| 00:02.53 | *** join/#edev RobotGuy (n=robotguy@dsl093-038-072.pdx1.dsl.speakeasy.net) |
| 00:07.42 | RobotGuy | I have a new opportunity to give out information about Hammer and help people get started with it. |
| 00:08.37 | Sienna | :arch |
| 00:09.20 | Sienna | i have been lazy today... i meant to go edit the RDP board prototype.. but never got around to it yet :look |
| 00:10.01 | Sienna | i did get a triple axis accel. that uses SPI that I need to start using with a hammer to try that whole software side of my bots :( |
| 00:12.48 | RobotGuy | Sienna: Hey there! |
| 00:13.13 | RobotGuy | Sienna: Could you resend your stuff to me please? I lost all my stuff on /home when I switched to Gentoo. :( |
| 00:13.30 | Sienna | what stuff? |
| 00:13.45 | RobotGuy | Sienna: Which one did you get? I have the ADXL330 that needs three ADCs. |
| 00:14.00 | RobotGuy | Sienna: The stuff for the Hammer-RDP board design |
| 00:14.33 | RobotGuy | There is another accelerometer SFE came out with that is 3-axis and I2C interface. :) |
| 00:14.46 | Sienna | that one (its I2C and SPI) |
| 00:14.58 | RobotGuy | Sienna: Ah, OK. I want to get one of those. |
| 00:15.47 | RobotGuy | Sienna: Have you setup your Buildroot and built your toolchain? |
| 00:15.55 | Sienna | haha... no... |
| 00:16.03 | Sienna | haven't even started software stuff yet |
| 00:16.11 | RobotGuy | I am working on tutorials on the TR boards for that. |
| 00:16.19 | RobotGuy | They are also on the Hammer Wiki |
| 00:16.46 | RobotGuy | I am still waiting for delivery of my replacement Hammer. I haven't been able to work with it for a few weeks now. :( |
| 00:17.53 | RobotGuy | Alex asked me if I could put some tutorials up. |
| 00:25.18 | g1powermac_PB | hey RobotGuy |
| 00:25.30 | RobotGuy | g1powermac_PB: Greetings |
| 00:25.49 | RobotGuy | I got buildroot setup and working under Gentoo now. |
| 00:26.14 | g1powermac_PB | has been looking more further into the cc2431's zigbee subsystem |
| 00:26.26 | RobotGuy | g1powermac_PB: That is a cool little chip. |
| 00:26.54 | RobotGuy | I'll be getting another package from TI soon. |
| 00:26.55 | g1powermac_PB | it seems it gives you a basic hardware interface that, though at first, seems complicated, it really isn't |
| 00:27.33 | g1powermac_PB | of course if you use just the hardware interface, you won't get all the niceties of a full blown zigbee stack |
| 00:27.40 | RobotGuy | g1powermac_PB: That cc2431 could be an entire sensor node processor. |
| 00:27.54 | g1powermac_PB | but I think to just move data around, it should work |
| 00:28.05 | g1powermac_PB | yupper |
| 00:28.10 | RobotGuy | It has no complete ZigBee stack? |
| 00:28.30 | g1powermac_PB | well, thats where things get tricky |
| 00:28.43 | g1powermac_PB | TI has a think called the zstack |
| 00:29.26 | g1powermac_PB | its a windows app utility which seems to have at least some of the source code of the zstack, if not all |
| 00:29.43 | g1powermac_PB | however, the code is meant to be compiled with the expensive IAR compiler |
| 00:30.10 | g1powermac_PB | porting the zstack to compile on sdcc could be interesting, assuming all the source code is there |
| 00:30.42 | g1powermac_PB | but then you also have that tinyOS thing that'll run on the cc2431 |
| 00:31.16 | g1powermac_PB | it has a simpleMAC thing built in, but to do any modifications to it, you'll need to learn nesC |
| 00:31.54 | RobotGuy | Hmmmmm, maybe the cc2431 won't be as useful as I thought it would be. |
| 00:32.52 | RobotGuy | I love it when companies give away free code, but it requires serious conversion or compilation with expensive commercial tools. |
| 00:34.09 | g1powermac_PB | yea |
| 00:34.52 | g1powermac_PB | but, I think you really might not need all the stuff found in a zigbee stack for what we want to do |
| 00:37.38 | RobotGuy | It looks like they intend for you to use an MPS30 with the cc2431 |
| 00:37.51 | RobotGuy | I should not be looking at the TI site.. |
| 00:38.10 | g1powermac_PB | no, not exactly |
| 00:38.16 | RobotGuy | I want to be able to do the mesh networking. That's very important to me. |
| 00:38.30 | g1powermac_PB | they want you to use a mps30 with some of the other products |
| 00:38.52 | g1powermac_PB | ahh |
| 00:38.53 | g1powermac_PB | hmm |
| 00:39.37 | g1powermac_PB | still can prolly homebrew your own mesh protocol built ontop of the hardware. . .I know I might :-0 |
| 00:40.59 | RobotGuy | Hmmmmm, interesting possibility, but I'd rather not reinvent the wheel. |
| 00:41.23 | RobotGuy | Even though it would probably not be too hard to do using sockets. |
| 00:41.47 | g1powermac_PB | heh, as a former CS major, we were thoroughly trained in reinventing the wheel ;-) |
| 00:43.23 | RobotGuy | I am sure! What would you use to create your own mesh protocol? |
| 00:43.36 | g1powermac_PB | not sure, would have to do some research |
| 00:43.49 | RobotGuy | Might be interesting because it may be usable over any wireless link. |
| 00:43.50 | g1powermac_PB | obviously the first thing would be abstracting the hardware stuff first |
| 00:44.04 | g1powermac_PB | basically create a custom MAC |
| 00:44.27 | RobotGuy | Media Access Controllers? |
| 00:45.13 | g1powermac_PB | yea |
| 00:45.51 | g1powermac_PB | err, actually more like Medium Access Control |
| 00:46.37 | g1powermac_PB | but I think we could do something |
| 00:48.03 | RobotGuy | Sienna: What is the p/n of the level shifters you are using on Hammer-RDP? |
| 00:48.38 | RobotGuy | Is this something that would require kernel drivers/mods? |
| 00:48.53 | Sienna | no, they are voltage only |
| 00:48.59 | Sienna | TI TXB0104 |
| 00:49.13 | RobotGuy | I wish they had a cc2480 with the location module. |
| 00:49.17 | RobotGuy | Sienna: Thanks |
| 00:49.49 | g1powermac_PB | RobotGuy: oh, we might have to do a bit of hacking to sdcc to handle the cc2431 |
| 00:49.58 | g1powermac_PB | mainly because it has 128k flash |
| 00:50.24 | RobotGuy | Sienna: Any other TI parts I should get? |
| 00:50.37 | Sienna | sec... |
| 00:50.43 | g1powermac_PB | and its all pages |
| 00:51.24 | RobotGuy | Could a software location module be done? |
| 00:51.41 | g1powermac_PB | that I'm not sure |
| 00:52.15 | SpeedEvil | location is a problem |
| 00:52.21 | SpeedEvil | it relies on very precise timing |
| 00:52.30 | g1powermac_PB | yea |
| 00:52.31 | Sienna | RobotGuy, ADS7844 |
| 00:52.32 | SpeedEvil | Signal strength - sure |
| 00:54.11 | g1powermac_PB | now this project has a hacked up sdcc for cc2430 support: http://sourceforge.net/project/showfiles.php?group_id=205346 |
| 00:54.31 | g1powermac_PB | it also has a flasher program, but it assumes using the TI board |
| 00:54.36 | RobotGuy | Sienna: Got it. Anything else? Got space to get four more parts. :) |
| 00:55.16 | Sienna | Those are the only two parts i used |
| 00:55.28 | RobotGuy | Sienna: OK, great. :) |
| 00:55.55 | RobotGuy | g1powermac_PB: Interesting. If that supports the cc2430, it seems it might be easy to extend that to the cc2431. |
| 00:56.20 | g1powermac_PB | yea, no sweat for the cc2431, since it really is the same chip+location module |
| 01:00.12 | RobotGuy | I'm downloading the SDCC 2.7 patched sources now |
| 01:00.35 | g1powermac_PB | is looking at the patch, seeing what it does |
| 01:00.46 | RobotGuy | What does the 805x CPU require for flashing software form Linux? |
| 01:01.02 | g1powermac_PB | its specific for each chip |
| 01:01.04 | RobotGuy | Standard JTAG? |
| 01:01.13 | g1powermac_PB | this one does some spi interface |
| 01:01.22 | g1powermac_PB | no, almost none use jtag |
| 01:01.28 | GPSFan | g1powermac_PB: hey, not much new witht the z2 stuff, I've been reading the blob init sequences in prep for moving it to APEX. blob is a maze of twisty little passages....... |
| 01:01.29 | RobotGuy | Hmmmm |
| 01:01.42 | g1powermac_PB | GPSFan: k |
| 01:01.44 | RobotGuy | is away: Gone away for now. |
| 01:02.13 | GPSFan | g1powermac_PB: been real busy with farm stuff. maybe i'll get a chanc e to do some more z2 stuff later this week. |
| 01:02.20 | g1powermac_PB | RobotGuy: yea, many use the uart to load the code, others are parallel only, and then others do their own thing like this one |
| 01:02.30 | g1powermac_PB | GPSFan: k, thats good |
| 01:02.36 | RobotGuy | All alike! |
| 01:02.53 | g1powermac_PB | is going to be insanely busy with moving very shortly |
| 01:03.25 | GPSFan | g1powermac_PB: well summer is here, 75+ today. nice werther, timeto work ma a$$ off. |
| 01:03.28 | RobotGuy-Out | returns. |
| 01:03.38 | g1powermac_PB | err, that *sounds good |
| 01:04.01 | RobotGuy | I'm not sure I want to mess with the 805x processors, even for the cc2431. |
| 01:04.08 | g1powermac_PB | GPSFan: ya, I here ya |
| 01:04.43 | g1powermac_PB | RobotGuy: once you got a dev system working, they're easy to use |
| 01:05.14 | RobotGuy | g1powermac_PB: I'm sure they are. It's getting the built software to the chip that has me worried. |
| 01:05.29 | RobotGuy | I don't have any problems programming in C |
| 01:06.04 | g1powermac_PB | don't be, I'll be setting one up :-) |
| 01:06.27 | g1powermac_PB | once I get the hard stuff done, I'll put instructions up online on how to do it |
| 01:06.55 | RobotGuy | g1powermac_PB: OK, that will be good, since you already know the chips. I just want to do software. :) |
| 01:07.27 | g1powermac_PB | it really shouldn't be hard to load stuff on this chip with the flyswatter. . .it has spi on it right? |
| 01:07.53 | RobotGuy | g1powermac_PB: Yes, I believe Flyswatter does SPI. |
| 01:08.07 | RobotGuy | And I2C |
| 01:08.28 | g1powermac_PB | yea, with that then, the only thing would be to write a simple loader program |
| 01:09.04 | RobotGuy | g1powermac_PB: I leave the loader program stuff to you. :) |
| 01:09.40 | g1powermac_PB | hehe :-) |
| 01:10.28 | g1powermac_PB | of course I need to get a flyswatter ;-) |
| 01:11.11 | g1powermac_PB | oh, the one really nice thing this chip has is a whole debug subsystem, something I've never seen in an 8051 |
| 01:11.28 | g1powermac_PB | it reminds me of what you can do with JTAG |
| 01:12.02 | g1powermac_PB | you can breakstep code, access all the registers, and program the flash from it |
| 01:16.30 | RobotGuy | Hmmmm, OK, I am still looking at these 805x thingies.. :) |
| 01:16.50 | RobotGuy | The CC2431 really is too interesting to pass on. |
| 01:17.01 | g1powermac_PB | yea |
| 01:17.05 | RobotGuy | It would be nice to pair one with Hammer. |
| 01:17.17 | g1powermac_PB | it really is the most complicated 8051 I've seen |
| 01:27.47 | RobotGuy | g1powermac_PB: I can see a communications board for Hammer shaping up here. |
| 01:27.56 | g1powermac_PB | yea |
| 01:27.58 | g1powermac_PB | hmm |
| 01:28.08 | g1powermac_PB | is seeing a bootloader for this chip. . . |
| 01:28.16 | RobotGuy | Something to stack with the Hammer-RDP |
| 01:28.40 | g1powermac_PB | though, meh, still going to need a loader on the computer side to initially load the bootloader |
| 01:28.54 | RobotGuy | Sienna: How is the stacking going to work.. With the Hammer-RDP be the top board always, given the height of the Hammer JTAG? |
| 01:29.26 | Sienna | um, actually i have an RFQ in with SAMTEC for .635" high sockets |
| 01:30.13 | RobotGuy | Sienna: Ahhh, OK, so that would clear the Hammer-JTAG then. Dave says he is considering a right angle or edge connector for JTAG on later versions of Hammer. |
| 01:30.34 | Sienna | oh, thank goodness! |
| 01:30.40 | Sienna | edge please! |
| 01:30.49 | Sienna | right angle is still too high |
| 01:30.57 | RobotGuy | Sienna: I have been talking to him about that. I also like the edge connector idea. |
| 01:31.03 | Sienna | and no, .635 won't clear the JTAG, just those pins that stick up |
| 01:31.56 | RobotGuy | Hmmmmm, that will not be good for stacking on top of Hammer. Maybe Hammer-RDP will just have to be the top board (too bad) all the time and others stack below it? |
| 01:32.24 | Sienna | thats ok for stacking... just means the stack can't cover the JTAG, which has always been my plan |
| 01:32.51 | RobotGuy | Sienna: A cut out in stacked boards to clear the JTAG? |
| 01:33.19 | Sienna | no, just restrict the dimension of the stack boards |
| 01:33.44 | RobotGuy | Sienna: Ahhh, OK, so the stacked boards would be a little shorter than Hammer-RDP. |
| 01:33.44 | Sienna | also, i am considering cutting out the power regs from the RDP, and making those a 'half width' stacking board |
| 01:34.04 | RobotGuy | Sienna: I don't think removing the regulators is a good idea. |
| 01:34.48 | Sienna | then you could use the 'default' low power regs, or swap in a 5A reg for USB / heavy draw. its actually a good idea |
| 01:34.52 | Sienna | (i think) |
| 01:34.52 | g1powermac_PB | man this chip is complicated. . . |
| 01:35.23 | g1powermac_PB | its far from a at89c2051 that I started with :-) |
| 01:35.28 | RobotGuy | Sienna: Ah, I have it! Combine the regulator board with a proto board! |
| 01:35.42 | RobotGuy | g1powermac_PB: It's interesting though, yes? |
| 01:35.52 | g1powermac_PB | heck yea |
| 01:36.08 | RobotGuy | Sienna: The regulator board would also include the breakout/protoboard area. |
| 01:36.19 | Sienna | if you built it maybe |
| 01:36.39 | Sienna | i think it more appropriate that a proto board be an entire stack area |
| 01:36.48 | RobotGuy | Sienna: Isn't it a good idea though? |
| 01:37.26 | RobotGuy | Sienna: I don't think it can be worked both ways. I think it's necessary to have the full size of the board used, not half size it. |
| 01:37.42 | RobotGuy | Sienna: OK, how about a regulator/usb hub board then? |
| 01:37.55 | Sienna | I wasn't planning on giving the tiny regulator a full stack connection, just power |
| 01:38.14 | Sienna | no, usb will require switching regulators, i don't know if the size is right |
| 01:38.21 | RobotGuy | A half size stack board would interfere with other stacking. |
| 01:38.28 | Sienna | there will be a hub stacking board too, don't worry |
| 01:38.36 | Sienna | not if it was always the top |
| 01:38.56 | Sienna | and I plan to make everything stand-off able |
| 01:38.58 | RobotGuy | Sienna: I don't know. I prefer that all stack boards be full sized. |
| 01:39.18 | Sienna | and remember, these are only my prototypes... TCT can completely ignore me! |
| 01:39.54 | RobotGuy | Sienna: I don't think TCT is going to ignore your design. Dave was telling me Rusty is likely going to use your design with few changes. |
| 01:40.46 | Sienna | anyway, i am off to bed |
| 01:40.50 | Sienna | night! |
| 01:40.52 | RobotGuy | If we are going to a stack board for power regulation, we might as well use the rest of the real estate for something. |
| 01:41.44 | RobotGuy | I should go get some food. Haven |
| 01:41.50 | RobotGuy | t eaten all day. |
| 01:47.31 | RobotGuy | BBIAB |
| 03:00.06 | *** join/#edev jdav_gone (n=james@d75-157-13-250.bchsia.telus.net) |
| 03:10.30 | *** join/#edev roxfan2 (n=dunno@202.55-136-217.adsl-dyn.isp.belgacom.be) |
| 03:55.47 | *** join/#edev DrZimmerman (n=theo@tbaumgartner.broker.freenet6.net) |
| 04:07.06 | *** join/#edev jayakumar (n=jayakuma@14.29.111.218.brf01-home.tm.net.my) |
| 04:10.02 | *** join/#edev Crofton (n=balister@66-207-66-26.black.dmt.ntelos.net) |
| 04:24.04 | *** join/#edev mbuf (n=shakthim@61.16.248.242) |
| 04:28.33 | *** join/#edev purl (i=ibot@pdpc/supporter/active/TimRiker/bot/apt) |
| 04:28.33 | *** topic/#edev is a channel for the discussion of hacking small consumer electronic devices, mostly ARM based; Hammer Development boards now available at http://www.tincantools.com/ |
| 04:50.40 | *** join/#edev RobotGuy (n=robotguy@dsl093-038-072.pdx1.dsl.speakeasy.net) |
| 05:15.03 | *** join/#edev ejdav_gon (n=james@d75-157-13-250.bchsia.telus.net) |
| 05:25.37 | RobotGuy | thinks about going back to Kubuntu. |
| 05:41.19 | ds2 | \ |
| 05:44.10 | RobotGuy | <PROTECTED> |
| 06:13.22 | *** join/#edev drath_ (i=vmaster@p5B07CEE9.dip.t-dialin.net) |
| 06:14.41 | Mirell | Hmm... |
| 06:16.02 | RobotGuy | Mirell: Hi |
| 06:16.11 | Mirell | 'allo RobotGuy |
| 06:16.23 | Mirell | is debating about where to go for vacation |
| 06:16.45 | RobotGuy | is debating whether he can make it to the new month |
| 06:16.54 | Mirell | ? |
| 06:18.27 | RobotGuy | I'm broke and almost no food. |
| 06:19.08 | Mirell | Ouch. |
| 06:19.28 | RobotGuy | I'll call for a food box tomorrow and maybe that will be enough. |
| 06:27.11 | *** join/#edev mbuf (n=shakthim@61.16.248.242) |
| 06:35.18 | *** join/#edev hw_ (n=hw@p578b3905.dip0.t-ipconnect.de) |
| 06:40.02 | *** join/#edev RobotGUI (n=robotguy@dsl093-038-072.pdx1.dsl.speakeasy.net) |
| 06:51.36 | RobotGuy | is away: Gone away for now. |
| 06:52.23 | RobotGuy | returns. |
| 07:03.53 | *** join/#edev howtoo (n=howtoo@195.159.62.66) |
| 07:14.41 | *** join/#edev [AD]Turbo (n=turbo@host24-241-dynamic.22-79-r.retail.telecomitalia.it) |
| 07:16.56 | [AD]Turbo | hola |
| 07:22.54 | *** join/#edev RobotGuy (n=robotguy@dsl093-038-072.pdx1.dsl.speakeasy.net) |
| 07:42.18 | *** join/#edev steliosk (n=Stelios@athedsl-284816.home.otenet.gr) |
| 07:57.38 | *** join/#edev likewise (n=Leon_Woe@atwork-186.r-212.178.104.atwork.nl) |
| 08:14.35 | *** join/#edev RobotGuy (n=robotguy@dsl093-038-072.pdx1.dsl.speakeasy.net) |
| 08:15.48 | *** join/#edev chouimat|busy (n=dieu@r2351064.cidc.net) |
| 08:21.17 | *** join/#edev RobotGuy (n=robotguy@dsl093-038-072.pdx1.dsl.speakeasy.net) |
| 08:22.01 | *** join/#edev mbuf (n=shakthim@61.16.248.242) |
| 08:23.45 | Mirell | I went ahead and got this shirt: http://www.bustedtees.com/welldoitlive |
| 08:27.37 | *** join/#edev samuelr (n=samuelr@dsl093-061-094.pit1.dsl.speakeasy.net) |
| 08:27.59 | *** join/#edev RobotGuy (n=robotguy@dsl093-038-072.pdx1.dsl.speakeasy.net) |
| 08:37.05 | *** join/#edev RobotGuy (n=robotguy@dsl093-038-072.pdx1.dsl.speakeasy.net) |
| 08:39.31 | *** join/#edev lyakh (n=lyakh@p57BD248A.dip0.t-ipconnect.de) |
| 08:47.39 | *** join/#edev chouimat|busy (n=dieu@r2351064.cidc.net) |
| 08:53.58 | *** join/#edev steliosk (n=Stelios@athedsl-284816.home.otenet.gr) [NETSPLIT VICTIM] |
| 09:00.49 | *** join/#edev hw_ (n=hw@p578b3905.dip0.t-ipconnect.de) |
| 09:01.27 | *** join/#edev _trine (n=trine@38.100.90.3) |
| 09:11.19 | *** join/#edev RobotGuy (n=robotguy@dsl093-038-072.pdx1.dsl.speakeasy.net) |
| 09:12.51 | *** join/#edev steliosk (n=Stelios@athedsl-284816.home.otenet.gr) |
| 09:20.12 | *** join/#edev linac (n=lin@117.24.1.22) |
| 09:54.33 | *** join/#edev steliosk (n=Stelios@athedsl-284816.home.otenet.gr) [NETSPLIT VICTIM] |
| 09:58.19 | *** join/#edev steliosk (n=Stelios@athedsl-284816.home.otenet.gr) [NETSPLIT VICTIM] |
| 09:59.23 | *** join/#edev samuelr (n=samuelr@dsl093-061-094.pit1.dsl.speakeasy.net) |
| 10:16.34 | *** join/#edev steliosk (n=Stelios@79.131.123.2) |
| 10:16.41 | *** join/#edev steliosk (n=Stelios@athedsl-284816.home.otenet.gr) |
| 10:22.19 | *** join/#edev steliosk (n=Stelios@79.131.123.2) |
| 10:40.25 | *** join/#edev samuel1 (n=samuelr@dsl093-061-094.pit1.dsl.speakeasy.net) |
| 11:24.54 | *** join/#edev g1powermac_PB (n=g1powerm@unaffiliated/g1powermac) |
| 11:35.52 | *** join/#edev KazaLite (n=kazaLite@203.215.180.205) |
| 12:12.03 | *** join/#edev GPSFan (n=kenm@12.10.255.246) |
| 12:12.03 | *** mode/#edev [+o GPSFan] by ChanServ |
| 12:15.04 | g1powermac_PB | morning GPSFan |
| 12:15.17 | GPSFan | g1powermac_PB: g'day |
| 12:19.39 | g1powermac_PB | needs to trace some C++ code, figure out what exactly its doing. . . |
| 12:24.58 | *** join/#edev sjhill (n=sjhill@real.realitydiluted.com) |
| 12:24.58 | *** mode/#edev [+o sjhill] by ChanServ |
| 12:50.09 | *** join/#edev prpplague (n=dave@mail.americanmicrosystems.com) |
| 12:54.31 | prpplague | ho ho hum |
| 12:54.39 | prpplague | yet another monday |
| 12:55.44 | g1powermac_PB | hey prpplague |
| 12:55.54 | prpplague | g1powermac_PB: greetings |
| 12:56.14 | g1powermac_PB | found a really interesting chip over the weekend |
| 12:56.21 | g1powermac_PB | its the TI cc2431 |
| 12:57.47 | g1powermac_PB | the interesting part is the 8051+zigbee integration plus a location module that allows you to estimate the physical location of nodes in the network |
| 13:00.17 | g1powermac_PB | is now trying to figure out how to load code into the chip |
| 13:00.39 | g1powermac_PB | its not simple it seems, since this chip is quite complicated |
| 13:01.31 | prpplague | g1powermac_PB: indeed that is interesting, however.... |
| 13:01.48 | prpplague | g1powermac_PB: i find doing alot of that very dedicated dev doesn't increase your embedded knowledge |
| 13:02.04 | prpplague | g1powermac_PB: granted knowing 8051 programming is of some value |
| 13:02.12 | g1powermac_PB | meh, maybe true |
| 13:02.17 | prpplague | g1powermac_PB: but these days avr is far more important |
| 13:02.20 | g1powermac_PB | but I'm an 8051 guy to begin with |
| 13:02.33 | g1powermac_PB | grew up on 8051s :-) |
| 13:03.20 | g1powermac_PB | so its not like I'm learning a new architecture |
| 13:03.48 | g1powermac_PB | but I do agree avrs are prolly more pertinent today |
| 13:04.12 | prpplague | g1powermac_PB: ahh |
| 13:04.49 | g1powermac_PB | my first mcu I worked on was the lowly at89c2051 :-) |
| 13:05.17 | prpplague | ahh |
| 13:06.06 | prpplague | ho ho hum |
| 13:06.12 | g1powermac_PB | had bought a microcontroller kit from iguana labs back quite a many years ago that was based on the at89c2051 |
| 13:06.21 | g1powermac_PB | thats what started me on electronics :-) |
| 13:10.34 | g1powermac_PB | man this chip is complicated to load code into |
| 13:11.14 | prpplague | ahh |
| 13:11.23 | g1powermac_PB | I'm going to need a flyswatter to do it too I think |
| 13:11.43 | g1powermac_PB | it wants a spi like interface to gain access to its debug interface |
| 13:11.50 | g1powermac_PB | which is then used to load code |
| 13:11.53 | g1powermac_PB | kinda like JTAG |
| 13:12.27 | g1powermac_PB | but I'm unsure exactly on the spi interface it wants |
| 13:14.23 | SpeedEvil | Anyone happen to know if if a device cannot work on USB1.1 due to insufficient bandwidth, if it's still supposed to supply config information? I've got a device that does not show anythin on syslog on plugin on usb1.1 systems |
| 13:14.31 | SpeedEvil | (crappy USB webcam) |
| 13:18.21 | prpplague | SpeedEvil: a usb2.0 device should always identify itself |
| 13:18.34 | SpeedEvil | K, thought so. |
| 13:18.41 | SpeedEvil | Thanks. |
| 13:18.41 | purl | pas de quoi, SpeedEvil |
| 13:30.56 | g1powermac_PB | wonders if he could reuse his parallel port to jtag cable to interface with this chip |
| 13:31.33 | prpplague | g1powermac_PB: probably |
| 13:32.06 | g1powermac_PB | http://www.digilentinc.com/Products/Catalog.cfm?Nav1=Products&Nav2=Cables&Cat=Cable |
| 13:32.11 | g1powermac_PB | mines the JTAG3 cable |
| 13:33.26 | prpplague | probably just a big bang parport via a 244 |
| 13:34.38 | g1powermac_PB | http://www.avrfreaks.net/index.php?name=PNphpBB2&file=viewtopic&p=201869&sid=1e517e00d8f0be2061871b0afea6610e#201869 |
| 13:34.46 | g1powermac_PB | there's the pinout for the cable |
| 13:36.47 | prpplague | g1powermac_PB: yea, you proably could use that but you;d need to figure out what data lines and which input line is used on the parport side |
| 13:36.56 | g1powermac_PB | do you know of a good parport bit bang spi example? |
| 13:36.58 | prpplague | g1powermac_PB: then you have to write the software to bit bang that |
| 13:37.13 | prpplague | g1powermac_PB: you can look at jflash on the lart page |
| 13:37.18 | g1powermac_PB | k |
| 13:37.33 | g1powermac_PB | that link does show what parallel port lines it uses |
| 13:37.51 | g1powermac_PB | "The signals are routed as follows on the parallel cable end?P2=TDI, P3= TCK, P4=TMS, P13=TDO and P18=Gnd (to provide common grd between the PC and the board." |
| 13:39.18 | prpplague | ahh, i read it as the pinout on the jtag end |
| 13:42.08 | ce_geek | prpplague: Mornin' |
| 13:42.16 | prpplague | ce_geek: greetings |
| 13:42.23 | prpplague | ce_geek: how did it go with rick on friday? |
| 13:42.55 | ce_geek | Pretty good. I ran through some stuff with him. Tried as an example to compile the kernel without those debug messages but apparently didn't quite get the .config right. |
| 13:43.16 | prpplague | ce_geek: ahh |
| 13:43.27 | ce_geek | prpplague: Basically just ran through all the things that I had to deal with to get to the point where I had a web-based temperature server |
| 13:43.58 | prpplague | ce_geek: so what was Rick's impression of the kit and/or your work? |
| 13:44.00 | g1powermac_PB | prpplague: btw, hows the zipit stuff? |
| 13:44.14 | prpplague | g1powermac_PB: nada, been swampped |
| 13:44.27 | ce_geek | prpplague: He was impressed, thought it's all really neat. |
| 13:44.28 | prpplague | g1powermac_PB: i'm actually working on the 8686 sdio and gspi stuff this morning |
| 13:44.33 | prpplague | g1powermac_PB: for another project |
| 13:44.39 | prpplague | ce_geek: great |
| 13:44.51 | prpplague | ce_geek: think he'll want to use it in the future? |
| 13:45.19 | g1powermac_PB | prpplague: oh cool |
| 13:45.39 | g1powermac_PB | prpplague: just let me know when ya ready to send it back |
| 13:45.56 | prpplague | g1powermac_PB: yea, i know you are chomping at the bits to get it |
| 13:46.11 | prpplague | g1powermac_PB: sorry for the delay, but gotta pay the bills first |
| 13:47.43 | prpplague | ce_geek: atleast it worked out well to help finish up your final class work |
| 13:48.19 | ce_geek | prpplague: Yup! |
| 13:48.36 | g1powermac_PB | prpplague: yup, totally understand |
| 13:49.03 | g1powermac_PB | is going to be in the process of moving shortly anyway |
| 13:49.16 | ce_geek | prpplague: I do think he'll use it, he's just going to have to spend some time with it and get familiar before he teaches it. I would imagine that at first it will be something he'll offer to students to do as a project in the 400 level class he teaches so they can learn something new perhaps. |
| 13:49.43 | prpplague | ce_geek: ahh |
| 13:49.51 | ce_geek | prpplague: I'm taking the robot that I built to NH with me so I have that to play with and make better for department demos. I'd like to get it Hammer-driven, that will be huge. |
| 13:50.00 | prpplague | ce_geek: well i hope you leave him with all our contact info |
| 13:50.25 | g1powermac_PB | mmm |
| 13:50.43 | ce_geek | prpplague: Oh gosh yes. I'll be in constant contact with him for a variety of reasons. He's the kind of guy I'll be visiting when I come back up to Orono to visit and stuff. |
| 13:50.54 | ce_geek | prpplague: Plus I enjoy going for flights in his Cessna :-P |
| 13:51.00 | g1powermac_PB | it seems avrs use basically an spi interface to program them as well, and there's quite a few programs doing it over the parport out there |
| 13:51.48 | prpplague | ce_geek: hehe |
| 13:51.56 | prpplague | g1powermac_PB: yea |
| 13:52.04 | prpplague | g1powermac_PB: its not true SPI |
| 13:52.14 | prpplague | g1powermac_PB: they call it In System Programmer |
| 13:52.19 | prpplague | g1powermac_PB: ISP |
| 13:52.19 | g1powermac_PB | ahh |
| 13:52.22 | ce_geek | prpplague: Did you need any information to get the other kit to me? |
| 13:52.44 | g1powermac_PB | well, not sure how true the spi is on this ti chip is either |
| 13:52.54 | prpplague | ce_geek: yea, i think the boss hasn't got your get allocated yet, it probably will be a week or so |
| 13:53.10 | prpplague | ce_geek: so i'll need to get an address other than the university iirc |
| 13:55.03 | ce_geek | prpplague: No problem. |
| 13:56.25 | ce_geek | prpplague: Sent to your email. |
| 13:56.32 | prpplague | ce_geek: ok thanks |
| 13:59.02 | prpplague | ho ho hum |
| 13:59.11 | prpplague | lots of debugging today |
| 14:05.17 | *** part/#edev samuel1 (n=samuelr@dsl093-061-094.pit1.dsl.speakeasy.net) |
| 14:18.40 | prpplague | cbrake: still tinkering with the libertas driver? |
| 14:19.06 | cbrake | prpplague: probably tomorrow I'll pick it back up |
| 14:19.27 | prpplague | cbrake: ahh ok |
| 14:19.29 | cbrake | prpplague: I'm going to set up an unsecured network and get that going first instead of doing WPA first |
| 14:19.38 | cbrake | prpplague: walk and then run ... |
| 14:19.38 | prpplague | cbrake: i'll be doing some tinkering for the next couple of days |
| 14:19.47 | cbrake | prpplague: cool |
| 14:19.54 | prpplague | cbrake: i'm having some problems with it on my sdio controller |
| 14:39.24 | *** join/#edev lyakh (n=lyakh@p57BD248A.dip0.t-ipconnect.de) |
| 14:54.50 | *** join/#edev MothaWork (n=monmotha@66.208.186.250) |
| 14:56.02 | MothaWork | anybody know the expected BogoMIPS/MHz for a ARM9EJS? |
| 14:56.38 | MothaWork | ARM926EJS, in fact |
| 14:56.43 | prpplague | MonMotha: the bogo mips will roughly be 1/2 of the clock rating |
| 14:56.53 | MothaWork | ok, that's what I'm seeing |
| 14:57.04 | MothaWork | wanted to make sure my PLL was in fact getting set up properly |
| 14:57.29 | MothaWork | there must be some floating point or something sneaking into this application |
| 14:57.48 | MothaWork | is there a good way to detect when that happens (it's likely in a library)? |
| 14:59.11 | prpplague | MonMotha: no real good way |
| 14:59.16 | MothaWork | I figured... |
| 14:59.18 | prpplague | MonMotha: you using soft or hard float? |
| 14:59.58 | MothaWork | hard |
| 15:00.23 | prpplague | MonMotha: you can turn on some debugging in the FPE |
| 15:00.34 | prpplague | MonMotha: that might give you some clues |
| 15:07.29 | MothaWork | ok |
| 15:08.13 | *** join/#edev linac (n=lin@121.41.39.129) |
| 15:14.19 | MothaWork | hum, I don't see any FPE debugging options in Kconfig |
| 15:14.45 | prpplague | MonMotha: yea probably aren't any, you'll have to hack a define or something |
| 15:14.50 | MothaWork | awesome |
| 15:22.31 | *** join/#edev CosmicPenguin (n=nobody@163.181.251.103) |
| 15:22.31 | *** mode/#edev [+o CosmicPenguin] by ChanServ |
| 15:26.40 | prpplague | am i the only one that finds git totally unusable? |
| 15:28.17 | Crofton | no :) |
| 15:30.17 | *** join/#edev bbradley (n=bbradley@92-235-133-146.cable.ubr17.azte.blueyonder.co.uk) |
| 15:32.37 | *** join/#edev purl (i=ibot@pdpc/supporter/active/TimRiker/bot/apt) |
| 15:32.37 | *** topic/#edev is a channel for the discussion of hacking small consumer electronic devices, mostly ARM based; Hammer Development boards now available at http://www.tincantools.com/ |
| 15:40.50 | *** join/#edev purl (i=ibot@pdpc/supporter/active/TimRiker/bot/apt) |
| 15:40.50 | *** topic/#edev is a channel for the discussion of hacking small consumer electronic devices, mostly ARM based; Hammer Development boards now available at http://www.tincantools.com/ |
| 15:49.19 | prpplague | anyone familiar with gitweb? |
| 15:52.19 | *** join/#edev adyer (n=adyer@adsl-69-223-13-78.dsl.chcgil.ameritech.net) |
| 15:53.41 | prpplague | bangs head on desk |
| 15:58.51 | prpplague | i just can't seem to get the information i want from git |
| 16:04.07 | adyer | . o O (ve haf vays ov making you talk) |
| 16:07.59 | CosmicPenguin | What are you trying to get? |
| 16:14.20 | MothaWork | gah, anybody know any good profiling tools that I could actually run on my ARM system? |
| 16:14.42 | MothaWork | this app runs blazing fast on even the most lowly old PC I can find for it, but takes several seconds to do anything on this ARM machine |
| 16:14.51 | MothaWork | but since it runs so fast on a PC, it doesn't accumulate any time in a profiler |
| 16:15.08 | Crofton | oprofile |
| 16:17.38 | SpeedEvil | MothaWork: I'll sell you a slow laptop. |
| 16:17.49 | SpeedEvil | I have a 486/75 somewhere :) |
| 16:18.46 | MothaWork | I've run it on similar hardware |
| 16:18.49 | MothaWork | blazing fast |
| 16:19.16 | MothaWork | which is why I suspected floating point (my 486 is a DX), but I don't see any floating point in my debug stuff, though I'm not sure it's outputting where it should |
| 16:24.57 | MothaWork | great, so I guess I get to cross-build oprofile |
| 16:24.59 | MothaWork | that's sure to be fun |
| 16:29.53 | *** join/#edev kaza_lite (n=danishJa@116.71.171.107) |
| 16:31.45 | *** join/#edev lyakh (n=lyakh@p57BD248A.dip0.t-ipconnect.de) |
| 16:49.13 | prpplague | CosmicPenguin: was trying to get a snapshot from a specific tag and head |
| 16:50.01 | prpplague | CosmicPenguin: i think i was able to pull the right one now |
| 16:50.08 | CosmicPenguin | using git-archive? |
| 16:50.20 | prpplague | CosmicPenguin: gitweb |
| 16:50.26 | CosmicPenguin | oh, even easier.. :) |
| 16:50.36 | prpplague | CosmicPenguin: doesn't seem so |
| 16:51.59 | prpplague | CosmicPenguin: http://opensource.wolfsonmicro.com/cgi-bin/gitweb.cgi?p=linux-2.6-asoc.git;a=summary |
| 16:52.10 | prpplague | CosmicPenguin: trying to get the 2.6.25 tag from the dev head |
| 16:52.48 | *** mode/#edev [+o prpplague] by ChanServ |
| 16:53.21 | CosmicPenguin | http://opensource.wolfsonmicro.com/cgi-bin/gitweb.cgi?p=linux-2.6-asoc.git;a=tree;h=735d446d1c39671146ca96cf5dea59ec35e433f2;hb=4b119e21d0c66c22e8ca03df05d9de623d0eb50f |
| 16:54.00 | prpplague | CosmicPenguin: and that is from dev head? |
| 16:54.08 | CosmicPenguin | I don't understand your question |
| 16:54.16 | CosmicPenguin | either you want HEAD or a tag |
| 16:54.23 | CosmicPenguin | they are exclusive operations |
| 16:54.27 | prpplague | ahh |
| 16:54.43 | CosmicPenguin | that is the commit that was tagged as 2.6.25 |
| 16:55.20 | prpplague | i guess i don't understand git, which is probably why i'm having trouble using it |
| 16:55.54 | prpplague | CosmicPenguin: under cvs i could select a branch and then a specific tag |
| 17:07.17 | *** join/#edev MothaWork (n=monmotha@66.208.186.250) |
| 17:16.10 | prpplague | CosmicPenguin: whats the reasoning behind having the sound support in the main dir rather than under the drivers section? |
| 17:16.19 | prpplague | CosmicPenguin: for the linux kernel that is |
| 17:18.01 | adyer | prpplague: I thought that was a leftover from long ago in the OSS days |
| 17:18.25 | CosmicPenguin | I'm not sure |
| 17:18.41 | CosmicPenguin | if I had to guess it was that they were going for a similar model to net/ |
| 17:19.06 | CosmicPenguin | and then for wahtever reason kept the drivers in the sound/ directory - either to make it easier to maintain or sync with the external tree |
| 17:20.42 | prpplague | ahh |
| 17:39.28 | *** join/#edev like2wise (n=likewise@82-171-51-231.ip.telfort.nl) |
| 17:44.09 | Lars_G | I'll see you guys latter, too little attention span for this channel today |
| 17:44.10 | Lars_G | bows |
| 17:44.11 | *** part/#edev Lars_G (n=Lars@unaffiliated/lars-g/x-000001) |
| 17:44.22 | MothaWork | gah, oprofile is just not being nice |
| 17:47.51 | prpplague | CosmicPenguin: still think MS + OLPC is a good thing? |
| 17:48.16 | CosmicPenguin | absolutely |
| 17:48.49 | chouimat | hi |
| 17:50.12 | prpplague | CosmicPenguin: funny how marketing puts a spin on things |
| 17:50.41 | chouimat | is bored ... |
| 17:51.51 | prpplague | CosmicPenguin: most of the PHB's i've spoken with say something to the effect, "well looks like open source couldn't do the job so olpc went crawling back to MS" |
| 17:52.06 | CosmicPenguin | Is there anything wrong with that statement? |
| 17:52.33 | prpplague | CosmicPenguin: you feel that the OSS didn't do the job? |
| 17:52.38 | CosmicPenguin | I |
| 17:52.46 | chouimat | CosmicPenguin: depend if you're an oss zealot or a pragmatic ;) |
| 17:52.54 | CosmicPenguin | I am not saying that - but do you think that statement is false in any way? |
| 17:53.08 | CosmicPenguin | or that it is a misguided opnion? |
| 17:53.37 | prpplague | CosmicPenguin: yes i feel it is inaccurate, but its based on people who don't know the truth of the situation, its more of what people "believe" |
| 17:53.53 | CosmicPenguin | certainly there haven't been any glowing reviews of the UI anywhere |
| 17:55.02 | *** join/#edev g1powermac (n=g1powerm@unaffiliated/g1powermac) |
| 17:56.06 | MothaWork | near as I can tell, OLPC has been a miserable failure |
| 17:56.17 | MothaWork | they did a bunch of neat things, but couldn't bring it together |
| 17:56.51 | CosmicPenguin | I'm not saying it was or it wasn't - I have an opinion, but I know enough to keep it to myself |
| 17:57.19 | CosmicPenguin | What I am saying is that you shouldn't assume that if OLPC starts loading Microsoft that it is because of evil reasons |
| 17:57.43 | CosmicPenguin | OSS needs to be willing to accept the fact that it might have failed |
| 17:57.49 | prpplague | CosmicPenguin: not venting your opinion eh? |
| 17:59.06 | prpplague | CosmicPenguin: i'm with MothaWork , from my little experience with the OLPC, i'd consider it a failure |
| 17:59.52 | CosmicPenguin | So based on that - the statement you mentioned earlier seems to be well founded |
| 18:02.26 | MothaWork | I'm not sure that OSS failed so much as OLPC failed to let it succeed, and now are looking for a commercial entity to effectively bail the project out |
| 18:02.34 | prpplague | CosmicPenguin: i think it was more of a failure of the management and organization of the OSS projects than i do the capabilities |
| 18:02.53 | MothaWork | which many would be happy to do in order to get the exposure |
| 18:03.30 | CosmicPenguin | failure is failure, regardless of where you put the blame |
| 18:03.47 | CosmicPenguin | if you read the average blog, then you'll see that most people think that sugar is unusable |
| 18:04.16 | MothaWork | it suffers from the same thing many commercial products of its nature do: it was released before it was ready |
| 18:04.19 | CosmicPenguin | So who is to blame? management? the UI designers? The individual programmers? |
| 18:04.37 | MothaWork | one thing opensource doesn't do well is deliver custom applications on a timetable |
| 18:05.12 | MothaWork | I don't think that's something one can expect it to do, at least not a timetable that isn't determined by a large assortment of people with an interest in the software project itself and not any of the business interests behind it |
| 18:05.13 | prpplague | CosmicPenguin: did sugar do well with the target audience of kids? |
| 18:06.00 | CosmicPenguin | I don't inow |
| 18:06.02 | CosmicPenguin | I haven't heard |
| 18:06.45 | CosmicPenguin | But regardless of who failed - the executive summary is that an open source project was given every advantage over other competitors to the tune of a 2 year lead time |
| 18:07.26 | prpplague | CosmicPenguin: interesting |
| 18:07.33 | CosmicPenguin | And they suffered what all large and innovative projects do - feature creep and failure to meet schedules |
| 18:07.44 | CosmicPenguin | Which we don't mind pointing out when a commercial entity is behind it |
| 18:07.48 | prpplague | CosmicPenguin: thats what i was seeing |
| 18:07.56 | CosmicPenguin | but it is very difficult for us to stomach when it is our own kind |
| 18:08.04 | prpplague | indeed |
| 18:08.17 | CosmicPenguin | especially since every time a new Microsoft operating system is delayed, the blogs catch on fire with "this is why open source is better" |
| 18:08.31 | CosmicPenguin | Today, open source was not better. It might not have been worse, but we have to learn from our mistakes |
| 18:08.35 | chouimat | hmmm so the scope of the project was too big or the timetable was unrealistic ... |
| 18:08.38 | CosmicPenguin | The problem is - nobody is willing to do that |
| 18:08.57 | CosmicPenguin | everybody would rather believe wild theories |
| 18:09.03 | CosmicPenguin | its easier to be the victim then the failure |
| 18:09.16 | prpplague | CosmicPenguin: well, my answer to that would a question, has anyone really pointed out what the mistakes were? |
| 18:09.25 | MothaWork | I'm not sure why anybody has any reason to believe that open source projects are "good" at delivering "on time" |
| 18:09.40 | CosmicPenguin | prpplague: I think people have, but they have been lost in the fan-boy noise |
| 18:09.40 | SpeedEvil | They are - if it's one committed guy in a shed. |
| 18:09.41 | MothaWork | I'd in fact venture to say that open source projects are noutorious for delivering late |
| 18:09.59 | chouimat | MothaWork: getting software on time is a myth |
| 18:10.09 | MothaWork | well, that is true to an extent, yes |
| 18:10.45 | CosmicPenguin | Personally, I think all of this is a good thing |
| 18:10.56 | prpplague | MonMotha: in general i do find OSS projects to be non-schedule oriented, however, when using OSS in business applications, with correct management, can be done on schedule, i do it daily |
| 18:11.14 | SpeedEvil | OSS isn't the problem. |
| 18:11.18 | CosmicPenguin | The community took the high road and leveled the playing field for everbody, even our most bitter rival |
| 18:11.21 | SpeedEvil | Managment is the problem |
| 18:11.40 | chouimat | prpplague: it's the same where I work but you need good managers |
| 18:11.41 | CosmicPenguin | And even though we may have lost on the field of battle, we are smarter for it |
| 18:11.46 | prpplague | CosmicPenguin: yea, maybe because i wasn't looking, but i've not really seen any authoritative statements about OSS mistakes with OLPC |
| 18:11.48 | MothaWork | prpplague: well, the lack of a schedule does not necessarily mean that OSS is not suitable for use in business applications. You just have to make sure that your business interests don't require that a schedule be met by the OSS project. |
| 18:12.06 | MothaWork | For example, if the project already has delivered what you need, then you don't have to worry about when the new version will com eout |
| 18:12.14 | prpplague | MonMotha: agreed |
| 18:12.26 | CosmicPenguin | prpplague: indeed - those people usually get shouted down |
| 18:12.57 | MothaWork | but if you develop something that relies on the next up-and-coming "best thing since laser sliced bread" feature that some project promises, don't be surprised if you get delayed by the project not getting that feature out the door, or even in to <insert VCS of the day here> |
| 18:13.10 | prpplague | CosmicPenguin: sounds like you are a little frustrated with that aspect of the OSS community behavior |
| 18:13.20 | prpplague | MonMotha: yea we never do that |
| 18:13.46 | MothaWork | no, you'd be a fool to |
| 18:13.52 | MothaWork | and that's just about what OLPC did |
| 18:13.58 | CosmicPenguin | prpplague: here's what i want - I want people to stop saying "We need to treet Microsoft like they treated us" and I want us to learn from our mistakes |
| 18:15.14 | CosmicPenguin | From an open source development perspective, OLPC was a success |
| 18:15.32 | CosmicPenguin | We developed open hardware and lots of great code |
| 18:15.48 | CosmicPenguin | It was so open and it was so great, in fact, that other operating systems came up on it in a matter of months, not years |
| 18:16.15 | CosmicPenguin | And it has spawned multiple subprojects and milestones that the rest of the community will be feeding from for years |
| 18:16.47 | CosmicPenguin | Case in point - there is now a free and open firmware stack that will boot Windows XP |
| 18:17.14 | CosmicPenguin | This makes a free BIOS solution a legitimate option for commercial vendors |
| 18:18.11 | CosmicPenguin | And I think, that in the end, some children will be educated |
| 18:18.36 | prpplague | CosmicPenguin: i can understand your point |
| 18:19.31 | CosmicPenguin | And AMD released dozens of documents and source code that wouldn't have been an option to release before this project came along |
| 18:19.55 | CosmicPenguin | So if anything, we can serve as an example for other vendors |
| 18:19.55 | prpplague | indeed and the libertas stuff as well |
| 18:20.51 | CosmicPenguin | So, there are many wins here that /. isn't going to report |
| 18:21.03 | prpplague | CosmicPenguin: yea |
| 18:21.08 | CosmicPenguin | And, if the open source leaders are smart, there are some loses that can be learned from |
| 18:21.22 | CosmicPenguin | I am encouraged though, because I think that there are smart folks out there paying attention |
| 18:21.26 | CosmicPenguin | even if they aren't being vocal |
| 18:22.05 | prpplague | CosmicPenguin: for our new M8050 pda, we'll be handing over to the contractors, working hardware, code examples, and complete functional specs, and they are still going to take 6 months to get wince up and running |
| 18:22.25 | prpplague | CosmicPenguin: i can only imagin if there were getting something untested |
| 18:22.31 | prpplague | CosmicPenguin: my guess would be a year |
| 18:23.24 | MothaWork | jeesh, I've gotten linux booting on such hardware in a matter of hours |
| 18:23.54 | CosmicPenguin | Thats part of the problem |
| 18:24.07 | CosmicPenguin | I had a picture of Linux booting on the XO in May of 2006 |
| 18:24.18 | CosmicPenguin | Its not the "getting it to boot" part that we struggle with |
| 18:24.26 | prpplague | getting it completed and production ready, thats a different story |
| 18:24.29 | CosmicPenguin | its the "making it useful" part |
| 18:24.43 | CosmicPenguin | 6 months is about right for a port + validation |
| 18:24.56 | prpplague | CosmicPenguin: for linux or for wince? |
| 18:25.05 | MothaWork | oh you were saying 6 months for complete usability |
| 18:25.08 | MothaWork | that's reasonable, I guess |
| 18:25.21 | *** join/#edev samuelr (n=samuelr@dsl093-061-094.pit1.dsl.speakeasy.net) |
| 18:25.24 | CosmicPenguin | Well - lets put it this way - the enterprise editions also run on a 6 to 8 month cycle |
| 18:25.24 | MothaWork | still, seems long |
| 18:25.57 | MothaWork | most of the time seems to be app development, which is not OS related |
| 18:26.03 | prpplague | CosmicPenguin: our linux port stuff is usally on a 90 days cycle |
| 19:12.49 | MothaWork | gah, I just cannot get oprofile to run on this system |
| 19:24.31 | prpplague | lordy lordy lordy |
| 19:25.30 | prpplague | interesting bit of info |
| 19:26.04 | prpplague | on aa.com if you select "price" as the primary search results, you get one set of prices |
| 19:26.27 | prpplague | and if you select "schedule" you get a set of prices that are about 20% less costly |
| 19:26.46 | prpplague | you can even select the same flights listed on the "price" section |
| 19:26.52 | prpplague | and they are still 20% less |
| 19:28.52 | ds2 | fun... penality to try to save time by searching by price |
| 19:30.48 | prpplague | $1900 by selecting the same flights unders schedule verses $2400 by selecting by price |
| 19:31.37 | ds2 | wonder what will they quote if you call them up |
| 19:33.04 | prpplague | ds2: it's an additional $25 per ticket if you order by phone |
| 19:33.25 | ds2 | but if they quote you $1800... ;) |
| 19:33.41 | ds2 | or maybe even $1600 |
| 19:33.43 | prpplague | ds2: yea, i doubt it would be that |
| 19:33.51 | prpplague | ds2: but i guess i could call and check |
| 19:34.41 | ds2 | these pricing games are rediculous |
| 19:34.53 | prpplague | one the phone now |
| 20:08.52 | MothaWork | anybody know oprofile very well? Is it possible to run the report tool on a set of data collected from another machine (that's a different arch)? |
| 20:09.03 | MothaWork | The profiler seems to work, but I can't get the reporting tool to work |
| 20:24.21 | prpplague | ds2: phone was alot more expensive |
| 20:24.38 | prpplague | ds2: $2700 plus $75 for the booking charge |
| 20:25.03 | Crofton | I think so |
| 20:25.18 | *** join/#edev markus (n=asdufg@endata.olf.sgsnet.se) |
| 20:29.10 | Crofton | MothaWork, also look for oprofileui |
| 20:29.49 | Crofton | http://labs.o-hand.com/oprofileui/ |
| 20:29.56 | MothaWork | Crofton: well, the report tool is failing with "invalid argument" on the profile data files |
| 20:30.00 | Crofton | I haven't had any luck with it, but maybe you will |
| 20:30.03 | MothaWork | which appear to contain nothing but null bytes |
| 20:30.17 | Crofton | I know you should be able to do this |
| 20:30.22 | Crofton | but I haven't done it |
| 20:30.31 | MothaWork | so I'm no longer so sure that the profiler is correctly running |
| 20:30.59 | markus | Is this the right channel to ask buildroot related questions? |
| 20:31.38 | MothaWork | there is #buildroot, but I don't know how active it is |
| 20:31.45 | MothaWork | #uclibc and #busybox are also useful for that |
| 20:31.51 | markus | sometimes it's good to have somebody to point in the right direction |
| 20:32.10 | markus | there is no #buildroot |
| 20:32.30 | markus | and no #busybox, not at this network. |
| 20:32.46 | MothaWork | really, I thought there was at least a #busybox |
| 20:33.17 | markus | there is a #uclibc |
| 20:33.23 | markus | that's how i got here :) |
| 20:34.46 | g1powermac | prpplague, you around? |
| 20:34.49 | prpplague | markus: #edev is just kind of a default buildroot channel as alot of buildroot users are here |
| 20:34.53 | prpplague | g1powermac: yea whats up? |
| 20:35.04 | g1powermac | prpplague, found the schematic to my jtag cable: http://www.g1powermac.com/jtag3.jpg |
| 20:35.20 | prpplague | looks |
| 20:35.55 | g1powermac | I think this should work to make a bit banged spi interface for my ti chip |
| 20:36.35 | g1powermac | the only thing is getting data back to the pc, I figure I'm going to have to connect what would be TDI and TDO together for the two wire spi interface |
| 20:38.54 | g1powermac | btw, when doing some google searches for parport spi stuff, I seen quite a bit of stuff come up for the linux kernel. . .is there a bit bang parport spi driver in the kernel already? |
| 20:41.14 | prpplague | g1powermac: http://www.ln.com.ua/~real/avreal/adapters_e.html#STK |
| 20:41.29 | prpplague | g1powermac: you can see the stk200 is almost identical to your jtag device |
| 20:41.47 | prpplague | g1powermac: you can use the avrdude to program avr's via ISP with a very minimal setup |
| 20:42.23 | *** join/#edev RobotGuy (n=robotguy@dsl093-038-072.pdx1.dsl.speakeasy.net) |
| 20:43.10 | g1powermac | prpplague, hmm, so you think I could prolly just modify avrdude to program this here ti chip? |
| 20:43.27 | RobotGuy | Greetings |
| 20:43.33 | g1powermac | heya RobotGuy |
| 20:43.48 | prpplague | g1powermac: no, avrdude isn't really designed for anything other than avr |
| 20:44.05 | prpplague | g1powermac: but you could get an idea of whats involved by looking at avrdude |
| 20:44.14 | prpplague | RobotGuy: hey |
| 20:44.36 | g1powermac | prpplague, ahh |
| 20:45.46 | g1powermac | sigh, looks like firefox 3 rc1 has some major issues |
| 20:46.00 | prpplague | g1powermac: whats the part number again? |
| 20:46.26 | g1powermac | the thing was using around 60% of my cpu and it dropped down to 20% by restarting it with the exact same tabs open |
| 20:46.41 | g1powermac | prpplague, cc2431 |
| 20:47.15 | g1powermac | though the datasheet for it is really for the cc2430, since its the exact same chip as the cc2431 minus the location module |
| 20:49.35 | RobotGuy | That location module is what has my interest. |
| 20:50.31 | prpplague | g1powermac: http://sourceforge.net/projects/ccflasher |
| 20:50.34 | RobotGuy | My internet is being flaky too. :( Speakeasy is sending me a new DSL modem. |
| 20:50.37 | ds2 | prpplague: I have had prices on the phone vary by 200% depending on which agent answered (this was with UA) |
| 20:50.44 | g1powermac | prpplague, I know of that program |
| 20:50.52 | g1powermac | however, it won't work in this case |
| 20:50.57 | prpplague | g1powermac: why not? |
| 20:51.25 | g1powermac | its assuming using ti's eval board which is using a usb port with a, I think, avr acting as the usb controller |
| 20:51.49 | prpplague | looks |
| 20:51.53 | g1powermac | so its telling the avr over usb to do stuff to the ti 8051 |
| 20:52.34 | g1powermac | specifically, look in the schematic included in the source code of cc_flasher |
| 20:53.24 | prpplague | g1powermac: looking now |
| 20:54.24 | prpplague | hmm i don't see a schematic included |
| 20:54.52 | g1powermac | its cc_flasher.png |
| 20:55.02 | g1powermac | in the tld |
| 20:56.27 | prpplague | yea found it, just wasn't expecting a png |
| 20:57.44 | prpplague | g1powermac: from what i am seeing |
| 20:57.51 | prpplague | g1powermac: this code is very very well written |
| 20:58.20 | prpplague | g1powermac: you can replace ccdebuginterfaceusb.cpp with ccdebuginterfacelpt.cpp |
| 20:58.24 | prpplague | g1powermac: very modular |
| 20:58.39 | g1powermac | was figuring he could prolly modify the thing |
| 21:00.26 | RobotGuy | prpplague: Can Flyswatter be used in some way to program the 8051? SPI maybe? |
| 21:00.48 | prpplague | RobotGuy: yea, it could program the cc2430 |
| 21:01.11 | RobotGuy | Then it should be able to program the cc2431 too. |
| 21:01.28 | g1powermac | RobotGuy, if I can hack the parport spi stuff into this flasher app, the flyswatter should work too |
| 21:01.54 | RobotGuy | g1powermac: Hack away! Get that Flyswatter. :) |
| 21:03.20 | RobotGuy | (Flies not included) |
| 21:03.26 | g1powermac | hehe |
| 21:03.42 | ds2 | what? no coupons for a free packet of live maggots?! |
| 21:04.04 | RobotGuy | ds2: Isn't it your nap time? ;) ;) |
| 21:04.17 | ds2 | uh no :P |
| 21:04.22 | prpplague | g1powermac: if you use the flyswatter, you could almost use the code with very little changes |
| 21:04.56 | g1powermac | prpplague, hmm, depends on getting the money to buy one :-) |
| 21:05.23 | g1powermac | has a couple of things on the plate for purchasing. . .always hard to choose |
| 21:05.40 | prpplague | g1powermac: indeed |
| 21:06.11 | g1powermac | for me to even use the chip, I need to get some schmartboards and a halfway decent soldering iron |
| 21:06.34 | g1powermac | of course I could really use the flyswatter for the zipit hacking |
| 21:07.11 | g1powermac | sigh, too many things to buy |
| 21:09.12 | RobotGuy | I already have the soldering station, but need to find out what schmartboard to get. I still need some for another chip or two. |
| 21:09.46 | prpplague | g1powermac: that code for ccflasher looks very nice |
| 21:09.54 | prpplague | g1powermac: you should be able to modify that easily |
| 21:10.55 | g1powermac | yea, just got to get my head around C++, I've been doing C like coding for way too long ;-) |
| 21:11.10 | CosmicPenguin | you say that like its a bad thing |
| 21:11.28 | RobotGuy | I need to get a good C++ book, and am looking at a couple other languages also. |
| 21:12.21 | g1powermac | was originally trained in basic C++, but then mostly did C like stuff these past few years |
| 21:12.37 | RobotGuy | I still need some connectors for a battery PS for my Hammer Carrier Board. |
| 21:13.43 | *** join/#edev MothaWork (n=monmotha@66.208.186.250) |
| 21:13.52 | RobotGuy | MothaWork: Greets |
| 21:14.06 | prpplague | RobotGuy: you get a replacement yet? |
| 21:14.19 | RobotGuy | prpplague: Not yet. |
| 21:15.08 | g1powermac | meh, I hate tracing through other peoples code |
| 21:15.32 | RobotGuy | g1powermac: I did that for four years, mostly old BASIC code. |
| 21:16.41 | g1powermac | tends to avoid it |
| 21:16.56 | RobotGuy | Well written code can be a good teacher. |
| 21:18.36 | g1powermac | true |
| 21:20.19 | RobotGuy | prpplague: When was it shipped? |
| 21:25.02 | RobotGuy | prpplague: I have a new opportunity to talk about Hammer now. :) |
| 21:26.17 | MothaWork | prpplague: you by chance ever had to deal with GTK/glib being REALLY, REALLY slow on ARM, but reasonably fast on even modest PC systems (though it does seem to scale with approximate MIPS) |
| 21:27.10 | g1powermac | ok, I think once I get the bit banging spi thing working, I'll hack through this program |
| 21:32.42 | prpplague | RobotGuy: sorry i don't know, mr herod was taking care of it |
| 21:33.21 | prpplague | MothaWork: i'm guessing you probably have some floating point issue |
| 21:34.05 | prpplague | MonMotha: the only time i've ever had performance issues was where i was running hardfloat and had a ton of float operations going on |
| 21:34.36 | g1powermac | prpplague, can I just use the SPI_BITBANG kernel driver to do what I need? |
| 21:34.50 | prpplague | g1powermac: uh for what? |
| 21:35.08 | g1powermac | to do the parport spi stuff for programming the ti chip |
| 21:35.24 | prpplague | g1powermac: no need for the kernel driver |
| 21:35.38 | g1powermac | hmm |
| 21:35.44 | prpplague | g1powermac: getting an application to bit bang spi is not an issue |
| 21:35.55 | prpplague | g1powermac: getting it to do the properly protocol for flashing is the issue |
| 21:36.09 | g1powermac | both are an issue for me atm :-) |
| 21:36.41 | prpplague | g1powermac: look at jflash from the lart project |
| 21:36.48 | prpplague | g1powermac: it is very easy to bit bang the lpt port |
| 21:37.07 | prpplague | g1powermac: there even detailed books on projects with the lpt port |
| 21:38.13 | prpplague | remembers the good ol' days of using his holly-gates jtag dongle that he purchased from TimRiker |
| 21:38.53 | prpplague | ~seen landley |
| 21:39.02 | purl | landley <n=landley@216-188-225-16.dyn.grandenetworks.net> was last seen on IRC in channel #edev, 2d 2h 26m 13s ago, saying: 'Coffee shop.'. |
| 21:46.53 | g1powermac | hmm |
| 21:46.59 | g1powermac | found parapin. . . |
| 21:47.05 | g1powermac | http://parapin.sourceforge.net/ |
| 21:53.11 | g1powermac | well, this simplifies things |
| 21:55.02 | prpplague | no need for that its pretty simple anyway |
| 21:55.08 | prpplague | g1powermac: did you look at jflash? |
| 21:55.14 | g1powermac | not yet |
| 22:00.29 | prpplague | anyone using fedora 9? |
| 22:01.16 | g1powermac | you know, one thing the datasheet for the cc2430 doesn't mention is the timing for the spi interface |
| 22:01.48 | g1powermac | I wonder how fast the spi interface can go on it |
| 22:02.02 | MothaWork | gah, what is .plt? Is that the table that links into shared libs? |
| 22:03.02 | roxfan2 | procedure linkage table or something |
| 22:03.26 | g1powermac | prpplague, is that something I need to look further into? |
| 22:03.26 | roxfan2 | it contains stubs for dynamic imports |
| 22:04.06 | prpplague | g1powermac: yea you need to know the exact protocol for transfering data |
| 22:04.15 | prpplague | g1powermac: some of that probably can be found in ccflash |
| 22:04.23 | prpplague | g1powermac: but you'll need to dig into it |
| 22:04.28 | g1powermac | maybe |
| 22:04.36 | prpplague | g1powermac: probably find something on the TI pages |
| 22:04.42 | g1powermac | remember its talking to an avr, not the TI chip |
| 22:04.48 | g1powermac | directly anyway |
| 22:04.58 | roxfan2 | MonMotha: let me know if you want to know more :) |
| 22:05.05 | prpplague | g1powermac: yes, but all it is doing is transfering the data from the usb to spi |
| 22:05.22 | MothaWork | ok, so basically yeah, linking in stuff |
| 22:05.28 | g1powermac | prpplague, and thats the thing, the avr is doing the spi stuff |
| 22:05.42 | prpplague | g1powermac: like i said the spi is easy, you need the protocol |
| 22:06.02 | g1powermac | yea, but the question is the timing of the spi stuff for the chip |
| 22:06.23 | prpplague | spi timming is easy |
| 22:06.28 | g1powermac | how fast the spi debug port on the chip can handle the spi data coming in |
| 22:06.40 | g1powermac | which the ti datasheet doesn't mention |
| 22:06.42 | prpplague | g1powermac: the par port will be well within the limits |
| 22:06.55 | g1powermac | k, that was my original question :-) |
| 22:07.46 | g1powermac | the protocol and such of sending the data is all in the datasheet, except for the specs on the spi/debug interface itself |
| 22:08.09 | g1powermac | which now I don't need to worry about |
| 22:08.17 | prpplague | g1powermac: http://focus.ti.com/analog/docs/techdocsabstract.tsp?familyId=933&abstractName=swra124 |
| 22:09.13 | g1powermac | ahh, crap, oh well, didn't see that one ;-) |
| 22:10.22 | *** join/#edev CosmicPenguin (n=nobody@163.181.251.103) |
| 22:10.23 | *** mode/#edev [+o CosmicPenguin] by ChanServ |
| 22:11.18 | g1powermac | thanks for finding that :-) |
| 22:13.56 | prpplague | g1powermac: all the info you need to program it is in that doc |
| 22:14.08 | g1powermac | yup, seems that way |
| 22:18.00 | g1powermac | hmm, this works nicely |
| 22:18.10 | g1powermac | thanks again for finding that doc :-) |
| 22:18.34 | prpplague | np |
| 22:18.46 | g1powermac | should be able to whip up his own simple flashing program quite quickly now with that doc |
| 22:19.04 | prpplague | g1powermac: all i did was go to ti.com , type is cc2430, and scroll to the documents section, heheh |
| 22:19.10 | prpplague | jokes with g1powermac |
| 22:19.40 | g1powermac | heh, yea, I just never saw the thing ;-) |
| 22:20.09 | prpplague | heads home |
| 22:23.03 | g1powermac | RobotGuy, alright, this chip is looking better now :-) |
| 22:26.20 | *** part/#edev like2wise (n=likewise@82-171-51-231.ip.telfort.nl) |
| 22:27.06 | RobotGuy | :D |
| 22:30.16 | MothaWork | so we have determined that there must be a floating point mutex hidden somewhere in this stuff |
| 22:35.15 | g1powermac | MothaWork, heh |
| 22:54.29 | g1powermac | RobotGuy, hmm, this is interesting |
| 22:54.36 | g1powermac | the location engine only does 2d |
| 22:54.52 | g1powermac | however, with some software, you can get some sort of 3d support |
| 23:21.17 | RobotGuy | Well, 2d should be all you really need for location. |
| 23:30.22 | *** join/#edev linac (n=lin@121.41.39.129) |
| 23:45.02 | *** join/#edev zoidb (n=jarv@96.245.16.205) |
