.53 | prpplague | RobotGuy:you flash it from apex or from openocd? |
| 19:29.01 | RobotGuy | prpplague: From APEX |
| 19:29.11 | prpplague | RobotGuy: are you flashing the kernel from apex too? |
| 19:29.30 | RobotGuy | prpplague: Flashing boot loaded and kernel using openocd. |
| 19:30.00 | prpplague | RobotGuy: and are you erasing the jffs2 area from openocd or from apex? |
| 19:30.15 | RobotGuy | prpplague: Erasing flash from APEX. |
| 19:30.57 | prpplague | RobotGuy: i can only guess that you either did a typo or something happened when transfering the file |
| 19:31.16 | prpplague | RobotGuy: a common mistake is for people to use 0x3080000 instead of 0x30800000 |
| 19:31.36 | RobotGuy | prpplague: copy 0x30800000+10074460 nor:1280k |
| 19:31.57 | RobotGuy | prpplague: That is the only place a typo might enter into things. |
| 19:31.58 | prpplague | RobotGuy: and the xreceive command? |
| 19:32.15 | RobotGuy | prpplague: xreceive 0x30800000 |
| 19:32.42 | prpplague | well there is definetly something in there creating the problem as part of the procedure |
| 19:32.42 | RobotGuy | prpplague: I copy and paste to reduce the chance of typing errors. |
| 19:33.07 | prpplague | RobotGuy: might try doing it by hand to make sure your cut-n-paste isn't creating a problem |
| 19:33.13 | RobotGuy | prpplague: erase nor:1280k+0xec0000 |
| 19:34.26 | RobotGuy | prpplague: These commands are the same ones I use when the kernel was fine. I don't care to type all that because it will make things worse due to errors. That's why I copy and paste to be sure I have it right. |
| 19:34.26 | ds2 | where does the erase command stop at? end of flash? |
| 19:34.34 | prpplague | ds2: yea |
| 19:34.47 | *** join/#edev ico2_ (n=ico2@modem-1578.chimpanzee.dialup.pol.co.uk) |
| 19:35.03 | prpplague | has to go |
| 19:35.03 | ds2 | so... if the lay out is APEX - KERNEL - JFFS2 and your flash order is JFFS2, then KERNEL and you do a erase for each one... doesn't that destory the JFFS2? |
| 19:35.05 | prpplague | bbiab |
| 19:36.53 | ds2 | RobotGuy: did you flash the kernel first or jffs2 first? |
| 19:38.49 | RobotGuy | ds2: Loader first. JFFS2 is done in APEX. |
| 19:39.03 | RobotGuy | ds2: I mean, loader then kernel |
| 19:39.36 | ds2 | RobotGuy: hmm that should be okay... cuz the erase for the kernel will take care of erasing things for JFFS2 |
| 19:39.55 | ds2 | RobotGuy: what are the image sizes? (kernel size and the JFFS2 image size) |
| 19:40.04 | RobotGuy | ds2: I don't need to do the erase in APEX for the jffs2 then? |
| 19:40.38 | ds2 | RobotGuy: I don't think so but it should hurt. and to fully answer that, I would need to ask a few more detailed questions |
| 19:40.45 | ds2 | should NOT hurt |
| 19:40.46 | RobotGuy | ds2: Kernel: 1018560, JFFS2: 3204212 |
| 19:41.12 | ds2 | hmmm that looks fine |
| 19:42.58 | RobotGuy | ds2: I only erase before I copy the jffs2 image to flash (in APEX). That is after flashing the bootloader and kernel (openocd) |
| 19:43.35 | ds2 | procedure wise, that looks fine |
| 19:44.09 | ds2 | if I were assigned to look at it, I'd get lines 107/108 decoded from the System.map |
| 19:45.31 | RobotGuy | ds2: I do not know how to do that or what that means. The problem I have is sometimes everything works fine, but then I make one change in my kernel and everything breaks. This does not seem reasonable. |
| 19:45.39 | RobotGuy | What will doing all that show us? |
| 19:45.50 | ds2 | RobotGuy: it will tell me where the kernel is crashing |
| 19:46.09 | RobotGuy | ds2: OK. Do I need that kernel oops page? |
| 19:46.16 | ds2 | those 2 lines are a stack trace but since every kernel has different addresses, those numbers are unique to your build |
| 19:46.29 | ds2 | RobotGuy: just post your System.map and I can decode it quickly |
| 19:46.45 | ds2 | and if that is done before the next 15mins is over, I can even explain how to do that ;) |
| 19:47.33 | RobotGuy | ds2: Can you give me a tag of the lines I should be looking at? |
| 19:47.33 | ds2 | RobotGuy: i am looking at this from purely a kernel developer point of view; so I resort to basics |
| 19:48.15 | ds2 | your System.map file will have lines that look like this: |
| 19:48.16 | ds2 | c0004000 A swapper_pg_dir |
| 19:48.16 | ds2 | c0008000 T __init_begin |
| 19:48.16 | ds2 | c0008000 T _sinittext |
| 19:48.23 | RobotGuy | ds2: Right |
| 19:48.37 | ds2 | those numbers in the first column are the entry points into the functions listed on the last column |
| 19:49.11 | ds2 | what you need to do is look for the closest hex address from line 107 and 108 of your pastebin. it should be the closest address that is less then the one in the pastebin |
| 19:49.50 | ds2 | once you have the names, we can find out who or what is looking at address 0xf7000201 |
| 19:51.25 | RobotGuy | ds2: c0012c88 t ns558_init, c0008858 t kernel_init |
| 19:51.32 | ds2 | whoa |
| 19:51.42 | ds2 | you have the joystick support turned on and it is probing for stuff |
| 19:51.45 | ds2 | please turn it off! |
| 19:52.01 | ds2 | (NS558 is the joystick driver) |
| 19:52.23 | ds2 | oops, gameport driver I mean |
| 19:52.38 | ds2 | do a grep for "NS558" in your .config |
| 19:53.03 | RobotGuy | ds2: I should not have any gameport turned on in my kernel, but will check. |
| 19:53.23 | ds2 | RobotGuy: you must. otherwise, ns558_init won't be mentioned in your System.map! |
| 19:53.51 | RobotGuy | ds2: It was set so I turned it off. |
| 19:54.11 | ds2 | but System.map is rebuilt each time you build the kernel |
| 19:54.47 | RobotGuy | I'm rebuilding my kernel now. I am learning this new debugging tool gradually. |
| 19:55.01 | ds2 | it takes time |
| 19:55.25 | ds2 | if you see another one of these dumps, you can do a similar decode to find out what is unhappy |
| 19:57.31 | RobotGuy | ds2: Thanks. Now I have at least some idea how to use that information. :D |
| 20:17.07 | *** mode/#edev [+o prpplague] by ChanServ |
| 20:28.33 | RobotGuy | ds2: It boots now! :D Many thanks. |
| 20:29.17 | RobotGuy | Now I am trying to break it again. :D:D Muahahahahah ;) |
| 20:31.31 | prpplague | returns |
| 20:31.51 | prpplague | RobotGuy: got stuck again with turning on stuff in the kernel eh? |
| 20:32.28 | RobotGuy | prpplague: Something was turned on that I didn't remember turning on. |
| 20:35.12 | prpplague | ds2: hehe i wouldn't have thought to look for a joystick driver in a million years |
| 20:41.28 | RobotGuy | prpplague: Is it safe to assume that I need to cross the RX/TX lines when connecting to my BT module? |
| 20:41.57 | prpplague | RobotGuy: no clue, not familiar with that module |
| 20:42.21 | RobotGuy | prpplague: Wouldn't one normall cross RX and TX? |
| 20:42.31 | prpplague | RobotGuy: normally, yes |
| 20:42.39 | RobotGuy | That's what I thought. |
| 20:44.41 | RobotGuy | Now I go back to breaking things. :D:D:D:D |
| 20:44.52 | RobotGuy | hides |
| 20:45.41 | RobotGuy | I could make my own router now. :) Boots from flash so nothing else needed. :) |
| 20:46.16 | RobotGuy | Oh, I found an MCU I would love to use as a companion to Hammer. |
| 20:46.54 | RobotGuy | It has 3x uarts, 2x quadrature encoder modules, etc. |
| 20:47.03 | prpplague | wmat: any word ? |
| 20:48.10 | prpplague | CosmicPenguin: name the top three reasons why a call to schedule() would kp with null pointer? |
| 20:52.08 | prpplague | hmm, i'm getting a similiar problem as this - http://groups.google.com/group/fa.linux.kernel/msg/95d2e9b29347e110 |
| 20:52.50 | ds2 | prpplague: that's why I wanted the dump decoded |
| 20:53.00 | prpplague | ds2: yea |
| 20:53.05 | prpplague | ds2: good catch |
| 20:54.25 | adyer | RobotGuy: I would look up the datasheets for both tx/rx on both chips. /usually/ tx indicates and output, but not always |
| 20:54.58 | RobotGuy | adyer: I have done that and I can not tell so assume I have to cross RX and TX. |
| 20:55.13 | RobotGuy | I also need to add a 1K or so pull up to the reset line. |
| 20:55.32 | ds2 | RobotGuy: tack solder the 2 together with a 470ohm resistor on each line and see if it works |
| 20:55.37 | ds2 | if not, swap things around |
| 20:55.51 | RobotGuy | ds2: Ah, OK. |
| 20:56.30 | ds2 | 470ohm should be sufficient to prevent any major damage with fighting xmitters |
| 20:56.38 | RobotGuy | I so want to get the eval board for this http://www.luminarymicro.com/products/lm3s6965.html which is only $69.95. I want to use it with Hammer. |
| 20:56.57 | ds2 | prpplague: what kernel version is your problem occurring on? |
| 20:57.08 | prpplague | 2.6.25-rc8 |
| 20:57.27 | ds2 | is this a samsung proc? |
| 20:58.55 | RobotGuy | Have you seen the Virtual COGs? |
| 20:59.35 | RobotGuy | They stack. :D |
| 20:59.56 | ds2 | so does PC104 ;) |
| 21:00.22 | RobotGuy | I know |
| 21:01.04 | prpplague | ds2: yea |
| 21:01.27 | ds2 | prpplague: checked to make sure you have all the errata work arounds for the cache? |
| 21:01.42 | ds2 | or if you got the patience, disable caches and try? |
| 21:01.49 | prpplague | ds2: hehe not my job, hehe, since 2.6.24 works |
| 21:01.58 | ds2 | oh :P |
| 21:06.15 | CosmicPenguin | prpplague: are you sure are in schedule, or are you just ending up in there? |
| 21:06.57 | prpplague | CosmicPenguin: yea i've printk'd it to hell and the last call before it kp's is to schedule() |
| 21:07.34 | prpplague | CosmicPenguin: it never returns from schedule() |
| 21:10.03 | prpplague | i'm beginning to suspect that something beign declared as __init and it really shouldn't be |
| 21:12.13 | CosmicPenguin | this is in init, then? |
| 21:12.55 | prpplague | yea, init/main.c |
| 21:13.16 | prpplague | CosmicPenguin: the rest_init() function |
| 21:13.46 | CosmicPenguin | keep in mind that when you call schedule, then something else might be running and faulting |
| 21:14.33 | ds2 | are you allowed to call schedule that early in the init? |
| 21:15.08 | prpplague | oops i think i found it, looks like i have a recursion |
| 21:18.17 | prpplague | a s3c24xx function that calls the s3c2440 function which calls the s3c2410 function which is defined as the s3c24xx function |
| 21:18.30 | RobotGuy | CosmicPenguin: You work with x86 stuff, right? I would like to find out the size of your boards. |
| 21:19.00 | prpplague | CosmicPenguin: they don't generally make their boards available retail mainly as reference to manufacturers |
| 21:19.10 | prpplague | CosmicPenguin: but you can get some pretty small geode boards |
| 21:19.26 | prpplague | bing that was it |
| 21:19.29 | prpplague | bingo |
| 21:19.29 | CosmicPenguin | prpplague: says you |
| 21:19.36 | RobotGuy | prpplague: I'm interested in something small that might be used on a robot, x86 compatible. |
| 21:19.38 | CosmicPenguin | RobotGuy: we can go down to mini-ITX |
| 21:19.49 | prpplague | CosmicPenguin: when did AMD start selling boards? |
| 21:20.04 | RobotGuy | CosmicPenguin: I think a mini-ITX would fit on an 11 1/2" diameter robot. |
| 21:20.10 | CosmicPenguin | prpplague: you know, at some point, somebody is going to buy our processor and make it available to the general public |
| 21:20.33 | prpplague | isn't that what i said? |
| 21:21.08 | CosmicPenguin | http://www.soekris.com/net5501.htm |
| 21:21.23 | CosmicPenguin | http://www.pcengines.ch/alix2.htm |
| 21:21.30 | RobotGuy | I'd like to get something like a geode board to try on a robot (not W.A.L.T.E.R.). |
| 21:21.42 | adyer | CosmicPenguin: you work at AMD? |
| 21:21.42 | CosmicPenguin | http://www.artecgroup.com/thincan/ |
| 21:21.45 | CosmicPenguin | Among others |
| 21:21.54 | CosmicPenguin | Actually, I think maybe the alix would work perfect for you |
| 21:22.20 | CosmicPenguin | it is headless, has lots of gozinas and gozoutas |
| 21:22.26 | ds2 | just put a OLPC on there |
| 21:22.29 | ds2 | :) |
| 21:22.40 | ds2 | probally has more complete docs then the other boards ;) |
| 21:22.53 | CosmicPenguin | Also, coreboot works more or less flawlessly with the alix boards |
| 21:23.02 | RobotGuy | I'm thinking a distributed processing arrangement, of course. |
| 21:24.13 | adyer | CosmicPenguin: can you get register level programming docs for the SB600 southbridge? |
| 21:24.24 | RobotGuy | CosmicPenguin: Oooooo, shiny! That Alix looks very interesting. |
| 21:24.33 | CosmicPenguin | adyer: when they come out, you'll be the 10th to know |
| 21:25.09 | adyer | CosmicPenguin: salesdroids wont sell me parts cause I want to use it with a non x86 host proc |
| 21:25.10 | RobotGuy | I would have to have the I2C option, for sure. :) |
| 21:25.16 | prpplague | CosmicPenguin: indeed that alix looks really cool |
| 21:25.17 | CosmicPenguin | You would be first, but there are other people in the queue |
| 21:25.37 | CosmicPenguin | RobotGuy: I don't know if they break out the I2C or not |
| 21:25.42 | ds2 | doesn't that assume there will be a doc team working on it when it is available ? ;) |
| 21:25.47 | CosmicPenguin | I assume you could solder on to it though - you have the power |
| 21:25.53 | ds2 | heard they are sheding people left and right }:-) |
| 21:25.56 | RobotGuy | CosmicPenguin: It says I2C is an option, so I am guessing they do. |
| 21:26.18 | RobotGuy | CosmicPenguin: My Kung-Fu is getting stronger each day. :) |
| 21:26.31 | CosmicPenguin | RobotGuy: ah - okay - I see that |
| 21:26.42 | ds2 | or maybe they'll be photocopies of green napkins with scribblings from the test group? |
| 21:26.46 | CosmicPenguin | I saw one of these boards this weekend - I was pretty impressed |
| 21:27.14 | RobotGuy | CosmicPenguin: I like the alix3c2. |
| 21:27.51 | CosmicPenguin | headless |
| 21:29.13 | RobotGuy | CosmicPenguin: The pricing is very reasonable too! |
| 21:31.17 | CosmicPenguin | yeah - the I2C is broken out right by the processor |
| 21:31.24 | CosmicPenguin | And like i said, coreboot works pretty well on this board |
| 21:31.34 | CosmicPenguin | so you can have free BIOS, which is always ++ |
| 21:32.07 | RobotGuy | CosmicPenguin: FREE is good in this case. :) |
| 21:33.07 | RobotGuy | CosmicPenguin: I'm interested in something I can potentially use for audio and video/vision processing on a robot. |
| 21:33.27 | RobotGuy | That Alix.3c2 looks very nice. |
| 21:33.39 | ds2 | blackfin! |
| 21:34.05 | RobotGuy | ds2: I am looking at the Blackfin Camera Board for the actual camera/processor on W.A.L.T.E.R. :) |
| 21:34.38 | ds2 | RobotGuy: have you looked at the dsPICs? |
| 21:35.21 | RobotGuy | ds2: I will be using dsPICs to process quadrature encoder signals, analog inputs, etc. Yes, I use PICs and dsPICs. :) |
| 21:35.27 | prpplague | feels silly now missing the recursion calls |
| 21:35.35 | RobotGuy | I have a small box full of PICs and dsPICs. :) |
| 21:35.41 | *** join/#edev ico2__ (n=ico2@modem-965.cougar.dialup.pol.co.uk) |
| 21:36.00 | CosmicPenguin | yeah - you can probably get better video processors then an plain x86 processor |
| 21:36.00 | ds2 | RobotGuy: couldn't you do the vision processing on the dsPICs? |
| 21:36.06 | CosmicPenguin | especially one that doesn't do floating point very well |
| 21:36.20 | ds2 | long live NexGen |
| 21:36.33 | RobotGuy | ds2: Not likely. |
| 21:37.48 | RobotGuy | CosmicPenguin: I'm looking for something that has multimedia capabilities. Ummm, I may even consider Windows if the Alix can run it. RoboRealm is very good for vision processing, but Windows only. :(:( |
| 21:37.55 | prpplague | wmat: ping |
| 21:38.08 | CosmicPenguin | Not sure if it can run windows, actually |
| 21:38.14 | CosmicPenguin | maybe with the Award BIOS |
| 21:38.18 | CosmicPenguin | not with coreboot forsure |
| 21:38.18 | RobotGuy | This will be a completely different robot than W.A.L.T.E.R. |
| 21:39.12 | ds2 | vmware inside Linux |
| 21:39.13 | RobotGuy | CosmicPenguin: Naturally, I would much prefer Linux, but vision processing software seems to not exist, although the components do. Hmmmmmmmm. |
| 21:39.32 | RobotGuy | ds2: Not enough memory on an alix for that |
| 21:39.34 | ds2 | CosmicPenguin: did the Geodes ever get the virtualization stuff like the athlons? |
| 21:39.38 | CosmicPenguin | no |
| 21:39.54 | CosmicPenguin | Geode development ended well before the virtualization craze took off |
| 21:40.07 | ds2 | it "ended"? :( |
| 21:40.58 | prpplague | oh how pretty |
| 21:40.59 | prpplague | http://groups.google.com/group/fa.linux.kernel/msg/95d2e9b29347e110 |
| 21:41.02 | prpplague | oops |
| 21:41.05 | prpplague | Memory: 128MB = 128MB total |
| 21:41.05 | prpplague | Memory: 128384KB available (1240K code, 164K data, 76K init) |
| 21:41.10 | prpplague | 128mb sdram |
| 21:41.16 | prpplague | no idea what to do with it all |
| 21:41.31 | prpplague | thats the most ram i've ever had on a system |
| 21:41.32 | ds2 | run windows via qemu ;) |
| 21:41.57 | prpplague | ds2: i didn't think qemu supported windows ce |
| 21:42.28 | ds2 | no, win xp |
| 21:43.37 | prpplague | hmm, that would be a challenge, to get that on arm, not sure it would have enough horsepower |
| 21:44.00 | RobotGuy | Interesting, my kernel does not seem to like my Ralink USB WiFi adapter. :( |
| 21:44.03 | CosmicPenguin | why wouldn't qemu be able to run windows ce? |
| 21:44.08 | ds2 | some guy did it on a N800 using bochs |
| 21:44.16 | ds2 | took him 30minutes to just boot! |
| 21:44.28 | prpplague | CosmicPenguin: no reason, just that i'd not heard or seen anyone doing it |
| 21:44.37 | prpplague | ds2: hehe |
| 21:44.42 | CosmicPenguin | I wouldn't be suprised if it didn't just work |
| 21:45.00 | RobotGuy | prpplague: I would love to have a board with 128 MB Ram and more flash. :) I love the Technologic TS-7800 board and want to get one eventually. |
| 21:45.30 | prpplague | RobotGuy: actually this one will be populated with 256mb sdram and 512mb flash |
| 21:45.53 | RobotGuy | I also like the stacking arrangement with the Virtual COGs and will probably end up using these to build my wheelchair control computer. |
| 21:46.01 | RobotGuy | prpplague: What board is this?? |
| 21:46.27 | prpplague | RobotGuy: new handheld design |
| 21:46.42 | RobotGuy | prpplague: Ah, OK. Something I won't ever see then. :( |
| 21:47.20 | RobotGuy | I'd be truly happy with 128 MB RAM/128 MB or 256 MB Flash. |
| 21:47.21 | prpplague | RobotGuy: this one will be publically available, but i'm very sure most individuals won't like the price |
| 21:47.37 | RobotGuy | prpplague: I am sure that will be the case, unfortunately. :( |
| 21:48.04 | RobotGuy | I like my Hammer though and am getting more excited about having it control W.A.L.T.E.R. :D |
| 21:52.10 | RobotGuy | CosmicPenguin: I might get one of those Alix.3C2 boards with the I2C. I like tinkering with different hardware. :) |
| 21:53.28 | CosmicPenguin | go for it |
| 22:01.09 | prpplague | ho ho hum |
| 22:01.11 | RobotGuy | prpplague: Please let me know when you have 2.6.25-rc8 working. I would like to try it. There appears to be a problem with the Ralink driver I need. |
| 22:01.33 | prpplague | RobotGuy: the -rc7 patch applies cleanly to rc8 |
| 22:02.03 | RobotGuy | prpplague: Cool! I will try it. |
| 22:04.49 | prpplague | RobotGuy: now how's this...... |
| 22:04.51 | prpplague | Welcome to the TinCanTools Hammer Development System |
| 22:04.52 | prpplague | Hammer login: root |
| 22:04.52 | prpplague | Dec 31 17:00:06 login[700]: root login on 'ttySAC0' |
| 22:04.52 | prpplague | # cat /proc/meminfo |
| 22:04.52 | prpplague | MemTotal: 128504 kB |
| 22:04.54 | prpplague | MemFree: 121980 kB |
| 22:05.17 | RobotGuy | prpplague: You should be paste binning.. :P |
| 22:05.34 | prpplague | RobotGuy: indeed i should |
| 22:05.41 | prpplague | RobotGuy: i was thinking it was less likes |
| 22:05.42 | RobotGuy | prpplague: I am jealous of the memory. |
| 22:05.43 | prpplague | lines |
| 22:05.51 | prpplague | but i got the issue text too |
| 22:06.14 | RobotGuy | Is that the new product board you were talking about? |
| 22:06.15 | prpplague | only reason we have it is for when they put windows ce and it |
| 22:06.28 | RobotGuy | prpplague: Samsung? |
| 22:06.34 | prpplague | RobotGuy: no AML |
| 22:06.41 | RobotGuy | I mean the processor. |
| 22:06.48 | prpplague | yea s3c2440 |
| 22:06.55 | prpplague | 400mhz |
| 22:06.59 | RobotGuy | Oh, I would so like to have one of those. |
| 22:07.08 | Redhatter | 0.4Hz? |
| 22:07.20 | prpplague | Redhatter: yea yea |
| 22:07.27 | prpplague | 400Mhz |
| 22:07.47 | Redhatter | That isn't bad at all |
| 22:07.55 | RobotGuy | prpplague: That's right in the area for doing most of what I want to do that I don't think I can do with Hammer. |
| 22:08.24 | Redhatter | is supposed to pick up an AT91SAM board today |
| 22:08.34 | prpplague | RobotGuy: well our build will have all of our software in a 16mb partition with the remander set asside for customer usage and backup files |
| 22:09.43 | RobotGuy | prpplague: In flash then? |
| 22:09.59 | RobotGuy | I really don't need a lot of flash, but more is always nice. |
| 22:10.11 | adyer | prpplague: LOL, it's MHz, not Mhz :-) |
| 22:10.46 | RobotGuy | prpplague: Is that S3C2440 board the one with 128 MB Flash also? |
| 22:11.12 | RobotGuy | Hmmm, the S3C2440 with Hammers as slaves.. Hmmmmmmm.. :D |
| 22:11.37 | RobotGuy | Oooooo, shiny. :D |
| 22:12.27 | prpplague | adyer: yea i know, just really lazy with that, i should try harder but i'm too tired today |
| 22:12.52 | adyer | prpplague: we all know, just tweaking you :-) |
| 22:13.05 | prpplague | RobotGuy: yea, its got 128MB on it now, but it will have 256MB |
| 22:13.09 | prpplague | adyer: ahh ok |
| 22:13.19 | prpplague | laughs .... ha ha ha |
| 22:13.36 | prpplague | i think its beer thirty time |
| 22:13.46 | RobotGuy | prpplague: Ooooooooooooooooooooooooooo, very shiny.. :D I bet it does lots of colors on an LCD too. |
| 22:14.03 | prpplague | RobotGuy: ? |
| 22:14.25 | RobotGuy | The s3c2440 board is what I am referring to with oooooooooo, shiny. :D |
| 22:14.34 | prpplague | RobotGuy: oh ok |
| 22:14.45 | prpplague | RobotGuy: yea 256MB sdram and 512MB nand flash |
| 22:15.05 | RobotGuy | prpplague: Can I see full specs, by chance? |
| 22:15.20 | prpplague | RobotGuy: uh no |
| 22:15.30 | RobotGuy | prpplague: Drats, it figures. |
| 22:16.03 | prpplague | ok so i'm used to all the spam for viagra and such, but this is a new twist: buy your urinal cakes and toilet paper from us wholesale |
| 22:16.18 | RobotGuy | I hate it when I find something so cool and I can't get one. |
| 22:16.48 | prpplague | RobotGuy: hehe you can get one when they are available, just gotta shell out alot of money, hehe |
| 22:17.12 | RobotGuy | prpplague: That's the catch though, they want oodles of bucks for the whole thing. I just want the board. |
| 22:17.42 | RobotGuy | prpplague: I'll have to go with the Virtual COGs to get what I want. |
| 22:17.55 | adyer | s3c2440 is close to the proc used in the iPhone, right? That spec is floating around the net |
| 22:18.22 | prpplague | CosmicPenguin: they don't want to stock two different boards, one with 32mb sdram and 16mb flash, and the other with 256mb sdram and 512mb flash, so we are going to just use the sam 256/512 board for linux |
| 22:18.23 | RobotGuy | adyer: I can get the s3c2440 datasheet. I want the friggin cool s3c2440 board. :D |
| 22:18.35 | prpplague | CosmicPenguin: is that a shame or what? |
| 22:18.53 | prpplague | adyer: indeed |
| 22:19.08 | prpplague | adyer: openoce will even read the jtag |
| 22:19.13 | CosmicPenguin | I dunno - depends on how much the 256/512 costs |
| 22:19.13 | prpplague | openocd |
| 22:19.46 | prpplague | CosmicPenguin: its about a $12 difference |
| 22:19.48 | RobotGuy | prpplague: How much would a Hammer with 256MB RAM/512MB Flash cost to make? Maybe a Sledge Hammer with an s3c2440 for instance. |
| 22:20.14 | CosmicPenguin | $12 isn't too bad - and that delta is only going to go down |
| 22:20.18 | prpplague | RobotGuy: that would be too expensive to do for the hammer |
| 22:21.03 | prpplague | CosmicPenguin: yea, still its funny that we'll have 256/512 and we will only be using 32/16 |
| 22:21.17 | adyer | wonders when the sickle board is coming out. would sell well with hammer among those linux commies :-) |
| 22:21.21 | CosmicPenguin | more memory doesn't ever hurt |
| 22:21.28 | prpplague | RobotGuy: sledge hammer will probably be populated with 64mb sdram and 32mb flash |
| 22:22.00 | adyer | prpplague: come on, you know code must expand to overflow the space available |
| 22:22.37 | prpplague | CosmicPenguin: yea, if we ever changed the software it might be handy, but 99% of our customers are using nothing more than a ssh or telnet session over encrypted tunnel |
| 22:23.16 | RobotGuy | prpplague: No chance of getting Sledge to 128 MB RAM and 128 MB Flash?? |
| 22:23.43 | prpplague | RobotGuy: nothing is set in stone since only pre-production boards have been made |
| 22:23.48 | CosmicPenguin | With the cost of producting lower sized chips and such - I try not to get worked up about it |
| 22:23.51 | RobotGuy | Even 128/64 or 128/32 would be nice. |
| 22:23.59 | CosmicPenguin | the way I see it is that if it costs me less to ship more unused bits, then so be it |
| 22:24.37 | prpplague | CosmicPenguin: i know, i was mainly talking about the comparison between whats need to job X under linux and the job X under windows ce |
| 22:25.11 | CosmicPenguin | you boot to a graphical engine too? |
| 22:25.13 | CosmicPenguin | kdrive? |
| 22:25.31 | prpplague | CosmicPenguin: not for the base software its just a small fb library |
| 22:26.06 | CosmicPenguin | so its not really a fair comparison with CE, is it? :) |
| 22:26.09 | RobotGuy | prpplague: Is the current rc7 hammer patch workable with rc8 kernel? |
| 22:26.24 | prpplague | RobotGuy: yep, booted it fine yesterday |
| 22:27.10 | prpplague | CosmicPenguin: yes and no, since they are doing the exact same thing, they just think they need windows ce |
| 22:27.30 | prpplague | CosmicPenguin: all they are doing is booting the device and opening something like *cough* tinyterm |
| 22:27.40 | prpplague | CosmicPenguin: under windows ce that is |
| 22:28.01 | CosmicPenguin | sounds like fishy sales guys to me |
| 22:28.07 | prpplague | CosmicPenguin: yep |
| 22:28.25 | CosmicPenguin | but if you had kdrive + a toolkit, you would probably be closer to 128 |
| 22:28.27 | prpplague | CosmicPenguin: our industry is ripe with BS sales data |
| 22:28.30 | RobotGuy | sighs |
| 22:28.46 | prpplague | CosmicPenguin: actually we did a small build like that on the M7220 |
| 22:29.07 | prpplague | CosmicPenguin: we did it under 32mb/32mb |
| 22:29.36 | prpplague | gets ready to go home for the day |
| 22:29.44 | prpplague | wmat: any word on the web site? |
| 22:31.17 | prpplague | later folks |
| 22:32.58 | ds2 | ayder: you expecting a easter egg where the a banner pops up with 'C C C P'? |
| 22:34.31 | *** join/#edev LyosNorezel (n=LyosNore@unaffiliated/lyosnorezel) |
| 22:35.00 | adyer | ds2: da! |
| 22:44.03 | RobotGuy | The Hammer patch for rc7 does not apply cleanly after patching to rc8 |
| 22:51.43 | *** part/#edev adyer (n=adyer@adsl-69-223-13-78.dsl.chcgil.ameritech.net) |
| 23:02.20 | landley | RobotGuy: no reason it shouldn't... |
| 23:02.28 | landley | Did you already have it applied before patching to -rc8? |
| 23:02.35 | RobotGuy | landley: It didn't for me |
| 23:02.57 | landley | goes to check... |
| 23:03.34 | RobotGuy | landley: I applied the stock rc8 patch to the 2.6.24 kernel, then the rc7-hammer patch to that after renaming to 2.6.25-rc7 |
| 23:04.22 | landley | Ah. |
| 23:04.31 | landley | Kernel patches are designed to be applied with -p1. |
| 23:04.37 | landley | cd into the linux directory and apply with -p1. |
| 23:04.51 | landley | patch -p1 -i blah.patch |
| 23:07.04 | RobotGuy | landley: I did that |
| 23:07.22 | RobotGuy | landley: Well, actually I did patch -p1 < blah.patch |
| 23:09.29 | ds2 | use git; it would make life so much easier |
| 23:09.54 | RobotGuy | git what? |
| 23:09.59 | landley | ds2: nah. hg clone http://kernel.org/hg/linux-2.6 |
| 23:10.34 | ds2 | landley: how is the branch management in hg? |
| 23:11.48 | landley | ds2: I don't really play with it much. |
| 23:11.52 | landley | I'm told it's there, and it works fine. |
| 23:12.15 | landley | Generally I just clone repositories when I need to. It hardlinks all the files if you do that locally, and breaks links when you commit... |
| 23:12.37 | landley | (It has actual branch management, but anything with "management" in it sounds way too much like work.) |
| 23:12.38 | ds2 | i know with git, you can either do a git pull to pickup the latest stuff into a local branch, then goto thehammer branch and just rebase it |
| 23:12.57 | landley | I do "hg pull -u". |
| 23:13.02 | landley | Pull, and update to tip. |
| 23:13.17 | ds2 | handles the patching/unpatching really nice + it lets you quickly figure out what got changed with the local history |
| 23:13.30 | ds2 | problem is you don't always want your stuff at the tip |
| 23:13.59 | landley | I know that mercurial _can_ do what you're talking about, I'm just saying that I've never needed to do it. |
| 23:14.09 | landley | You're talking to someone who wrote a git tutorial aimed at people who don't want to deal with branching. |
| 23:14.20 | ds2 | ah I see. hehe |
| 23:14.23 | landley | http://kernel.org/doc/local/git-quick.html |
| 23:14.45 | ds2 | after years of CVS torture... I _LIKE_ real branching and real ability to have a change list |
| 23:14.48 | landley | (It's basically "the minimum you need to be able to know about git to do a git-bisect without cargo cult programming.) |
| 23:15.05 | landley | And then hg merged bisect into mainline instead of as an extension, and I stopped caring again. |
| 23:15.22 | landley | ds2: mercurial is light years ahead of cvs. |
| 23:15.32 | landley | ANY modern source control system is light years ahead of CVS. :) |
| 23:15.45 | ds2 | the other nice thing is being able to do 13324324234234 things at once and let git keep track of my state by keeping things in a branch |
| 23:15.54 | ds2 | landley: yep, but people still like using cvs |
| 23:16.03 | landley | ds2: "like" isn't the right word there. |
| 23:16.09 | landley | The qemu repo recently moved to svn (FINALLY!) |
| 23:16.43 | landley | And the main reason I forked tinycc off of tcc instead of trying to take over the old project was the insistence on keeping the old repository, in CVS. |
| 23:17.04 | ds2 | heh... religious reasons for wanting to keep history intact? |
| 23:17.14 | landley | Not sure. |
| 23:17.35 | landley | I could dig up Fabrice's old message, but it was the main condition he put on a successor maintainer: he didn't want the project moving its repository elsewhere. |
| 23:17.45 | landley | And so I forked it instead. (Got his permission to do so, even.) |
| 23:17.48 | landley | Go figure. |
| 23:17.54 | ds2 | nice |
| 23:18.07 | landley | shrugs. |
| 23:18.15 | ds2 | btw, is TCC ready to generate bare metal code? |
| 23:18.17 | landley | I need about a year to sit down and revamp the entire thing. |
| 23:18.23 | ds2 | I should say TinyCC |
| 23:18.24 | landley | ds2: define bare metal. |
| 23:18.49 | ds2 | no OS. crt0 is part of the CPU startup vector that sets up stack and everything else in the proc. |
| 23:19.21 | landley | ds2: http://fabrice.bellard.free.fr/tcc/tccboot.html is from 2004. |
| 23:19.44 | landley | The problem is it doesn't compile an _unmodified_ Linux kernel. (And the one he modified was 2.4. And it was only a subset of the full kernel.) |
| 23:20.04 | landley | My definition of a 1.0 release is "builds an complete unmodified LInux kernel for x86 target". |
| 23:20.43 | landley | Unfortunately, shortly after that is when the project stalled. |
| 23:20.50 | landley | QEMU took off, and sucked up all Fabrice's time. |
| 23:21.11 | landley | (Ironic considering that qemu started as a conceptual fork off of tcc.) |
| 23:21.39 | landley | Once he had multiple code generators and it worked fast enough to treat C as an executable format, he started looking at multiple source formats... |
| 23:21.51 | ds2 | hmmm |
| 23:21.58 | ds2 | maybe an alternative is to fork the linux kernel |
| 23:22.12 | landley | And machine code was an interesting source format to dynamically recompile... |
| 23:22.16 | landley | ds2: er, no. |
| 23:22.42 | landley | I might come up with my own build system to replace kbuild, but it would compile the same C/assembly files. |
| 23:22.59 | landley | And a new build system for the kernel is as big a project as the ~5 I've got now... |
| 23:23.18 | ds2 | why not fork the linux kernel? |
| 23:23.24 | landley | No. |
| 23:23.56 | ds2 | but why? |
| 23:27.55 | RobotGuy | Hammer patch does not apply to rc8 clean, at least not for me. It figures. |
| 23:28.27 | ds2 | where are the rejects? |
| 23:29.48 | Dr{Afk} | still does not know why anyone would touch 2.6.25.. |
| 23:30.50 | RobotGuy | Dr{Afk}: Better drivers not supported in 2.6.22 |
| 23:32.46 | Dr{Afk} | all odd numbered releases of linux are not for production ever. That is where linus and friends break everything.. |
| 23:33.47 | RobotGuy | Dr{Afk}: I know that. |
| 23:35.17 | Dr{Afk} | they are doing some serious refactoring on 25 in prep for 26 but most driver support wont be ready till late late 25 or early 26 some driver developers wont even touch odd # releases.. because they know that the core of linux will change in the odd releases and will be in constant flux. |
| 23:36.11 | *** join/#edev ico2_ (n=ico2@modem-3401.chameleon.dialup.pol.co.uk) |
| 23:36.17 | Dr{Afk} | im sticking with 2.6.24 for most of my production stuff |
| 23:37.36 | Dr{Afk} | or for the hammer 2.6.22.10 seems stable. |
| 23:40.25 | RobotGuy | Dr{Afk}: Well, I don't have a patch for just 2.6.24. |
| 23:41.04 | Dr{Afk} | what is the patch for? |
| 23:41.58 | RobotGuy | Dr{Afk}: 2.6.25-rc7 |
| 23:42.07 | Dr{Afk} | sorry i ment what is in the patch. |
| 23:42.25 | RobotGuy | Dr{Afk}: Goes from 2.6.24 to 2.6.25-rc8 (not rc7). |
| 23:43.12 | RobotGuy | Dr{Afk}: Drivers for Ralink based USB WiFi adapters. |
| 23:45.08 | landley | drafk: er, no. The odd numbered release thing went away years ago. |
| 23:45.08 | *** join/#edev prpplague (n=dave123_@ppp-70-242-113-58.dsl.rcsntx.swbell.net) |
| 23:45.14 | prpplague | wmat ping |
| 23:45.19 | landley | Now they're all broken. That's why they have a fourth level of dot releases. :) |
| 23:45.41 | prpplague | landley what are we talking about? |
| 23:45.47 | prpplague | landley kernel? |
| 23:46.07 | ds2 | FORK THE KERNEL FORK THE KERNEL ;) |
| 23:46.14 | RobotGuy | can't get the hammer patch applied after patching to 2.6.25-rc8 |
| 23:46.16 | landley | prpplague: no. What? |
| 23:46.37 | landley | No, I was responding to DrWho's thing about odd numbered releases being broken. |
| 23:46.52 | landley | That went away in... 2003? |
| 23:46.57 | landley | Whenever 2.6.0 shipped. |
| 23:47.02 | RobotGuy | landley: Odd numbered releases are just development releases. |
| 23:47.10 | landley | RobotGuy: nope. |
| 23:47.30 | landley | Each release has a -rc1 period that accepts arbitrary new changes. |
| 23:47.40 | landley | 2.6.24 had one of these, 2.6.22 had one of these, and 2.6.26 will too. |
| 23:47.51 | landley | I'm hoping to get the hammer patch into mainline via 2.6.26-rc1. |
| 23:48.08 | RobotGuy | I don't care about all that though. I just care about the drivers I need right now. |
| 23:48.27 | landley | http://lwn.net/Articles/94386/ |
| 23:48.38 | *** join/#edev prpplague (n=dave123_@ppp-70-242-113-58.dsl.rcsntx.swbell.net) |
| 23:48.50 | prpplague | grumbles |
| 23:49.03 | landley | ds2: why do you want a kernel fork? |
| 23:49.35 | prpplague | wmat ping |
| 23:49.43 | RobotGuy | I can not get the hammer patch applied to 2.6.25-rc8 |
| 23:49.48 | Dr{Afk} | um landley just read the lkml and linus's chatter he is doing it again then. they openly are talking about the mejor changes that they hope to stabalize only in 2.4.26 |
| 23:49.59 | prpplague | RobotGuy why? |
| 23:50.12 | ds2 | landley: Cuz the mainline is getting too bloated |
| 23:50.17 | RobotGuy | prpplague: No idea. It gets a rej at the end. |
| 23:50.23 | ds2 | I want a smaller more manageable kernel |
| 23:50.24 | landley | DrAFK: do you mean he's doing another -rc1? |
| 23:50.37 | RobotGuy | People say it should apply but I can't get it to apply. |
| 23:51.02 | prpplague | RobotGuy you are probably using the patch posted on the wiki and not the posted to LAK |
| 23:51.02 | landley | ds2: ds2: http://www.selenic.com/linux-tiny/ |
| 23:51.31 | landley | Although these days it moved to http://elinux.org/Linux_Tiny when it got a new maintainer, but that site's still down. |
| 23:51.33 | RobotGuy | prpplague: This confusing me. I never know what to take from where. |
| 23:51.35 | ds2 | landley: I am aware of the little projects but it seems for every little thing the project does... there are 10 other things the main line will add to bloat it more |
| 23:51.48 | landley | ds2: I wont' fork the kernel, but I do have a related project. |
| 23:51.52 | landley | I want to build a "hello world" kernel. |
| 23:52.08 | landley | Outside the normal build system, compile just enough files to boot up the kernel and say "hello world". Nothing more. |
| 23:52.14 | ds2 | a fork could potentially combine all these small system projects into one tree |
| 23:52.34 | landley | Grab some early_printk() stuff and the start.S bits, and run a hello world function instead of init/main.c. |
| 23:52.38 | prpplague | RobotGuy just forwarded an email with the patch |
| 23:52.43 | RobotGuy | ds2: You lose a major base of software if you do that. |
| 23:52.44 | landley | Alas, doing this is nontrivial. |
| 23:52.48 | RobotGuy | prpplague: OK |
| 23:52.57 | prpplague | RobotGuy you have to realize therer are almost always going to be two trees |
| 23:53.09 | landley | Once I've got a hello world kernel, I can start adding in other bits as I isolate them and understand what each one does. |
| 23:53.09 | ds2 | RobotGuy: I think it is worth while given the amount and pace of bloat |
| 23:53.24 | prpplague | RobotGuy one being the one "make it easy" and the other to add "core support" |
| 23:53.36 | landley | ds2: might I interest you in one of the BSDs, or a 2.2 linux kernel? |
| 23:53.54 | ds2 | landley: 2.4 or 2.0 is sufficent |
| 23:54.10 | ds2 | don't like 2.2...bad SMP behavior |
| 23:54.32 | landley | You want a small streamlined kernel, and you want SMP. |
| 23:54.34 | landley | Uh-huh. |
| 23:54.45 | ds2 | sure :) |
| 23:54.48 | landley | wanders away from this conversation to do something more productive. |
| 23:54.56 | ds2 | otherwise, I don't see a reason go beyond 2.0 |
| 23:55.17 | prpplague | landley hehe |
| 23:55.18 | landley | Specifically, reading Wil Wheaton's most recent star trek episode review on tvsquad. |
| 23:55.34 | prpplague | landley oh ? url? |
| 23:56.08 | prpplague | i can't believe i wasted 1/2 a day on that darn recursion error |
| 23:57.10 | Dr{Afk} | landley rc1,2 etc are to stabalize the release they are on i get that but that does not change the fact that they do "mejor" changes on odd release.. |
| 23:58.22 | prpplague | ho ho hum |
| 23:58.28 | prpplague | lcd fun tomorrow |