| 00:23.57 | *** join/#edev prpplague (n=dave123_@ppp-70-244-175-101.dsl.rcsntx.swbell.net) |
| 00:25.40 | *** join/#edev joe_bleau (n=joe_blea@adsl-69-152-210-234.dsl.fyvlar.swbell.net) |
| 00:29.49 | prpplague | joe_bleau hey joe |
| 00:29.56 | prpplague | joe_bleau whats cookin? |
| 00:29.57 | joe_bleau | hola! |
| 00:30.04 | joe_bleau | Not a darn thing. |
| 00:30.09 | prpplague | bummer |
| 00:30.48 | joe_bleau | I'll tell you what, this here internet pipe system is pretty spiffy at times. |
| 00:31.14 | joe_bleau | I had no idea you could buy grey-market japanese writing instruments so easily. |
| 00:34.32 | CosmicPenguin | be sure to send a nice thank you note to Al Gore |
| 00:36.31 | joe_bleau | Grrr. |
| 00:37.20 | prpplague | hehe |
| 00:37.34 | prpplague | CosmicPenguin had a chance to look at the photo of the nail board? |
| 00:37.44 | prpplague | CosmicPenguin i'm curious as to your thoughts |
| 00:37.47 | CosmicPenguin | the who now? |
| 00:38.07 | prpplague | CosmicPenguin btw sstallion appears to be doing geode dev |
| 00:38.17 | prpplague | CosmicPenguin http://www.elinux.org/Nail_Board |
| 00:38.26 | prpplague | CosmicPenguin usb powered hammer board |
| 00:38.35 | CosmicPenguin | well - isnt that purty |
| 00:38.38 | prpplague | CosmicPenguin has jtag and serial console via the usb port |
| 00:38.46 | prpplague | CosmicPenguin yea yea rough proto |
| 00:39.09 | CosmicPenguin | how long do the caps keep charged after the usb is removed? |
| 00:39.25 | prpplague | about .00000001 seconds |
| 00:39.26 | prpplague | hehe |
| 00:39.32 | CosmicPenguin | doh |
| 00:39.38 | prpplague | CosmicPenguin why? |
| 00:40.02 | CosmicPenguin | just wondering if there was enough time to grok the removal and do emergency things |
| 00:40.16 | CosmicPenguin | I guess that would be a pretty hefty cap |
| 00:40.22 | prpplague | ahh |
| 00:40.30 | prpplague | CosmicPenguin i could add a small super cap |
| 00:40.48 | prpplague | get about 20 seconds |
| 00:40.57 | CosmicPenguin | 20 seconds would be more then enough |
| 00:41.08 | CosmicPenguin | maybe even too much |
| 00:41.41 | prpplague | CosmicPenguin the big question is whether to include a 10baseT on the end or a usb host connector |
| 00:42.25 | prpplague | CosmicPenguin the back side, as of right now, has a 8 position dip switch, a 7 segment display, 2 leds, a push button, and a buzzer |
| 00:43.35 | CosmicPenguin | and a partridge in a pear tree |
| 00:43.54 | CosmicPenguin | I say stick with the host connector |
| 00:43.55 | prpplague | too much? |
| 00:44.33 | CosmicPenguin | The USB is way more useful then a network connnection |
| 00:44.40 | CosmicPenguin | worse comes to worse, you plug a NIC into that |
| 00:45.00 | prpplague | CosmicPenguin we thats the thing, if its usb powered i only have 500mA to work with |
| 00:45.07 | CosmicPenguin | the client side is a real UDC? |
| 00:45.24 | prpplague | CosmicPenguin the client side is a ft2232 chipset |
| 00:45.46 | prpplague | CosmicPenguin dual purpose, one section is jtag, the other is rs-232 connected to the console |
| 00:46.14 | prpplague | CosmicPenguin the ft2232 takes about 30mA, the hammer 100mA |
| 00:46.25 | CosmicPenguin | oh - that runs with the ft_sio module on the host side, I guess |
| 00:46.32 | CosmicPenguin | hmmm - useful |
| 00:46.38 | prpplague | CosmicPenguin: rough figure would be about 150mA leaving about 350mA |
| 00:46.43 | CosmicPenguin | though having usb gadget connections isn't without its charms |
| 00:46.44 | prpplague | CosmicPenguin yea |
| 00:46.54 | prpplague | CosmicPenguin such as? |
| 00:47.11 | CosmicPenguin | intercepting USB streams for one |
| 00:47.18 | CosmicPenguin | in this case |
| 00:47.36 | prpplague | i.e. a usb sniffer? |
| 00:47.40 | CosmicPenguin | sure |
| 00:47.50 | prpplague | hmm, thats beyond my skills |
| 00:47.55 | CosmicPenguin | I could see the value in a USB gadget to wiggle GPIOs |
| 00:48.12 | prpplague | CosmicPenguin you can do that thru the console |
| 00:48.19 | CosmicPenguin | true |
| 00:48.28 | CosmicPenguin | but sadly, USB is probably more universal |
| 00:48.39 | CosmicPenguin | especially when talking about other operating systems |
| 00:48.51 | prpplague | more universal than rs-232? |
| 00:49.09 | CosmicPenguin | not real rs232 - its cooked |
| 00:49.23 | CosmicPenguin | so if you don't have the chef, you are at sea |
| 00:50.04 | prpplague | hmm, as far as i know its suppose to be fully rs-232, we've never had a problem implementing stuff via the interface |
| 00:50.11 | CosmicPenguin | no - cooked on the host side |
| 00:50.15 | prpplague | we do ppp, xmodem, zmodem |
| 00:50.32 | CosmicPenguin | anyway |
| 00:50.35 | CosmicPenguin | USB gadgets are fun |
| 00:50.44 | CosmicPenguin | but thats cool - ft_sio interfaces are fun too |
| 00:50.52 | prpplague | hmm |
| 00:51.05 | CosmicPenguin | I have a LPC break out card that has a FT chip on it |
| 00:51.18 | CosmicPenguin | I use that more then i use the DB9 attachment |
| 00:51.22 | prpplague | yea i've seen some of those |
| 00:52.08 | prpplague | CosmicPenguin idea is to basically have an ipaq minus the lcd that you can plug into your laptop and dev on |
| 00:52.26 | prpplague | CosmicPenguin kinda like you are using the laptops screen and keyboard for the device |
| 00:52.46 | prpplague | i.e. got to starbucks and dev with you arm board dongle |
| 00:52.58 | CosmicPenguin | sure |
| 00:54.59 | prpplague | CosmicPenguin think the 7-seg and dip switches are too much? |
| 00:56.33 | CosmicPenguin | 7 seg no |
| 00:56.40 | CosmicPenguin | cant think what I would use the dip switches for |
| 00:58.11 | prpplague | config different startup options |
| 00:59.00 | CosmicPenguin | does the regular hammer have them? |
| 00:59.46 | prpplague | yea you can configure them with the i/os |
| 00:59.56 | CosmicPenguin | I guess its not bad - probably taking the place of resistors or fuses on a production style system |
| 01:00.11 | prpplague | yea |
| 01:00.29 | CosmicPenguin | if you are teaching people to depend on straps like that, then yeah, you'll need to keep them through the line |
| 01:01.18 | prpplague | yea, for doing basic dev it makes sense |
| 01:08.43 | CosmicPenguin | time for home I reckon |
| 01:08.51 | joe_bleau | prpplague: got a link to the AML dev download? Just curious... |
| 01:09.13 | prpplague | joe_bleau what? |
| 01:09.27 | joe_bleau | AML uses linux in some of their products, right? |
| 01:09.52 | joe_bleau | Don't y'all have a download for people to write programs for your scanners? |
| 01:10.10 | joe_bleau | (Followed the M5900 stuff from linuxdevices) |
| 01:11.26 | *** join/#edev flyback (n=ubuntu@c-24-131-197-176.hsd1.pa.comcast.net) |
| 01:11.40 | prpplague | joe_bleau only about 10% of our retailers design custom software, and we ship them a sdk cd, we also follow the GPL and include an offer for all source in the box |
| 01:11.55 | joe_bleau | So, that would be a no? |
| 01:12.39 | prpplague | joe_bleau we generally don't provide the url for the downloads, its on the paper included in the box |
| 01:12.49 | prpplague | joe_bleau why? |
| 01:12.55 | joe_bleau | I ended up here: http://www.barcodepower.com/software_downloads.asp, but that didn't look right. |
| 01:13.10 | joe_bleau | Why? You were telling me some time ago how y'all went open source. |
| 01:13.22 | joe_bleau | I was just curious what your end users got to work with. |
| 01:13.45 | prpplague | joe_bleau yea the downloads page is just firmware images |
| 01:13.58 | joe_bleau | Wasn't sure if it would be public or not. Guess it's not. That's cool. |
| 01:14.03 | prpplague | joe_bleau not "went" we've always been open |
| 01:14.37 | joe_bleau | Oh, I read something in a FAQ about the M5500 using a prop. OS. |
| 01:14.47 | prpplague | uh where? |
| 01:14.48 | joe_bleau | Anyway, I'm not trying to pick at you or anything. |
| 01:14.52 | prpplague | uh where? |
| 01:15.16 | prpplague | oh you said 5500 |
| 01:15.27 | joe_bleau | I'ts a pdf: "The M5500 used a proprietary operating system running on a 16 MHz processor, and limited memory (256K RAM |
| 01:15.27 | joe_bleau | It's in a pdf: "The M5500 used a proprietary operating system running on a 16 MHz processor, and limited memory (256K RAM" |
| 01:15.47 | prpplague | yea that was indeed closed source but that was years and years ago |
| 01:15.52 | prpplague | more than 10 years ago |
| 01:16.07 | prpplague | that was one of the few closed source device AML has done |
| 01:16.45 | joe_bleau | Gotcha. |
| 01:18.52 | ds2 | ah /dev/rmem |
| 01:19.22 | prpplague | joe_bleau looking for something specific? |
| 01:19.51 | joe_bleau | Nope. Like I said, I was curious as to the end-user experience. |
| 01:20.07 | joe_bleau | If I were to go with an open source OS for our products in the future (!)... |
| 01:20.23 | joe_bleau | I don't know what to expect in terms of supporting outside development. |
| 01:21.01 | joe_bleau | I thought just maybe that there would be a download link to a tarball. |
| 01:21.18 | prpplague | joe_bleau generally speaking we don't provide support for third party dev. it's shipped with what we support |
| 01:21.42 | joe_bleau | So, "here's the SDK link. Don't call us, good luck?" |
| 01:22.17 | joe_bleau | Hey, if this is something you shouldn't be discussing in the open, we can drop it. |
| 01:22.21 | MonMotha | that's better than what many companies using linux do :) |
| 01:22.21 | prpplague | joe_bleau yep! we aren't in the business of doing software support |
| 01:22.25 | joe_bleau | Not trying to cause waves, you know? |
| 01:22.39 | MonMotha | many companies ship binaries for GPL stuff and don't even provide source, after all |
| 01:22.42 | prpplague | joe_bleau no problems discussing what we do |
| 01:23.06 | prpplague | joe_bleau in the last 10 years i think we've had less than 100 requests for source cd's |
| 01:23.34 | joe_bleau | So, hardly anyone outside AML is compiling code for your devices? |
| 01:23.45 | prpplague | joe_bleau yep |
| 01:23.54 | prpplague | joe_bleau but almost all the code is mainlined |
| 01:23.57 | joe_bleau | OK. Nevermind me then. |
| 01:24.51 | prpplague | joe_bleau http://git.kernel.org/?p=linux/kernel/git/torvalds/linux-2.6.git;a=blob;f=arch/arm/mach-s3c2410/mach-amlm5900.c;h=a67a0685664d0ca258b6bbc2b35ed6ea8f6156a2;hb=HEAD |
| 01:25.24 | prpplague | joe_bleau just a note, that will probably change with future products |
| 01:25.50 | prpplague | joe_bleau we've had such hell getting patches even looked at, that we have a moratorium on submitting them |
| 01:26.26 | ds2 | prpplague: does that mean there will be git.elinux.org soon to host it? |
| 01:26.29 | prpplague | joe_bleau if you want a copy of our sdk, i'll be happy to have one shipped |
| 01:26.48 | prpplague | ds2 uh no, why would elinux.org have a git? |
| 01:26.52 | joe_bleau | No thanks. |
| 01:27.11 | ds2 | prpplague: that seems to be the method in vogue to host kernel variants |
| 01:27.30 | prpplague | so why would AML have a git on elinux.org? |
| 01:27.46 | ds2 | elinux.org has the appearance of being somewhat neutral |
| 01:28.01 | prpplague | joe_bleau ok, guess i'm missing what you are asking about then |
| 01:28.17 | prpplague | ds2 you have me confused |
| 01:28.37 | prpplague | aml has no association with elinux.org |
| 01:28.52 | ds2 | prpplague: vendors seems to like to give the illusion that submitted their changes to the world by having it hosted on a "neutral" site |
| 01:28.55 | prpplague | elinux.org is a CELF sponsored web site |
| 01:29.04 | prpplague | ahh |
| 01:29.22 | ds2 | one of those marketing illusion they try to push... seems to be a trend I am seeing |
| 01:29.35 | prpplague | ds2 no if we were to have an open repo we'd post it on our own servers |
| 01:29.39 | ds2 | if you host it yourself, it must be "bad" and didn't bother to submit it |
| 01:30.27 | prpplague | ds2 we are bad at this stage, but the jerk who are doing repo and patch control are worse |
| 01:30.44 | ds2 | heh |
| 01:31.04 | prpplague | we simply can't waste time preparing patches when the simply are ignored |
| 01:31.13 | prpplague | s/the/they |
| 01:31.58 | ds2 | the OMAP tree folks seems pretty reasonable |
| 01:32.39 | prpplague | yea it all depends on the maintainer and if you are "known" to the group |
| 01:33.11 | prpplague | in addition if the item has anything out of ordinary that is business related, it gets dumped |
| 01:33.25 | ds2 | *nod* which is why I try to use someone else to interface with K.O. submissions |
| 01:34.04 | ds2 | be happy to give the author of the code that isbeing patched as much credit as he/she wants to get code in |
| 01:34.33 | *** join/#edev prpplague (n=dave123_@ppp-70-244-175-101.dsl.rcsntx.swbell.net) |
| 01:34.37 | prpplague | doh |
| 01:34.43 | prpplague | lost my wifi for some reason |
| 01:34.59 | prpplague | joe_bleau is there something specific you need info on? |
| 01:35.56 | prpplague | joe_bleau curious since you are the first in 4 years of my employement with AML to ask about such things |
| 01:36.52 | joe_bleau | I was under the impression y'all had more customers writing code. |
| 01:37.03 | prpplague | ahh |
| 01:37.22 | joe_bleau | If we were to run some form of real time linux on future products, |
| 01:37.30 | joe_bleau | I'll need to allow for end users to do things. |
| 01:37.38 | joe_bleau | And "no support" won't be an option. |
| 01:37.55 | prpplague | joe_bleau as in using AML's products to do that? |
| 01:38.03 | joe_bleau | No! |
| 01:38.27 | joe_bleau | We build hardware. |
| 01:38.32 | joe_bleau | Scared to go to linux. |
| 01:38.32 | prpplague | as do we |
| 01:38.41 | joe_bleau | Someone else is building hardware, shipping linux. |
| 01:38.50 | joe_bleau | I thought they had end users programming said hardware. |
| 01:38.59 | joe_bleau | Was curious what the end-users had to work with. |
| 01:39.00 | prpplague | ahhhhh |
| 01:39.18 | prpplague | using us as a guide |
| 01:39.21 | joe_bleau | Man, I suck at explaining things. |
| 01:39.21 | ds2 | joe_bleau: isn't that kind of like all the semiconductor vendors? |
| 01:39.28 | prpplague | we are a verticle market, very very very very odd |
| 01:39.52 | prpplague | most people have no clue and don't care what OS the device is running |
| 01:39.59 | prpplague | as long as it does what they need it to do |
| 01:40.10 | prpplague | only the very very large companies do custom applications |
| 01:40.27 | prpplague | the rest generally use off the shelf apps |
| 01:41.07 | joe_bleau | End user dev would be rare for us too. |
| 01:41.10 | prpplague | those that do create custom apps are generally entry level developers |
| 01:41.19 | prpplague | and generally have low skills as to linux |
| 01:41.30 | joe_bleau | But that 2% is very influential. They have power to lock in or out a product. |
| 01:41.30 | prpplague | which is why we have a whole host of those evil proc entries |
| 01:41.39 | prpplague | joe_bleau indeed |
| 01:42.00 | prpplague | joe_bleau so for those we have a complete set of environment to get them running very quick with custom apps |
| 01:42.05 | joe_bleau | (Speaking about 2% of our customer base, not yours.) |
| 01:42.35 | prpplague | its the same in our customer base |
| 01:43.15 | joe_bleau | Anyway, that also explains why I was interested in the hammer dev environment, too. |
| 01:43.43 | prpplague | well don't be confused, AML and TinCanTools are two very distinct businesses |
| 01:44.19 | joe_bleau | I don't *feel* confused...but that doesn't mean that I'm not, in fact, confused. |
| 01:44.26 | prpplague | hehe |
| 01:44.54 | prpplague | TinCanTools basically is getting hardware designs that AML didn't want to use, then selling them with software |
| 01:45.04 | prpplague | targeting a whole different set of customers |
| 01:45.16 | joe_bleau | Correct. |
| 01:46.09 | prpplague | joe_bleau software wise, TCT would definetly provide software support for a fee |
| 01:46.27 | prpplague | joe_bleau the best example of hardware companies that ship linux is simtec |
| 01:46.41 | prpplague | joe_bleau you can look at their business model as a good example |
| 01:46.51 | prpplague | joe_bleau http://www.simtec.co.uk iirc |
| 01:47.11 | prpplague | joe_bleau they have several levels of tech support |
| 01:47.26 | prpplague | joe_bleau can choose to purchase the product with the associated support level |
| 01:48.29 | prpplague | joe_bleau if you buy the product as cheap as possible without support don't come a calling |
| 01:49.01 | joe_bleau | Yeah, that's probably not an option for us, either. Little too extreme. |
| 01:49.09 | joe_bleau | But thanks, I'm looking over their site now. |
| 01:53.21 | prpplague | joe_bleau okie dokie |
| 01:53.33 | prpplague | joe_bleau well i'll be happy to answer any question you have |
| 01:53.41 | joe_bleau | Seriously, I appreciate it. |
| 01:54.11 | prpplague | joe_bleau shipping linux is a big deal, it usually is a pretty big shock for the management |
| 01:54.30 | prpplague | joe_bleau because it is a very non-traditional way of doing business |
| 02:00.52 | ds2 | tradition is obsolete. |
| 02:01.58 | prpplague | the lack of tradition is what allowed the US to speed ahead of alot of other countries |
| 02:02.28 | prpplague | but now that the US has been around for awhile, it is starting to develop certain traditional business thinking that is limiting growth |
| 02:02.41 | T0mW | heh |
| 02:02.42 | Redhatter | Yet, despite this, they're one of the few countries I know that stick by the imperial system of measurement. :-P |
| 02:03.15 | T0mW | prpplague: yeah, it is called "intelletual property" and businesses suddenly don't like to share |
| 02:03.32 | T0mW | "what's mine is mine and yours is mine" |
| 02:03.59 | Redhatter | Not just that, but also trade secrets...etc. |
| 02:04.42 | T0mW | IMHO, the problem is with the new management: they come from Burger King to work at a technology company and have no clue as to what engineers do. |
| 02:04.43 | Redhatter | It's like stepping onto a nudist beach... it'll take them a while to get comfortable with stripping their kit off. |
| 02:07.08 | prpplague | hehe |
| 02:07.38 | prpplague | Redhatter remaining with the imperial measurements is exactly the type of tradition that we are talking about |
| 02:08.07 | T0mW | joe_bleau: I have one customer that I've been talking linux to for several years. They had been resistant to it. Something changed and now they are receptive. |
| 02:08.31 | joe_bleau | I like imperial. Use it every day. |
| 02:08.55 | Redhatter | I use imperial if I need a rough measurement. IF I'm going to measure properly, I may as well use metric. |
| 02:08.56 | joe_bleau | Besides, imperial unit using countries did pretty good in WW2, didn't they? ;-) |
| 02:08.57 | T0mW | maybe the owner sat next to someone on a first-classs airline flight and learned what Linux is? |
| 02:09.05 | Redhatter | That, and if I need to do calculations, I use metric. |
| 02:10.48 | T0mW | prpplague: ethernet and mmc working on the digi2410. Now, I need to move onto SPI stuff (SC16IS752 uarts). Is there any howto's on the SPI layer? |
| 02:11.35 | T0mW | e.g. starting a block on it's way, waiting for completion, etc? |
| 02:13.03 | prpplague | T0mW mmc via the sdio controller? |
| 02:13.40 | T0mW | prpplague: yeah, that four bit data bus thingy |
| 02:14.08 | T0mW | prpplague: you want the patch? |
| 02:16.45 | prpplague | naw, i have that working |
| 02:16.55 | prpplague | its the spi that's giving me hell |
| 02:17.06 | T0mW | prpplague: mind if I see that code? |
| 02:17.19 | prpplague | hell ya |
| 02:17.25 | T0mW | ok, thanks |
| 02:17.30 | prpplague | hehe |
| 02:17.33 | prpplague | its at the office |
| 02:17.42 | prpplague | i'll have to get it in the morning |
| 02:17.56 | T0mW | ~lart dave for not having a vpn |
| 02:17.56 | purl | lowers dave's priority for not having a vpn |
| 02:18.25 | T0mW | prpplague: what?! you go *home* ? |
| 02:19.05 | ds2 | I happen to like the imperial system of measurement.. it actually makes sense! |
| 02:27.24 | flyback | http://www.youtube.com/watch?v=K2GLrZSTsI8# |
| 02:27.25 | flyback | OW? |
| 02:27.27 | MonMotha | the only thing I really use imperial units for anymore (other than causal day-to-day stuff) is designing PCBs |
| 02:28.18 | MonMotha | it's currently very annoying that about half the parts are designed in inches and half in metric |
| 02:28.34 | MonMotha | if it gets to where almost everything is in metric, I'll switch |
| 02:29.32 | prpplague | later folks |
| 02:32.02 | ds2 | 0.040 = 1mm ;) |
| 02:35.51 | T0mW | MonMotha: yeah, parts are in metric and the boards are inches |
| 02:36.22 | MonMotha | T0mW: but SOICs are still common and are in inches, and many connectors, despite being specified in metric, are actually designed in inches and then converted to non-nice metric numbers |
| 02:36.42 | T0mW | lol, yeah, btdt |
| 02:39.36 | MonMotha | and then of cours all the metric parts are on odd spacings: 1, 0.8, 0.75, 0.65, 0.5 |
| 02:52.16 | *** join/#edev linac (n=lin@222.76.82.93) |
| 03:38.12 | *** join/#edev prpplague (n=dave123_@ppp-70-244-175-101.dsl.rcsntx.swbell.net) |
| 03:52.56 | prpplague | ok well this definetly the oddest match up of headlines i've ever seen: |
| 03:52.58 | prpplague | # |
| 03:52.58 | prpplague | Make Love, Not War |
| 03:52.58 | prpplague | Group encouraging planet get busy on winter solstice for synchronized 'Global Orgasm for World Peace' |
| 03:52.58 | prpplague | • Study Shows Why Monkeys Shout During Sex |
| 03:53.59 | cajaka | lol |
| 04:11.08 | cajaka | prpplague you still around? |
| 04:12.41 | prpplague | cajaka jsut barely |
| 04:13.56 | cajaka | I'll ask you tomorrow Its nothing that I would need an answer today. |
| 04:14.08 | prpplague | cajaka go ahead |
| 04:14.28 | prpplague | boss lady is still watching tv, i can't go to sleep with the tv on hehe |
| 04:14.57 | cajaka | Ok the hammer has 2.26.2x kernel and support for PPP right? |
| 04:15.23 | prpplague | cajaka yea 2.6.22 kernel, you can enable ppp in the kernel |
| 04:15.27 | prpplague | yep |
| 04:15.50 | cajaka | What are the chances of getting this is work on it: http://www.sierrawireless.com/faq/ShowFAQ.aspx?ID=607 |
| 04:16.10 | cajaka | With a powered hub |
| 04:17.25 | prpplague | shouldn't be a problem at all |
| 04:18.27 | prpplague | cajaka the 2.6.23 release appears to have the drivers in already |
| 04:18.54 | prpplague | cajaka if it were me, i'd wait to use the 2.6.23 or 24 kernel for the hammer |
| 04:19.05 | prpplague | cajaka but you could use their posted driver |
| 04:19.53 | cajaka | I'll have to up my budget my over all plan is to make another ROV better then the one I did in college. Which in all honesty it wouldn't be to hard |
| 04:20.02 | prpplague | hehe |
| 04:20.16 | prpplague | cajaka really need one with that kind of communication? |
| 04:20.54 | cajaka | No, but it would be nice. They offer a good data plan at work for it |
| 04:20.55 | prpplague | sigwife |
| 04:21.09 | prpplague | cajaka you've been nice'd to a lower status |
| 04:21.11 | prpplague | later |
| 04:37.28 | *** join/#edev redblue (i=star@ppp186.109-253-207.mtl.mt.videotron.ca) |
| 04:43.27 | *** join/#edev DrZimmerman (n=theo@2001:5c0:8fff:fffe:0:0:0:6af1) |
| 05:26.34 | *** join/#edev landley (n=landley@cpe-72-177-114-214.austin.res.rr.com) |
| 10:51.58 | *** join/#edev purl (i=ibot@pdpc/supporter/active/TimRiker/bot/apt) |
| 10:51.58 | *** 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/ |
| 11:29.17 | *** join/#edev DrZimmerman (n=theo@2001:5c0:8fff:fffe:0:0:0:6af1) |
| 11:37.53 | *** join/#edev steliosk (n=Stelios@ipa226.211.tellas.gr) |
| 11:47.36 | *** join/#edev linac (n=lin@121.41.53.24) |
| 12:47.32 | *** join/#edev Redhatter (n=stuartl@203.94.56.252) |
| 13:11.15 | *** join/#edev howtoo (n=howtoo@195.159.62.66) |
| 13:12.27 | *** join/#edev MothaWork__ (n=monmotha@149.164.193.61) |
| 14:00.51 | *** join/#edev GPSFan (n=kenm@12.10.255.246) |
| 14:00.51 | *** mode/#edev [+o GPSFan] by ChanServ |
| 14:12.02 | *** join/#edev prpplague (n=dave@mail.americanmicrosystems.com) |
| 14:30.50 | *** join/#edev dijenerate (n=dijenera@66.54.118.16) |
| 14:31.46 | *** join/#edev chouimat|work (n=mchouina@209.217.106.98) |
| 14:38.26 | *** mode/#edev [+o prpplague] by ChanServ |
| 14:39.26 | chouimat|work | morning |
| 15:00.12 | *** join/#edev dijenerate (n=dijenera@66.54.118.16) |
| 15:28.19 | *** join/#edev redblue (i=star@ppp032.108-253-207.mtl.mt.videotron.ca) |
| 15:36.44 | hapchap_ | so i have decided to change careers... I saw a truck yesterday with yellow flashing lights on top... the side of it had "Mobile Rotary" on the side... |
| 15:37.59 | hapchap_ | "Oh No! That needs to be notarized! But where will we ever find a Notary Public?" |
| 15:38.17 | hapchap_ | Im going to be a billionaire |
| 15:38.38 | hapchap_ | whoops: "Mobile Notary" |
| 15:53.11 | SpeedEvil | Rotary notary! |
| 15:57.06 | *** join/#edev CosmicPenguin (n=nobody@163.181.251.103) |
| 15:57.06 | *** mode/#edev [+o CosmicPenguin] by ChanServ |
| 16:16.26 | *** join/#edev ico2 (n=ico2@modem-17.cougar.dialup.pol.co.uk) |
| 16:32.30 | hapchap_ | SpeedEvil: hehehe |
| 16:50.48 | *** join/#edev nelson (n=nelson@rrcs-72-43-17-50.nys.biz.rr.com) |
| 17:02.16 | Mirell | hapchap_: They already have those, too. |
| 17:45.02 | *** join/#edev ico2 (n=ico2@modem-17.cougar.dialup.pol.co.uk) |
| 17:59.09 | *** join/#edev dijenerate (n=dijenera@66.54.118.16) |
| 18:08.13 | *** join/#edev lyakh (n=lyakh@p57BD2877.dip0.t-ipconnect.de) |
| 18:16.22 | *** join/#edev hapchap (n=hapchap@mail.americanmicrosystems.com) |
| 19:13.42 | *** join/#edev purl (i=ibot@pdpc/supporter/active/TimRiker/bot/apt) |
| 19:13.42 | *** 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/ |
| 19:28.08 | *** join/#edev dijenerate (n=dijenera@66.54.118.16) |
| 19:48.11 | *** join/#edev dstuxo (n=dstuxo@212.234.51.91) |
| 19:54.06 | *** join/#edev dijenerate (n=dijenera@66.54.118.16) |
| 20:08.26 | *** join/#edev ico2 (n=ico2@modem-2507.cheetah.dialup.pol.co.uk) |
| 20:10.50 | *** join/#edev T0mW (n=Tom@24.238.86.158.res-cmts.sth.ptd.net) |
| 21:17.54 | *** join/#edev redblue (i=star@ppp147.108-253-207.mtl.mt.videotron.ca) |
| 21:28.38 | prpplague | wmat: man, or forum is being over run with spammers |
| 21:42.52 | ds2 | overgrown zipit? - http://www.ogo.com/ogoFeaturesCT25.aspx |
| 21:52.11 | prpplague | interesting |
| 21:52.16 | prpplague | ds2: what does it cost? |
| 21:54.10 | ds2 | not available in US speaking countries yet |
| 21:56.15 | prpplague | hmm |
| 21:56.36 | ds2 | the exterior specs sure look a lot like a zipit with t0mw's mods |
| 21:57.29 | chouimat|work | is US speaking countries is the bad version of english spoken south of the border? |
| 21:57.47 | ds2 | =) |
| 22:08.45 | SpeedEvil | hwo much are thost thingys? |
| 22:08.51 | *** join/#edev ico2_ (n=ico2@modem-3414.chameleon.dialup.pol.co.uk) |
| 22:09.07 | T0mW | prpplague: that sysfs looks to be pretty slick idea. I think it will help me in building/testing kernel modules |
| 22:09.48 | prpplague | T0mW: indeed sysfs has it's uses |
| 22:10.10 | ds2 | this is annoying... a device actually following USB specs :/ |
| 22:11.18 | T0mW | prpplague: any reason not to use sysfs to control / interface to a module rather than use IOCTL? |
| 22:11.35 | T0mW | is sysfs faster? |
| 22:11.55 | ds2 | ioctls are more natural if there is afilehandle already |
| 22:12.05 | prpplague | T0mW: i'd guess no |
| 22:12.12 | prpplague | T0mW: might pester CosmicPenguin |
| 22:12.21 | T0mW | ds2: yeah, ioctl has been around for years |
| 22:12.29 | prpplague | T0mW: i'm still an old procfs fuddy duddy |
| 22:12.33 | T0mW | lol |
| 22:12.39 | CosmicPenguin | eh? |
| 22:12.51 | T0mW | CosmicPenguin: : any reason not to use sysfs to control / interface to a module rather than use IOCTL? |
| 22:13.05 | T0mW | CosmicPenguin: is sysfs faster? |
| 22:13.06 | CosmicPenguin | -enoparse |
| 22:13.19 | prpplague | CosmicPenguin: T0mW has some custom i/o modules |
| 22:13.38 | prpplague | CosmicPenguin: he wants to know if it would be easier/faster to talk to them via sysfs |
| 22:13.59 | CosmicPenguin | it depends on what you are telling them |
| 22:14.03 | prpplague | CosmicPenguin: custom hardware modules i mean |
| 22:14.07 | ds2 | also, sysfs won't work til it gets mounted so if you need access early on... |
| 22:14.08 | CosmicPenguin | sysfs is for single bits of information |
| 22:14.33 | CosmicPenguin | meant to be human parsable |
| 22:14.38 | T0mW | CosmicPenguin: I was looking at sysfs and it is pretty cool. I've got some kernel modules to build (array of 16 LEDs) and am deciding whether sysfs is merely a debug interface or meant to replace ioctl calls. |
| 22:14.39 | prpplague | i suspect T0mW mainly wants to do quick checks of status |
| 22:14.58 | CosmicPenguin | ioctl() is still valid for sending in more complex chunks of data |
| 22:15.20 | prpplague | T0mW: what are you doing with 16 LEDS? |
| 22:15.26 | CosmicPenguin | it sort of all depends on what you have and what you are doing |
| 22:15.36 | T0mW | CosmicPenguin: ok, so I should use ioctl for production modules and sysfs for dev purposes? |
| 22:15.45 | CosmicPenguin | its not an either or thing |
| 22:16.11 | CosmicPenguin | ioctl() is as useful as it ever was in the classic posix sense |
| 22:16.33 | CosmicPenguin | and sysfs is useful for checking and tweaking kernel and device behavior |
| 22:16.47 | T0mW | CosmicPenguin: ok, thanks |
| 22:16.49 | CosmicPenguin | sometimes you can use both |
| 22:17.03 | CosmicPenguin | sysfs works best when you have human parsable values |
| 22:17.07 | CosmicPenguin | like say - backlight level |
| 22:17.13 | T0mW | CosmicPenguin: so, there is no common consensus on the use / misuse of sysfs then? |
| 22:17.16 | CosmicPenguin | or to read PCI device information |
| 22:17.27 | CosmicPenguin | or to tweak kernel behavior like power managmenet |
| 22:17.27 | ds2 | the firmware loader does use sysfs.... |
| 22:18.06 | CosmicPenguin | T0mW: well - there are some pretty solid guidelines for the sysfs in the kernel documentation |
| 22:18.23 | T0mW | CosmicPenguin: so, you would say that sysfs is a more formal approach than hacking into proc fs to get stuff done like the LCD backlight control |
| 22:18.29 | CosmicPenguin | yes |
| 22:18.32 | T0mW | aha |
| 22:18.33 | CosmicPenguin | its a formalized infrastructure |
| 22:18.36 | T0mW | good |
| 22:18.39 | CosmicPenguin | not adhoc like proc |
| 22:18.43 | T0mW | yeah |
| 22:19.17 | T0mW | like: echo '1' > /proc/net/ipv4/ip_forward |
| 22:20.10 | T0mW | you could expose a more formalized set of control / info points (files) within sysfs |
| 22:20.56 | CosmicPenguin | indeed |
| 22:21.06 | CosmicPenguin | you might want to check out drivers/leds |
| 22:21.15 | CosmicPenguin | there is already a start of a LED class for sysfs |
| 22:21.48 | T0mW | I was reading prpplague's code for the tct-hammmer-beeper and playing with that CLASS_DEVICE_ATTR stuff |
| 22:23.01 | T0mW | I can already see that I could dev a kernel module more quickly by exposing some vars with sysfs rather than write a quick program to do an ioctl |
| 22:31.07 | prpplague | T0mW: the tct-hammer also has example driver for the leds class |
| 22:31.20 | prpplague | T0mW: the driver for the onboard led as well as the carrier board led use it |
| 22:31.54 | prpplague | T0mW: if the led controller supports dimming you can control it with a value between 0 and 255 |
| 22:32.13 | T0mW | prpplague: yeah, I have 16 LEDs to blink. They are on a pair of 8bit drivers and I will need to bit-bang then via GPIO pins. |
| 22:32.29 | prpplague | T0mW: ahh |
| 22:32.33 | prpplague | T0mW: very easy to do then |
| 22:32.50 | prpplague | T0mW: just enumberate them in your i/o init, similiar to what i do on the hammer code |
| 22:33.11 | T0mW | prpplague: I haven't written a 2.6 kernel module yet, I have the 3rd edition book. Looking at the code for the beeper + enc28j60 + hammer-LED, it doesn't look imposing. |
| 22:33.40 | prpplague | T0mW: no, the biggest problem for me is re-learning alot of items such as the block driver support |
| 22:33.48 | T0mW | prpplague: what will be interesting is that I will have to schedule a timer within the kernel to service the LED blinking. |
| 22:33.49 | prpplague | T0mW: (coming from 2.4 i mean) |
| 22:34.21 | CosmicPenguin | thats not so bad - the timers are pretty straight forward |
| 22:35.02 | T0mW | I'm still messing with details in OE for my basic filesystem and looking over the kernel sources while OE is building. |
| 22:35.56 | wmat | prpplague: damn spammers |
| 22:36.44 | prpplague | T0mW: probably the best bet on that is to use the led driver, but add an option that uses the brightness in conjunction with the timmer |
| 22:39.26 | T0mW | prpplague: I pulse the LEDs for, what is it, 50ms. This way they imitate activity occuring on the serial ports rather than stay on. It also saves power if I run them at a 50% duty cycle, they are visible but don't chew up a lot of milliamps |
| 22:39.27 | *** join/#edev hapchap (n=hapchap@mail.americanmicrosystems.com) |
| 22:39.28 | prpplague | T0mW: actually seems like there is an option for led flashing in the led class |
| 22:39.49 | prpplague | T0mW: indeed |
| 22:40.36 | prpplague | T0mW: hehe you could pwm the ground connection |
| 22:40.51 | T0mW | 16 LEDs X 8ma-per == 128ma for LEDs |
| 22:41.29 | T0mW | prpplague: I do have an enable line! You know, I just didn't think of that! |
| 22:41.30 | T0mW | heh |
| 22:41.34 | prpplague | T0mW: just set up the pwm at 1hz |
| 22:41.37 | T0mW | doh |
| 22:42.37 | prpplague | hehe |
| 22:44.24 | T0mW | prpplague: that would work, far simpler than what I was thinking. Pulse the enable line to make them blink. When I "fire" an LED, set a count interval for how long it should stay lit. Then, repeated "firings" would simply put that counter back to max value. eventually, no more "firing" (reloads) and the count will run out extinguishing it. |
| 22:44.54 | T0mW | each LED would fire asynchronously |
| 22:45.12 | T0mW | prpplague: thank you DAve |
| 22:45.15 | prpplague | hehe |
| 22:45.34 | T0mW | sometimes you just cannot see it, that is why I need "co-workers" |
| 22:45.48 | prpplague | indeed |
| 23:06.29 | prpplague | Mirell: hehe, landley went to the nudie bar with you yesterday? |
| 23:20.27 | *** join/#edev FireEgl (i=FireEgl@Sebastian.Tcldrop.US) |
| 23:21.21 | flyback | THIS IS COOL |
| 23:21.44 | flyback | i got 4 16 mb usb circular flash drives for $7, being made by verbatium, I figured they would pick a good quality oem |
| 23:21.50 | flyback | they did -> m-systems/sandisk :) |
| 23:21.52 | flyback | score! |
| 23:34.11 | *** join/#edev CosmicPe1guin (n=nobody@163.181.251.103) |
| 23:34.41 | *** mode/#edev [+o CosmicPenguin] by ChanServ |
| 23:56.22 | CosmicPenguin | we heard you the first time |