| 00:45.08 | *** join/#edev Sienna (n=Sienna@pool-71-246-90-23.bltmmd.east.verizon.net) |
| 00:46.25 | Mirell | Hehe. |
| 01:05.50 | *** join/#edev cajaka (n=none@cpe-24-58-193-200.twcny.res.rr.com) |
| 01:21.27 | *** join/#edev Madsy (n=madsy@ti500710a342-1985.bb.online.no) |
| 01:28.06 | *** join/#edev purl (i=ibot@pdpc/supporter/active/TimRiker/bot/apt) |
| 01:28.06 | *** 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/ |
| 01:47.20 | *** join/#edev cajaka (n=none@cpe-24-58-193-200.twcny.res.rr.com) |
| 01:56.24 | *** join/#edev prpplague (n=dave123_@ppp-70-242-124-21.dsl.rcsntx.swbell.net) |
| 01:57.58 | *** join/#edev prpplague^2 (n=dave123_@ppp-70-242-124-21.dsl.rcsntx.swbell.net) |
| 01:58.11 | prpplague^2 | ho ho hum |
| 02:05.42 | *** join/#edev prpplague^2 (n=dave123_@ppp-70-244-166-139.dsl.rcsntx.swbell.net) |
| 02:48.24 | *** join/#edev rodserling (n=rodserli@unaffiliated/rodserling) |
| 02:52.45 | *** part/#edev LyosNorezel (n=LyosNore@unaffiliated/lyosnorezel) |
| 02:55.47 | *** join/#edev LyosNoreze1 (n=LyosNore@96-24-178-62.day.clearwire-dns.net) |
| 03:01.47 | *** join/#edev rodserling (n=rodserli@unaffiliated/rodserling) |
| 03:10.17 | *** join/#edev rodserling (n=rodserli@unaffiliated/rodserling) |
| 03:21.08 | *** join/#edev chouimat (n=dieu@r2351064.cidc.net) |
| 03:25.23 | *** join/#edev Madsy (n=madsy@ti500710a342-1985.bb.online.no) |
| 03:45.50 | *** join/#edev dfarnsworth (n=dfarnswo@65.200.49.156) [NETSPLIT VICTIM] |
| 04:01.13 | *** join/#edev DrZimmerman (n=theo@77-56-246-76.dclient.hispeed.ch) |
| 04:12.10 | *** part/#edev LyosNorezel (n=LyosNore@unaffiliated/lyosnorezel) |
| 05:41.42 | *** join/#edev RobotGuy (n=robotguy@dsl093-038-072.pdx1.dsl.speakeasy.net) |
| 05:43.21 | *** join/#edev harpal (n=harpal@121.246.75.165) |
| 05:54.26 | *** join/#edev chouimat (n=dieu@r2351064.cidc.net) |
| 06:05.01 | *** join/#edev drath_ (i=vmaster@p5B07F722.dip.t-dialin.net) |
| 06:09.11 | *** join/#edev howtoo (n=howtoo@195.159.62.66) |
| 06:45.55 | *** join/#edev magnetron (n=magnetro@unaffiliated/magnetron) |
| 07:12.48 | *** join/#edev hw_ (n=hw@p578b3905.dip0.t-ipconnect.de) |
| 07:21.04 | *** join/#edev chouimat (n=dieu@r2351064.cidc.net) |
| 07:25.40 | *** join/#edev [AD]Turbo (n=turbo@host24-241-dynamic.22-79-r.retail.telecomitalia.it) |
| 07:26.40 | [AD]Turbo | yo all |
| 07:27.13 | bsdfox_ | anyone programmed a tiny11? |
| 07:36.49 | *** join/#edev ds2 (i=noinf@netblock-66-245-251-24.dslextreme.com) |
| 08:04.13 | *** join/#edev Appleman1234 (n=Appleman@CPE-121-223-90-178.qld.bigpond.net.au) |
| 09:20.00 | *** join/#edev lyakh (n=lyakh@p57BD22E0.dip0.t-ipconnect.de) |
| 11:35.06 | *** join/#edev ico2 (n=ico2@modem-171.cheetah.dialup.pol.co.uk) |
| 11:38.00 | *** join/#edev bbradley (n=bbradley@92-235-133-146.cable.ubr17.azte.blueyonder.co.uk) |
| 11:56.12 | *** join/#edev sjhill (n=sjhill@real.realitydiluted.com) |
| 11:56.12 | *** mode/#edev [+o sjhill] by ChanServ |
| 12:38.22 | *** join/#edev harpal (n=harpal@121.246.75.165) |
| 12:52.23 | *** join/#edev chouimat|work (n=dieu@209.217.106.98) |
| 12:55.07 | *** join/#edev prpplague (n=dave@mail.americanmicrosystems.com) |
| 13:02.34 | Chocobo | Anyone know of any good web rescourse on stand-alone ARM programming under linux (not putting Linux on ARM, but just developing stand alone applications using linux as a host OS) Google takes me to "Put Linux on ARM!" and that isn't what I am really looking for :) |
| 13:02.35 | *** join/#edev GPSFan (n=kenm@12.10.255.246) |
| 13:02.35 | *** mode/#edev [+o GPSFan] by ChanServ |
| 13:04.48 | prpplague | GPSFan: greetings |
| 13:05.30 | prpplague | Chocobo: running an application on an arm processor is without an os is basically the same as doing a bootloader |
| 13:05.47 | prpplague | Chocobo: you can look at something like apex or uboot as examples |
| 13:05.48 | GPSFan | prpplague: g'day, seen 2.6.25 yet? |
| 13:05.59 | prpplague | GPSFan: yep, booted it yesterday morning |
| 13:06.32 | GPSFan | prpplague: swell, looks like a few s3c24xx changes in the last patchset. |
| 13:06.42 | Chocobo | Ok, thanks prpplague. Would the toolchain the buildroot creates work? I guess I am mostly confused between arm-elf and arm-linux. |
| 13:07.02 | prpplague | GPSFan: yea, some reorganizing since the the 24xx series has been widened |
| 13:07.10 | prpplague | Chocobo: yep |
| 13:07.42 | GPSFan | prpplague: did mmc via spi make it in, and does it work? |
| 13:07.58 | prpplague | GPSFan: iirc yes, just haven't tried it yet on the hammer carrier board |
| 13:08.15 | Chocobo | Ok, now I just need to figure out how to get buildroot to only build the toolchain :) thanks prpplague |
| 13:08.30 | GPSFan | prpplague: you survive the "heavy weather" yesterday? |
| 13:08.35 | prpplague | Chocobo: just don't select any packages |
| 13:08.45 | prpplague | GPSFan: yea, it was pretty bad though |
| 13:09.00 | prpplague | GPSFan: kids were scared, and the tornando sirens didn't help |
| 13:09.08 | GPSFan | prpplague: the weather radah had some truely impressive looking stuff. |
| 13:09.26 | prpplague | GPSFan: yea, it was a very interesting evening |
| 13:09.33 | chouimat|work | morning |
| 13:09.38 | prpplague | greetings |
| 13:09.53 | GPSFan | chouimat: g'day |
| 13:13.40 | *** mode/#edev [+o prpplague] by ChanServ |
| 13:17.58 | prpplague | ho ho ho |
| 13:18.02 | prpplague | merry freakin friday |
| 13:18.19 | prpplague | Chocobo: you picked up a hammer board yet? |
| 13:19.57 | Chocobo | No, I am looking at the one ce_geek got. There is lots of interest in the department. We are both trying to convince the higher-ups that PIC chips are lame and we should be teaching students ARM. |
| 13:20.47 | prpplague | Chocobo: oh oh, yea, sorry i forgot that you guys were in there at the same place |
| 13:21.56 | prpplague | Chocobo: how is that working out? convincing the higher-ups that is |
| 13:22.25 | Chocobo | What we really need to do is come up with a project that demonstrates the benefits of ARM. We have a fairly decent mobile robot that the hammer might make a nice brain for. What would be really handy is WiFi. |
| 13:22.57 | *** join/#edev ico2 (n=ico2@modem-171.cheetah.dialup.pol.co.uk) |
| 13:22.58 | chouimat|work | pic are easy to use :) |
| 13:23.36 | Chocobo | It is a slow process. I have been trying to convince them to ditch pic and go avr for a year now. Very little interest. I think an AVR (for smaller projects) and ARM (for larger projects) would be great for the students. |
| 13:23.37 | prpplague | chouimat|work: but don't provide a large amount practical knowledge for use in a business environment |
| 13:23.54 | prpplague | Chocobo: indeed |
| 13:23.59 | chouimat|work | true |
| 13:24.11 | Chocobo | What we have found lately is that PIC is buggy as hell. |
| 13:24.15 | prpplague | Chocobo: actually, you could even use smaller arm devices for the smaller projects |
| 13:24.22 | chouimat|work | prpplague: who said school were teaching pratical stuff ;) |
| 13:24.46 | prpplague | chouimat|work: indeed, however thats what Chocobo and ce_geek are trying to overhaul |
| 13:25.05 | chouimat|work | I wish them good luck ... |
| 13:25.26 | prpplague | Chocobo: we use alot more avr's like the atmega48 that i'd like |
| 13:25.47 | Chocobo | Ahh, prpplague the only problem there is that ARM comes in inconvienient packages for prototyping/quick and dirty projects. I suppose we could whip up some breakout boards.... PCB layout is one of my passions :) |
| 13:25.54 | prpplague | Chocobo: too many times when we can't find a specific chip for a job, we implement it in something like the atmega48 |
| 13:26.46 | prpplague | http://www.sparkfun.com/commerce/product_info.php?products_id=263 |
| 13:26.52 | Chocobo | Yeah, I really like the AVR's. I just got a Dragon yesterday and have been playing with it. I was using a PP programmer I whipped up, but with the PP fading and the limited cable length, it was time to upgrade. |
| 13:27.20 | prpplague | http://www.sparkfun.com/commerce/product_info.php?products_id=775 |
| 13:28.00 | prpplague | Chocobo: the flyswatter can actually do ISP programming, just that no one has added it to avrdude yet |
| 13:28.47 | prpplague | Chocobo: both of those arm boards from sfe can be ordered in bulk |
| 13:29.37 | Chocobo | prpplague, the students here are incredibly thrifty. The deparment has no problem paying for development kits for a class, but students won't want to dish out 50 or 60 bucks for a chip. If we made our own breakouts we could probably reduce the cost by a lot. I recently whipped out a FT232RL - DIP breakout and we can hand them out for like 4 bucks a piece. They are FLYING off the shelves here. |
| 13:30.13 | prpplague | Chocobo: nice |
| 13:30.37 | prpplague | Chocobo: well if you can handle the ft232 you can handle the lpc or sam series pcbs |
| 13:30.41 | prpplague | Chocobo: very easy to implement |
| 13:31.04 | Chocobo | For something like that I would really have to figure out the BOM and see if we could compete. It would be impossible to convince a student to pay that money if their project could be done with a free pic/avr sample. |
| 13:31.28 | prpplague | Chocobo: you can, the ones at sfe are marked up very high |
| 13:31.47 | Chocobo | oh yeah, the ft232 was simple. I have never done any high frequency stuff before... so I would avoid those :) |
| 13:32.25 | prpplague | Chocobo: naw, the smaller arm chips are fairly low freq and easy to implement |
| 13:32.38 | prpplague | Chocobo: if you can do the ft232 you can do the lpc and sam |
| 13:32.46 | prpplague | Chocobo: the schematics are on the page |
| 13:32.56 | Chocobo | What page? |
| 13:33.01 | Chocobo | Ohh the SFE |
| 13:34.29 | Chocobo | Wow, that is pretty simple. Not much more than any other mcu. For some reason I picture the ARM as being big and scary. I think it has to do with the remapping commands. I don't have enough experience with it yet to really say that though :) |
| 13:37.46 | prpplague | Chocobo: remapping? |
| 13:39.25 | Chocobo | Yeah, like I said I have only read about it. remapping memory because address 0 is the start of the vectors, and really needs to be RAM.... somethign strange like that. For ARM's without MMU. |
| 13:40.07 | prpplague | Chocobo: actually its the same with all arm's since when they start up the mmu is off |
| 13:40.25 | prpplague | Chocobo: its not so bad when you get the hang of it |
| 13:40.35 | prpplague | Chocobo: each arm chip has its own quirks |
| 13:41.57 | Chocobo | prpplague, in your honest opinion is the hammer board a good one to start with, or do you think the lower frequency LPC, SAM chips would be a better stating place? |
| 13:48.04 | Chocobo | Shoot, I need to go... be back in a while. |
| 13:54.56 | SpeedEvil | It's different. |
| 13:55.07 | SpeedEvil | It's a fundamentally different space. |
| 13:55.19 | SpeedEvil | On one hand you've got what's basically a small linux system. |
| 13:55.26 | SpeedEvil | You program it in C or python or ... |
| 13:55.58 | SpeedEvil | On the other hand you have a 10-20K ROM system with a few hundred bytes of RAM, that you program in tightly constrianed C or asm. |
| 13:56.55 | prpplague | Choco_away: sorry was away from my desk |
| 13:57.31 | prpplague | Choco_away: SpeedEvil basically is correct |
| 13:57.40 | prpplague | Choco_away: those are two different arenas |
| 13:57.56 | SpeedEvil | On one hand - you want a JPEG compressor, you write it yoursel,f, in tightly packed assembler. |
| 13:58.09 | SpeedEvil | Ont he other, you download libjpeg and install it. |
| 13:58.44 | SpeedEvil | Of course - the other thing is that the hammer can be 10-20* the price and power use of a chip that you program in asm that is that constrained. |
| 13:59.54 | prpplague | SpeedEvil: that price variable is only relevant when considering pre-assembled systems |
| 14:10.48 | SpeedEvil | Or if manufacturing something. |
| 14:13.05 | *** join/#edev magnetron (n=magnetro@unaffiliated/magnetron) |
| 14:17.50 | *** join/#edev CosmicPenguin (n=nobody@163.181.251.103) |
| 14:17.50 | *** mode/#edev [+o CosmicPenguin] by ChanServ |
| 14:18.06 | chouimat|work | hi CosmicPenguin |
| 14:18.07 | prpplague | SpeedEvil: well the cost difference between the lpc2148 and the s3c2410 is about $4usd |
| 14:19.32 | prpplague | SpeedEvil: actually just checked, the lpc2148 in qty is actually more expensive than the s3c2410 in qty 500 |
| 14:19.49 | prpplague | http://www.mouser.com/Search/ProductDetail.aspx?qs=Q7xBXtx7xQhP%252bComh%252b%252byMA%3d%3d |
| 14:21.09 | prpplague | SpeedEvil: yea just double checked, the lpc2148 is actually $2 more expensive than the s3c2410 in qty 500 |
| 14:21.11 | chouimat|work | prpplague: it's a nice chip ... but only 42KB of ram (when you don't use the usb) |
| 14:21.54 | prpplague | the s3c2410 has 16kb of TCM that can be used |
| 14:22.16 | prpplague | so techincally you could run the s3c2410 without mmu and just a little amount of flash |
| 14:23.10 | chouimat|work | prpplague: but why would you do this ? |
| 14:23.50 | prpplague | chouimat|work: the only reason i can think is if you need the 200mhz clock and or number of i/o's |
| 14:24.16 | chouimat|work | hmmm |
| 14:24.41 | prpplague | in class room setting you could use the hammer board to both, low level arm development as well as high level dev such as running a full os |
| 14:27.42 | prpplague | goes to really fun fun fun engineering meeting |
| 14:27.55 | chouimat|work | prpplague: have fun |
| 14:53.33 | Choco_away | Ok, here is a question that is out there.... what does it mean for a bluetooth controller chip to have a UART? could I connect a mcu up to that and then send commands to create a bluetooth rfcomm socket? |
| 15:06.47 | *** join/#edev Crofton (n=balister@66-207-66-26.black.dmt.ntelos.net) |
| 15:11.26 | SpeedEvil | It varies |
| 15:11.31 | SpeedEvil | It means it has a UART on it |
| 15:11.37 | SpeedEvil | what that means is complex |
| 15:11.53 | SpeedEvil | It can mean something as simpe as yes, it's a virtual serial port. |
| 15:12.22 | SpeedEvil | It can mean that the serial interface is basically to varying portions of the bluetooth stack integrated on the chip. |
| 15:12.45 | SpeedEvil | Which may or may not be configurable as a simple serial interface you can talk to with rfcomm. |
| 15:13.26 | SpeedEvil | Or it may require you to do some or most of the heavy lifting in software, but be more flexible, and able to for example act like an A2DP host, or serial, or ... |
| 15:23.27 | *** join/#edev ico2_ (n=ico2@modem-2161.chimpanzee.dialup.pol.co.uk) |
| 15:24.49 | *** join/#edev RobotGuy (n=robotguy@dsl093-038-072.pdx1.dsl.speakeasy.net) |
| 15:28.41 | ds2 | morning |
| 15:35.41 | RobotGuy | Alloooooooooooooo |
| 15:36.31 | *** join/#edev Shingoshi (n=shingosh@c-67-169-195-234.hsd1.or.comcast.net) |
| 15:36.39 | RobotGuy | prpplague: When you get a chance, please have a look at www.hybotics.info - I've been working on a new Wiki format and would like your feedback. I don't know how to do tables of contents yet though. |
| 15:41.05 | *** join/#edev emil|k (i=emil@e179233080.adsl.alicedsl.de) |
| 15:42.01 | RobotGuy | I have all of my stuff ported over to the new Wiki now. |
| 15:43.24 | *** part/#edev ferrix (n=ferrix@ferrix.fi) |
| 15:52.13 | *** join/#edev chouimat (n=dieu@r2351064.cidc.net) |
| 15:54.12 | *** join/#edev KeyserSoze (n=Miles@67.107.206.170.ptr.us.xo.net) |
| 16:02.57 | RobotGuy | Well, I am going out for a bit, but will be back later. |
| 16:14.57 | cbrake | good for Atheros: http://lwn.net/Articles/278132/ |
| 16:20.30 | Choco_away | Thanks SpeedEvil. |
| 16:21.31 | SpeedEvil | np |
| 16:22.27 | SpeedEvil | :) |
| 16:22.30 | SpeedEvil | ->cbra |
| 16:24.24 | Chocobo | We took apart a USB bluetooth dongle and are looking up data sheets on the bluetooth controller. Trying to figure out if there may be a simple way to interface with it. The sparkfun modules are outrageously expensive. |
| 16:24.57 | *** join/#edev CosmicPenguin (n=nobody@163.181.251.103) |
| 16:24.57 | *** mode/#edev [+o CosmicPenguin] by ChanServ |
| 16:32.36 | SpeedEvil | what do you mean by 'interface with it' |
| 16:32.54 | *** join/#edev sjhill (n=sjhill@real.realitydiluted.com) |
| 16:32.54 | *** mode/#edev [+o sjhill] by ChanServ |
| 16:33.06 | SpeedEvil | If you've got something like the hammer, you simply configure bluez appropriately |
| 16:34.00 | Chocobo | yeah, we are thinking much more low level than that. Maybe even AVR. Hopefully we can open up an rfcomm connection by sending it AT commands or something like that. The datasheets, however, are not much use. |
| 16:34.33 | prpplague | Chocobo: thats because the "AT"commands you are refering to are part of a "firmware" |
| 16:35.02 | prpplague | Chocobo: the chipsets that you are looking at in the bt dongle are just the transceiver portion |
| 16:35.11 | prpplague | Chocobo: you have to have some "brains" attached |
| 16:35.22 | prpplague | Chocobo: for instance the modules you are refering to at sparkfun |
| 16:35.38 | prpplague | Chocobo: they actually have a arm core with a firmware to give you the "AT" commands |
| 16:35.47 | Chocobo | Well, it also talks about the 8051 cpu on the chip. It is all one thing. It has the mcu and the TX/RX all on the same package. |
| 16:35.50 | prpplague | Chocobo: so what you are paying for in those modules is the firmware |
| 16:36.28 | prpplague | Chocobo: ahh yea, if indeed it does include a 8051 core, you would need the "firmware" to add the AT commands |
| 16:36.36 | prpplague | Chocobo: and usually thats what costs money |
| 16:36.58 | Chocobo | That is EXACTLY what I told my advisor but he did not believe me. |
| 16:38.14 | Chocobo | My guess is if we hook that thing up to the FT232 we won't get much of any results. I am not really in the mood to try soldering to those tiny pins either. lol |
| 16:39.04 | prpplague | hehe |
| 16:39.06 | prpplague | indeed |
| 16:39.34 | Chocobo | However, if it did work... $5 is a lot better than $50. :) |
| 16:39.47 | *** join/#edev steliosk (n=Stelios@athedsl-25580.home.otenet.gr) |
| 16:41.02 | prpplague | Chocobo: indeed |
| 16:41.21 | SpeedEvil | Chocobo: broadly what prpplague said is right - with the caveat that some 'cheap' bluetooth keys are in fact the exact same chips - but with different preloaded firmware. However, in many cases you can't use them without a $5K dev package. Also - they may not be flashable - and may be actual ROM ones. |
| 16:41.52 | Chocobo | Yes, this chip seems to have a ROM in it :/ |
| 16:42.02 | SpeedEvil | In some cases you can get these to boot code from RAM, but you will usually need to remove them from the board, and do strange thigns to them on init. |
| 16:42.05 | prpplague | Chocobo: hehe, i researched the heck out of them, we are actually paying around $1.50 for the BT and $35 for the firmware |
| 16:42.20 | SpeedEvil | digikey does have some slightly cheaper BT chips that may suit. |
| 16:42.27 | SpeedEvil | I've never used them. |
| 16:42.29 | Chocobo | It's the IS1002... if you are familiar with such things. |
| 16:43.24 | Chocobo | Well thanks much, very helpful. I think I will suggest that we don't destroy the dongle by soldering to it :) |
| 16:43.26 | SpeedEvil | CSRs bluetooth line. |
| 16:43.31 | SpeedEvil | Err |
| 16:43.33 | SpeedEvil | Bluecore |
| 16:44.00 | SpeedEvil | The SDK is $$$$ - I also noticed it while searching on P2P - but I don't know anymore than that. |
| 16:44.37 | SpeedEvil | A 'hire purchase' - where you could buy it at $5 per BT item would be nice. |
| 16:44.49 | SpeedEvil | With no support until you've paid it in full. |
| 16:45.24 | Chocobo | Hmmm, for our low volume stuff that would be nice. |
| 16:45.47 | Chocobo | The SFE modules seem like they might be our best bet still. |
| 16:46.15 | SpeedEvil | I noticed someone on ebay selling for 50 quid (?) a GPS module with serial and bluetooth onboard. |
| 16:46.29 | SpeedEvil | But they diddn't have the BT module seperately, I asked. |
| 16:46.49 | prpplague | Chocobo: yea probably |
| 16:47.03 | Chocobo | Ahhh bluecore! That is what the linksys bluetooth dongle has in it... maybe I should say had... I basically destroyed that one in the name of science. :P |
| 16:47.14 | prpplague | hehe |
| 16:47.24 | prpplague | Chocobo: yea, bluecore is what we use |
| 16:47.29 | SpeedEvil | They have ROM, and Flash ones. |
| 16:47.44 | SpeedEvil | Though I suspect the ROM may be actually flash that they won't unlock. |
| 16:49.07 | *** join/#edev lyakh (n=lyakh@p57BD22E0.dip0.t-ipconnect.de) |
| 16:49.38 | Chocobo | jerks :p |
| 16:55.40 | SpeedEvil | These http://search.stores.ebay.co.uk/Sure-Electronics_gps_W0QQfciZQ2d1QQfclZ3QQfsnZSureQ20ElectronicsQQfsooZ1QQfsopZ1QQsaselZ208644246QQsofpZ0 |
| 17:04.19 | Chocobo | Bookmarked, those look awesome. I will have to show those to my advisor. |
| 17:17.48 | SpeedEvil | I haven't ordered them - but I have gotten other stuff - the GPS module works quite well. |
| 17:32.21 | prpplague | http://www.linuxdevices.com/news/NS2083975586.html |
| 17:32.36 | prpplague | glad to see someone at the top that "gets it" |
| 17:34.40 | prpplague | CosmicPenguin: looks like the hammer patch should be clear for 2.6.26 |
| 17:35.08 | prpplague | CosmicPenguin: http://ftp.arm.linux.org.uk/pub/armlinux/kernel/git-cur/arm:devel.diff |
| 17:48.48 | *** join/#edev LyosNorezel (n=Ralph@unaffiliated/lyosnorezel) |
| 17:52.53 | *** join/#edev Shingoshi (n=shingosh@c-67-169-195-234.hsd1.or.comcast.net) |
| 17:54.03 | Chocobo | Ugh, I need to find a book on technical writing. Writing papers is such a painful thing for me. |
| 17:56.50 | SpeedEvil | Just treat it like IRC. |
| 17:58.09 | prpplague | CosmicPenguin: andrew mortons presentation at CELF basically states everything i've been complaining about on the embedded development arena |
| 17:59.48 | CosmicPenguin | prpplague: do you now understand that the problem is localized to the individual that you have to deal with? |
| 17:59.54 | CosmicPenguin | Linux is not broken |
| 17:59.58 | CosmicPenguin | Open source is not broken |
| 18:00.47 | prpplague | CosmicPenguin: never said either of those, my problem is primarily with the patch submission system no being business friendly |
| 18:01.07 | CosmicPenguin | but again, you understand that is unique to your situation? |
| 18:01.19 | CosmicPenguin | What andrew does is very business friendly |
| 18:01.29 | CosmicPenguin | he has always given me way more credit then I deserved |
| 18:01.32 | prpplague | CosmicPenguin: from andrew morton's presentation, its not just me |
| 18:02.02 | CosmicPenguin | My business disagrees |
| 18:02.06 | CosmicPenguin | And we generate a lot of code |
| 18:02.21 | SpeedEvil | tries to remember who CosmicPenguin works for. |
| 18:02.28 | prpplague | so you disagree with the statements morton made in the presentation? |
| 18:02.46 | CosmicPenguin | I don't think the problem is as bad as he states |
| 18:02.59 | CosmicPenguin | I think that generally developers that work for company's are unwilling to play the open source game |
| 18:03.12 | CosmicPenguin | they either are unfamiliar with the community or believe they deserve special treatment |
| 18:04.00 | CosmicPenguin | An anology would be somebody who is an expert in say, geology who wants to edit a page on Wikipedia, but doesn't understand the policies |
| 18:04.27 | prpplague | but the policies for editing the pages on wikipedia are written and posted |
| 18:04.31 | prpplague | all they have to do is read |
| 18:04.44 | CosmicPenguin | All you have to do for the open source community is read too |
| 18:05.07 | CosmicPenguin | There are plenty of experts who come to wikipedia and get frustrated because the policies are unfamiliar or believe that they deserve special treatment |
| 18:05.12 | CosmicPenguin | anyway - more later |
| 18:05.29 | SpeedEvil | And in many cases they do. |
| 18:05.37 | prpplague | CosmicPenguin: so you imply that i expect special treatment when i submit patches? |
| 18:05.42 | SpeedEvil | Which is one of the reasons wikipedia is a bit broken. |
| 18:05.49 | CosmicPenguin | No - I'm talking about business in general |
| 18:06.30 | CosmicPenguin | none of this has to do with you - you have a specially frustrating barrier in your way |
| 18:06.40 | CosmicPenguin | I'm talking in general |
| 18:06.46 | CosmicPenguin | anyway, I'll be back |
| 18:06.51 | prpplague | CosmicPenguin: okie dokie, later |
| 18:07.18 | *** join/#edev ds2 (i=noinf@netblock-66-245-251-24.dslextreme.com) |
| 18:07.18 | *** join/#edev drath (i=vmaster@p5B07F722.dip.t-dialin.net) |
| 18:07.18 | *** join/#edev DrZimmerman (n=theo@77-56-246-76.dclient.hispeed.ch) |
| 18:07.27 | *** join/#edev DrZimmerman2 (n=theo@77-56-246-76.dclient.hispeed.ch) |
| 18:08.50 | prpplague | well i find mortons presentation to address exactly the items that are of concern to me, such as the fact that the vast majority of "rules" are not written down \ |
| 18:09.24 | SpeedEvil | Or that they are written down - spread everywhere on mailing lists, and there is no centralised policy like - say - wikipedia. |
| 18:09.45 | SpeedEvil | centralised place you can go and say you've read all the policy. |
| 18:09.49 | SpeedEvil | (at least in principle) |
| 18:09.52 | prpplague | SpeedEvil: indeed |
| 18:10.22 | prpplague | SpeedEvil: most of the coding rules are documented, but even with that, there are subsystems that have their own style |
| 18:11.08 | *** join/#edev drath_ (i=vmaster@p5B07F722.dip.t-dialin.net) |
| 18:14.16 | Dr{Who} | afternoon.. ok back to work. gdbserver needs a serial port open between the hammer and my pc.. i have the console port used up landley offered an interesting solution using netcat and some redirecting. Had me a little lost on all of the details. I am again wondering the hammer has usb ports cant I turn one into a serial port that both the hammer and the pc can talk on? |
| 18:16.20 | prpplague | Dr{Who}: there is also a second serial port on the carrier board, its just not brought out to a db9 header |
| 18:16.44 | Dr{Who} | perfect.. |
| 18:16.54 | Dr{Who} | ok ill review the docs and find it thanks. |
| 18:17.05 | prpplague | Dr{Who}: see the carrier board schematics for the location and connection |
| 18:17.19 | Dr{Who} | i did not want to use usbserial gadget just because i was not able to get the usbharddrive gadget to work yet as you know. |
| 18:17.43 | Dr{Who} | im needing some known good : c ) too many brick walls this week.. need some progress. |
| 18:17.46 | prpplague | Dr{Who}: header J6 is the second serial port |
| 18:18.19 | prpplague | Dr{Who}: is there something specific you are trying to debug? |
| 18:19.20 | Dr{Who} | ya i am trying to get tinycc to build a hello world |
| 18:19.33 | Dr{Who} | it builds and will run but if i call "printf" form my hello it faults. |
| 18:19.49 | prpplague | ahh |
| 18:19.50 | Dr{Who} | so its some issue in the stack or alignment with the libs so basicly i want to get tinycc working.. |
| 18:20.12 | Dr{Who} | i want to compile on the hammer. |
| 18:20.27 | prpplague | Dr{Who}: you can just add a db9 connection t the J6 connector |
| 18:20.31 | Dr{Who} | it will make my dev time for projects much shorter. |
| 18:20.47 | prpplague | Dr{Who}: yea thats something i want as well with tinycc |
| 18:24.00 | *** join/#edev DrZimmerman2 (n=theo@77-56-246-76.dclient.hispeed.ch) |
| 18:29.29 | *** join/#edev ds2 (i=noinf@netblock-66-245-251-24.dslextreme.com) |
| 18:36.03 | *** join/#edev solar (n=solar@smtp.gentoo.org) |
| 18:40.42 | SpeedEvil | gcc works just fine. |
| 18:40.57 | SpeedEvil | Actually - I've not used it on a platform with as little memory |
| 18:41.22 | SpeedEvil | Only 128M |
| 18:41.55 | Dr{Who} | been reading up on slimming down gcc. its a bit of work. |
| 18:42.13 | Dr{Who} | also tincc can run as an interpreter. |
| 18:42.16 | chouimat|work | SpeedEvil: I used gcc on a 16M platform ... lot of swap and lot of patience :) |
| 18:48.53 | SpeedEvil | I've used it on a 3M platform. |
| 18:48.58 | SpeedEvil | Though that was some time ago. |
| 18:50.55 | chouimat|work | SpeedEvil: I remember using TurboC on a 640K 8088 |
| 18:51.58 | SpeedEvil | Yeah - I was only talking about gcc. |
| 18:52.20 | SpeedEvil | I think I started on a 128K CPM machine, with a Hisoft compiler. |
| 18:52.36 | SpeedEvil | It was an example copy, that had a subtle bug. |
| 18:52.52 | SpeedEvil | It deleted the output file, so you couldn't really use it. |
| 18:52.59 | SpeedEvil | undelete worked though :) |
| 18:53.38 | *** join/#edev ico2_ (n=ico2@modem-2161.chimpanzee.dialup.pol.co.uk) |
| 19:23.21 | prpplague | ~seen davidm_ |
| 19:23.39 | purl | davidm_ <n=davidm@209.144.24.2> was last seen on IRC in channel #debian, 503d 21h 10m 16s ago, saying: 'Any Ideas on what to do with this system?'. |
| 19:25.09 | *** join/#edev ico2_ (n=ico2@modem-2161.chimpanzee.dialup.pol.co.uk) |
| 19:27.28 | *** join/#edev emil|k (i=emil@e179233080.adsl.alicedsl.de) |
| 19:30.36 | prpplague | ~seen davidm__ |
| 19:30.39 | purl | davidm__ <n=davidm@209.144.24.2> was last seen on IRC in channel #debian, 503d 20h 55m 26s ago, saying: 'bye'. |
| 19:31.59 | *** join/#edev prpplague^2 (n=dave@mail.americanmicrosystems.com) |
| 19:32.50 | CosmicPenguin | real work sucks |
| 19:34.02 | prpplague | CosmicPenguin: hehe |
| 19:35.55 | *** mode/#edev [+o prpplague] by ChanServ |
| 20:12.38 | CosmicPenguin | Tiny bit of spam, but you'll forgive me |
| 20:12.40 | CosmicPenguin | http://www.siliconmechanics.com/i17581/news/april-18-2008.php |
| 20:12.55 | CosmicPenguin | This is one of the first vendors to ship coreboot from the factory, so if you are in need of a server, I recommend looking into them |
| 20:13.10 | CosmicPenguin | I return you to your previously scheduled tincan hacking |
| 20:28.01 | prpplague | CosmicPenguin: hehe |
| 20:28.25 | prpplague | ho ho hum |
| 20:28.59 | prpplague | CosmicPenguin: the gentoo guys seem pretty excited about using the nail board |
| 20:29.19 | prpplague | CosmicPenguin: haven't heard from either david mandala or andrew morton yet |
| 20:30.22 | CosmicPenguin | I would expect the Gentoo guys have been casting around for a board to call their own |
| 20:30.30 | CosmicPenguin | everything else of popularity uses a different distribution |
| 20:30.51 | prpplague | i really wish there was kernel arg to turn of screen blanking |
| 20:30.58 | prpplague | its a real pain when you are doing lcd dev |
| 20:31.38 | prpplague | CosmicPenguin: i think the freebie items is starting to pay off |
| 20:39.11 | prpplague | ugh, i think wife is trying to poison me, lunch isn't sitting well today |
| 21:15.17 | Dr{Who} | prpplague.. what devices should that other serial port show up as? /dev/ttySAC1 ? |
| 21:19.34 | prpplague | Dr{Who}: /dev/ttySAC2 |
| 21:21.42 | Dr{Who} | ok ill have to set that up i only have sac0 and 1 setup |
| 21:25.27 | *** join/#edev ico2__ (n=ico2@modem-1947.chameleon.dialup.pol.co.uk) |
| 21:37.13 | Dr{Who} | ok thats minor device 66. |
| 21:40.46 | SpeedEvil | SAC2 can either be a serial port, or the pins can be used as flow control for SAC1 |
| 21:43.03 | Dr{Who} | does that require any settings? |
| 21:43.37 | SpeedEvil | yes |
| 21:43.44 | SpeedEvil | I don't know how it's setup on the hammer |
| 21:47.09 | *** join/#edev LyosNorezel (n=LyosNore@unaffiliated/lyosnorezel) |
| 21:48.02 | *** join/#edev ico2_ (n=ico2@modem-2646.chimpanzee.dialup.pol.co.uk) |
| 21:53.35 | prpplague | SpeedEvil: default configuration for the s3c2410 and the hammer is that those are configured as serial ports |
| 21:54.31 | Dr{Who} | DXD2 and RXD2 are the pin's im presuming that the pins can be configured in various ways via flags? |
| 21:57.33 | prpplague | that is correct |
| 21:57.43 | prpplague | by default they are configured as ttySAC2 |
| 21:58.41 | Dr{Who} | is this simmilaar to the mega128 or other processors flags that are programmed via the jtag to configure how pins work? or is this a kernel driver software only setup? |
| 22:00.01 | *** join/#edev CosmicPe1guin (n=nobody@163.181.251.103) |
| 22:00.14 | prpplague | Dr{Who}: just how the kernel is configured |
| 22:00.31 | Dr{Who} | ok thanks. that makes sense. |
| 22:00.36 | *** mode/#edev [+o CosmicPenguin] by ChanServ |
| 22:00.52 | Dr{Who} | to the kernel driver its just DIO pins.. and a software serial driver |
| 22:03.01 | prpplague | Dr{Who}: the init for the device defines those pins as a uart and it tells the serial driver that they are configured as a uart |
| 22:05.30 | prpplague | heads home |
| 22:05.31 | Dr{Who} | so the kernel command line can change this as the driver's init is looking for options? or is this hard coded in the driver's init() |
| 22:05.44 | Dr{Who} | ok thanks for the data.. |
| 22:06.01 | Dr{Who} | have a good weekend. |
| 22:06.04 | prpplague | its hardcoded, have a look at the arch/arm/mach-s3c2410/mach-tct_hammer.c file |
| 22:06.16 | prpplague | you'll see in that file where the uarts are defined |
| 22:06.25 | Dr{Who} | yep.. just being lazy and asking : c ) |
| 22:07.02 | prpplague | there is a struct called tct_hammer_uartcfgs |
| 22:07.06 | prpplague | later |
| 22:54.29 | *** join/#edev RobotGuy (n=robotguy@dsl093-038-072.pdx1.dsl.speakeasy.net) |
| 22:56.45 | *** join/#edev ico2__ (n=ico2@modem-1412.cougar.dialup.pol.co.uk) |
| 23:04.33 | Dr{Who} | hmm not convinced that rxd2 and txd2 are alive.. im getting ~4v on txd2 but 0v on rxd2 |
| 23:04.59 | Dr{Who} | on rxd1 i am getting ~4v |
| 23:05.40 | SpeedEvil | what happens if you play with the flow control on SAC1 |
| 23:05.58 | SpeedEvil | I.E. are they in that role - rather than the serial port |
| 23:06.12 | SpeedEvil | I.E. are they in that role - rather than the serial port 2's RXd/TXD |
| 23:06.28 | Dr{Who} | ya so thinking they are used as rts and cts |
| 23:07.40 | RobotGuy | There isn't any hardware flow control on ttySAC1 |
| 23:07.42 | Dr{Who} | now whats the best way to do this. I have some c code that can adjust the flow but it wont build on the box. |
| 23:07.52 | RobotGuy | Only for ttySAC0 |
| 23:09.21 | SpeedEvil | I know one of the UARTs txd/rxd pins can be flow control instead for one of the others. |
| 23:09.30 | SpeedEvil | I'm not quite sure which on it was. |
| 23:10.07 | RobotGuy | SpeedEvil: Where is this documented? |
| 23:10.20 | Dr{Who} | it would be part of the driver code itself |
| 23:10.28 | RobotGuy | CTS/RTS is only brought out for ttySAC0 |
| 23:10.29 | SpeedEvil | RobotGuy: in the SC* datasheet |
| 23:11.02 | RobotGuy | SpeedEvil: You need to look at the Hammer pinout to see what is brought out on the module. |
| 23:11.18 | SpeedEvil | Sure. |
| 23:11.33 | RobotGuy | CTS/RTS only on ttySAC0, not for ttySAC1 |
| 23:11.51 | SpeedEvil | But SAC2 was mentioned - and I was pretty sure that SAC2 is the rts/cts pins for SAC1 |
| 23:11.59 | SpeedEvil | in its alternate role |
| 23:13.24 | RobotGuy | Interesting, there is RXD2/TXD2. No CTS/RTS on ttySAC1, SpeedEvil. Not in a primary or alternate role. |
| 23:13.36 | RobotGuy | Look at the module pinout. |
| 23:13.54 | SpeedEvil | the coloured table? |
| 23:14.26 | RobotGuy | SpeedEvil: The module pinout above the colored table |
| 23:15.32 | RobotGuy | The table just reflects what the module pinout shows, with a bit more description. |
| 23:16.51 | RobotGuy | It's odd that there is RXD2/TXD2 when there are just two serial ports available. I wonder if this is an error. |
| 23:18.11 | Dr{Who} | well anyway rxd2 is showing 0v and that is different that RXD0 that is at ~4v so i can be very sure that the cpu itself has configured that pin differently so that tells me the serial driver is not setup or initialzing correctly |
| 23:18.44 | RobotGuy | Dr{Who}: I am not sure the second serial port is initialized at all. |
| 23:19.37 | Dr{Who} | will have to see. i am going to read over mach-tct_hammer.c and other parts of the kernel. |
| 23:20.07 | RobotGuy | I've only looked at a bit of that file. |
| 23:20.35 | RobotGuy | I can't do any testing here right now. |
| 23:20.54 | SpeedEvil | GPH7 and GPH6 |
| 23:21.20 | SpeedEvil | either RXD2 and TXD2 or |
| 23:21.34 | SpeedEvil | nCTS1 and nRTS1 |
| 23:21.53 | SpeedEvil | p260 of the datasheet |
| 23:22.03 | RobotGuy | SpeedEvil: That is just the second serial port RX/TX lines. Nothing about CTS/RTS there. |
| 23:22.10 | RobotGuy | SpeedEvil: Where are you seeing this? |
| 23:22.22 | Dr{Who} | i see a typo on hammer_carrier_rev-c |
| 23:22.33 | RobotGuy | Where? |
| 23:22.40 | Dr{Who} | it shows RXD2 and TXD2 as GPH7 |
| 23:22.51 | RobotGuy | True. |
| 23:23.12 | RobotGuy | GPH6 and GPH7 |
| 23:23.20 | SpeedEvil | GPHCON 12-15 - the register |
| 23:23.34 | SpeedEvil | sets them up |
| 23:23.47 | RobotGuy | SpeedEvil: Sets what up? You can |
| 23:23.56 | RobotGuy | t use pins that are not brought out on the module. |
| 23:23.59 | Dr{Who} | sets up what type of pin's they are |
| 23:24.37 | Dr{Who} | im presuming this is like other uP and that you can general purpose pins that can be used for different things.. Input or output sink or float. |
| 23:24.46 | RobotGuy | Show me where there is CTS/RTS for a second serial port on the hammer module. |
| 23:24.59 | SpeedEvil | you can set GPH7 - for example - to be GPIO, RXD2, or nCTS1 |
| 23:25.14 | SpeedEvil | Dependant on the settings of the GPHCON register |
| 23:25.21 | Dr{Who} | yep. |
| 23:25.50 | SpeedEvil | page 273 of the datasheet |
| 23:25.52 | RobotGuy | There has to be a change to the pinout and table then if that is true. |
| 23:26.23 | SpeedEvil | I just did the table quickly based on the pinout, and diddn't bother going into tertiary pin functions IIRC |
| 23:26.43 | RobotGuy | SpeedEvil: We need to see ALL functions a pin can perform. |
| 23:27.13 | SpeedEvil | Sure - I diddn't look deeply at all the pins - I forgot that case. |
| 23:28.17 | RobotGuy | Now I need to find out what pins are the second serial port (ttySAC1) and if the RXD2/TXD2 is correct and whether we address it as ttySAC2. |
| 23:28.50 | Dr{Who} | supposed to be ttySAC2 is J6 |
| 23:29.05 | RobotGuy | There must be an RXD1/TXD1 somewhere. |
| 23:30.57 | SpeedEvil | GPH5 and 4 |
| 23:32.06 | RobotGuy | GPSFan: ping |
| 23:32.30 | GPSFan | RobotGuy: pong.. |
| 23:32.56 | RobotGuy | GPSFan: Hey there! Do you have the User Manual for the S3C2440 and the datasheet for the S3C2410? |
| 23:33.00 | Dr{Who} | robot.. are you going to relay the mislable of that doc to prp? |
| 23:33.13 | GPSFan | RobotGuy: yes |
| 23:33.17 | RobotGuy | Dr{Who}: Of course I am. |
| 23:33.17 | SpeedEvil | google um_s3c2410a_rev11.pdf |
| 23:33.23 | RobotGuy | GPSFan: Could I have them please? |
| 23:33.27 | Dr{Who} | k so ill just drop it from my stack.. : c ) |
| 23:33.35 | SpeedEvil | that's the 2410 |
| 23:33.45 | RobotGuy | I know. the 2410. |
| 23:34.00 | GPSFan | RobotGuy: I sent you a link for the 2440 user manual, why can't you dl it from there? |
| 23:34.01 | RobotGuy | I will verify the data and write up the needed changes. |
| 23:34.51 | Dr{Who} | not seeing gph4/5 on the hamnmer pin assinemnt pdf |
| 23:35.02 | RobotGuy | OK, sorry. It's the datasheets for both that I need |
| 23:35.02 | SpeedEvil | Is GPH there at all? |
| 23:35.23 | RobotGuy | I have the user manuals. |
| 23:35.48 | GPSFan | ah datasheets. let me look.. |
| 23:35.50 | SpeedEvil | why? the user manuals have the pinout details |
| 23:36.02 | SpeedEvil | and the register settings |
| 23:37.17 | Dr{Who} | so presumably its not tied to anything external? its on K16 |
| 23:40.00 | Dr{Who} | this is a s3c2410 right? |
| 23:40.34 | SpeedEvil | The hammer is, yes. |
| 23:43.02 | GPSFan | RobotGuy: the usermanuals for both have electrical characteristics, pinnouts, timing info, register bit descriptions. what else would a datasheet have? That's all I have, and I can't find anything else other than marketing BS flysheets. |
| 23:43.54 | RobotGuy | GPSFan: OK. Most chips have a datasheet, but if all that is in the User Manuals, cool. |
| 23:48.04 | GPSFan | RobotGuy: in some cases that is true, the PXA270 datasheet is only 126 pages while its' user manual is 1246 pages. one has stuff usefull for a hardware designer, one has stuff usefull for a programmer. |
| 23:53.39 | RobotGuy | SpeedEvil: nCTS1 and nRTS1 won't do anyone any good, unless TXD2/RXD2 is supposed to be TXD1/RXD1, which would mean a mislabel. |
| 23:54.36 | SpeedEvil | Possibly. |
| 23:54.53 | SpeedEvil | RobotGuy: the GPH pins that are brought out are the important bit |
| 23:56.45 | SpeedEvil | Anyway - txd2/rxd2 is the third serial port |
| 23:57.01 | SpeedEvil | txd1/rxd1 is the second |
| 23:57.36 | SpeedEvil | Oh - right |
| 23:57.54 | RobotGuy | There is no GPH4 or GPH5 on the Hammer pinout. |
| 23:59.14 | SpeedEvil | Ok - txd1/rxd1 are not brought out |
| 23:59.26 | SpeedEvil | rdx2 and txd2 are |
| 23:59.30 | RobotGuy | I'm too tired to get my head around this right now, so will try and get this done tomorrow. I'm trying to create a table of Hammer Pin, Chip Ball, 1st Func, 2nd Func, 3rd Func. |
| 23:59.52 | SpeedEvil | Ah right |
| 23:59.54 | SpeedEvil | I remember now |