| 00:23.01 | *** join/#edev redblue (i=star@ppp175.108-253-207.mtl.mt.videotron.ca) |
| 00:42.09 | Mirell | RobotGuy: Yep |
| 00:42.27 | Mirell | First they thought they were secure with analog trunking |
| 00:42.30 | Mirell | Now with digital |
| 00:42.31 | RobotGuy | They give SSN's and everything |
| 00:43.25 | SpeedEvil | idn't the digital pretty secure? |
| 00:43.35 | Mirell | SpeedEvil: It's not encrypted. |
| 00:44.21 | SpeedEvil | oh |
| 00:45.56 | SpeedEvil | you're listening? |
| 00:47.21 | RobotGuy | I listen sometimes, but I only pick up on specific locations and such |
| 00:49.27 | RobotGuy | I just listen for locations where friends live, and the area around me mostly. |
| 00:49.38 | SpeedEvil | Ah. |
| 00:49.45 | SpeedEvil | Want to get one up on the zombies. |
| 00:50.27 | RobotGuy | Zombies? |
| 00:50.53 | SpeedEvil | Well - when the zombies arrive, you'll be forewarned. |
| 00:51.04 | RobotGuy | I am having trouble with a Makefile. |
| 00:51.33 | RobotGuy | SpeedEvil: Oh, uh huh, sure, and what is it you have been smoking and/or drinking tonight? |
| 00:51.47 | SpeedEvil | is suffering from a lack of sleep. |
| 00:52.19 | MonMotha | well, in theory, knowing an SSN and the name/address that goes with it shouldn't do you any good |
| 00:52.28 | MonMotha | sadly, that's not how practice behaves |
| 00:52.29 | SpeedEvil | yeah. |
| 00:52.47 | SpeedEvil | It's way too convenient as a UUID |
| 00:52.56 | MonMotha | even though it is neither universal nor unique |
| 00:53.22 | MonMotha | there are several documented instances of SSN collisions happening, and it only applies to US citizens (and companies, if you include TINs) |
| 00:53.41 | MonMotha | well, and people in the US who have otherwise attained an SSN without being citizens for work purposes |
| 00:57.09 | RobotGuy | Can somebody explain to me what this segment of Makefile code is doing? http://www.pastebin.ca/964110 |
| 00:58.07 | RobotGuy | I hate giving my SSN out. Who actually has legal right to have and use it? It's an ACCOUNT number. |
| 00:58.23 | *** join/#edev rodserling (n=rodserli@unaffiliated/rodserling) |
| 01:00.37 | RobotGuy | Many places want to use SSN's as account numbers, but this is not legal, but they often refuse services if you do not give them the SSN. |
| 01:02.32 | RobotGuy | Everybody pretty much wants your SSN, so how to you prevent somebody from misusing it? |
| 01:05.31 | RobotGuy | http://www.pastebin.ca/964110 |
| 01:05.43 | RobotGuy | Anybody have a clue what this Makefile code is doing? |
| 01:05.51 | RobotGuy | It's frustrating me. |
| 01:06.22 | RobotGuy | I'm about to toss buildroot out the window and forget I ever knew how to use it. |
| 01:06.40 | RobotGuy | It won't let me put stuff where I need it. :( |
| 01:07.52 | RobotGuy | I'm just going to give it up for now. I can't get anything to go where I need it. Bye for now. |
| 01:07.56 | *** part/#edev RobotGuy (n=robotguy@dsl093-038-072.pdx1.dsl.speakeasy.net) |
| 01:40.26 | *** join/#edev ccjoe (n=ccjoe@75-169-37-101.slkc.qwest.net) |
| 02:36.34 | *** join/#edev chouimat|tv (n=dieu@r2351064.cidc.net) |
| 02:39.13 | *** join/#edev redblue (i=star@ppp127.108-253-207.mtl.mt.videotron.ca) |
| 02:40.40 | *** join/#edev redblue (i=star@ppp127.108-253-207.mtl.mt.videotron.ca) |
| 04:00.41 | *** join/#edev DrZimmerman (n=theo@tbaumgartner.broker.freenet6.net) |
| 04:17.50 | *** join/#edev rodserling (n=rodserli@unaffiliated/rodserling) |
| 05:10.38 | *** join/#edev harpal (n=harpal@121.246.75.165) |
| 06:13.44 | *** join/#edev Shingoshi (n=shingosh@c-67-169-195-234.hsd1.or.comcast.net) |
| 06:13.44 | *** join/#edev ohsix (i=ohsix@66.220.111.222) [NETSPLIT VICTIM] |
| 06:37.36 | *** join/#edev howtoo (n=howtoo@195.159.62.66) |
| 07:49.22 | *** join/#edev steliosk (n=Stelios@79.131.72.44) |
| 07:52.35 | *** join/#edev lyakh (n=lyakh@p57BD2CC9.dip0.t-ipconnect.de) |
| 07:53.05 | *** join/#edev redblue (i=star@ppp091.108-253-207.mtl.mt.videotron.ca) |
| 07:53.43 | *** join/#edev dijenerate (n=dijenera@69.73.220.145) |
| 07:55.43 | *** join/#edev ejdav_gon (n=james@d75-157-13-250.bchsia.telus.net) |
| 07:57.56 | *** join/#edev _trine (n=trine@38.100.90.3) [NETSPLIT VICTIM] |
| 08:06.30 | *** join/#edev linac (n=lin@121.41.37.219) |
| 08:09.11 | *** join/#edev jdav_gone (n=james@d75-157-13-250.bchsia.telus.net) [NETSPLIT VICTIM] |
| 08:13.37 | *** join/#edev wmat (n=btraynor@gromit.mixdown.ca) |
| 10:00.59 | *** join/#edev samuel1 (n=samuelr@dsl093-061-094.pit1.dsl.speakeasy.net) |
| 10:42.08 | *** join/#edev samuelr (n=samuelr@dsl093-061-094.pit1.dsl.speakeasy.net) |
| 11:07.35 | *** join/#edev hw_ (n=hw@p578b3905.dip0.t-ipconnect.de) |
| 11:27.39 | *** join/#edev RobotGuy (n=robotguy@dsl093-038-072.pdx1.dsl.speakeasy.net) |
| 11:49.34 | *** join/#edev bbradley (n=bbradley@82-32-35-42.cable.ubr05.azte.blueyonder.co.uk) |
| 12:12.08 | *** join/#edev sjhill (n=sjhill@real.realitydiluted.com) |
| 12:12.08 | *** mode/#edev [+o sjhill] by ChanServ |
| 12:20.53 | RobotGuy | sjhill: Greetings |
| 12:25.02 | *** join/#edev [AD]Turbo (n=turbo@host131-128-dynamic.57-82-r.retail.telecomitalia.it) |
| 12:25.18 | [AD]Turbo | yo |
| 12:28.14 | RobotGuy | Hello |
| 12:37.37 | Mirell | *yawn* |
| 12:37.50 | RobotGuy | Mirell: G'Morning |
| 13:22.38 | *** join/#edev prpplague (n=dave@mail.americanmicrosystems.com) |
| 13:23.08 | *** mode/#edev [+o prpplague] by ChanServ |
| 13:26.43 | sjhill | RobotGuy: hello |
| 13:26.56 | RobotGuy | Hi |
| 13:26.57 | sjhill | RobotGuy: you still interested in those batteries? |
| 13:27.22 | RobotGuy | sjhill: I don't think so, sorry. It looks like the required charger would be expensive. |
| 13:27.35 | sjhill | RobotGuy: really? how much is the charger? |
| 13:27.57 | RobotGuy | sjhill: Truth is, I can't tell what charger would work with them. |
| 13:28.20 | RobotGuy | I'm just going based on what I have seen for chargers for similar batteries. |
| 13:28.28 | RobotGuy | I could be way off base though. |
| 13:28.59 | sjhill | hmm |
| 13:29.19 | sjhill | well, the price of each battery is free, so you'd just have to pay shipping |
| 13:29.34 | sjhill | i'll research the chargers too |
| 13:29.44 | RobotGuy | sjhill: Yes, I know, and it is still tempting. I need to find a charger for them though. |
| 13:30.02 | RobotGuy | If I could find a charger I would take 2 of them. |
| 13:30.15 | sjhill | k |
| 13:30.53 | RobotGuy | There are six threads on the forums that need to be deleted. |
| 13:36.33 | sjhill | which forums? |
| 13:37.16 | RobotGuy | The TCT forums |
| 13:45.49 | *** join/#edev Chocobo (n=swinchen@strongbad.eece.maine.edu) |
| 13:45.51 | *** join/#edev swinchen (n=swinchen@strongbad.eece.maine.edu) |
| 13:47.14 | chouimat|work | morning |
| 13:56.27 | prpplague | RobotGuy: done |
| 13:57.01 | RobotGuy | prpplague: Cool. |
| 13:58.50 | prpplague | RobotGuy: seems to have a good thread going on the HBB stuff |
| 13:59.15 | RobotGuy | prpplague: Yes, and I am trying to get more people involved. |
| 13:59.57 | RobotGuy | prpplague: There really are just three things that are super important to me. |
| 14:00.09 | prpplague | the real question is can you get a concensis(sp?) on the features |
| 14:00.56 | RobotGuy | prpplague: I believe I can. |
| 14:01.06 | RobotGuy | consensus |
| 14:01.10 | RobotGuy | grins |
| 14:01.31 | RobotGuy | Sienna and I are nearly in agreement now. |
| 14:02.27 | RobotGuy | At least the two of us are going for a more general configuration usable in many ways, not just robotics. Form factor is important. |
| 14:03.46 | sjhill | prpplague: wth is my nail board?! |
| 14:03.53 | sjhill | :) |
| 14:03.57 | prpplague | sjhill: hehe |
| 14:04.08 | prpplague | sjhill: did i promise you a beta unit? |
| 14:04.27 | sjhill | you just told me two weeks, two weeks ago i think |
| 14:04.53 | prpplague | sjhill: ahh, we are running a little behind, we found a couple minor layout issues |
| 14:05.00 | sjhill | k |
| 14:05.10 | prpplague | sjhill: assuming no additional changes, we will go to production a week from today |
| 14:05.10 | sjhill | but sure, i can be a beta tester |
| 14:05.28 | prpplague | sjhill: i'll see if we have any additional beta units |
| 14:06.12 | sjhill | cool, thx |
| 14:06.40 | prpplague | we need some really cool killer apps for it |
| 14:07.43 | *** join/#edev adyer (n=adyer@adsl-69-223-13-78.dsl.chcgil.ameritech.net) |
| 14:09.00 | RobotGuy | prpplague: Would a real live robot running Python on Hammer be a good thing then? My goal is to have W.A.L.T.E.R. fully operational with Hammer before the Linux Plumber's Conference. |
| 14:09.02 | *** join/#edev howtoo (n=howtoo@195.159.62.66) |
| 14:09.23 | sjhill | Linux Plumber's Conference? wth? |
| 14:10.15 | prpplague | RobotGuy: no clue, i'm not a fan of python(or any of the others such as perl and php) and i don't know much about reboot design |
| 14:11.01 | RobotGuy | prpplague: I just thought a real robot running with Hammer would be a nice demo. The programming language used is irelevent. |
| 14:11.22 | sjhill | use LISP, real men use LISP |
| 14:11.35 | RobotGuy | I've already got a Python IRC bot working in standard configuration and am going to expand it to include a web interface. |
| 14:11.52 | chouimat|work | sjhill: cool a linux powered toilet ;) |
| 14:11.55 | RobotGuy | sjhill: I don't have brain room for another programming language. |
| 14:12.10 | sjhill | chouimat|work: :) |
| 14:18.59 | prpplague | ho ho hum |
| 14:19.05 | prpplague | has a case of the mondays |
| 14:19.21 | sjhill | i've had that everyday for the past couple weeks |
| 14:19.36 | RobotGuy | prpplague: I got /mnt/flash back. :) |
| 14:20.50 | *** join/#edev redblue (i=star@ppp150.108-253-207.mtl.mt.videotron.ca) |
| 14:21.17 | RobotGuy | prpplague: You just do regular C? |
| 14:22.20 | prpplague | RobotGuy: wonder why ol' hylands is using the smirf bt module instead of the dip or smd package |
| 14:22.24 | prpplague | RobotGuy: yea |
| 14:22.50 | prpplague | RobotGuy: no expert, just enough to fool people |
| 14:22.58 | RobotGuy | prpplague: No idea why he does that. I have the Bluetooth DIP module (NXP BGB203) |
| 14:23.18 | RobotGuy | prpplague: I'm far from an expert either, but I do pretty well. |
| 14:24.14 | prpplague | RobotGuy: there been much discussion about the HBB on any of the mailing lists? |
| 14:24.19 | prpplague | or other forums? |
| 14:26.41 | RobotGuy | prpplague: Not so far. I have started threads on Trossen Robotics and Lynxmotion forums, and the PARTS list so far. |
| 14:27.20 | prpplague | wonders where cosmic is |
| 14:27.27 | prpplague | sjhill: hehe |
| 14:27.39 | prpplague | sjhill: you not see the info on LPC before? |
| 14:27.53 | sjhill | is it real? |
| 14:27.55 | prpplague | sjhill: its basically where everyone is going instead of OLS this year |
| 14:27.59 | prpplague | sjhill: indeed |
| 14:28.09 | RobotGuy | prpplague: Will you be at LPC? |
| 14:28.36 | prpplague | RobotGuy: there is some discussion about going, but i doubt either AML or TCT will pay for it, so most likely no |
| 14:28.57 | RobotGuy | prpplague: Too bad. |
| 14:29.09 | prpplague | RobotGuy: managment for AML and TCT are not too keen on techincal conferences |
| 14:29.27 | RobotGuy | prpplague: That is really sad. |
| 14:29.36 | prpplague | indeed |
| 14:30.21 | chouimat|work | prpplague: I'm working about 5 minutes walk from the rideau center and I'm not sure if I will go |
| 14:30.44 | prpplague | chouimat|work: yea from what i hear, OLS has gotten very stale |
| 14:30.47 | RobotGuy | chouimat|work: LPC is in Portland, OR. |
| 14:31.04 | prpplague | chouimat|work: not that i'd know, since i've not been in years |
| 14:31.18 | prpplague | chouimat|work: not since i worked for abcs |
| 14:31.34 | chouimat|work | prpplague: like 99.5% of all technical conference ... big business get into them and the fun dies |
| 14:37.15 | RobotGuy | prpplague: I just got a PM from Head Honcho of Lynxmotion about the name Hammer Bot Board. He has a bit of an issue with that. |
| 14:37.34 | RobotGuy | is really starting to dislike people in general |
| 14:37.54 | chouimat|work | RobotGuy: welcome to the club |
| 14:38.29 | RobotGuy | I am going to terminate the BRAT Project I have been working on with Lynxmotion. |
| 14:38.39 | RobotGuy | Gonna ship everything back |
| 14:39.27 | RobotGuy | prpplague: http://www.pastebin.ca/964745 |
| 14:43.54 | prpplague | looks |
| 14:44.54 | prpplague | RobotGuy: thats not really an issue as we had no intentions of calling it bot board |
| 14:45.21 | RobotGuy | prpplague: Well, what do I call it then? |
| 14:45.34 | RobotGuy | prpplague: This is the next PM I got: You can call it whatever you want, but just don't link to it from my forum. |
| 14:46.03 | RobotGuy | I'll remain a Lynxmotion customer, but I am done working on projects with them and their forums. |
| 14:46.57 | prpplague | RobotGuy: was t his via email? |
| 14:47.18 | RobotGuy | prpplague: Lynxmotion forum PM |
| 14:47.19 | prpplague | RobotGuy: or via the forum? |
| 14:48.52 | prpplague | RobotGuy: ahh, ok, do me a favor and let him know two things, 1) you aren't an employee of TCT 2)that you are aware of TCT has a robotics development platform under development, but that a name has not been announced |
| 14:49.47 | RobotGuy | prpplague: I have always stated I am not an employee of TCT. |
| 14:50.02 | RobotGuy | prpplague: He knows I do not work for anyone. |
| 14:50.33 | prpplague | RobotGuy: yea, but based on what you posted, it sounds like he thinks you are developing the board |
| 14:51.01 | RobotGuy | prpplague: I'll make sure he knows in my next PM to him. |
| 14:51.27 | RobotGuy | prpplague: I don't want to do anything that might reflect badly on TCT. |
| 14:51.56 | RobotGuy | I just hit post #2300 on the Lynxmotion forums. |
| 14:55.03 | RobotGuy | prpplague: Done as you requested. |
| 14:55.48 | prpplague | RobotGuy: thanks, please let me know of any responses |
| 14:55.59 | RobotGuy | prpplague: I will. |
| 14:56.20 | RobotGuy | I'm not too happy with Head Honcho right now. |
| 14:56.57 | prpplague | RobotGuy: he's there to make money |
| 14:58.03 | *** join/#edev CosmicPenguin (n=nobody@163.181.251.103) |
| 14:58.03 | *** mode/#edev [+o CosmicPenguin] by ChanServ |
| 14:58.59 | prpplague | CosmicPenguin: greetings |
| 15:00.15 | CosmicPenguin | <PROTECTED> |
| 15:01.00 | RobotGuy | prpplague: I realize that, but I strongly disagree with him now on at least two things. He deteted my reference to the HBB being in production and linking to it. Yet he has not deleted any other reference I made to it as Hammer Bot Board. |
| 15:02.15 | prpplague | RobotGuy: well thats his choice |
| 15:02.41 | RobotGuy | prpplague: I have a real problem with selective censorship. |
| 15:06.09 | *** join/#edev gandhijee (n=root@host-66-202-34-165.spr.choiceone.net) |
| 15:08.05 | RobotGuy | Did I past the one where he says I am rude? |
| 15:08.19 | *** join/#edev redblue (i=star@ppp125.109-253-207.mtl.mt.videotron.ca) |
| 15:10.41 | RobotGuy | If I have done something wrong, I will do what needs to be done to correct it, even to the point of not being a Hammer evaluator anymore. I will do as you ask. I gotta get off the net for a bit. |
| 15:15.02 | *** join/#edev ico2 (n=ico2@modem-863.cheetah.dialup.pol.co.uk) |
| 15:22.04 | *** join/#edev ico2 (n=ico2@modem-863.cheetah.dialup.pol.co.uk) |
| 15:22.26 | prpplague | robotguy is a nice person, but he seems oblivious to the way the business world works |
| 15:22.45 | CosmicPenguin | As are most of the geeks of the world |
| 15:23.07 | CosmicPenguin | some by choice, some by ignorance |
| 15:23.18 | CosmicPenguin | Business isn't logical - thats why it sucks |
| 15:24.36 | adyer | CosmicPenguin: I would say it is mostly logical, but the underlying rules are a lot different |
| 15:24.58 | CosmicPenguin | Its definately not _binary_ logic |
| 15:27.58 | prpplague | CosmicPenguin: indeed |
| 15:29.51 | chouimat|work | seems we will have a free lunch soon here ... the drawback is you have to listen to a financial planner :( |
| 15:30.06 | *** join/#edev landley (n=landley@adsl-68-93-84-18.dsl.hstntx.swbell.net) |
| 15:30.10 | *** mode/#edev [+o landley] by ChanServ |
| 15:30.15 | CosmicPenguin | Gettinga little financial education never hurt anybody |
| 15:30.23 | adyer | tanstaafl :-) |
| 15:30.33 | prpplague | landley: greetings, hows the ear? |
| 15:30.53 | landley | Better. |
| 15:30.56 | landley | must finish OLS paper this evening. |
| 15:30.58 | landley | Due tomorrow. |
| 15:31.47 | prpplague | landley: doh |
| 15:36.05 | landley | Eh, I've known about it for a while. They're just a lot to write... |
| 15:36.06 | prpplague | landley: you'll have to get chouimat|work to show you all the good pubs at OLS |
| 15:36.22 | CosmicPenguin | good luck |
| 15:36.29 | CosmicPenguin | I've been trying to get him to show them to me for three years |
| 15:36.34 | prpplague | hehe |
| 15:36.34 | landley | prpplague: I don't drink. |
| 15:36.43 | chouimat|work | CosmicPenguin: now I can I have money :) |
| 15:36.44 | landley | I have Red Bulls at the blackthorn to show willing, though. |
| 15:36.51 | landley | Five two years ago. Six last year. |
| 15:36.51 | prpplague | hehe |
| 15:36.53 | *** join/#edev RobotGuy (n=robotguy@dsl093-038-072.pdx1.dsl.speakeasy.net) |
| 15:37.00 | CosmicPenguin | Ahh, the blackthorn |
| 15:37.06 | landley | Six is apparently the point at which I start slurring my words and my eyeball starts bleeding. |
| 15:37.09 | landley | Don't plan to do that again. |
| 15:37.09 | prpplague | RobotGuy: re |
| 15:37.20 | prpplague | landley: hehe |
| 15:37.34 | prpplague | CosmicPenguin: indeed, one of favorite places on the planet |
| 15:37.44 | landley | Alas, this is the last OLS. |
| 15:37.57 | landley | Like Atlanta Linux Showcase, they're moving out of the city they're named after. |
| 15:37.57 | samuelr | landley: Really? |
| 15:38.08 | landley | ALS didn't survive, I don't know what OLS will do. |
| 15:38.13 | CosmicPenguin | they are tearing down the conference centre |
| 15:38.34 | landley | I was waffling about whether or not to come this year anyway, but if it's going to be the last one... |
| 15:38.56 | landley | Most people I used to see there don't go anymore. Alan Cox wasn't there, Thomas Gleixner was last there in 2006... |
| 15:39.15 | CosmicPenguin | they all go down under now |
| 15:39.37 | CosmicPenguin | unfortunately, my fine employer takes a dim view of sending me half way around the world for a cut-rate operating system |
| 15:39.44 | prpplague | yea, rmk hasn't been to OLS in years |
| 15:40.25 | landley | There used to be four interesting conferences. The east and west LinuxWorld Expos, Atlanta LInux Showcase, and Ottawa Linux Symposium. |
| 15:40.41 | landley | ALS died, LWE turned pointy haired, and OLS is finally shuffling off into the sunset... |
| 15:41.09 | landley | There's still room for more than _one_, but the US still has an idiot in charge scaring people away and arresting foreign nationals without charges. |
| 15:41.26 | landley | Canada was a decent compromise, but still... |
| 15:42.10 | prpplague | indeed, no one in their right mind is going to have a tech conference in the US |
| 15:42.29 | CosmicPenguin | and they never will, regardless of who is in charge |
| 15:42.32 | prpplague | over half the people who would want to attend won't due to visa/passport issues |
| 15:43.51 | prpplague | CosmicPenguin: indeed |
| 15:43.56 | CosmicPenguin | Then again - the kernel summit is in Portland, so.... |
| 15:46.39 | landley | There was a big PPC conference in Austin a few months back. (I didn't go because it cost $500 to attend and this is a _hobby_ for me.) |
| 15:47.12 | landley | People still hold conferences here. I'm putting together an Austin version of Penguicon in fall 2009 because this is where I am. |
| 15:47.17 | landley | If people can't come, they can't come. |
| 15:47.24 | landley | But the ones who can should have a marvelous time. |
| 15:47.49 | landley | Besides, if I fly in Erik Andersen and Denis Vlasenko, we can have quite a fun panel on busybox. :) |
| 15:48.04 | landley | Maybe Mike Frysinger would be interested too, haven't asked him yet. |
| 15:48.42 | landley | Then again, I'm _not_ in my right mind and never was, so... |
| 15:48.48 | landley | Not exactly a counter-argument. |
| 15:52.03 | RobotGuy | I'm not even going to reply to his last PM. It would only make things worse, regardless of what I said. |
| 15:52.27 | RobotGuy | Ooops, wrong window. :( |
| 15:52.53 | RobotGuy | I'm screwing up too much today. Bye. |
| 16:09.58 | *** join/#edev ico2_ (n=ico2@modem-124.chimpanzee.dialup.pol.co.uk) |
| 16:25.25 | *** join/#edev RobotGuy (n=robotguy@dsl093-038-072.pdx1.dsl.speakeasy.net) |
| 16:28.59 | *** join/#edev landley (n=landley@adsl-68-93-84-18.dsl.hstntx.swbell.net) |
| 16:28.59 | *** mode/#edev [+o landley] by ChanServ |
| 16:29.46 | landley | Memo to self: rebooting the machine that's masquerading for you has side effects. |
| 16:29.55 | prpplague | landley: how ya feeling about the nail board these days? |
| 16:30.01 | prpplague | landley: hehe |
| 16:30.32 | prpplague | landley: i was ssh'd into a server once and did a shutdown thinking it was my local box |
| 16:30.58 | RobotGuy | landley: Did you get the 2.6.25-rc7 kernel to compile? |
| 16:31.28 | prpplague | RobotGuy: yea, he says so on his dev notes |
| 16:31.40 | prpplague | RobotGuy: the patch should be pretty solid |
| 16:31.52 | prpplague | RobotGuy: i pulled a couple of unused headers from the patch |
| 16:32.44 | RobotGuy | I could not get it to compile. |
| 16:33.16 | prpplague | RobotGuy: with what error? |
| 16:34.03 | RobotGuy | prpplague: I will paste the error if any after I do a fresh build. |
| 16:44.17 | RobotGuy | prpplague: I have it setup now, but do not see how to select ARM architecture. |
| 16:49.56 | RobotGuy | Interesting - I have to unplug and replug my WiFi adapter to get Hammer to recognize it. It will not configure since I have to do that. |
| 16:50.56 | RobotGuy | Am I stupid, or is there a secret way to select arm architecture with 2.6.25-rc7 kernel? |
| 16:53.15 | sjhill | landley: do you still work for TimeSys? |
| 16:55.08 | adyer | RobotGuy: usually you either override the ARCh setting in the main Makefile or add ARCH=arm on teh command line. Also need to set CROSS_COMPILE to the name of the cross compiler used |
| 16:55.44 | RobotGuy | adyer: It should be selectable in make menuconfig/ |
| 16:57.04 | prpplague | RobotGuy: all you should have to do is apply the patch |
| 16:57.08 | prpplague | RobotGuy: it sets the arm arch |
| 16:57.23 | prpplague | RobotGuy: then copy the hammer-config to .config and run make oldconfig |
| 16:57.35 | RobotGuy | OK, but what about the make menuconfig which is set for PC Compatible? |
| 16:57.52 | prpplague | RobotGuy: i am sure, i did not understand your question |
| 16:58.00 | prpplague | s/sure/sorry |
| 16:58.45 | RobotGuy | If I do not make any changes to the Makefile after patching, everything is set for PC Compatible when I do make menuconfig. |
| 16:59.13 | prpplague | RobotGuy: then something is wrong |
| 16:59.31 | prpplague | RobotGuy: where did you get the patch? |
| 16:59.48 | RobotGuy | prpplague: From www.kernel.org |
| 17:00.13 | prpplague | RobotGuy: hmm i wasn't aware that the hammer patch was available on kernel.org, you got a url? |
| 17:00.30 | RobotGuy | See above response. :) |
| 17:00.53 | prpplague | i am not seeing a hammer kernel patch at kernel.org |
| 17:00.57 | RobotGuy | I don't know that the hammer patch is in there. Where else would I get the patch from? |
| 17:01.11 | prpplague | RobotGuy: from elinux.org |
| 17:01.17 | RobotGuy | It's the patch to get to 2.6.25-rc7 |
| 17:01.29 | prpplague | http://www.elinux.org/Hammer_Board_Software_System |
| 17:02.01 | prpplague | RobotGuy: download 2.6.24, apply the 2.6.25-rc7 patch from kernel.org , then apply the 2.6.26-rc7-hammer patch |
| 17:03.06 | RobotGuy | OK, all I need is the hammer patch then. I already did the first two steps - unpacking 2.6.24 and applying the 2.6.25-rc7 patch from kernel.org |
| 17:03.31 | prpplague | RobotGuy: if you;ve already done the those, then yes all you ned is the hammer patch |
| 17:09.46 | *** join/#edev magnetron (n=magnetro@unaffiliated/magnetron) |
| 17:14.38 | landley | sjhill: no, not for a year now. |
| 17:15.13 | landley | RobotGuy: yes, I installed 2.6.25-rc7 on my nail board. |
| 17:15.20 | landley | That's checked into the Impact build, I think... |
| 17:15.35 | RobotGuy | landley: Cool. I am hoping to do the same on my Hammer. |
| 17:16.06 | RobotGuy | considers moving to the Firmware build system - at least to try it out. |
| 17:16.22 | landley | RobotGuy: menuconfig shows different stuff depending on what arch you selected. Selecting arch _in_ menuconfig is too late. |
| 17:16.37 | landley | They'd have to rearchitect the kconfig stuff a bit to let you select in there, and then you'd still have to specify your cross compiler. |
| 17:16.51 | landley | And then make sure your cross compiler was in the $PATH. Generally considered too confusing to do it that way, I think. |
| 17:17.23 | landley | RobotGuy: I'm happy to answer any questions about the stuff I maintain... :) |
| 17:17.28 | landley | If it doesn't work for you, I wanna _fix_ it. |
| 17:20.02 | landley | RobotGuy: I just checked in the hammer patch, so the kernel source it uses should be patched for hammer. |
| 17:20.22 | RobotGuy | Great! |
| 17:20.26 | landley | Keep in mind you still have to build your own hammer kernel because it doesn't have a sources/configs/hammer (maybe I should make one...) |
| 17:20.45 | landley | But you can use the source in the build/sources directory to do it after the build. |
| 17:20.51 | landley | And use the cross compiler it builds. |
| 17:20.55 | RobotGuy | landley: I think I am already building my own Hammer kernel. |
| 17:23.23 | landley | I just made an arm-hammer config file that should automatically build a hammer kernel instead of a qemu one. |
| 17:23.28 | landley | Testing it now. |
| 17:24.31 | landley | I need some kind of config syntax to say "only the kernel is different"... |
| 17:30.14 | RobotGuy | I finally got through the kernel configuration for 2.6.25-rc7-hammer. Had to turn a LOT of stuff off. |
| 17:31.36 | landley | RobotGuy: I just used the defconfig supplied in the patch. |
| 17:31.55 | RobotGuy | landley: You did not change ANYthing?? |
| 17:32.22 | RobotGuy | Hammer does not need a lot of the stuff that is turned on in the default. It would be cool if the Hammer patch handled this. |
| 17:33.10 | RobotGuy | Here is my error report: http://www.pastebin.ca/964916 |
| 17:36.11 | landley | RobotGuy: Ii was just trying to get it to boot. |
| 17:36.18 | RobotGuy | I ran make mrproper and restored my config file (glad I saved it first). Looks like it is working now. |
| 17:36.25 | landley | Also, keep in mind "make defconfig" != "cp arch/arm/blah/hammer_defconfig .config" |
| 17:36.33 | landley | Make defconfig is an anachronism. |
| 17:36.46 | RobotGuy | landley: I never use defconfig |
| 17:37.02 | landley | I trimmed it down to a miniconfig, but other than that I haven't tweaked it yet. Probably will later. |
| 17:37.19 | landley | At the time I was just trying to get something to _boot_, which I did, declared victory, and moved on to the other 30 things I had to do that weekend... |
| 17:38.00 | RobotGuy | If this works it will be great! Then I can use the new Ralink module. Maybe it will fix my WiFi. |
| 17:38.41 | *** join/#edev emil|k (i=emil@e179237105.adsl.alicedsl.de) |
| 17:53.43 | RobotGuy | prpplague: Last week you were talking about using JFFS2 and getting around 24 MB of free memory. I do not fully understand what you are talking about. |
| 17:56.48 | SpeedEvil | RobotGuy: the current images - as I understand it. Uncompress a filesystem from the flash into RAM and use it there. |
| 17:57.06 | SpeedEvil | jffs allows you to use a filesystem on the flash - without decompressing it into RAM. |
| 17:57.27 | RobotGuy | SpeedEvil: I know what JFFS2 allows. :) That is not my question. |
| 17:57.58 | SpeedEvil | this means that you don't have the RAM taken up by a filesystem image - so more RAM |
| 17:59.51 | RobotGuy | SpeedEvil: What I do not understand is how I am supposed to set it up. Does it have to do with my initrd? |
| 18:05.31 | SpeedEvil | In general - there are two options - it's been a long time since I touched this. |
| 18:05.49 | SpeedEvil | First - the bootloader loads the kernel + compressed RAMdisk into RAM. |
| 18:06.10 | SpeedEvil | Then the kernel boots, and uncompresses the compressed RAMdisk into RAM, and deletes the compressed image. |
| 18:06.18 | SpeedEvil | Second. |
| 18:07.07 | SpeedEvil | The kernel boots, and is configured with the use of the command line and other methods to understand how to mount the filesystem - jffs - direct from flash. |
| 18:07.12 | RobotGuy | I understand how that works. I want to know how JFFS2 affects everything if it is used for the RAMDISK and initrd. |
| 18:07.26 | prpplague | returns from lunch |
| 18:07.37 | SpeedEvil | Sorry, I don't understand what your problem is. |
| 18:07.45 | RobotGuy | sighs |
| 18:08.01 | RobotGuy | I guess I can not communicate. |
| 18:08.13 | SpeedEvil | Oh - and then there is third - a teeny initial filesystem which has the tools to load the larger filesystem using scripts and stuff. |
| 18:08.27 | SpeedEvil | This is typically loaded by the first method. |
| 18:08.27 | prpplague | RobotGuy: use JFFS2 instead of initrd does not affect your rootfs layout |
| 18:08.55 | prpplague | RobotGuy: picture it as the exact same contents of your initrd, but in a jffs2 instead |
| 18:09.07 | RobotGuy | prpplague: How do I do that? I am obviously not understanding something. I already have JFFS2 for the flash (/mnt/flash). |
| 18:09.37 | prpplague | RobotGuy: simply change you buildroot option to generate a jffs2 image instead of the initrd image |
| 18:09.48 | prpplague | RobotGuy: then flash your jffs2 image to the jffs2 partition |
| 18:11.16 | RobotGuy | prpplague: JFFS2 instead of ext2, right? |
| 18:11.52 | prpplague | RobotGuy: yep |
| 18:12.33 | *** join/#edev linac (n=lin@121.41.34.123) |
| 18:14.26 | RobotGuy | OK, I can do that. 1) Then I just put it on /mnt/flash? I do not know how to flash to that partition. 2) Can't I just copy everything to it? 3) How do I get the loader to load from /mnt/flash then? |
| 18:16.19 | prpplague | RobotGuy: to copy the jffs2 image to flash, simply boot to apex, erase the flash area, then transfer the file up to sdram , then copy it to flash |
| 18:16.36 | RobotGuy | 4) What happens if I make an image a little bit smaller than my /mnt/flash partition? |
| 18:16.54 | prpplague | RobotGuy: change the root= command line to boot to /dev/mtdblock2 |
| 18:17.09 | RobotGuy | OK, boot to APEX (Ctrl/C) to stop the boot process. |
| 18:17.12 | RobotGuy | Ahh, right. |
| 18:17.34 | prpplague | RobotGuy: you can also do it from openocd, but its much easier to work with apex |
| 18:17.55 | RobotGuy | prpplague: I will do it from APEX, which I am more familiar with at present. |
| 18:18.45 | RobotGuy | I am learning, if slowly. I eventually get it. |
| 18:18.45 | SpeedEvil | prpplague: somewhat off-topic. I'm currently looking for SDRAM in the 8M*8 range - would you have any suggested sources in the 100-1K range? |
| 18:19.13 | prpplague | SpeedEvil: is that 8 megabit or or 8megabyte? |
| 18:19.19 | SpeedEvil | 64 megabit |
| 18:19.25 | RobotGuy | prpplague: /dev/mntblock2? Wouldn't it be /dev/mntblock1? |
| 18:19.42 | prpplague | RobotGuy: no, /dev/mtdblock2 |
| 18:19.52 | prpplague | RobotGuy: same as the mount as /mnt/flash |
| 18:19.59 | RobotGuy | prpplague: I only have /dev/mntblock1 |
| 18:20.11 | prpplague | RobotGuy: are you sure about that? |
| 18:20.21 | prpplague | RobotGuy: oh sorry my bad |
| 18:20.30 | prpplague | RobotGuy: i forget that it has mtdbock0 |
| 18:20.44 | SpeedEvil | I'm bit-banging it with a cortex micro. |
| 18:21.02 | RobotGuy | prpplague: Yes, /mnt/mtdblock1 is what I have, nothing else for flash. |
| 18:21.26 | prpplague | RobotGuy: yea there is a mtdblock0, which is where the system is |
| 18:21.41 | prpplague | SpeedEvil: uh? bit banging sdram? |
| 18:22.02 | RobotGuy | prpplague: Ah, OK. Does this mean I could have an image as large as /mnt/mtdblock1 is (about 12 MB)? |
| 18:22.31 | SpeedEvil | prpplague: yes. It seems to work - according to the timing diagrams. And I'm using the auto-increment write to good use. |
| 18:22.39 | prpplague | RobotGuy: as you won't be using the initrd, you could have up to around 14mb |
| 18:22.58 | RobotGuy | prpplague: Ah, GOOD! :D:D Muahahahahahahaha ;) |
| 18:23.00 | prpplague | SpeedEvil: i would not have thought that it would be possible or practical |
| 18:24.08 | RobotGuy | prpplague: I will be able to put most of the really important utilities and libraries I need in /mnt/flash then. :D This will greatly help me. |
| 18:24.22 | RobotGuy | claps and makes strange noises |
| 18:24.44 | prpplague | SpeedEvil: http://www.micron.com/products/partdetail?part=MT48LC8M8A2P-75 |
| 18:25.02 | prpplague | RobotGuy: it won't be /mnt/flash any more it will be / |
| 18:25.07 | SpeedEvil | prpplague: It seems to be - I need to do maybe 10 assorted writes at the boundaries of blocks, and can just let the SD soak up the data from my camera the rest of the time using the auto-increment write feature. Other than that, yes, it's not particularly fast - and yes you need a fast micro to do it - the cortex ones seem well matched - and it's not too many pins. |
| 18:25.28 | RobotGuy | prpplague: That's perfectly fine with me. :D |
| 18:25.31 | prpplague | SpeedEvil: does the cortex device you have not have an external sdram controller? |
| 18:25.43 | RobotGuy | prpplague: This is what I have needed! :D |
| 18:25.48 | SpeedEvil | prpplague: no |
| 18:26.48 | RobotGuy | prpplague: Does this mean I can completely disable initrd/initramfs and ramdisk in the kernel? |
| 18:26.53 | SpeedEvil | prpplague: and yes, it's a hack. There is one reason for doing it this way. $ |
| 18:27.02 | prpplague | RobotGuy: yea |
| 18:27.07 | prpplague | SpeedEvil: sounds like a trade off |
| 18:27.11 | RobotGuy | Oooooooooooooooooooooooooooooooooooh! YAY!! |
| 18:29.00 | SpeedEvil | prpplague: well - $ and board area. I can I think get a 8M*8 RAM, a 32bit CPU, and a SD socket all in around $10 in dozens. |
| 18:29.06 | SpeedEvil | BOM |
| 18:29.16 | RobotGuy | prpplague: Now I think I see why I have to actually flash my image to the /dev/mtdblock1 |
| 18:29.53 | SpeedEvil | oops - that price was using my thousands price for RAM - $12 |
| 18:29.58 | prpplague | SpeedEvil: not sure what you are going to do with it, but the sdram i think is going to be a big performance hit |
| 18:30.13 | RobotGuy | Oh, and kernel 2.6.25-rc7-hammer supports my Quickcam Messenger webcam. :D |
| 18:30.31 | SpeedEvil | Sure. But I can use the auto-increment to do the fast write, and it doesn't need to be fast otherwise. |
| 18:30.56 | SpeedEvil | Sigh. I hate microns site. I seem to have missed that part. |
| 18:31.26 | SpeedEvil | and yes - read or write bytes is hideous. read or write 512 byte blocks isn't bad at all. |
| 18:31.50 | prpplague | ho ho hum |
| 18:32.06 | SpeedEvil | Thanks. |
| 18:32.57 | prpplague | np |
| 18:34.01 | SpeedEvil | The CPU can do DMA to internal SRAM - so once it's inside, it's painless. |
| 18:35.18 | prpplague | RobotGuy: i probably should do a config and instructions on using a jffs2 rootfs |
| 18:35.28 | prpplague | makes a note on his 6 mile long todo list |
| 18:35.55 | RobotGuy | prpplague: I was just thinking of trying to do that too. :) |
| 18:36.15 | SpeedEvil | feels lucky that his is only 1.5 miles. |
| 18:36.16 | RobotGuy | prpplague: I will start it off once I get this working. |
| 18:36.32 | RobotGuy | prpplague: You can correct my mistakes. :) |
| 18:42.06 | prpplague | does a quick jffs2 build for testing |
| 18:45.58 | adyer | RobotGuy: make sure you have mtd-tools installed on the cross compile buildroot machine - first mistake I made :-) |
| 18:46.53 | RobotGuy | quickly installs mtd-tools |
| 18:47.15 | ds2 | morning |
| 18:47.56 | *** join/#edev ico2__ (n=ico2@modem-124.chimpanzee.dialup.pol.co.uk) |
| 18:53.11 | RobotGuy | prpplague: Next question. :) Can I automate the flashing of the jffs2 image to the flash (/mnt/mtdblock1)? |
| 18:54.07 | RobotGuy | Uh oh, ds2 woke up.. :D |
| 18:55.14 | *** join/#edev GPSFan (n=kenm@12.10.255.246) |
| 18:55.14 | *** mode/#edev [+o GPSFan] by ChanServ |
| 18:57.23 | prpplague | RobotGuy: automate the flashing? |
| 18:58.30 | ds2 | depends on how well you could automate removal of a trenchcoat ;) |
| 18:58.49 | RobotGuy | prpplague: To mtdblock1 |
| 19:01.32 | adyer | RobotGuy: handy script for 'mounting' an mtd image on a PC: http://maemo.org/community/wiki/ModifyingRootImage/#mount_jffs2_mtdram (you might want to change the total_size parameter...) |
| 19:02.19 | prpplague | RobotGuy: well, you can setup scripts in both apex and openocd |
| 19:02.36 | ds2 | adyer: I'd be careful with that. mtdram uses non swappable memory |
| 19:02.59 | prpplague | i think he means to flash the jffs2 image on the hammer |
| 19:07.13 | ds2 | as long as it is not in use, no why you can't flash_eraseall /dev/mtd1; dd if=jffsimage of=/dev/mtd1 bs=128k |
| 19:08.32 | prpplague | ds2: thats assuming linux is already running |
| 19:09.31 | ds2 | prpplague: yes, but he said /dev/mtdblock1 which odesn't exist til Linux is running |
| 19:10.07 | ds2 | no reason why you can't tftp in a kernel +initramfs if Linux is preferred over jtag/bootloader flashing |
| 19:16.02 | RobotGuy | prpplague: Yes, flash the new JFFS2 image to the flash (mtdblock1). |
| 19:16.41 | RobotGuy | I am getting confused. |
| 19:17.55 | prpplague | RobotGuy: ok so how did you flash it? |
| 19:18.44 | RobotGuy | prpplague: I have not flashed it yet. I am trying to find out the proper procedure and if it can be automated, like with openocd the way the initrd is now. |
| 19:19.50 | *** join/#edev Shingoshi (n=shingosh@c-67-169-195-234.hsd1.or.comcast.net) |
| 19:20.35 | prpplague | RobotGuy: from apex, erase the flash partition, use xreceive to transfer the file up to sdram (you can use 0x30800000 ) and then copy to flash paritition |
| 19:21.42 | RobotGuy | prpplague: OK, right, because there is no system at this point. I am creating the root. :D I begin to understand |
| 19:21.58 | prpplague | RobotGuy: i'm running a fresh build now, i'll post the info on the wiki as soon as the build finishes |
| 19:22.20 | RobotGuy | prpplague: OK. I do believe I am understanding now. |
| 19:23.15 | RobotGuy | This would definitely be a good thing for the Wiki. Once I get a working 2.6.25-rc7-hammer kernel, I am hoping my WiFi will not be flaky and I can put that info on the Wiki also. |
| 19:39.29 | RobotGuy | Dang, now I have some error that seems to have to do with dma. |
| 19:39.51 | prpplague | RobotGuy: whats the error? |
| 19:40.06 | prpplague | RobotGuy: you got it flashed? |
| 19:40.17 | RobotGuy | prpplague: error: implicit declaration of function 'dma_sync_single_range_for_device' |
| 19:40.21 | prpplague | RobotGuy: did you change the root value in the bootloader? |
| 19:40.38 | prpplague | RobotGuy: uh, for what ? |
| 19:40.41 | RobotGuy | prpplague: Not yet. Have not got the 2.6.25-rc7 kernel configured and built yet. |
| 19:41.02 | prpplague | RobotGuy: are we talking about 2.6.25-rc7? |
| 19:41.06 | prpplague | RobotGuy: or 2.6.22? |
| 19:41.26 | RobotGuy | prpplague: Oh, wait, I didn't paste the whole error. --> drivers/net/b44.c:151: error: implicit declaration of function 'dma_sync_single_range_for_device' |
| 19:41.39 | prpplague | RobotGuy: are we talking about 2.6.25-rc7? |
| 19:41.40 | prpplague | RobotGuy: or 2.6.22? |
| 19:41.44 | RobotGuy | Yes, I am working on 2.6.25-rc7 now |
| 19:42.03 | prpplague | RobotGuy: ok sounds like you have some net device driver enabled that might be broken |
| 19:42.11 | prpplague | i have no idea what b44 is |
| 19:42.25 | RobotGuy | prpplague: Hmmm, could be. I don't know what b44 is either. |
| 19:43.00 | RobotGuy | prpplague: I think I know where it might be. |
| 19:43.01 | prpplague | RobotGuy: thats for a broadcom fast ethernet device |
| 19:43.25 | RobotGuy | prpplague: Hmmm, didn't think I had any of that enabled - I will check. |
| 19:43.30 | prpplague | thinks RobotGuy wastes alot of time tracking down stuff that shouldn't be enabled |
| 19:44.51 | RobotGuy | prpplague: Lots of stuff was enabled in the stock kernel. I just didn't get everything turned off that isn't necessary. |
| 19:45.08 | prpplague | RobotGuy: not in my standard hammer config |
| 19:45.35 | RobotGuy | prpplague: I found lots of stuff turned on that shouldn't be or isn't necessary. |
| 19:45.43 | prpplague | RobotGuy: such as? |
| 19:46.04 | RobotGuy | prpplague: Like that broadcom and similar stuff. |
| 19:46.18 | prpplague | those aren't on in my config |
| 19:46.24 | RobotGuy | prpplague: I can't remember if I turned that on or if it was already on. |
| 19:47.02 | prpplague | RobotGuy: no network devices are enabled in my default config |
| 19:47.08 | RobotGuy | prpplague: I can't remember all the stuff I turned off from the stock config. |
| 19:47.32 | RobotGuy | prpplague: I must have turned the broadcom stuff on then - can't remember why though. |
| 19:50.04 | prpplague | http://idle.slashdot.org/idle/08/03/31/131235.shtml |
| 19:50.12 | prpplague | yep, says alot about tech |
| 19:53.27 | prpplague | is totally stressed out |
| 19:53.39 | RobotGuy | prpplague: Why? |
| 19:54.01 | prpplague | RobotGuy: do you have 72 hours for me to list everything? hehe |
| 19:54.16 | prpplague | mostly about the moron of an EE |
| 19:54.20 | RobotGuy | prpplague: Ummm, 3 days.. Well, yeah, I do. :D |
| 19:54.54 | RobotGuy | prpplague: Him again. Just ummm, well, that would not be appropriate. :D |
| 19:55.15 | RobotGuy | prpplague: Why do they not get rid of that guy if he is so stupid? |
| 19:55.25 | prpplague | we have three giant projects that all going to implode |
| 19:55.42 | prpplague | RobotGuy: because management fails to see the extend of the problem |
| 19:55.46 | SpeedEvil | Giant enough to be company damaging? |
| 19:55.59 | prpplague | RobotGuy: they won;t until the projects die a horrible death |
| 19:56.01 | prpplague | SpeedEvil: yea |
| 19:56.05 | SpeedEvil | Damn. |
| 19:57.10 | prpplague | SpeedEvil: all three projects she be gearing up for production now, but instead, i doubt we will be doing that until this time next year, if at all |
| 19:57.38 | SpeedEvil | Is it a desire for everything to be as polished as possible - rather than working enough to ship out the door? |
| 19:59.03 | prpplague | SpeedEvil: well in one aspect, yes, the guy doesn't even want to do prototypes, he wants to go straight to formfactor and he doesn't want to do even that until he's got the board "perfect" |
| 19:59.16 | SpeedEvil | ... |
| 19:59.25 | prpplague | which in both cases is totally unrealistic |
| 19:59.34 | SpeedEvil | I mean - sure if your prototype costs are $10M per... |
| 20:00.16 | prpplague | goes to get kids |
| 20:00.19 | prpplague | bbiaf |
| 20:00.20 | SpeedEvil | wave. |
| 20:00.45 | RobotGuy | prpplague: So project(s) must fail before they see the problem with the EE? |
| 20:01.54 | SpeedEvil | I can see why non-techy managment would like a 'sure - no need for a lengthy iterative design cycle - we can just get it right first time'. |
| 20:03.07 | RobotGuy | SpeedEvil: I've had to work with people like that, including one Boss, and he got me fired. System's Analyst told him my idea was the only way to do something and his would never work. |
| 20:08.29 | RobotGuy | I have another question. I've been just using openocd to flash the bootloader, kernel (zImage), and rootfs to Hammer. Now with switching to using JFFS2 for the mtdblock1 of Flash, can I just edit the script file to not flash the rootfs? |
| 20:09.35 | RobotGuy | I just commented the line that flashes the rootfs out of hammer.ocd. |
| 20:11.03 | RobotGuy | This will be my first 2.6.25-rc7 kernel |
| 20:11.15 | *** join/#edev ico2_ (n=ico2@modem-3550.cougar.dialup.pol.co.uk) |
| 20:15.20 | RobotGuy | -sigh- |
| 20:28.49 | prpplague | RobotGuy: yea, the projects will have to implode before it will become an issue |
| 20:29.13 | RobotGuy | That sucks. Do any of those projects affect TCT? |
| 20:29.22 | prpplague | none |
| 20:29.40 | RobotGuy | Good, but still bad in a few ways probably. |
| 20:30.14 | prpplague | RobotGuy: yes you can make the change to the openocd script to flash the jffs2 image, you have to make sure you get the right address |
| 20:33.25 | SpeedEvil | wanders around the secondary DRAM market. |
| 20:38.53 | *** join/#edev emil|k (i=emil@f054012181.adsl.alicedsl.de) |
| 20:41.34 | RobotGuy | prpplague: How do I get the right address? |
| 20:42.14 | RobotGuy | I don't see where to change the root mount point either. |
| 20:42.29 | prpplague | RobotGuy: the address of the jffs2 partition should be the same as the size you mark for your system |
| 20:42.42 | prpplague | RobotGuy: for example if you have .size=0x340000 |
| 20:42.54 | prpplague | you need to start flashing at 0x340000 |
| 20:43.49 | RobotGuy | prpplague: .size is set to 0x240000 at present |
| 20:44.16 | RobotGuy | I am not sure that size is appropriate now though |
| 20:44.38 | RobotGuy | Bootloader + Kernel only now, right? |
| 20:44.56 | prpplague | RobotGuy: yep |
| 20:45.34 | RobotGuy | My Kernel is 936 KB |
| 20:51.18 | prpplague | hmm getting a weird jffs2 error i've never seen before :( |
| 20:58.37 | landley | And back. :) |
| 20:58.59 | prpplague | and front :( |
| 20:59.26 | landley | prpplague: is the EEdiot likely to crater the other company badly enough to affect TCT's funding? |
| 20:59.41 | landley | Or are they likely to give you more time to work on TCT after the crater, due to there being nothing else? |
| 21:00.01 | prpplague | landley: no, TCT is completely seperate |
| 21:00.15 | landley | Is it self-funding yet? Or in the vc/angel cocoon? |
| 21:00.15 | RobotGuy | So should I set it to 1 MB? |
| 21:00.33 | prpplague | landley: well, i'm guessing that in october/november when everything comes to a head, 1 of 2 things is going to happen |
| 21:00.45 | landley | RobotGuy: In theory you could set it to bootloader size (256k I think) plus kernel size. |
| 21:00.51 | chouimat|work | hmmm EEdiot that must look good on a business card |
| 21:00.54 | landley | Call it 1256k. |
| 21:01.12 | prpplague | landley: 1) the fire the EE and try to start over in the engineering department or 2) they fire everyone and outsource the engineering |
| 21:01.27 | RobotGuy | landley: 1280 KB should work then, but I think it has to be on a page boundary. |
| 21:01.33 | landley | I'd guess #2. |
| 21:01.50 | landley | Because #1 hits the whole "when lost, never turn around and go back" standard failure mode. |
| 21:01.59 | landley | Admitting defeat is worse than driving off a cliff, you see. |
| 21:02.23 | landley | They'll do the cognitive dissonance thing and decide that outsourcing was the way to go all along. |
| 21:02.29 | RobotGuy | 1024 + 256 = 1280 :) |
| 21:02.46 | RobotGuy | That should be on a page boundary |
| 21:03.01 | landley | RobotGuy: your kernel is <1024k. But yeah, page boundary good... |
| 21:03.22 | RobotGuy | landley: My kernel is 936 KB |
| 21:03.31 | landley | I think the bootloader has 256k reserved for it. (Includes the environment variable space.) |
| 21:03.52 | prpplague | landley: indeed |
| 21:04.13 | prpplague | landley: so i'm making sure my resume is uptodate and i've got some personal networking going on |
| 21:04.18 | RobotGuy | prpplague: Can I set .size to 1280KB and not have any complaints or errors? |
| 21:04.31 | prpplague | RobotGuy: should |
| 21:04.41 | RobotGuy | prpplague: If #2 happens, will TCT go under? |
| 21:05.17 | *** join/#edev gandhijee (n=root@host-66-202-34-165.spr.choiceone.net) |
| 21:05.22 | prpplague | RobotGuy: no, as i said, TCT is completely independent of AML |
| 21:05.50 | RobotGuy | Oh, I know. But sometimes stuff happening at the parent company can affect children. |
| 21:06.13 | prpplague | RobotGuy: AML isn't the parent compay of TCT |
| 21:06.28 | prpplague | RobotGuy: they just happen to be owned by the same person |
| 21:07.06 | RobotGuy | Ahhh, OK. So no downward roll from AML then. That's good. |
| 21:10.13 | landley | prpplague: I take it TCT is unlikely to become a full-time gig any time soon? |
| 21:10.45 | prpplague | landley: probably not |
| 21:13.51 | adyer | prpplague: my old boss used to say "it's time to shoot (the engineer) and ship (the product)" |
| 21:16.44 | RobotGuy | OK, setting .size to 0x140000 |
| 21:17.49 | prpplague | RobotGuy: hmm, i'm having troubles with my jffs image |
| 21:18.07 | RobotGuy | prpplague: I didn't have trouble building mine. |
| 21:18.22 | prpplague | RobotGuy: not building, booting |
| 21:18.29 | RobotGuy | prpplague: I have not built the final image though. |
| 21:18.44 | prpplague | List of all partitions: |
| 21:18.44 | prpplague | 1f00 2304 mtdblock0 (driver?) |
| 21:18.44 | prpplague | 1f01 14080 mtdblock1 (driver?) |
| 21:18.44 | prpplague | No filesystem could mount root, tried: ext2 msdos vfat |
| 21:18.44 | prpplague | Kernel panic - not syncing: VFS: Unable to mount root fs on unknown-block(31,1) |
| 21:18.49 | RobotGuy | I am having trouble loading my jffs2 image to hammer. |
| 21:19.05 | RobotGuy | prpplague: xreceive is xmodem, right? |
| 21:19.16 | adyer | prpplague: are you using a static map for the mtdparts or passing them on the kernel cmd line? |
| 21:19.51 | RobotGuy | I don't have any xmodem |
| 21:20.05 | prpplague | adyer: static |
| 21:20.16 | prpplague | RobotGuy: how have you been uploading via apex? |
| 21:20.22 | adyer | prpplague: and rootfstype=jffs2 in cmd line |
| 21:20.28 | RobotGuy | prpplague: I have not been uploading via apex |
| 21:20.34 | RobotGuy | Never have |
| 21:20.49 | prpplague | RobotGuy: really need to add xmodem |
| 21:21.00 | RobotGuy | prpplague: I do not know what package to add |
| 21:21.08 | prpplague | adyer: hmm |
| 21:21.17 | prpplague | adyer: is that documented somewhere? |
| 21:21.19 | RobotGuy | I can never find what I need on the various package lists |
| 21:21.24 | prpplague | mainly uses initrd |
| 21:21.38 | prpplague | RobotGuy: might be time to get a different distro then |
| 21:21.45 | adyer | prpplague: I think I picked it up from the denx wiki about using u-boot and jffs2 |
| 21:21.55 | RobotGuy | prpplague: I've run all the major distros. |
| 21:21.56 | prpplague | ahh |
| 21:24.07 | RobotGuy | Kubuntu is the best I have run so far, at least as far as multi-media goes. |
| 21:25.03 | adyer | prpplague: http://www.linux-mtd.infradead.org/faq/jffs2.html search for rootfstype |
| 21:25.28 | prpplague | adyer: ahh thanks |
| 21:25.47 | RobotGuy | Oh, wait, I do have xmodem. |
| 21:25.58 | prpplague | adyer: ahh new to 2.6, no wonder i didn't know about it |
| 21:27.22 | adyer | prpplague: 2.6 was released in late 2003 :-) |
| 21:28.05 | prpplague | adyer: i've not done a jffs2 rootfs since 2.4 |
| 21:28.53 | prpplague | wheee that did the trick |
| 21:29.00 | prpplague | took awhile on first boot |
| 21:29.38 | adyer | prpplague: yeah, I think it's writing clean markers or something in the unused part of the partition |
| 21:30.24 | prpplague | adyer: yea |
| 21:30.25 | adyer | prpplague: if you make the jffs2 image the size of the partition I think it doesn;t do that |
| 21:30.47 | prpplague | RobotGuy: hows this: |
| 21:30.52 | prpplague | MemTotal: 30576 kB |
| 21:30.52 | prpplague | MemFree: 27404 kB |
| 21:31.18 | ds2 | jffs2 always wants the rootfstype |
| 21:31.24 | ds2 | use yaffs if you need autodetection |
| 21:31.41 | RobotGuy | prpplague: I like! |
| 21:31.43 | prpplague | ds2: yea, no need, just wasn't aware of that item |
| 21:31.55 | RobotGuy | prpplague: I am uploading my first jffs2 image now |
| 21:32.17 | ds2 | prpplague: yaffs also supports memmap'ing in case you runinto that with jffs2 |
| 21:32.49 | prpplague | RobotGuy: need to make sure you remove the item in the inittab to mount /dev/mtdblock1 on /mnt/flash |
| 21:33.10 | prpplague | Filesystem Size Used Available Use% Mounted on |
| 21:33.10 | prpplague | /dev/mtdblock1 13.8M 1.4M 12.3M 11% / |
| 21:33.10 | RobotGuy | prpplague: I don't know where to do that |
| 21:33.29 | RobotGuy | Oh yeah, wait, you mean fstab |
| 21:33.32 | prpplague | RobotGuy: in the skeleton file system |
| 21:33.40 | prpplague | RobotGuy: no its in inittab |
| 21:34.02 | RobotGuy | prpplague: I know but also in my fstab |
| 21:34.18 | prpplague | RobotGuy: if its in fstab, you added it |
| 21:37.00 | RobotGuy | prpplague: I did |
| 21:37.11 | landley | prpplague: xmodem is not a real protocol. Do zmodem if you're going that route. |
| 21:37.18 | landley | Although uuencode is probably plenty... |
| 21:37.27 | ds2 | kermit! |
| 21:37.30 | RobotGuy | apex does not grok zmodem |
| 21:37.42 | RobotGuy | Kermit is our head maintenance guy |
| 21:37.44 | CosmicPenguin | the horror! |
| 21:38.46 | landley | prpplague: do you just go "rootfstype=jffs2" or do you need to pass some weird "evil:/blah" thing? |
| 21:39.05 | RobotGuy | My first upload did not work |
| 21:39.22 | RobotGuy | Do I need to change a root option for the kernel? |
| 21:40.02 | adyer | landley: our jffs2 boot cmd line is: root=/dev/mtdblock3 console=ttyS0 rootfstype=jffs2 quiet mtdparts=flash00:0x40000(uboot),0x20000(ubootenv),0x100000(kernel),-(root) |
| 21:40.07 | prpplague | RobotGuy: what are you saying aobut kermit? |
| 21:40.18 | adyer | landley: the mtdparts stuff at the end is not 100% necessary |
| 21:40.24 | landley | mtdparts? |
| 21:40.27 | RobotGuy | prpplague: nvm, nothing important |
| 21:40.27 | landley | *boggle* |
| 21:40.32 | prpplague | landley: the mtdparts isn't needed for the hammer |
| 21:40.46 | RobotGuy | prpplague: my first upload got an error though |
| 21:40.50 | prpplague | landley: we specify the mtdpartitions in the tct hammer startup code |
| 21:41.01 | landley | ah. |
| 21:41.02 | adyer | landley: allows overriding the static mtd partitions in the kernel from boot |
| 21:41.10 | prpplague | CosmicPenguin: don't like xmodem? |
| 21:41.14 | landley | So you're hardwiring in something that didn't actually need hardwiring. |
| 21:41.14 | ds2 | normally, it is done in the board* file |
| 21:41.24 | CosmicPenguin | prefer zmodem |
| 21:41.34 | prpplague | landley: it is acceptable to do either |
| 21:41.50 | prpplague | landley: but most do it in the board startup code |
| 21:41.57 | landley | Those partitions don't seem to have a length. I guess they always go to next partition or end of device? |
| 21:42.03 | adyer | landley: yes, handy for dev work where the kernel can be a lot bigger than for production (network, etc) |
| 21:42.06 | prpplague | landley: correct |
| 21:42.11 | landley | is wondering what it would take to get qemu to emulate a hammer... |
| 21:42.23 | SpeedEvil | Not that much. |
| 21:42.25 | RobotGuy | OK, upload worked this time |
| 21:42.26 | prpplague | landley: irrc qemu already supports s3c2410 |
| 21:42.32 | SpeedEvil | Just rip out most of the neo1973 hardware |
| 21:42.37 | prpplague | landley: it would only require some very basic support |
| 21:42.43 | SpeedEvil | (virtual hardwar) |
| 21:42.53 | landley | very basic for you != very basic for me. |
| 21:42.59 | landley | "Oh, simple, just patch qemu!" |
| 21:43.02 | prpplague | landley: hehe |
| 21:43.04 | RobotGuy | Now do I use the copy command to copy to the NOR flash? |
| 21:43.09 | adyer | landley: the first item 'flash00' for us specs a base address, the numbers are sizes, and the '-' means all remaining space |
| 21:43.19 | RobotGuy | I have no clue what values to use. |
| 21:43.26 | RobotGuy | Other than 1280k |
| 21:43.28 | ds2 | robotguy: if you haven't, erase the partition first |
| 21:43.32 | ds2 | then copy |
| 21:43.51 | RobotGuy | Do I just erase the whole flash? |
| 21:43.52 | SpeedEvil | As long as you don't touch the peripehrals - I think actually the neo1973 version should be just fine. It'll have too much ROM and RAM |
| 21:43.56 | landley | How does it turn flash00 into something useful? |
| 21:43.56 | prpplague | RobotGuy: if you transfered the file to 0x30800000 then use: copy 0x30800000+size_of_image nor:1280k |
| 21:44.07 | ds2 | RobotGuy: just hte partition - i.e. flash_eraseall /dev/mtd1 |
| 21:44.18 | prpplague | ds2: no |
| 21:44.24 | prpplague | ds2: he's doing it from apex |
| 21:44.27 | prpplague | ds2: not from linux |
| 21:44.28 | ds2 | oh |
| 21:44.31 | ds2 | bah |
| 21:44.47 | RobotGuy | prpplague: So copy 0x30800000+1280k nor:1280k Is that right? |
| 21:44.49 | ds2 | life is so much easier if one just tftp/nfs boota kernel and do everything fromt ehre |
| 21:44.51 | ds2 | there |
| 21:45.00 | prpplague | RobotGuy: is your jffs2 image 1280k? |
| 21:45.16 | prpplague | landley: i assume flash00 is a struct for the board init file |
| 21:45.19 | RobotGuy | prpplague: No, not exactly |
| 21:45.27 | RobotGuy | This too confusing to me |
| 21:45.29 | prpplague | RobotGuy: you need the exact size |
| 21:45.43 | adyer | ds2; cool, I'll just ship a server with each battery operated handheld :-) |
| 21:46.04 | RobotGuy | Current jffs2 image is 1.6 MB |
| 21:46.14 | prpplague | RobotGuy: you need the size down to the last byte |
| 21:46.17 | ds2 | adyer: no, I mean do that instead of doing it from the bootloader... mtdutils are easier to the different bootloader variants |
| 21:46.41 | adyer | ds2: i know, just teasing :-) |
| 21:46.46 | RobotGuy | 1673168 is what ls -l reports |
| 21:46.54 | prpplague | then do this |
| 21:46.55 | adyer | prpplague: flash00: is the MTDID |
| 21:47.07 | ds2 | adyer: Had to make sure. I have worked with people who really thought that! |
| 21:47.15 | prpplague | RobotGuy: copy 0x30800000+1673168 nor:1280k |
| 21:47.35 | prpplague | RobotGuy: assuming you transfered the file up to 0x30800000 |
| 21:47.43 | RobotGuy | I did |
| 21:47.44 | prpplague | RobotGuy: and also assuming you've already erase the flash |
| 21:47.51 | RobotGuy | I did not. Damn |
| 21:49.06 | adyer | ds2: I like doing the stuff from the bootloader because I grok it better and I have that up before a kernel, so habit sets in |
| 21:49.26 | RobotGuy | I don't know how to figure this out. It will be different next time |
| 21:49.29 | prpplague | RobotGuy: erase nor:1280k+0xec0000 |
| 21:49.40 | ds2 | adyer: but there are so many boot loaders one looses track of what can be done in what and how |
| 21:49.59 | prpplague | yea i much prefer using bootloader |
| 21:50.11 | adyer | ds2: yes, but we've mostly been on u-boot for linux stuff |
| 21:50.31 | ds2 | adyer: so you are a uboot expert? :) |
| 21:51.01 | adyer | ds2: expert, no, but I know the mips and arm parts OK |
| 21:51.42 | ds2 | adyer: every now and then I get stuck trying to figure out how to get uboot to do basic stuff like read from nand or erase nand |
| 21:52.43 | adyer | ds2: never used nand, systems all had nor for boot and a hard disk or one big NOR flash |
| 21:53.22 | adyer | ds2: or CF |
| 21:53.32 | prpplague | adyer: nand boot is soooooo much fun |
| 21:53.44 | prpplague | adyer: every SoC has its own stepping stone boot process |
| 21:54.38 | RobotGuy | tries to record all these steps |
| 21:54.38 | ds2 | adyer: do you use JFFS2 with NOR boot then? |
| 21:54.44 | adyer | prpplague: yeah, I've used nand, but only as block storage and not as boot or with u-boot |
| 21:54.48 | adyer | ds2: yes |
| 21:54.56 | ds2 | only way I can get NOR to boot is write a raw image and bootm where it is mapped |
| 21:55.05 | ds2 | adyer: what is the magic to get uboot to understand jffs2? |
| 21:55.47 | RobotGuy | prpplague: No boot. Got Kernel panic. |
| 21:55.59 | prpplague | RobotGuy: did you change the kernel args? |
| 21:56.17 | adyer | ds2: ahh, we have kernel ina raw partition, and jffs2 for root. I have booted off of jffs2 in the past though, have to dig up stuff |
| 21:56.47 | ds2 | adyer: oh heh.. jffs2 root is easy it is getting it to do the bootm that I couldn't figure out |
| 21:56.49 | *** join/#edev CosmicPenguin (n=nobody@163.181.251.103) |
| 21:56.49 | *** mode/#edev [+o CosmicPenguin] by ChanServ |
| 21:58.54 | adyer | ds2: you need CFG_CMD_JFFS2 and then you have to spec a valid mtdids and mtdparts env. variable. syntax is a bit weird |
| 21:59.40 | ds2 | adyer: I didn't build uboot; but the jffs help shows commands are there |
| 22:00.29 | *** join/#edev rodserling (n=rodserli@unaffiliated/rodserling) |
| 22:01.24 | RobotGuy | prpplague: Don't I have to tell APEX what the root is? |
| 22:01.52 | prpplague | RobotGuy: yes, hence my question about the arguments |
| 22:02.16 | RobotGuy | prpplague: I don't see where to change it. I am looking now. |
| 22:02.27 | prpplague | RobotGuy: you need to set the command line to this: cmdline = console=ttySAC0 rootfstype=jffs2 root=/dev/mtdblock1 |
| 22:02.37 | prpplague | RobotGuy: do this: printenv |
| 22:02.45 | prpplague | RobotGuy: that will show you the current settings |
| 22:03.08 | prpplague | RobotGuy: then do: setenv cmdline "console=ttySAC0 rootfstype=jffs2 root=/dev/mtdblock1" |
| 22:03.19 | prpplague | RobotGuy: then do a printenv again to verify the changes |
| 22:03.53 | adyer | hates IrDA |
| 22:04.29 | RobotGuy | prpplague: Can't I change that in APEX somewhere to make it permanent> |
| 22:05.06 | prpplague | RobotGuy: its permanent until you erase apex |
| 22:05.36 | prpplague | RobotGuy: or else you can edit apex and do a fresh build |
| 22:06.41 | RobotGuy | prpplague: I mean, can't I change it somewhere in the APEX config file? |
| 22:07.02 | prpplague | hehe, i thought thats what i just said, hehe |
| 22:08.03 | prpplague | RobotGuy: you can edit apex-1.5.6/src/mach-s3c2410/env.c |
| 22:08.11 | prpplague | RobotGuy: change the default command like |
| 22:08.24 | prpplague | RobotGuy: you can also disable the copy ramdisk on boot option as well |
| 22:08.39 | RobotGuy | prpplague: Why would I do that? |
| 22:08.57 | prpplague | RobotGuy: why do you need to copy the ramdisk on boot when you have no ramdisk? |
| 22:10.02 | RobotGuy | prpplague: OK. I am just confused. |
| 22:10.28 | RobotGuy | prpplague: I disabled the ramdisk options in APEX config. Is that what you mean? |
| 22:10.34 | prpplague | RobotGuy: yes |
| 22:10.58 | RobotGuy | prpplague: OK, good. These were just values, not commands. |
| 22:11.26 | prpplague | RobotGuy: they enable commands |
| 22:11.55 | prpplague | RobotGuy: under "environment" when ramdisk is enable you have a "copy ramdisk on boot" option |
| 22:12.43 | prpplague | RobotGuy: under environment you can also specify the default kernel arguments as well as the default startup command |
| 22:13.27 | *** join/#edev sneak (n=sneak@64.220.234.21) |
| 22:13.34 | RobotGuy | prpplague: Hey, I found a place in APEX config where I can override the default kernel command line! |
| 22:14.38 | RobotGuy | prpplague: The copy command must go away when I disabled ramdisk options |
| 22:16.58 | prpplague | RobotGuy: yea |
| 22:17.34 | prpplague | gets ready to go home |
| 22:17.50 | RobotGuy | gets ready to have a breakdown |
| 22:48.18 | *** join/#edev rodserling (n=rodserli@unaffiliated/rodserling) |
| 23:02.54 | *** join/#edev rodserling (n=rodserli@unaffiliated/rodserling) |
| 23:05.15 | *** join/#edev prpplague (n=dave123_@ppp-70-244-160-94.dsl.rcsntx.swbell.net) |
| 23:06.11 | prpplague | ho ho hum |
| 23:25.38 | RobotGuy | muh oh oh |
| 23:44.07 | *** join/#edev redblue (i=star@ppp084.109-253-207.mtl.mt.videotron.ca) |
| 23:48.01 | *** join/#edev Crofton (n=balister@66-207-66-26.black.dmt.ntelos.net) |
| 23:58.37 | Mirell | I'm tempted to start biking in to work come tomorrow. |