| 00:00.44 | RobotGuy | Has anyone else tried following the tutorial on how to use JFFS2 flash as the rootfs? |
| 00:00.51 | g1powermac_PB | http://cgi.ebay.com/850B-SMT-SMD-HOT-AIR-REWORK-STATION-SOLDER-STATION-NEW_W0QQitemZ360046593088QQihZ023QQcategoryZ36327QQssPageNameZWDVWQQrdZ1QQcmdZViewItem |
| 00:01.16 | RobotGuy | Sparkfun has what looks like a good one. |
| 00:01.29 | g1powermac_PB | pretty good looking for $100 |
| 00:03.30 | *** join/#edev ce_geek (n=ce_geek@roger-blanchette-4.umelst.maine.edu) |
| 00:03.36 | RobotGuy | ce_geek: Greetings |
| 00:03.44 | ce_geek | RobotGuy: Howdy |
| 00:03.46 | ce_geek | prpplague: Ping |
| 00:04.02 | RobotGuy | SpeedEvil: ping |
| 00:04.02 | ce_geek | How're things? |
| 00:04.37 | RobotGuy | ce_geek: I'm wrestling with some tutorials I think are very badly written. Did you try the JFFS2 tutorial? |
| 00:04.56 | *** join/#edev Crofton (n=balister@66-207-66-26.black.dmt.ntelos.net) |
| 00:05.01 | ce_geek | RobotGuy: No. I'm not sure I'll get to it right off, I need to focus on some hardware stuff. |
| 00:05.03 | RobotGuy | Crofton: Greetings |
| 00:05.12 | ce_geek | RobotGuy; Though JFFS2 would make that a bit easier. Hmmn... |
| 00:05.16 | ce_geek | RobotGuy: I might get to. |
| 00:05.29 | ce_geek | Is prp around at all? |
| 00:05.34 | Crofton | gm |
| 00:05.45 | g1powermac_PB | he was just a few minutes ago |
| 00:05.46 | RobotGuy | ce_geek: My next step is to get Hammer to use a Thumb Drive as rootfs. |
| 00:07.01 | RobotGuy | Although I would rather use an SD/MMC card. |
| 00:08.04 | RobotGuy | I found a 32 GB SD card. |
| 00:11.16 | ce_geek | RobotGuy: That'd be cool. |
| 00:11.18 | *** join/#edev dijenerate (n=dijenera@72.51.92.180) |
| 00:15.19 | ce_geek | RobotGuy: I've got to blink some LEDs and talk to a 16x2 character LCD. |
| 00:15.22 | ce_geek | RobotGuy: Done any of that? |
| 00:15.28 | ce_geek | RobotGuy: With the hammer, that is. |
| 00:15.32 | RobotGuy | ce_geek: No, I haven't. |
| 00:17.08 | prpplague | ce_geek give me 10 minutes and i'll be back |
| 00:17.13 | ce_geek | prpplague: Cool |
| 00:21.46 | prpplague | back |
| 00:21.50 | prpplague | ce_geek whats up? |
| 00:22.20 | ce_geek | prpplague: I need to blink an LED. Gotta make that happen. Just curious what I'm going to have to do. |
| 00:22.30 | ce_geek | prpplague; If I can do it with calls to devmem, that'd be fine. |
| 00:22.57 | ce_geek | prpplague: Looking for a quick and even filthy way to toggle a GPIO pin in user space. |
| 00:23.45 | prpplague | ce_geek depends on "where" you want to do the blinking |
| 00:24.00 | ce_geek | prpplague: Probably not in a kernel driver, not the first run around. |
| 00:24.42 | prpplague | http://www.elinux.org/Hammer_LED_Driver |
| 00:24.59 | prpplague | from userland the easiest is to use the led driver |
| 00:25.35 | ce_geek | prpplague: Let's say I want to pick 4 GPIO pins and be able to toggle them or display 4 bits of information or something like that. |
| 00:26.21 | prpplague | ce_geek ok without using an existing kernel driver, you'd need to use a memory map similiar to what devmem does |
| 00:26.26 | ce_geek | prpplague: And ultimately use those 4 pins along with a few other control pins to bit-bang an interface to a 16x2 character LCD. |
| 00:26.53 | ce_geek | prpplague: Okay, cool. If I were writing in C, what would the tool be? |
| 00:26.56 | prpplague | again it all depends on where you want to do it |
| 00:27.12 | prpplague | you can look at the source for devmem2.c |
| 00:27.22 | ce_geek | prpplague: ok |
| 00:27.38 | prpplague | ce_geek you can use the core of devmem2.c to create a simple program to blink a led |
| 00:27.43 | ce_geek | prpplague: One place to start might be to pick the GPIO pins and then check to see how they're configured with devmem? |
| 00:27.56 | prpplague | ce_geek once you have that working, you can move on to the the larger items such as the lcd |
| 00:28.05 | prpplague | ce_geek bingo |
| 00:28.08 | ce_geek | prpplague: Right. I need to master control of one LED, then worry about how to do several. |
| 00:28.15 | prpplague | exactly |
| 00:28.37 | ce_geek | prpplague: All the while acknowledging that it should be done as a kernel driver ;-) |
| 00:31.16 | ce_geek | prpplague; I want to do the kernel driver type stuff soon, I'm just trying to get some useful example stuff together for Rick |
| 00:32.19 | prpplague | ce_geek yea, what i would do is use devmem2.c to do your blinking led from userland, then do a blinking led module |
| 00:32.26 | prpplague | kernel module that is |
| 00:32.46 | ce_geek | ok |
| 00:33.52 | prpplague | a small led blinking kernel module can be done pretty easy |
| 00:34.04 | ce_geek | prpplague: Excellent. |
| 00:34.09 | ce_geek | prpplague: Looking at devmem2.c now. |
| 00:34.35 | ce_geek | prpplague: Does this exist as a software package somewhere on the interweb other than the hammer wiki? |
| 00:34.52 | ce_geek | prpplague: I see the comments where it talks of LART and whatever. |
| 00:35.11 | ce_geek | prpplague: Google turns it up I guess. k. |
| 00:35.48 | prpplague | hehe yea LART was where they did the original code |
| 00:36.04 | prpplague | http://www.elinux.org/Hammer_Board_Software_System |
| 00:36.13 | prpplague | ce_geek source is at the bottom of the page |
| 00:36.31 | ce_geek | prpplague: Yeah I found it there. Thanks for putting it there. |
| 00:36.37 | prpplague | ce_geek no |
| 00:36.43 | prpplague | ce_geek problem |
| 00:36.45 | prpplague | hehe |
| 00:37.27 | ce_geek | prpplague: I can sorta understand it, but it could be commented a bit better :-P |
| 00:38.53 | prpplague | ce_geek indeed, not really doc'd\ |
| 00:39.08 | prpplague | ce_geek basically it uses mmap to create a virtual to physcal mapping |
| 00:39.23 | prpplague | ce_geek so when you pass it a physical address as an argument |
| 00:39.43 | prpplague | ce_geek it maps some of the virtual memory address space to that physical address |
| 00:40.46 | ce_geek | prpplague; So how would the software that does the memory stuff be different in a kernel module? Use something other than mmap? |
| 00:41.34 | prpplague | ce_geek the kernel has methods for doing the physical to virtual via function calls |
| 00:41.42 | ce_geek | ah |
| 00:42.05 | prpplague | ce_geek for instance you can look at how we configured the second serial port during the machine init |
| 00:42.11 | ce_geek | Yea |
| 00:44.26 | *** join/#edev sunrunner20 (n=jamesdoe@pool-70-104-23-68.dllstx.fios.verizon.net) |
| 00:44.31 | *** join/#edev robbat2 (n=robbat2@gentoo/developer/robbat2) |
| 00:44.43 | prpplague | ce_geek best thing is to just use devmem first to get the address and configurations right |
| 00:44.51 | ce_geek | prpplague: Well I guess I'll try to blink one by using devmem by hand first, just for giggles. |
| 00:44.57 | ce_geek | prpplague.... hah |
| 00:44.59 | prpplague | hehe |
| 00:45.10 | prpplague | ce_geek when you going to start on it? |
| 00:45.13 | solar | ce_geek: the devmem code should be in busybox w/in the month hopefully. |
| 00:45.29 | ce_geek | What is busybox, just exactly? |
| 00:45.42 | prpplague | solar actually i'm gonna post the patch for it on the wiki and in buildroot shortly |
| 00:45.46 | solar | the embedded swiss army knife. |
| 00:45.52 | robbat2 | prpplague, bunch of issues, I realize I wrote em in #elinux instead of #edev |
| 00:46.00 | solar | heh. |
| 00:46.02 | robbat2 | 1. the cd with the nail mentions hammer-reflash.cfg but that file isn't on the CD at all |
| 00:46.06 | prpplague | ce_geek http://busybox.net/about.html |
| 00:46.16 | ce_geek | prpplague; thanks :) |
| 00:46.19 | robbat2 | 2. using the Hammer 2.6.22 sources, even after loading the g_ether into the kernel while it shows up on the nail properly, it doesn't show up on the host worked around that for now with a externally-powered USB docking bay |
| 00:46.39 | robbat2 | 3. the apex.bin shipped is not built with the same config that is posted on the wiki |
| 00:46.48 | ce_geek | prpplague; Okay so you've tweaked busybox with things appropriate for the hammer board |
| 00:47.06 | prpplague | ce_geek actually no |
| 00:47.13 | robbat2 | http://tinyurl.com/5555bh <---- photos of Nail and the docking bay |
| 00:47.15 | prpplague | ce_geek busybox is used with no mods |
| 00:47.24 | ce_geek | prpplague; oh |
| 00:48.16 | prpplague | robbat2 http://www.elinux.org/upload/4/44/S3c2410_udc.patch |
| 00:48.19 | solar | robbat2: what about the beep really being volume? |
| 00:48.37 | prpplague | robbat2 sounds like you got one of the older source cds |
| 00:48.44 | prpplague | makes a note to check on that |
| 00:48.59 | prpplague | robbat2 you need to apply the udc patch |
| 00:49.06 | robbat2 | one sec, just rebuilding with it |
| 00:50.25 | robbat2 | now just to wait for the flashing :-) |
| 00:50.30 | prpplague | robbat2 np |
| 00:51.07 | robbat2 | there was one other weird thing. I rebuilt apex to copy nor:256k+2048kb for the kernel, and it just went into doing that in an infinite loop |
| 00:51.46 | robbat2 | nor:256k+2048k for the kernel, nor:2304k+4096k for the rootfs |
| 00:52.32 | prpplague | robbat2 when making those changes you have to change to mtd flash partition settings as well to match |
| 00:52.43 | robbat2 | where is that actually changable? |
| 00:52.46 | robbat2 | i couldn't find it |
| 00:54.00 | prpplague | robbat2 the parition info is in the arch/arm/mach-s3c2410/mach-tct_hammer.c |
| 00:54.15 | *** join/#edev linac (n=lin@222.76.82.48) |
| 00:54.15 | robbat2 | ah, great |
| 00:54.39 | robbat2 | the udc patch works too |
| 00:54.47 | prpplague | robbat2 good deal |
| 00:54.51 | solar | prpplague: is it worth it to give the .25 you are working on a whack? |
| 00:55.20 | prpplague | robbat2 you'll have to forgive us, we are still getting the hang of dealing with retail and making sure things like the cd's are correct |
| 00:55.50 | robbat2 | not a problem there, it's just stuff that you'd want to have right for people that can't figure it out that well on their own |
| 00:55.55 | robbat2 | else you're stuck hand-holding them forever |
| 00:56.25 | prpplague | solar sure you can test it - http://www.arm.linux.org.uk/developer/patches/viewpatch.php?id=4963/2 |
| 00:56.30 | prpplague | robbat2 indeed |
| 00:56.58 | prpplague | robbat2 we normally do B2B stuff and its taken a little getting used to on the retail end |
| 00:57.13 | robbat2 | i'll test that kernel patch in a moment, just trying the MTD partition stuff first |
| 00:57.18 | prpplague | solar it only provides very core support |
| 00:57.52 | prpplague | it doesn't include any of the led or push button support |
| 00:58.03 | prpplague | but the rest such as the udc and such should be there and working |
| 00:58.40 | robbat2 | is it worthwhile to promote tct_hammer_mtd_partitions[0].size to a Kconfig option instead? |
| 00:59.17 | prpplague | robbat2 yea based the problems RobotGuy had, i've been considering that |
| 00:59.58 | prpplague | robbat2 i think i'd have trouble getting into the main tree, but we could keep it as an add on patch |
| 01:00.16 | RobotGuy | prpplague: If you mean .SIZE, I have that under control now. However, it is somewhat of a stumbling block that maybe could be removed. |
| 01:00.32 | robbat2 | i'm pretty certain I have seen similar stuff in the tree before |
| 01:00.47 | prpplague | RobotGuy indeed i just meant i learned from you about how difficuly it was to get clear on it |
| 01:00.59 | RobotGuy | Cool |
| 01:01.21 | RobotGuy | That was one of the harder parts for me, maybe the hardest. |
| 01:03.27 | prpplague | robbat2 so i assume you are still having fun with the nail? |
| 01:04.02 | robbat2 | prpplague, yup, it mostly works, just trying to make something nice we can throw up for users |
| 01:04.29 | ce_geek | prpplague: Often 22 bits in registers like GPHCON? Seems weird that there are labels for PAIRS of bits, like GPH0 in GPHCON. I suppose if there's more than two options for a single pin that's necessary. |
| 01:05.35 | prpplague | ce_geek correct, normally you have something like this: 00=input 01=output 10=txd2 11=cts1 |
| 01:05.46 | prpplague | ce_geek so that you can choose which function |
| 01:05.51 | ce_geek | Yea |
| 01:06.09 | ce_geek | prpplague; Good good. Makes sense. |
| 01:06.32 | ce_geek | prpplague: Making sure I draw the same conclusion you drew the other day with the state of the RXD2 and TXD2 pins before fixing them. |
| 01:06.50 | prpplague | ce_geek yep got it |
| 01:08.03 | prpplague | solar / robbat2 i've got 3 nail boards allocated for you guys |
| 01:11.52 | solar | are there any redesign plans in mind for the pcb itself? |
| 01:12.08 | solar | it would be nice to shove one of these in an altoids can |
| 01:12.58 | prpplague | solar hehe, well we've discussed if people are interested in a software dev only version to redo the board in a much smaller configuration |
| 01:13.30 | prpplague | solar we tried to have a good balance between size and hackability |
| 01:13.38 | solar | understandable |
| 01:14.33 | prpplague | if we remove the vast majority of hardware expansion, we can get it all down to the size of a slightly oversized thumb drive |
| 01:15.43 | prpplague | wonders what happend to robbat2 |
| 01:16.29 | solar | he comes and goes |
| 01:16.50 | prpplague | much like my sanity (i have kids) |
| 01:17.17 | *** join/#edev robbat2 (n=robbat2@gentoo/developer/robbat2) |
| 01:17.31 | prpplague | ce_geek you starting on the led blinking tonight? |
| 01:27.53 | prpplague | solar if we can get a larger number of sales we can drop the overall quantity price with larger builds |
| 01:28.32 | solar | this unit is somewhat compat with the openmoko right? |
| 01:29.50 | prpplague | solar yea, the nail is basically the neo1973 minus all the external devices |
| 01:31.11 | prpplague | solar there are some differences like the fact that the neo uses nand flash and the hammer uses nor |
| 01:31.57 | prpplague | solar we are considering doing a small lcd module for the nail board |
| 01:33.04 | *** join/#edev Appleman1234 (n=Appleman@131.181.98.112) |
| 01:36.46 | RobotGuy | prpplague: You could run a text LCD the length of Nail. Graphics might be hard to do that way though. |
| 01:36.52 | ce_geek | prpplague: Yeah |
| 01:37.32 | ce_geek | prpplague: Just got back from a laundry break. Looking at the data sheet now trying to figure out what port would be good to try messing with. |
| 01:40.29 | prpplague | RobotGuy we have a 96x64 oled that fits fine |
| 01:40.51 | RobotGuy | prpplague: Is that the one that is shown on the Wiki? |
| 01:40.55 | prpplague | yea |
| 01:41.00 | RobotGuy | Ahh, cool. |
| 01:41.18 | RobotGuy | I forgot that OLEDs go that small. |
| 01:42.03 | prpplague | http://www.femacorp.com/Products/LCD/OLED/OLED%20Datasheets/GM9664A-OLX2NLF.pdf |
| 01:43.14 | robbat2 | prpplague, changing .SIZE didn't help my infinite APEX loop |
| 01:43.22 | robbat2 | APEX Boot Loader 1.5.13 -- Copyright (c) 2004,2005,2006,2007 Marc Singer |
| 01:43.22 | robbat2 | <PROTECTED> |
| 01:43.22 | robbat2 | <PROTECTED> |
| 01:43.22 | robbat2 | <PROTECTED> |
| 01:43.22 | robbat2 | <PROTECTED> |
| 01:43.22 | robbat2 | <PROTECTED> |
| 01:43.24 | robbat2 | <PROTECTED> |
| 01:43.26 | robbat2 | <PROTECTED> |
| 01:43.28 | robbat2 | # copy $kernelsrc $bootaddr |
| 01:43.30 | robbat2 | # copy nor:256k+2048k 0x30008000 |
| 01:43.32 | robbat2 | and repeats |
| 01:44.15 | prpplague | robbat2 hmm thats odd, so the system reboots after the copy of the kernel? |
| 01:44.29 | robbat2 | after or during, i cannot differentiate |
| 01:45.01 | robbat2 | it's the latest APEX, with the apex changed forward-ported |
| 01:45.06 | robbat2 | s/changed/changes/ |
| 01:45.55 | prpplague | robbat2 uh wait, thats copying over the apex bootloader in sdram |
| 01:46.28 | robbat2 | lol, ok |
| 01:46.31 | robbat2 | that makes sense now |
| 01:46.50 | prpplague | robbat2 might reduce the kernel size by 128k and you should be ok |
| 01:47.20 | robbat2 | the kernel isn't that big yet, but i'll allocate less blocks sure |
| 01:48.41 | prpplague | robbat2 just out of curiosity, you've had a kernel that big before? |
| 01:48.59 | RobotGuy | My current kernel is 998 Kb. |
| 01:49.07 | robbat2 | earlier when mucking, i had one that was 1.3Mb. then I trimmed it down to work on stuff |
| 01:49.19 | prpplague | ahh |
| 01:50.00 | robbat2 | if I get right what i'm working on, it's build kernel, build rootfs, tune+build apex to correct size/locations, flash |
| 01:50.03 | prpplague | robbat2 one of the techniques for doing quick boot time is to have the kernel as small as possible and load the rest of the drivers via modules after the system boots |
| 01:50.21 | robbat2 | i was just avoiding modules for now |
| 01:50.26 | prpplague | robbat2 np |
| 01:50.27 | RobotGuy | prpplague: That's what I do. |
| 01:50.29 | agaffney | wow, solar *and* robbat2? :P |
| 01:50.49 | robbat2 | but once i've got the dynamic-sizing for build+flash going, then a lot of stuff will go modular |
| 01:50.54 | robbat2 | because a lot of it won't be needed |
| 01:51.09 | prpplague | ahh gotcha |
| 01:51.44 | robbat2 | speaking of which, any reason not to switch to using a cmdline partition table? |
| 01:52.02 | robbat2 | i'm wondering about building the env block for apex and flashing that at the same time |
| 01:52.10 | robbat2 | since it can have the right partition table |
| 01:52.16 | robbat2 | without building it at kernel time |
| 01:52.23 | RobotGuy | prpplague: I am designing a really small robot - 3.75" x 3.75" that is going to get a Hammer-RDP. |
| 01:52.43 | prpplague | robbat2 there is no reason you can't |
| 01:53.02 | robbat2 | that avoids the .SIZE problem entirely |
| 01:53.26 | robbat2 | agaffney, you think we're going to smoke this board with both of us putting time into it? |
| 01:53.32 | prpplague | robbat2 the only catch to command line partitioning is if you have a jffs2 partition you have to be careful about changing the boundries |
| 01:53.37 | solar | agaffney: yeah we are both poking at the nailboard. http://www.gentoo.org/proj/en/base/embedded/handbook/?part=4&chap=4 |
| 01:54.22 | agaffney | sweet |
| 01:54.52 | solar | question is what do you do here that you don't do in -embedded? |
| 01:55.34 | prpplague | solar sorry not sure i understand the question |
| 01:56.01 | solar | prpplague: sorry that was directed at agaffney |
| 01:56.31 | ce_geek | prpplague: Okay. Finally got a chance to look at the data sheet and understand what you had me to to change GPH6 and GPH7 to make them TX and RX instead of RTS and CTS. |
| 01:57.03 | prpplague | ce_geek good deal, things a little clearer? |
| 01:57.36 | prpplague | solar ahh ok |
| 01:58.13 | RobotGuy | SpeedEvil: ping |
| 01:58.25 | agaffney | solar: eh, I was originally in here when I picked up a pair of DHT walnut boards |
| 01:58.33 | agaffney | and then I've just stuck around :P |
| 01:58.44 | prpplague | jokes with agaffney |
| 01:58.54 | RobotGuy | I can't remember who gave me the part number of the PDIP8 level I2C translators |
| 01:59.08 | prpplague | yea its kind like herpes, once you got it, it seems to hang around |
| 01:59.19 | ce_geek | prpplague: Yeah now that I've finally taken a minute to look at things. Kind of just trying to get something cool to happen before, it's detail time now. |
| 01:59.41 | ce_geek | prpplague: I think I might have to wait until tomorrow to hook up the LED, I need to make more ribbon cables and that's a PIA. |
| 01:59.44 | prpplague | ce_geek yep, just remember the gpio's on the hammer can't sink alot of current |
| 01:59.54 | ce_geek | prpplague: How much? :-/ |
| 02:00.04 | ce_geek | looks at datasheet |
| 02:00.08 | prpplague | ce_geek no more than about 10mA |
| 02:00.21 | prpplague | ce_geek see the led driver page for an example |
| 02:00.29 | prpplague | ce_geek or look at the schematic for the carrier board |
| 02:00.35 | ce_geek | prpplague; K. I'll size the resistor for 8mA or so. |
| 02:00.46 | ce_geek | prpplague: Hopefully these LEDs I have will light up :-/ |
| 02:00.47 | prpplague | ce_geek you can still turn the onboard led on/of with devmem if you want |
| 02:00.59 | ce_geek | Ooh, good call. I'll just do that. :) |
| 02:04.13 | robbat2 | http://dev.gentoo.org/~robbat2/tincan-nail/ <--- anybody interested in Gentoo on there. ./flashme is what you want to put it on. |
| 02:04.51 | prpplague | nice |
| 02:04.57 | prpplague | bookmarks for testing |
| 02:05.12 | prpplague | ce_geek the onboard led for the hammer is on GPF0 |
| 02:05.25 | ce_geek | prpplague: Hehe I was just digging that up on the wiki / schematic. Thanks :) |
| 02:05.50 | ce_geek | prpplague: Just for giggles going to check GPFCON first |
| 02:06.18 | robbat2 | prpplague, thanks for the tip about the APEX loadpoint being overwritten. sizing down solved that |
| 02:07.46 | prpplague | robbat2 ahh good deal |
| 02:08.30 | ce_geek | prpplague: Okay it looks like GPFCON is all zeros. I'll have to change that so I can output a value on GPF0. |
| 02:08.57 | prpplague | ce_geek you running a kernel without the LED driver configured? |
| 02:09.58 | ce_geek | prpplague: aaaaaaaand I can blink the LED now. yay. |
| 02:10.10 | ce_geek | prpplague: uhhhh maybe? Somehow? Let me go see if it's there. |
| 02:10.19 | prpplague | ce_geek no biggy i was just curious |
| 02:11.12 | ce_geek | prpplague; I can echo 0 and 1 to /sys/devices/platform/s3c24xx_led.0/leds:led0/brightness |
| 02:11.20 | ce_geek | prpplague: to change the status of the LED |
| 02:11.43 | ce_geek | prpplague: Lemme clear GPFCON and see what it does. |
| 02:11.50 | g1powermac_PB | just bought one of those visor magnifying lens things, should help me hack the zipit :-) |
| 02:12.37 | ce_geek | prpplague: Looks like the driver fixed GPFCON |
| 02:12.58 | ce_geek | prpplague: Cool. |
| 02:13.23 | ce_geek | prpplague; Nice to see how this works. Good good. |
| 02:13.24 | prpplague | ce_geek yep, should set GPFCON when you do a write for on/off |
| 02:13.34 | ce_geek | prpplague; yep! |
| 02:13.40 | ce_geek | prpplague: Nifty. |
| 02:14.01 | ce_geek | prpplague: I think that's about it for me tonight. Tomorrow is documentation day. I want to document the hell out of everything I've done with the hammer so far. |
| 02:14.31 | prpplague | ce_geek good deal |
| 02:14.32 | ce_geek | prpplague: Hopefully that will be a useful document for everyone. I want to do so much more stuff, but I need to have some sort of deliverable for Rick this week. |
| 02:14.48 | prpplague | ce_geek think rick will be happy with it? |
| 02:15.12 | prpplague | ce_geek should be able to do the led and then we can look at the lcd |
| 02:15.19 | prpplague | ce_geek you already got one picked out? |
| 02:15.19 | ce_geek | prpplague: I think so. I wish I had more time to do some other stuff, I may try to squeeze some in next week, but I'm going to give hinm whatever I write tomorrow anyway. |
| 02:15.38 | ce_geek | prpplague: I've got one I've used in another project, one of the HD8840asdbasdfasd hitachi standard ones. |
| 02:15.44 | ce_geek | Just a 16x2 character LCD. |
| 02:15.51 | prpplague | ce_geek ahh ok |
| 02:15.59 | ce_geek | prpplague: Maybe if I got it to display temperature or something. Should be reasonable. |
| 02:16.01 | prpplague | ce_geek let me know if i can help |
| 02:16.22 | prpplague | ce_geek the only thing that might be an issue with your lcd |
| 02:16.23 | ce_geek | prpplague: Will do, thank you. I'm anxious to get your feedback on the document when I'm getting closer to calling it finished for the purpose of ricks' class. |
| 02:16.31 | prpplague | ce_geek is that it might be 5v i/o |
| 02:16.33 | ce_geek | prpplague: And getting it GPL'd or whatever. |
| 02:16.38 | prpplague | ce_geek if so we'll need to level shift it |
| 02:16.52 | ce_geek | prpplague: SNAP I haven't thought that far ahead. Yeah, can level shift. Just annoying :-P. |
| 02:17.15 | ce_geek | prpplague: Damn I think it is 5V only too, I doubt it would work happily with 3.3v. Might be worth trying. Damn. |
| 02:17.29 | prpplague | ce_geek ok just as long as you are aware if it |
| 02:17.47 | ce_geek | prpplague: Yea. Silly 3.3v :) |
| 02:17.51 | robbat2 | prpplague, other random question. the 'reset' button next to R1, is that supposed to be hardreset or softreset? |
| 02:17.54 | ce_geek | prpplague; Alright. Talk to you tomorrow. |
| 02:18.48 | prpplague | ce_geek later |
| 02:19.09 | prpplague | robbat2 that does a hardreset of just the hammer board |
| 02:19.24 | robbat2 | ah, that explains why it wasn't using during flashing |
| 02:19.32 | robbat2 | i'll have to hack up an idea instead |
| 02:19.53 | prpplague | robbat2 ?? |
| 02:20.29 | robbat2 | as odd as it may seem, the largest time-eater in setting this up, was solar waiting for me to cycle power by disconnecting the USB each time |
| 02:20.50 | prpplague | robbat2 ahh ok, you can reset the board via software on the host pc |
| 02:21.16 | robbat2 | after programming via JTAG, the ttyUSB* interfaces vanish |
| 02:22.02 | prpplague | robbat2 right after the flashing it should reset the board and then boot back up |
| 02:22.42 | robbat2 | it might be how I modified the flashing stuff to work with a newer OpenOCD |
| 02:22.52 | robbat2 | using the stock stuff there, it just flashed infinitely |
| 02:23.09 | prpplague | robbat2 do you have you openocd script posted? |
| 02:23.38 | prpplague | http://dev.gentoo.org/~robbat2/tincan-nail/hammer.ocd.20080427_1840 |
| 02:23.41 | prpplague | that the one? |
| 02:23.57 | robbat2 | http://dev.gentoo.org/~robbat2/tincan-nail/hammer.cfg.20080427_1914 |
| 02:24.03 | robbat2 | http://dev.gentoo.org/~robbat2/tincan-nail/hammer.ocd.20080427_1914 |
| 02:24.27 | prpplague | yea you have the reset command commented out |
| 02:24.48 | prpplague | after the flash just use the reset command to restart the hammer |
| 02:25.01 | robbat2 | if I leave the reset there, it flashes infinitely |
| 02:25.18 | prpplague | flashed infinitely? |
| 02:25.30 | robbat2 | it flashes, then repeats |
| 02:25.35 | robbat2 | without exiting openocd |
| 02:25.53 | prpplague | hmmm something doesn't sound right |
| 02:25.56 | prpplague | looks |
| 02:26.16 | robbat2 | target_script 0 reset hammer.ocd.20080427_1914 |
| 02:26.21 | robbat2 | from my understanding of openocd |
| 02:26.26 | robbat2 | that means, on reset, run the script |
| 02:26.38 | robbat2 | so having the reset inside the script causes it to start again |
| 02:27.50 | prpplague | yep, just noticed that |
| 02:28.06 | prpplague | i don't recall having it set that way before |
| 02:28.14 | robbat2 | i think it's a change in openocd |
| 02:28.19 | prpplague | wonders who at the office changed that example |
| 02:29.42 | prpplague | robbat2 ok, so what you can do is do another cfg file calling out a reset.ocd script |
| 02:29.52 | prpplague | robbat2 i'll have to retest in the morning |
| 02:30.18 | prpplague | robbat2 i'll test and doc that first thing in the morning |
| 02:30.24 | robbat2 | ok |
| 02:31.15 | prpplague | robbat2 wonder why no one has complained about that before |
| 02:31.43 | prpplague | RobotGuy still around? |
| 02:31.43 | RobotGuy | Yes |
| 02:31.49 | robbat2 | prpplague, probably that they used the bundled openocd |
| 02:32.01 | robbat2 | whereas I just grabbed the latest gentoo one, since my host machine is PPC in this case |
| 02:33.23 | prpplague | robbat2 ahh ok |
| 02:33.35 | prpplague | robbat2 i'll grab a svn copy tomorrow and retest |
| 02:34.02 | prpplague | RobotGuy nm, you are using the openocd source i posted for the hammer right? |
| 02:34.23 | RobotGuy | prpplague: Yes, I am. |
| 02:35.20 | prpplague | goes to look for the openocd changelog |
| 02:40.48 | prpplague | robbat2 ahh yea looks like the reset function operation has been changed and added to |
| 02:41.00 | prpplague | robbat2 i'll grab a copy in the morning and post new scripts |
| 02:41.03 | prpplague | robbat2 sorry about that |
| 02:41.07 | robbat2 | not a problem |
| 02:41.17 | robbat2 | that makes life easier for solar while i'm away |
| 02:42.00 | prpplague | ok i'm off to bed, later folks |
| 02:42.09 | prpplague | robbat2 thanks for all the good work |
| 02:42.13 | robbat2 | np |
| 02:42.15 | prpplague | solar you too! |
| 02:42.16 | robbat2 | enjoy your sleep |
| 02:42.24 | solar | bye |
| 03:56.23 | *** join/#edev RobotGuy (n=robotguy@dsl093-038-072.pdx1.dsl.speakeasy.net) |
| 05:43.27 | *** join/#edev DrZimmerman (n=theo@77-56-246-76.dclient.hispeed.ch) |
| 06:00.12 | *** join/#edev harpal (n=harpal@121.246.75.165) |
| 06:10.07 | *** join/#edev drath_ (i=vmaster@p5B07D077.dip.t-dialin.net) |
| 07:03.29 | *** join/#edev Appleman1234 (n=Appleman@131.181.47.49) |
| 07:29.19 | *** join/#edev [AD]Turbo (n=turbo@host24-241-dynamic.22-79-r.retail.telecomitalia.it) |
| 07:30.40 | [AD]Turbo | hola |
| 07:34.21 | *** join/#edev hw_ (n=hw@p578b3905.dip0.t-ipconnect.de) |
| 08:45.24 | *** join/#edev lyakh (n=lyakh@p57BD297D.dip0.t-ipconnect.de) |
| 09:12.52 | *** join/#edev cubesat_webuser1 (n=481593f4@212.85.1.24) |
| 09:13.03 | *** part/#edev cubesat_webuser1 (n=481593f4@212.85.1.24) |
| 10:00.00 | *** join/#edev samuelr (n=samuelr@dsl093-061-094.pit1.dsl.speakeasy.net) |
| 10:41.43 | *** join/#edev samuelr (n=samuelr@dsl093-061-094.pit1.dsl.speakeasy.net) |
| 11:26.58 | *** join/#edev emil|k (i=emil@f054013226.adsl.alicedsl.de) |
| 11:31.30 | *** join/#edev g1powermac_PB (n=g1powerm@unaffiliated/g1powermac) |
| 11:41.12 | *** join/#edev sjhill (n=sjhill@real.realitydiluted.com) |
| 11:41.12 | *** mode/#edev [+o sjhill] by ChanServ |
| 11:58.03 | *** join/#edev GPSFan (n=kenm@12.10.255.246) |
| 11:58.03 | *** mode/#edev [+o GPSFan] by ChanServ |
| 12:45.08 | *** join/#edev roxfan (n=dunno@107.216-64-87.adsl-dyn.isp.belgacom.be) |
| 13:16.15 | *** join/#edev chouimat|work (n=dieu@209.217.106.98) |
| 13:17.12 | chouimat|work | morning |
| 13:18.19 | g1powermac_PB | heya chouimat|work |
| 13:19.52 | *** join/#edev prpplague (n=dave@mail.americanmicrosystems.com) |
| 13:20.05 | g1powermac_PB | morning prpplague |
| 13:20.15 | prpplague | greetings |
| 13:21.32 | g1powermac_PB | prpplague: got my dev machine fully working now, both windows and linux sides |
| 13:21.37 | *** mode/#edev [+o prpplague] by ChanServ |
| 13:21.45 | prpplague | g1powermac_PB: good deal |
| 13:21.58 | prpplague | Lars_G: ping |
| 13:22.24 | g1powermac_PB | prpplague: so now I'm ready to see if there's jtag on the board |
| 13:22.55 | prpplague | which jtag device are you using? |
| 13:23.00 | g1powermac_PB | what do you think of the area right above the chip? any good places I should look for? |
| 13:23.34 | prpplague | yea, i think so, but thats only a guess since i don't have a pcb to look at |
| 13:23.52 | g1powermac_PB | k |
| 13:25.11 | g1powermac_PB | gonna have to pull it apart again and really study the board |
| 13:25.37 | prpplague | which jtag device are you using? |
| 13:26.14 | g1powermac_PB | jtag cable? http://www.digilentinc.com/Products/Catalog.cfm?Nav1=Products&Nav2=Cables&Cat=Cable |
| 13:26.21 | g1powermac_PB | mines the parallel cable one |
| 13:27.12 | prpplague | ahh ok, so its a known working device |
| 13:28.31 | g1powermac_PB | yea, programmed my minila with it |
| 13:30.51 | prpplague | g1powermac_PB: and you've tested it with openocd? |
| 13:30.59 | g1powermac_PB | have not, yet |
| 13:31.15 | g1powermac_PB | how should I test it? hook it up the minila? |
| 13:31.27 | prpplague | no you'd need an arm device |
| 13:31.43 | g1powermac_PB | hmm |
| 13:31.59 | g1powermac_PB | do you know if the gumstix has a jtag port? |
| 13:32.11 | prpplague | g1powermac_PB: its not nessary that you test, but you might end up spending alot of time working on talking to the zipit before you realize that the jtag dongle doesn't work |
| 13:32.14 | prpplague | g1powermac_PB: yea it does |
| 13:32.15 | g1powermac_PB | I got one of them floating around |
| 13:32.38 | g1powermac_PB | I also got that wifi phone, not sure if it has jtag |
| 13:33.27 | g1powermac_PB | yea, that does make sense |
| 13:34.42 | g1powermac_PB | hmm, to do with the gumstix, I think I need their jtag board. . . |
| 13:37.10 | g1powermac_PB | yea, need to get a tweener board to get to the jtag on the gumstix |
| 14:03.44 | g1powermac_PB | prpplague: this pretty interesting: http://www.c3a.de/wiki/index.php/JTAG_Finder |
| 14:04.03 | prpplague | g1powermac_PB: yea i looked at that before |
| 14:04.20 | prpplague | g1powermac_PB: thats usefull if you know which lines are jtag, but not which pinout |
| 14:04.29 | g1powermac_PB | ahh |
| 14:11.46 | g1powermac_PB | bbiab, going to work |
| 14:30.28 | *** join/#edev LyosNorezel (n=Ralph@unaffiliated/lyosnorezel) |
| 14:31.32 | *** join/#edev g1powermac (n=g1powerm@unaffiliated/g1powermac) |
| 14:31.42 | g1powermac | woot, got my minisd to full sd adapter today |
| 14:40.49 | *** join/#edev ce_geek (n=ce_geek@roger-blanchette-4.umelst.maine.edu) |
| 14:43.27 | GPSFan | g1powermac: sothing can stop you now... |
| 14:43.53 | GPSFan | s/sothing/nothing |
| 14:45.02 | g1powermac | is on a roll! ;-) |
| 14:47.08 | SpeedEvil | Bacon? |
| 14:47.56 | SpeedEvil | Mmmm. http://images.google.co.uk/images?hl=en&rlz=&q=bacon+roll&um=1&ie=UTF-8 |
| 14:48.30 | GPSFan | yummm |
| 14:57.04 | *** join/#edev CosmicPenguin (n=nobody@163.181.251.103) |
| 14:57.04 | *** mode/#edev [+o CosmicPenguin] by ChanServ |
| 15:01.56 | g1powermac | hah |
| 15:04.49 | ce_geek | prpplague: Seems like it ought not to be too bad to write a kernel module that could make a... custom "port" out of an arbitrary number of GPIO pins (eight?) and have a function call that would write any 8-bit value to that port. |
| 15:05.15 | prpplague | ce_geek: nope, not a big deal at all |
| 15:05.28 | prpplague | ce_geek: you could use a basic char device easily |
| 15:05.51 | ce_geek | prpplague: the bits in GPCCON would have to get set (or whatever register) and then just write the masked bits to GPCDAT or whatever. |
| 15:05.59 | ce_geek | prpplague; Basic char device? |
| 15:06.21 | prpplague | ce_geek: there are two basic types of dev nodes, block and character |
| 15:06.29 | ce_geek | prpplague; Oh cool |
| 15:06.34 | prpplague | ce_geek: a basic character devices is very easy to create |
| 15:06.57 | ce_geek | prpplague; Do you recommend an example that I could go stare at for a little bit? |
| 15:07.15 | prpplague | ce_geek: the Linux Device Drivers book has alot |
| 15:07.24 | prpplague | ce_geek: there are also many tutorials on the net |
| 15:08.03 | prpplague | ce_geek: http://tldp.org/LDP/lkmpg/2.6/html/ |
| 15:08.53 | chouimat|work | http://lwn.net/Kernel/LDD3/ |
| 15:09.24 | prpplague | is really upset with the changes to openocd |
| 15:09.29 | ce_geek | prpplague: K. Yes, I need to get that book and read it cover to cover. |
| 15:09.49 | prpplague | ce_geek: that url i gave you has a good tutorial and some examples |
| 15:09.58 | ce_geek | prpplague: Cool, thx |
| 15:10.41 | ce_geek | prpplague: Look at that. "Character device drivers." |
| 15:10.55 | chouimat|work | prpplague: bad changes? |
| 15:11.11 | ce_geek | I noticed that the colorful chart on the wiki with the pin names doesn't match the names silkscreened on the hammer carrier board. |
| 15:11.42 | ce_geek | The LCD pins are labeled otherwise, for example. |
| 15:11.48 | prpplague | looks |
| 15:13.00 | prpplague | ce_geek: what items don't match? |
| 15:13.39 | ce_geek | prpplague: Pins 21 through 28 are labeled as either LCD or GPC pins, but they're not labeled that way on the single row next to the proto area on the hammer carrier. |
| 15:14.09 | ce_geek | prpplague: Looking to see if there are 8 consecutive GPIO pins on that single row header just to keep it simple. |
| 15:14.29 | prpplague | ce_geek: correct, since we only have enough room for one function label, and since we have specific uses for each pin on the carrier board, we label them with the function they are dedicated for |
| 15:15.02 | prpplague | ce_geek: the lcd section is the only 8-bit in a row iirc |
| 15:15.17 | ce_geek | prpplague; Yeah, that seems to make sense. If anything I'd just say that it would be helpful for the table to reflect the names printed on the hammer carrier is all. |
| 15:15.35 | prpplague | ce_geek: hehe, it does, imho, hehe |
| 15:15.45 | ce_geek | prpplague; They are numbered |
| 15:16.07 | prpplague | ce_geek: for instance pin 28 has both the lcd and gpio functions listed in the table |
| 15:17.14 | ce_geek | prpplague; Yeah, but on the hammer carrier pin 28 is labeled "EINT10" |
| 15:18.03 | ce_geek | prpplague; (I'm whining a little here... it's fine as long as I look at the pin numbers and forget what the labels are) |
| 15:18.20 | prpplague | ce_geek: are you sure about that? |
| 15:18.37 | prpplague | ce_geek: my pin 28 is labled GPC15 |
| 15:19.23 | ce_geek | prpplague: Heh, looks like I'm upside down or twisted or something. Sec. |
| 15:19.41 | prpplague | ce_geek: Pin1 on the carrier board is marked as RXD0 |
| 15:20.43 | ce_geek | prpplague: I see now. The picture on the wiki is clear, I just overlooked the position of the LED. Everything makes great sense now. After I said something about EINT10 I noticed something wasn't right because that's like pin 8 in the table. |
| 15:20.47 | ce_geek | prpplague: My bad, sorry. |
| 15:21.07 | ce_geek | How does that whaleslap thing work again? |
| 15:21.14 | prpplague | ce_geek: actually i see the way we have the labels makes it hard to understand |
| 15:21.20 | prpplague | ~whaleslap ce_geek |
| 15:21.21 | purl | ACTION beats ce_geek upside and over the head with a freakishly huge killer whale named Hugh |
| 15:21.49 | ce_geek | prpplague; It's fine if the person between the keyboard and the chair pays attention to detail.... which is sort of important with this kind of work :-/ |
| 15:22.25 | ce_geek | tries to whaleslap himself |
| 15:22.29 | ce_geek | ~whaleslap ce_geek |
| 15:22.30 | purl | ACTION beats ce_geek upside and over the head with a freakishly huge killer whale named Hugh |
| 15:27.29 | *** join/#edev sunrunner20 (n=jamesdoe@pool-70-104-23-68.dllstx.fios.verizon.net) |
| 15:34.18 | *** join/#edev RobotGuy (n=robotguy@dsl093-038-072.pdx1.dsl.speakeasy.net) |
| 15:54.43 | *** join/#edev roxfan2 (n=dunno@107.216-64-87.adsl-dyn.isp.belgacom.be) |
| 16:00.52 | *** join/#edev landley (n=landley@cpe-70-116-30-223.austin.res.rr.com) |
| 16:00.52 | *** mode/#edev [+o landley] by ChanServ |
| 16:29.01 | *** join/#edev roxfan (n=dunno@107.216-64-87.adsl-dyn.isp.belgacom.be) |
| 16:33.12 | *** join/#edev g1powermac (n=g1powerm@unaffiliated/g1powermac) |
| 16:35.52 | *** join/#edev bbradley (n=bbradley@92-235-133-146.cable.ubr17.azte.blueyonder.co.uk) |
| 16:38.12 | *** join/#edev g1powermac_ (n=g1powerm@rrcs-64-183-163-253.west.biz.rr.com) |
| 16:38.38 | *** join/#edev lyakh (n=lyakh@p57BD297D.dip0.t-ipconnect.de) |
| 16:45.38 | g1powermac_ | bwahaha |
| 16:45.57 | g1powermac | running the shell on my zipit :-) |
| 16:48.41 | SpeedEvil | :) |
| 16:48.58 | SpeedEvil | waits for g1zipit |
| 16:49.22 | g1powermac | hrm, it seems to be stuck at "Daemonizing. . ." |
| 16:49.34 | g1powermac | GPSFan, is that normal for the first run? |
| 16:49.59 | RobotGuy | SpeedEvil: Was it you who gave me the part number for the I2C level shifters? |
| 16:50.24 | RobotGuy | g1powermac: Congrats! |
| 16:50.53 | g1powermac | now if it only gets past this part. . . |
| 16:51.00 | g1powermac | so i can actually gain access to the shell |
| 16:53.09 | g1powermac | meh, its prolly the network stuff, been having trouble with it here |
| 16:54.31 | SpeedEvil | robbat2|na: perhaps - I don't remember |
| 16:54.35 | SpeedEvil | rob0: |
| 16:57.00 | g1powermac | woot, ok, I just commented out the wireless/ssh stuff for now, I now have the shell :-) |
| 16:57.59 | SpeedEvil | Hold it up to your ear. |
| 16:58.03 | SpeedEvil | Can you hear the sea? |
| 16:59.25 | GPSFan | g1powermac: the boot from sd card stuff always requires some tweaking. especially the wireless stuff. |
| 16:59.44 | GPSFan | g1powermac: I have a picture of the wireless chip if you would like it. |
| 16:59.52 | g1powermac | oh, awesome |
| 17:00.14 | g1powermac | one question, how do you get the dmesg on this sucker? seems there's no dmesg command |
| 17:00.39 | SpeedEvil | cat /proc/dmesg |
| 17:00.48 | g1powermac | k |
| 17:01.27 | g1powermac | no such file |
| 17:01.33 | SpeedEvil | err |
| 17:01.35 | SpeedEvil | kmsg |
| 17:01.41 | SpeedEvil | I think |
| 17:02.18 | GPSFan | SpeedEvil: no /proc/dmesg on my boxes. |
| 17:02.32 | SpeedEvil | <PROTECTED> |
| 17:03.09 | g1powermac | hmm, how do you do a ctrl-c on this sucker? |
| 17:08.46 | g1powermac | woot, ok, got the dmesg off the device |
| 17:08.47 | RobotGuy | I really need the part number of those level translator chips in PDIP. I want to order some in a few days. This is holding up some of my development with Hammer. |
| 17:11.24 | GPSFan | SpeedEvil: interesting, I never saw that before. that's not how I did it. I used a version of busybox statically compiled with dmesg enabled. IIRC that probably just reads kmsg. |
| 17:11.40 | RobotGuy | sighs |
| 17:11.57 | RobotGuy | I can't seem to find it on DigiKey. |
| 17:14.01 | *** join/#edev ce_geek (n=ce_geek@roger-blanchette-4.umelst.maine.edu) |
| 17:14.02 | RobotGuy | SpeedEvil: Do you know what I am talking about? |
| 17:14.18 | ce_geek | wonders why openocd seems to reliably crash his machine |
| 17:14.31 | RobotGuy | ce_geek: Are you running it as root? |
| 17:14.35 | ce_geek | ya |
| 17:14.43 | prpplague | RobotGuy: what are you trying to level shift? |
| 17:14.52 | RobotGuy | ce_geek: Built from sources on the Wiki? |
| 17:15.07 | g1powermac | GPSFan, do you have a place to host that pic or are you gonna try to send it to me? |
| 17:15.09 | ce_geek | RobotGuy: Dunno used the one on the CD |
| 17:15.14 | RobotGuy | prpplague: All levels, I2C, regular TTL. |
| 17:15.21 | RobotGuy | ce_geek: Use the one on the Wiki. |
| 17:15.25 | ce_geek | RobotGuy: It's intermittent |
| 17:15.36 | prpplague | RobotGuy: whats wrong with standard 244 and 245's ? |
| 17:15.44 | GPSFan | g1powermac: sending you mail has failed twice can I post it to your sf forum? |
| 17:15.45 | ce_geek | RobotGuy: Not quite reliable. |
| 17:16.05 | g1powermac | sure, do you want me to make you a dev at sf? |
| 17:16.13 | prpplague | ce_geek: might want to rebuild it from sources |
| 17:16.20 | prpplague | ce_geek: what distro? |
| 17:16.24 | ce_geek | prpplague: Arch Linux :-D |
| 17:16.26 | RobotGuy | prpplague: I need bi-dir shifting for I2C and SPI. I just lost the part number. It's a DIP form. |
| 17:16.39 | GPSFan | g1powermac: how about to just the regular discussion forum, can it take pictures? |
| 17:16.41 | prpplague | RobotGuy: spi is not bi-directional |
| 17:16.50 | g1powermac | hmm, not sure |
| 17:17.10 | ce_geek | prpplague: It probably had something to do with me disconnecting and reconnecting the USB cable from the flyswatter a bunch of times over the last week. First time openocd locked up really, it's just that between minicom and openocd I usuallyl end up with something funny. |
| 17:17.21 | RobotGuy | prpplague: How is SPI not bi-directional? |
| 17:17.36 | g1powermac | GPSFan, I could just make you a dev and then you could load it directly to the sf zipit site |
| 17:18.00 | prpplague | RobotGuy: miso is only one direction, mosi is only one direction, chip select is only one direction, clock is only one direction |
| 17:18.04 | GPSFan | g1powermac: well it won't let me log in anyway. I registered ok. |
| 17:18.17 | g1powermac | hrmm |
| 17:18.25 | g1powermac | did you confirm your email and such? |
| 17:18.29 | GPSFan | yes |
| 17:18.37 | g1powermac | thats odd |
| 17:19.24 | g1powermac | well, lemme know when you're able to login, I'll set you up as a dev |
| 17:20.05 | GPSFan | I've tried several times, I get some message about META somethingorother then it refreshes back to the login screen. |
| 17:20.20 | prpplague | RobotGuy: http://www.standardics.nxp.com/support/documents/i2c/pdf/an97055.pdf |
| 17:20.44 | g1powermac | huh |
| 17:24.44 | GPSFan | g1powermac: http://www.paonia.com/~kenm/Pix/images/Zipit/Z2/AGF00009.JPG |
| 17:25.36 | g1powermac | that'll do, thanks :-) |
| 17:25.45 | prpplague | hmm lots of test points |
| 17:25.56 | g1powermac | yea |
| 17:26.04 | RobotGuy | There is a DIP part that I need the part number for. I can't remember the part number, which I thought I had written down but apparently did not. |
| 17:26.21 | RobotGuy | It's an 8 pin DIP |
| 17:26.30 | prpplague | RobotGuy: and what is the part? |
| 17:26.51 | RobotGuy | prpplague: It's a level shifter, bi-directional, if I remember correctly. |
| 17:27.20 | SpeedEvil | ah |
| 17:27.23 | prpplague | you don't who made it? |
| 17:27.51 | prpplague | seems kind of a waste when you can do it with two transistors for i2c |
| 17:28.01 | SpeedEvil | or two resistors |
| 17:28.07 | prpplague | using a 244 is fine for spi |
| 17:28.09 | SpeedEvil | (sometimes) |
| 17:28.53 | RobotGuy | prpplague: Maybe it's a waste to you, but not for me. |
| 17:29.10 | RobotGuy | The part number started with P. |
| 17:29.11 | SpeedEvil | Pullup to the lowest VCC |
| 17:29.32 | prpplague | RobotGuy: its waste in the aspect that you are doing alot more work |
| 17:29.32 | SpeedEvil | works fine if the voltages aren't too different |
| 17:30.37 | g1powermac | bwahahaha |
| 17:30.44 | g1powermac | my test program works on my zipit |
| 17:30.58 | g1powermac | built with buildroot toolchain |
| 17:31.56 | GPSFan | g1powermac: so does mine: http://www.paonia.com/~kenm/Pix/?display=images/Zipit/Z2/PICT1097a.JPG&width=800&height=600 |
| 17:31.57 | prpplague | robbat2|na: http://focus.ti.com/docs/prod/folders/print/pca9306.html |
| 17:31.59 | prpplague | oops |
| 17:32.01 | prpplague | RobotGuy: http://focus.ti.com/docs/prod/folders/print/pca9306.html |
| 17:32.02 | RobotGuy | prpplague: How is it more work to connect up a DIP than two transistors? |
| 17:32.39 | SpeedEvil | sourcing the DIP |
| 17:32.44 | g1powermac | GPSFan, nice |
| 17:33.17 | g1powermac | GPSFan, of course I now have to figure out why my wireless ain't working, but then again, it could just be the wireless router I tried to resetup here |
| 17:33.18 | GPSFan | g1powermac: that's my re-customized version of wavemon. no fancy gui, just ncurses. |
| 17:33.25 | g1powermac | prolly best now if I did everything at home |
| 17:33.40 | prpplague | RobotGuy: i don;t see any of the pca series available in dip |
| 17:33.56 | g1powermac | only thing about doing it at home is that I need to setup wep on my zipit, not sure how to do that |
| 17:34.05 | g1powermac | don't have wpa running at home |
| 17:34.06 | SpeedEvil | g1powermac: iwconfig ... |
| 17:34.14 | SpeedEvil | g1powermac: wep is easy |
| 17:34.32 | RobotGuy | Maybe it was a Philips/NXP part. |
| 17:34.33 | g1powermac | SpeedEvil, k, its just been forever since I've done it |
| 17:34.33 | SpeedEvil | it's entirely done through iwconfig |
| 17:34.40 | prpplague | RobotGuy: http://focus.ti.com/docs/prod/folders/print/p82b96.html |
| 17:36.01 | prpplague | http://www.mouser.com/Search/ProductDetail.aspx?qs=aEuGZpxfbxV0WPFsO7nJmg%3d%3d |
| 17:37.03 | prpplague | http://search.digikey.com/scripts/DkSearch/dksus.dll?Detail?name=296-20943-5-ND |
| 17:37.07 | RobotGuy | This is it: http://search.digikey.com/scripts/DkSearch/dksus.dll?Detail?name=568-1008-5-ND |
| 17:37.44 | *** join/#edev emil|k_ (i=emil@f054012173.adsl.alicedsl.de) |
| 17:37.50 | prpplague | yea thats the cross part for the TI part |
| 17:38.26 | RobotGuy | prpplague: Is either better than the other? |
| 17:38.38 | prpplague | probably not, just price |
| 17:38.47 | RobotGuy | prpplague: Thanks for helping me re-find that. :) |
| 17:39.08 | prpplague | i've never used either, since we just do it with two $0.05 transistors |
| 17:40.51 | RobotGuy | I can get samples from TI. :) |
| 17:41.36 | RobotGuy | prpplague: That's your price, but what about for me who would just be ordering a few? |
| 17:43.28 | prpplague | RobotGuy: transistors are cheap |
| 17:46.25 | RobotGuy | I don't want to deal with transistors. |
| 17:47.32 | SpeedEvil | Were you attacked by a mob of them? |
| 17:48.31 | ce_geek | considers putting his root filesystem under revision control |
| 17:49.33 | prpplague | RobotGuy: iirc we use the 2n7002 |
| 17:49.46 | prpplague | RobotGuy: but you could use the thru hole version 2n7000 |
| 17:50.36 | prpplague | RobotGuy: http://www.mouser.com/Search/ProductDetail.aspx?qs=FOlmdCx%252bAA1YVT2qEPBJ%2fw%3d%3d |
| 17:50.46 | prpplague | RobotGuy: thats a cheapy at $0.11 in qty 1 |
| 17:52.20 | *** join/#edev roxfan2 (n=dunno@107.216-64-87.adsl-dyn.isp.belgacom.be) |
| 17:54.00 | RobotGuy | prpplague: $0.26 each at DigiKey |
| 17:54.36 | RobotGuy | Oh wait, there is one for $0.11 |
| 17:58.42 | g1powermac | should try to load the aufs module on the zipit. . . |
| 17:59.12 | RobotGuy | waits for TI to send him his password. |
| 18:04.47 | g1powermac | dang, it doesn't load |
| 18:05.17 | g1powermac | hmm, "Unknown symbol in module (-1): No such file or directory" |
| 18:05.36 | g1powermac | why do I have a feeling thats from not compiling it statically |
| 18:08.41 | prpplague | robbat2|na: |
| 18:08.48 | prpplague | robbat2|na: give me a shout later |
| 18:08.56 | RobotGuy | I'm figuring out what I need to order from DigiKey. |
| 18:09.17 | RobotGuy | prpplague: Are there any other items you can think of I might need for interfacing to Hammer? |
| 18:09.41 | g1powermac | ponders how to compile a kernel module statically |
| 18:10.33 | prpplague | RobotGuy: no clue what you want to do |
| 18:10.44 | prpplague | RobotGuy: the 2n7000 is a good one to have plenty of |
| 18:10.51 | prpplague | RobotGuy: you can do alot with that one |
| 18:10.51 | RobotGuy | prpplague: Is it possible, using those Mosfets, to make a bi-dir interface for GPIOs also? |
| 18:11.08 | RobotGuy | prpplague: OK, I will order a lot of those. :) |
| 18:11.24 | prpplague | RobotGuy: thats kind of an oxymoron, no need for bi-directional with gpios |
| 18:11.52 | prpplague | RobotGuy: GPIO's are either output or input |
| 18:12.50 | RobotGuy | prpplague: There is a need when interfacing to some 5V sensors such as http://www.trossenrobotics.com/parallax-ping-ultrasonic-range-sensor.aspx |
| 18:13.00 | prpplague | looks |
| 18:13.06 | ce_geek | prpplague: I added digitemp to my rootfs, having funky problems... |
| 18:13.10 | ce_geek | prpplague: http://www.pastebin.ca/1000667 |
| 18:13.47 | ce_geek | prpplague: I usually don't get those user.debug kernel messages, wondering if it would make sense that if my rootfs was too big if that could cause those to show up. |
| 18:14.27 | *** part/#edev LyosNorezel (n=Ralph@unaffiliated/lyosnorezel) |
| 18:14.40 | prpplague | ce_geek: no sounds like you've got some debugging turned on somewhere |
| 18:14.53 | prpplague | ce_geek: one sec, looking at something for RobotGuy |
| 18:15.45 | ce_geek | prpplague: Sure thing |
| 18:17.03 | RobotGuy | prpplague: For the PING, you pulse the pin to start a detection sequence, then read the pulse width coming back from it for the value. |
| 18:19.18 | prpplague | RobotGuy: yea, technically its not bi directional, since its configured as an output first, then as a input |
| 18:19.50 | RobotGuy | prpplague: True, but I still need input and output capability on the GPIOs. |
| 18:20.13 | g1powermac | GPSFan, alright, I put your image up on the sf site and also put up the dmesg |
| 18:21.03 | prpplague | RobotGuy: using a 245 would be easier |
| 18:21.15 | prpplague | RobotGuy: does require one more gpio, but its much easier |
| 18:21.54 | RobotGuy | prpplague: Would the extra GPIO be for direction control? |
| 18:21.54 | prpplague | RobotGuy: yea |
| 18:22.38 | RobotGuy | prpplague: Hmmm, I could control direction for several GPIOs using one or more of those. |
| 18:23.05 | prpplague | RobotGuy: yea or get another of the series with more direction controls |
| 18:23.42 | prpplague | RobotGuy: i.e. 4 gpios on 1 direction line and 4 gpios on 1 direction line for a total of 8 gpios |
| 18:23.51 | prpplague | RobotGuy: there are a number of combinations |
| 18:24.58 | RobotGuy | prpplague: Hmmm, if I used the dual dir setup, one chip would control 4 GPIOs, right? |
| 18:25.20 | prpplague | RobotGuy: sorry i did not understand your statement |
| 18:25.51 | RobotGuy | prpplague: If I want to have the dual direction for GPIOs, would one chip then control 4 pins instead a total of 8? |
| 18:26.27 | g1powermac | wonders what symbols this kernel module needs |
| 18:26.33 | g1powermac | that it can't find |
| 18:27.01 | prpplague | RobotGuy: the 245 has two groups of 4, each group of 4 is controlled with a direction enable line, if you want to control them seperately, use a seperate gpio for each direction control line, if you want to controll all eight together, just tie the two control lines together |
| 18:27.39 | RobotGuy | prpplague: Ah, so the dir bit controls whether it would be input or output, with respect to Hammer. |
| 18:27.47 | prpplague | correct |
| 18:28.16 | prpplague | ce_geek: can you pastebin you inittab |
| 18:28.19 | RobotGuy | prpplague: This seems like it would work well for ultrasonic sensors like the PING or SRF-05. |
| 18:28.29 | ce_geek | prpplague: Sure, uno momento |
| 18:28.44 | prpplague | RobotGuy: yea probably, although i have no experience with those devices |
| 18:29.08 | RobotGuy | The 245's do the level shifting also, right? |
| 18:29.22 | prpplague | yes they can |
| 18:29.34 | RobotGuy | prpplague: That's what I need. |
| 18:29.47 | ce_geek | http://www.pastebin.ca/1000691 |
| 18:29.55 | ce_geek | prpplague: oops http://www.pastebin.ca/1000691 |
| 18:29.58 | RobotGuy | Level shifting and direction control. |
| 18:30.48 | RobotGuy | I see how I could write code to go into a device driver for the PING. |
| 18:31.01 | prpplague | ce_geek: looks fine, i suspect you have low level debugging turned on in the kernel |
| 18:31.29 | ce_geek | prpplague: That must have happened in the patch for the source or something, because I've used the stuff from the wiki. |
| 18:31.43 | ce_geek | I just noticed that I don't get those messages when I'm SSH'd in, only when using minicom |
| 18:31.52 | RobotGuy | looks to see if there is already a device driver for GPIOs. |
| 18:31.53 | prpplague | ce_geek: yea thats the console |
| 18:32.06 | prpplague | ce_geek: if you check your kernel config under "kernel hacking" |
| 18:32.10 | ce_geek | prpplague; Easy to find where low level debuggign might be turned on? |
| 18:32.12 | ce_geek | Oh |
| 18:32.14 | prpplague | ce_geek: there is an option at the very bottom |
| 18:32.18 | ce_geek | ok |
| 18:32.23 | prpplague | ce_geek: that does lowlevel debugging via the s3c24xx uart |
| 18:32.48 | ce_geek | prpplague, CONFIG_DEBUG_KERNEL=y |
| 18:32.54 | ce_geek | heh |
| 18:34.12 | ce_geek | prpplague: CONFIG_DEBUG_S3C2410_PORT=y |
| 18:34.12 | prpplague | ce_geek: yea you can leave that on, just turn of the s3c2410 uart debugging |
| 18:34.12 | ce_geek | prpplague: Cool |
| 18:34.12 | ce_geek | prpplague: learn something new every day around this place :) |
| 18:34.13 | prpplague | hehe |
| 18:34.14 | prpplague | ce_geek: i say the same thing when i go to the nudie bar at 5:30 |
| 18:35.11 | ce_geek | prpplague: hahahaha |
| 18:35.15 | *** join/#edev MothaWork (n=monmotha@mail.forwardengineering.com) |
| 18:37.56 | *** join/#edev MothaWork (n=monmotha@mail.forwardengineering.com) |
| 18:38.01 | g1powermac | GPSFan, how did you compile those other kernel modules thats in the shell package? |
| 18:58.20 | RobotGuy | Has anyone used the TSOPs on breadboards? Are they easy to use? |
| 18:58.50 | g1powermac | with the proper adapter, it shouldn't be too hard |
| 18:59.22 | prpplague | RobotGuy: http://www.beldynsys.com/p513.htm |
| 18:59.34 | g1powermac | prpplague, I'm studying the large pic of the zipit, especially the area that would most likely be where the jtag should be |
| 19:00.02 | g1powermac | there's alot of 'corrections' done to that area |
| 19:00.30 | g1powermac | and four resistors |
| 19:02.27 | g1powermac | makes me wonder if they 'disabled' the jtag by doing those corrections |
| 19:05.48 | g1powermac | the solder bridge correction is the most intriguing at TP101? |
| 19:06.14 | g1powermac | that is right at the right row for the jtag |
| 19:12.13 | RobotGuy | prpplague: That's pretty cool. |
| 19:13.17 | g1powermac | RobotGuy, yea, I've used similar ones from http://www.schmartboard.com/ |
| 19:13.57 | GPSFan | g1powermac: apply the zipit keyboard patch from ethicalclone, enable the frame buffer and fb console stuff, make modules. Start my .config, it has most of that stuff enabled. http://www.pastebin.ca/1000750 |
| 19:14.05 | RobotGuy | g1powermac: I have four SchmartBoards I was going to use with PIC18F87J50 but apparently they are not the right ones for that. |
| 19:14.33 | g1powermac | GPSFan, ok |
| 19:14.45 | RobotGuy | There are some TSOP28 parts I would like to be able to use. |
| 19:15.48 | g1powermac | GPSFan, did you build the kernel within buildroot or out? |
| 19:16.07 | GPSFan | out |
| 19:16.10 | g1powermac | k |
| 19:16.28 | RobotGuy | I would like to be able to use http://focus.ti.com/docs/prod/folders/print/tlc5946.html |
| 19:16.48 | RobotGuy | Blinky lights and robots go together. :) |
| 19:16.58 | g1powermac | hehe |
| 19:18.20 | RobotGuy | I have some RGB leds that would be cool to pair with a few of those chips. |
| 19:18.32 | GPSFan | g1powermac: ah, the problem with sf login is that I hadn't received the e-mail and confirmed the new account yet, I can login now. |
| 19:19.14 | g1powermac | hmm, schmartboards finally came out with the large qfp boards, might be able to use the multitude of samples I got from freescale one day now |
| 19:19.29 | Lars_G | hey all. |
| 19:19.35 | g1powermac | heya Lars_G |
| 19:19.42 | Lars_G | prpplague: if you've answered my poll I can't see it until i get to job #2, sorry |
| 19:20.05 | g1powermac | GPSFan, ahh, whats your unix sf name? |
| 19:20.09 | g1powermac | I'll add you as a dev |
| 19:20.26 | GPSFan | uh, gpsfan |
| 19:20.46 | RobotGuy | g1powermac: I never could quite get the hang of using SF source control. |
| 19:20.55 | prpplague | Lars_G: we are indeed geared towards large production runs |
| 19:21.05 | g1powermac | RobotGuy, cvs? yea, its a pita |
| 19:21.06 | prpplague | Lars_G: we can do up to 10k per build right now |
| 19:21.32 | RobotGuy | Didn't SF convert to SVN? |
| 19:21.44 | g1powermac | RobotGuy, they have both |
| 19:21.53 | RobotGuy | Ahhhh |
| 19:22.18 | robbat2 | prpplague, yo |
| 19:22.47 | prpplague | robbat2: hey, yea, there were alot of changes as to how the reset calls work with openocd |
| 19:23.07 | prpplague | robbat2: a temp work around is to change the reset command in the script from just "reset" to "reset run" |
| 19:23.15 | g1powermac | GPSFan, alright, you're setup |
| 19:23.27 | robbat2 | prpplague, ok, will do later |
| 19:23.29 | prpplague | robbat2: you'll get a couple warnings during the execution of the script but it will work with |
| 19:23.29 | RobotGuy | prpplague: Is there any particular 'flavor' of the 245 you would recommend? |
| 19:23.34 | prpplague | robbat2: no problems |
| 19:23.38 | g1powermac | GPSFan, you have access to the web stuff, cvs, and other stuff |
| 19:23.53 | prpplague | RobotGuy: ahc is my choice, but MonMotha might be able to direct you better |
| 19:23.57 | GPSFan | g1powermac: thanks. I'm in... |
| 19:24.10 | RobotGuy | MonMotha: ping |
| 19:24.53 | g1powermac | GPSFan, feel free to edit the website and such |
| 19:26.26 | prpplague | RobotGuy: AC or AHC should be fine for what you are doing |
| 19:26.56 | RobotGuy | prpplague: OK, good. |
| 19:28.38 | RobotGuy | is chatting with flavioribeiro |
| 19:29.07 | RobotGuy | speaks Robotics. :) |
| 19:33.09 | g1powermac | heh |
| 19:38.29 | *** join/#edev ds2 (i=noinf@netblock-66-245-251-24.dslextreme.com) |
| 19:40.23 | RobotGuy | prpplague: Did you check with shipping today? |
| 19:40.35 | prpplague | not as of yet |
| 19:41.42 | g1powermac | GPSFan, interesting, reading the dmesg, I'm a bit confused with its output of the size of the onboard flash |
| 19:42.01 | g1powermac | it's saying it's 16MiB. . . |
| 19:44.09 | *** join/#edev flavioribeiro (n=flavio@189.43.63.147) |
| 19:44.22 | prpplague | ce_geek: ping |
| 19:44.37 | g1powermac | hrmm, no 8MiB. . . |
| 19:45.07 | prpplague | ce_geek: flavioribeiro is looking at using the hammer possible for some education stuff, might /msg him when you get a chance |
| 19:45.32 | RobotGuy | Heh heh, I found that nice I2C 3-axis accelerometer. |
| 19:46.22 | prpplague | flavioribeiro: there are a number of hammer/nail board users here, so if you have questions, someone can generally answer them or point you in the right direction |
| 19:46.38 | flavioribeiro | sure man, thanks :D |
| 19:48.49 | flavioribeiro | hey guys let me tell what im im planning to do.. first sorry me for the bad english, im brazilian |
| 19:49.43 | flavioribeiro | i need a platform that can runs linux and have some wireless communication way (bluetooth in this case) |
| 19:50.24 | flavioribeiro | and some pins to be used as an interface to people just plug some stuff like motors and leds |
| 19:51.21 | flavioribeiro | my point is create an education platform where guys will use the target to plug their stuff in the pins like red-led in pin1 and green-led in pin2 |
| 19:51.50 | flavioribeiro | and in server (a conventional computer) their will just open a terminal and put 'pin1.start(2)' |
| 19:52.07 | flavioribeiro | and the red one will be on about 2 seconds |
| 19:52.31 | Lars_G | prpplague: Good thanks a lot. |
| 19:52.49 | prpplague | Lars_G: got a project in mind? |
| 19:52.55 | Lars_G | prpplague: just to make stuff clear, the usb slave mode is a device mode? so the hammer looks like a device to a host? |
| 19:52.59 | Lars_G | prpplague: Yes, my boss does. |
| 19:53.15 | Lars_G | prpplague: Commissioned me to find viable sbcs to develope it. |
| 19:53.26 | prpplague | Lars_G: yes the second usb port can be used as a device or host, the first port is host only |
| 19:53.56 | Lars_G | is the use of the second por in device mode possible in parallel to using the first port in host mode? |
| 19:54.18 | Lars_G | And, I guess the device ID of the second port is configurable in the kernel/bootloader by code? |
| 19:54.27 | flavioribeiro | i dont know if you understood, but im wondering if some kid in their school plug 4 wheels in the pins and put it on, he will made a mobile car |
| 19:54.28 | prpplague | Lars_G: yes, you can use it as 1 host and 1 slave, or 2 hosts |
| 19:54.48 | Lars_G | Good, I think the hammer fits the profile very nicely :) |
| 19:54.48 | prpplague | Lars_G: correct, the ID is assigned by the gadget driver |
| 19:55.02 | Lars_G | I'll check a few tech details, and pass the info onto my boss. thanks |
| 19:55.09 | Lars_G | prpplague: Btw long time no see, has life been well? |
| 19:55.14 | prpplague | flavioribeiro: the hammer can do that with no problems |
| 19:55.25 | prpplague | Lars_G: reasonably well |
| 19:55.26 | flavioribeiro | good |
| 19:55.53 | *** join/#edev likewise (n=likewise@82-171-51-231.ip.telfort.nl) |
| 19:55.54 | prpplague | flavioribeiro: http://www.elinux.org/Hammer_Board_Example_Kernel_Drivers |
| 19:56.02 | prpplague | flavioribeiro: there are some example drivers there |
| 19:56.41 | flavioribeiro | ok prpplague, and how can i put the 'pins interface' in the board? i will need to write any device driver to make the pins on? |
| 19:56.42 | prpplague | goes to pickup kids from school |
| 19:56.44 | prpplague | bbiab |
| 19:56.51 | Lars_G | prpplague: do you know of anybody who has "bit banged" storage to an SD card using the GPIOs? |
| 19:57.07 | Lars_G | prpplague: have a safe trip |
| 19:57.07 | prpplague | be back in 20 minutes guys |
| 19:57.33 | Lars_G | don't worry the details I don't get from you, i can get from the source, howtos and example code in the site |
| 19:58.51 | ds2 | there is a SPI SD driver in the kernel |
| 20:02.18 | *** join/#edev monmotha_ (n=monmotha@mail.forwardengineering.com) |
| 20:03.27 | Lars_G | nice, very nice. |
| 20:06.43 | RobotGuy | I just don't get the DigiKey search engine. |
| 20:08.24 | SpeedEvil | what you looking for |
| 20:10.17 | *** join/#edev magnetron (n=magnetro@unaffiliated/magnetron) |
| 20:11.14 | Lars_G | for me it's always been better to use the online catalog rather than the search engine |
| 20:11.24 | Lars_G | with mouser too, in their case I always get too many results |
| 20:11.33 | SpeedEvil | Naah. Digikey is easy. |
| 20:11.54 | SpeedEvil | All you do is to scrape their entire catalog into a tab seperated databse, and parse it there. |
| 20:13.24 | RobotGuy | How do I do that? |
| 20:13.41 | SpeedEvil | you don't practically, that is a silly way. |
| 20:13.47 | SpeedEvil | what are you trying to do ? |
| 20:13.52 | SpeedEvil | search for that is |
| 20:14.05 | RobotGuy | All I want is some 14 pin and 20 pin female ribbon cable connectors |
| 20:14.25 | SpeedEvil | idc? |
| 20:14.31 | RobotGuy | For JTAG stuff |
| 20:14.40 | SpeedEvil | you mean 0.1" 7*2 and 10*2? |
| 20:14.47 | RobotGuy | I think I want IDC type connectors. 0.1" spacing. |
| 20:15.01 | RobotGuy | Yes, 2x7 and 2x10 |
| 20:16.15 | RobotGuy | I |
| 20:16.36 | RobotGuy | I'll probably make the cables with round 20 conductor. |
| 20:18.25 | RobotGuy | I'm not sure which is best. I usually avoid making cables if I can get something off the shelf. I don't like making cables. |
| 20:19.37 | GPSFan | g1powermac: the 3 flash partitions total up to about 8M so I would say that 8M is correct. |
| 20:22.15 | g1powermac | hmm, k |
| 20:22.28 | g1powermac | more documentation: http://zipit2system.sourceforge.net/board.html |
| 20:22.36 | g1powermac | did some research on some of the chips |
| 20:25.30 | g1powermac | also found this: http://www.seattlewireless.net/Zipit2Messenger |
| 20:25.35 | g1powermac | not sure who's page that is |
| 20:26.03 | g1powermac | which reminds me |
| 20:26.40 | g1powermac | what are the markings of that other largish chip under the wifi shield? |
| 20:27.08 | GPSFan | g1powermac: I did some probing of those test points near the wifi chip. none connect to ball A9 which is RXD for the STUART. it is set up in the blob source code as an an alt function 2 of gpio46 and is still setup that way after bootup to the sd card userland. |
| 20:28.34 | GPSFan | g1powermac: gee the seattle wireless folks have a Sept rev cpu board, no yellow wires!! |
| 20:28.45 | *** join/#edev jarv (n=jarv@96.245.16.205) |
| 20:31.07 | SpeedEvil | RobotGuy: http://search.digikey.com/scripts/DkSearch/dksus.dll?Detail?name=CSC10G-ND |
| 20:31.47 | SpeedEvil | go to the homepage |
| 20:32.06 | SpeedEvil | hit 'go' in the searchbox with no text |
| 20:32.33 | SpeedEvil | go to connectors - rectangular - free hanging, panel mount |
| 20:32.43 | RobotGuy | SpeedEvil: That's a 2x5. I need 2x7 and 2x10 for JTAG. |
| 20:32.50 | RobotGuy | I will try finding that |
| 20:33.31 | SpeedEvil | number of contacts 10, rows 2, pitch .1" |
| 20:33.53 | SpeedEvil | and receptacle-socket |
| 20:34.02 | SpeedEvil | in the parametric selector |
| 20:34.46 | SpeedEvil | that should get you lots |
| 20:38.02 | RobotGuy | I doing something wrong. Not getting any results |
| 20:41.28 | SpeedEvil | check how many items you have at each selection |
| 20:41.34 | prpplague | returns |
| 20:41.34 | SpeedEvil | do one or two at once |
| 20:41.44 | prpplague | doh, just missed him |
| 20:41.53 | SpeedEvil | to avoid creating impossible combinations |
| 20:42.31 | prpplague | Lars_G: the spi/bitbang sd/mmc driver in the 2.6.25 kernel is fully functional with the hammer |
| 20:44.30 | RobotGuy | There is no numbers on any selections |
| 20:45.09 | RobotGuy | I can't deal with it anymore right now - too frustrating |
| 20:45.12 | SpeedEvil | when you click apply - it shows the number of matching parts at the bottom |
| 20:46.47 | RobotGuy | No matter what I do, I get no matches |
| 20:47.10 | RobotGuy | Giving up for now |
| 20:58.18 | Lars_G | prpplague: yeah stuppid me, i even saw you already have an sd slot over SPI in the development board |
| 20:58.41 | prpplague | CosmicPenguin: http://useopensource.blogspot.com/2008/04/synching-open-source-release-schedule.html |
| 20:58.58 | SpeedEvil | prpplague: what speed? |
| 20:59.05 | prpplague | CosmicPenguin: i know its not specific to the kernel, but imho that shows i'm not the only one thinking about schedules |
| 20:59.16 | prpplague | SpeedEvil: speed of the spi mmc/sd ? |
| 20:59.29 | SpeedEvil | transfet speed I mean |
| 20:59.31 | SpeedEvil | r |
| 20:59.43 | prpplague | SpeedEvil: iirc i have it running at 16mhz |
| 21:00.31 | Lars_G | well with any luck we might get a couple hammer kits to prototype, but I can't promisse, it's up to my boss |
| 21:00.36 | Lars_G | personally I'll love to play with them |
| 21:00.40 | SpeedEvil | prpplague: 1 bit? |
| 21:00.41 | prpplague | np |
| 21:00.49 | Lars_G | And once the project goes thorough I hope I can keep one for myself |
| 21:00.51 | prpplague | SpeedEvil: hehe, yea spi is one bit |
| 21:01.08 | Lars_G | serial buses are generally 1 bit |
| 21:01.09 | Lars_G | :P |
| 21:01.21 | SpeedEvil | true. |
| 21:01.22 | Lars_G | hence the serial part |
| 21:01.32 | SpeedEvil | But SD cards can be run at 4 bits |
| 21:01.52 | SpeedEvil | oh - somehow I misread over SPI |
| 21:02.19 | Lars_G | I think those are "parallel serial interfaces for speed" kind of stuff, like the MS's magicgate |
| 21:02.22 | prpplague | SpeedEvil: yea, if you utilize the SD protocol, you can do 4 bit |
| 21:02.30 | prpplague | SpeedEvil: but only sd cards can be used that way |
| 21:02.55 | Lars_G | Anyhow this application won't need such a speed to merit trying to bitbang a 4 bit bus |
| 21:02.56 | prpplague | SpeedEvil: 4-bit interface is not supported via psi |
| 21:02.58 | prpplague | spi |
| 21:03.10 | SpeedEvil | Oh - right - the SPI mode, yes. |
| 21:03.15 | SpeedEvil | forgot that difference. |
| 21:12.10 | RobotGuy | prpplague: How hard is it to do a full SDIO interface? |
| 21:13.42 | RobotGuy | There is a 32 GB SD card |
| 21:14.06 | SpeedEvil | sd cards can be used in spi mode |
| 21:14.12 | SpeedEvil | just slower |
| 21:14.22 | RobotGuy | I know. |
| 21:14.33 | RobotGuy | How fast in relation to a thumb drive? |
| 21:14.59 | RobotGuy | Is there any particular type of SD/MMC card I should get for Hammer? |
| 21:15.40 | SpeedEvil | 3,3Mbytes/s in SD mode - with some kingston flash |
| 21:15.50 | SpeedEvil | Not on a hammer - but same chip |
| 21:16.00 | RobotGuy | Why not on Hammer? |
| 21:16.13 | SpeedEvil | Because I do not have one. |
| 21:16.31 | RobotGuy | Oh, so it would be the same speed then for hammer. |
| 21:16.40 | prpplague | RobotGuy: not sure i understand your question |
| 21:16.53 | SpeedEvil | robbat2|na: if the pins are brought out. |
| 21:17.29 | RobotGuy | It seem like a direct question to me. |
| 21:17.37 | RobotGuy | Can hammer do full SDIO? |
| 21:17.40 | prpplague | RobotGuy: no |
| 21:17.52 | RobotGuy | OK, then other question doesn't matter |
| 21:17.53 | prpplague | RobotGuy: the pins nessary for full SDIO were not brought out |
| 21:18.26 | RobotGuy | prpplague: How many more pins for full SDIO? |
| 21:19.00 | prpplague | RobotGuy: 8 iirc |
| 21:19.16 | RobotGuy | Will they be on Sledge? |
| 21:19.41 | prpplague | yea at this stage they are brought out to the sd slot |
| 21:19.48 | RobotGuy | Cool |
| 21:20.18 | RobotGuy | That would include for potentially using SDIO cards for stuff like WiFi? |
| 21:20.24 | prpplague | http://www.elinux.org/SledgeHammer_Board |
| 21:21.01 | RobotGuy | Picture does not give me pinouts. |
| 21:21.04 | ce_geek | prpplague: Looks like I missed him :-/ |
| 21:21.09 | prpplague | some of the current questions we have about the sledge is whether we should change the slot to micro sd, and headers for a sdio wifi module, and placing a mini-usb connector on board |
| 21:21.14 | prpplague | ce_geek: np |
| 21:21.22 | prpplague | ce_geek: hopefully he'll be back around |
| 21:21.30 | ds2 | RobotGuy: would you like to volunteer some reliable and cheap sources of SDIO WLAN cards? :) |
| 21:21.42 | ce_geek | prpplague: Good good. |
| 21:21.56 | RobotGuy | ds2: Go back to sleep. :) |
| 21:21.57 | prpplague | ds2: zcomac has a lovely 8385 and 8686 modules |
| 21:22.09 | prpplague | ds2: both of which support sdio and spi |
| 21:22.24 | prpplague | ds2: however only spi firmware for the 8686 is available |
| 21:22.28 | prpplague | xcomax that is |
| 21:22.29 | ds2 | prpplague: are they really orderable? |
| 21:22.46 | prpplague | better be, since thats what we put in out handhelds |
| 21:23.03 | RobotGuy | prpplague: I don't see why a mini-USB connector couldn't be used on the Sledge module - would that be for Host or Slave? |
| 21:23.08 | SpeedEvil | Hmm. SDIO wifi cards are getting cheaper. |
| 21:23.23 | prpplague | RobotGuy: slave port |
| 21:23.30 | ds2 | prpplague: got a link? getting a lot of random noise on google |
| 21:23.42 | *** part/#edev RobotGuy (n=robotguy@dsl093-038-072.pdx1.dsl.speakeasy.net) |
| 21:23.53 | *** join/#edev RobotGuy (n=robotguy@dsl093-038-072.pdx1.dsl.speakeasy.net) |
| 21:24.25 | RobotGuy | Why even put USB on the Sledge module? |
| 21:24.36 | prpplague | http://www.zcomax.com/embedded.htm |
| 21:24.51 | prpplague | RobotGuy: just that some people have asked for it onboard |
| 21:25.09 | ds2 | ah they are not in the SDIO formfactor |
| 21:25.10 | RobotGuy | Oh. |
| 21:25.32 | SpeedEvil | Hmm. SDIO wifi linux search on ebay works. |
| 21:25.53 | SpeedEvil | http://cgi.ebay.co.uk/Spectec-802-11b-g-SDIO-WiFi-WLAN-Wireless-LAN-SD-Card_W0QQitemZ170213567041QQihZ007QQcategoryZ58542QQssPageNameZWDVWQQrdZ1QQcmdZViewItem |
| 21:26.04 | SpeedEvil | That's not _too_ spendy. |
| 21:26.10 | ds2 | Spectec's are Marvell? |
| 21:26.25 | SpeedEvil | says Linux V2.4.18 on Samsung 2410/2440 |
| 21:26.31 | SpeedEvil | dunno what that means |
| 21:26.35 | prpplague | ds2: the spectec device i got are not marvell |
| 21:26.43 | ds2 | oh :/ |
| 21:26.43 | prpplague | SpeedEvil: what says that?? |
| 21:26.50 | ds2 | geeks.com had those |
| 21:27.01 | SpeedEvil | prpplague: the above ebay auction. |
| 21:27.14 | SpeedEvil | takes no responsability for random ebay sellers. |
| 21:27.39 | SpeedEvil | 60mA recieve at 3V isn't bad. |
| 21:27.40 | ds2 | anyone figured out what chip are the Sandisk ones? |
| 21:27.40 | prpplague | SpeedEvil: ahh, i ordered a wide range of spectec cards, none were marvell |
| 21:28.18 | SpeedEvil | sdw-821 sound familiar? |
| 21:28.40 | SpeedEvil | ponders. |
| 21:28.57 | SpeedEvil | Oh - I can't use SDIO cards over SPI can I :( |
| 21:29.11 | ds2 | nope |
| 21:29.36 | SpeedEvil | And I can'd use a wire from a microSD socket either. |
| 21:29.39 | ds2 | but there (or will be) SDIO capable boards available at reasonable cost - beagle and probally some of the existing PXA270's |
| 21:29.40 | SpeedEvil | :/ |
| 21:29.44 | SpeedEvil | Oh well. |
| 21:29.59 | SpeedEvil | ds2: yeah - I was wondering about it for upgrading my phone. |
| 21:30.21 | ds2 | SpeedEvil: which phone? |
| 21:30.25 | SpeedEvil | ds2: it's an alternative to shaving down a USB card and an internal hub. |
| 21:30.29 | SpeedEvil | Neo1973 |
| 21:30.40 | ds2 | ah |
| 21:30.48 | SpeedEvil | Which is the same chip as the hammer, but with more stuff onboard. |
| 21:30.54 | ds2 | does the neo bring out a parallel interface? |
| 21:31.02 | ds2 | i.e. CF |
| 21:31.04 | SpeedEvil | no. I2C, SPI |
| 21:31.05 | prpplague | doesn't mean you can't use SPI based wifi |
| 21:31.11 | SpeedEvil | Sure. |
| 21:31.26 | SpeedEvil | However, finding a physically small enough module is a challenge |
| 21:31.28 | prpplague | the zipit is using the 8686 via spi |
| 21:31.43 | ds2 | prpplague: Z2 right? |
| 21:31.44 | prpplague | the xg modules are pretty small |
| 21:31.46 | prpplague | ds2: y ea |
| 21:31.59 | SpeedEvil | http://www.zcomax.com/embedded.htm |
| 21:32.01 | SpeedEvil | oops |
| 21:32.55 | prpplague | the marvell cf cards can actually be converted into spi or sdio |
| 21:33.01 | SpeedEvil | xg182m I suppose |
| 21:33.04 | ds2 | they can? |
| 21:33.09 | ds2 | Hmmmm |
| 21:34.15 | SpeedEvil | Hmm. 20mm might be tight. |
| 21:34.17 | SpeedEvil | measures |
| 21:35.25 | SpeedEvil | 20mm means it pokes out the back :/ |
| 21:35.39 | ds2 | bah the zcomax site is annoying... "forms" to order?! :( |
| 21:35.43 | g1powermac | GPSFan, interesting, didn't notice they had a new model |
| 21:36.29 | SpeedEvil | prpplague: do they sell the modules in ones? |
| 21:36.45 | ds2 | SpeedEvil: http://gumstix.com/store/catalog/product_info.php?cPath=31&products_id=191 |
| 21:36.53 | ds2 | 8385's in singles |
| 21:37.00 | prpplague | SpeedEvil: yea |
| 21:37.06 | SpeedEvil | prpplague: and drivers? |
| 21:37.16 | ds2 | you might need this one instead - |
| 21:37.18 | ds2 | http://gumstix.com/store/catalog/product_info.php?cPath=31&products_id=192 |
| 21:37.19 | prpplague | oh yea forgot gumstix had them |
| 21:37.24 | prpplague | SpeedEvil: in main kernel tree |
| 21:37.30 | ds2 | prpplague: is there another place that has them? |
| 21:37.33 | SpeedEvil | Hmm. |
| 21:37.57 | prpplague | ds2: yea. just google xg-180 |
| 21:38.06 | SpeedEvil | ponders. |
| 21:38.39 | SpeedEvil | wishes those were 17mm wide. |
| 21:39.25 | ds2 | hmmm nothing useful... gumstix will do |
| 21:39.41 | ds2 | prpplague: does it need a reflashing to use the SPI interface? I think the gumstix ones are setup for CF |
| 21:40.14 | prpplague | ds2: firmware is uploaded to the device, but it does have a resistor bank to control operations |
| 21:40.20 | prpplague | ds2: i can tell you any more than that |
| 21:40.30 | ds2 | you can or cannot? |
| 21:40.34 | prpplague | nda |
| 21:40.40 | ds2 | Okay |
| 21:41.05 | ds2 | so the gumstix ones will be troublesome for SPI :/ |
| 21:43.50 | SpeedEvil | Yow. 400mA at 3.3V typ. |
| 21:44.00 | SpeedEvil | (xg182) |
| 21:44.24 | ds2 | SpeedEvil: thought the Neo's had a built in WiFi interface? |
| 21:47.53 | SpeedEvil | not yet |
| 21:50.43 | *** join/#edev sunrunner20 (n=jamesdoe@pool-70-104-23-68.dllstx.fios.verizon.net) |
| 21:51.28 | prpplague | http://www.embeddedworks.net/wlan/oem_sdio_80211g.html |
| 21:51.31 | prpplague | look familiar |
| 21:52.19 | g1powermac | prpplague, hows the spi driver coming for the marvell 88W8686, btw? |
| 21:52.33 | prpplague | g1powermac: haven't checked lately |
| 21:52.37 | g1powermac | k |
| 21:52.42 | g1powermac | was just wonderin |
| 21:52.43 | prpplague | g1powermac: i've been meaning to , but i've been swampped |
| 21:52.55 | prpplague | g1powermac: i'm not directly working on it |
| 21:53.06 | g1powermac | ahh |
| 21:53.39 | prpplague | i personally like these embedded usb modules |
| 21:53.54 | prpplague | they can be configured for +3.3v or +5v operations |
| 21:56.29 | prpplague | you can bitbang 8-bit cf interface, but its combersom |
| 21:58.36 | g1powermac | wonders what these other chips are on the zipit, there's a few I can't find any data to |
| 21:59.18 | roxfan | which ones? |
| 21:59.27 | g1powermac | http://zipit2system.sourceforge.net/board.html |
| 21:59.31 | prpplague | g1powermac: got a list going? |
| 21:59.38 | g1powermac | see the list, there's a couple marked unknown |
| 21:59.41 | g1powermac | prpplague, yupper |
| 22:01.10 | prpplague | g1powermac: the AER915 GPSFan suspects is the Aeronix rom version of the LPC915 |
| 22:01.46 | prpplague | g1powermac: the LP35 is a backlight driver |
| 22:01.53 | g1powermac | ahh, awesome |
| 22:04.00 | prpplague | g1powermac: is there a manufacturer mark on the sdram? |
| 22:04.40 | g1powermac | prpplague, nothing more than what you can see in the photos |
| 22:05.07 | prpplague | ~lart g1powermac |
| 22:05.07 | purl | chops g1powermac in half with a free AOL CD |
| 22:05.10 | roxfan | there's something that looks like M and T superscript in the corner |
| 22:05.13 | prpplague | g1powermac: Micron |
| 22:05.28 | g1powermac | roxfan, yea |
| 22:06.46 | prpplague | g1powermac: http://www.micron.com/products/partdetail.aspx?part=MT48H8M16LFB4-8+IT |
| 22:07.51 | g1powermac | prpplague, how did you find that? |
| 22:08.18 | prpplague | g1powermac: http://www.micron.com/support/part_info/fbga/decoder.aspx then enter D9HCX |
| 22:08.30 | g1powermac | hah, cool |
| 22:08.37 | SpeedEvil | wonders that he can actually understand micron part numbers. |
| 22:08.44 | SpeedEvil | too much staring at lists. |
| 22:09.18 | prpplague | SpeedEvil: hehe me too |
| 22:10.19 | prpplague | ~seen T0mW |
| 22:10.22 | purl | t0mw <n=Tom@24.115.218.135.res-cmts.sth.ptd.net> was last seen on IRC in channel #edev, 21d 56m 17s ago, saying: 'this is the first time I've built debian with debootstrap'. |
| 22:10.44 | SpeedEvil | needs to get ordering stuff for the bit-banging sdram thingy. |
| 22:10.54 | g1powermac | prpplague, he was here not too long ago |
| 22:10.58 | g1powermac | but didn't say anything |
| 22:11.03 | g1powermac | just a couple of days ago |
| 22:11.19 | g1powermac | prpplague, btw, that one chip is LP53, no LP35 |
| 22:11.43 | g1powermac | *not |
| 22:12.04 | ds2 | whoa bitbang SDRAM? what's that? |
| 22:12.51 | SpeedEvil | I noticed that I can bitbang SDRAM on a processor without too much pain. |
| 22:13.12 | SpeedEvil | So I'm going to try. |
| 22:14.37 | ds2 | what is bitbanging an SDRAM |
| 22:15.14 | prpplague | more to the point, why? |
| 22:15.20 | SpeedEvil | to bitbang is to drive something directly under processor control, using general purpose IO, rather than special purpose memory. |
| 22:15.25 | SpeedEvil | to bitbang is to drive something directly under processor control, using general purpose IO, rather than special purpose drivers. |
| 22:15.38 | Lars_G | o.O one thing I wouldn't bitbang is something as fast or time sensitive as ram |
| 22:15.40 | Lars_G | personally. |
| 22:16.05 | SpeedEvil | prpplague: cheapness. And that I can add 8M of memory to a STM32 without much apparent pain. (though terrible single-cycle performance) |
| 22:16.44 | ds2 | are you implying this as a means of adding SDRAM to a busless chip? |
| 22:16.50 | SpeedEvil | yes. |
| 22:16.58 | ds2 | ah okay, that makes sense |
| 22:17.16 | g1powermac | alright, list has been updated |
| 22:19.36 | g1powermac | ok, there is still at least one more chip that didn't make it onto GPSFan's shot but is on this one: http://www.seattlewireless.net/Zipit2Messenger |
| 22:19.37 | SpeedEvil | ds2: though not 'properly' - you have to do calls to accessbyte() or copyblock() |
| 22:20.03 | ds2 | SpeedEvil: yeah, I figure that much... it is bascially a giant ramdisk |
| 22:20.30 | SpeedEvil | If it had an MMU that trapped, it would be easy (in software) |
| 22:20.33 | SpeedEvil | but nope. |
| 22:21.03 | ds2 | wonder if it is that much cheaper then adding 8M of SRAM instead |
| 22:21.18 | SpeedEvil | SDRAM is both cheaper, and has less wires |
| 22:21.33 | ds2 | performance is got to be horrible with the software refreshing |
| 22:21.40 | roxfan | o.o this guy doesn't have yellow wires? |
| 22:21.44 | SpeedEvil | Sure. But it's adequate, and I don't care. |
| 22:21.55 | prpplague | g1powermac: looks like i was wrong on the LP53 |
| 22:22.08 | prpplague | g1powermac: thats a LP2981 ldo |
| 22:22.43 | SpeedEvil | ds2: I figure I can read a 512 byte block in in 60us - which is fine. |
| 22:23.12 | g1powermac | so the lp53 is this: http://focus.ti.com/docs/prod/folders/print/lp2981-28.html ? |
| 22:23.18 | prpplague | http://focus.ti.com/docs/prod/folders/print/lp2981-28.html |
| 22:23.20 | prpplague | yea |
| 22:23.24 | g1powermac | alrighty |
| 22:23.25 | roxfan | SpeedEvil, don't you have to keep the refresh running? |
| 22:23.30 | prpplague | if you look the topside marking is LP53 |
| 22:23.37 | ds2 | SpeedEvil:but can you still do other useful things on the CPU while running the refresh? |
| 22:23.49 | SpeedEvil | ds2: oh - misread. |
| 22:23.59 | SpeedEvil | ds2: no - I put it in self-refresh most of hte time. |
| 22:24.07 | ds2 | ah |
| 22:25.21 | g1powermac | prpplague, sounds good |
| 22:25.37 | g1powermac | list updated |
| 22:25.41 | prpplague | http://focus.ti.com/general/docs/partmarking/partmarkinglookup.tsp?startPosition=0&navigationId=&templateId=&siteFlag=&keyword=LP53&searchType=partmark |
| 22:26.47 | ds2 | prpplague: you on the ARM list? |
| 22:26.48 | g1powermac | alright, so the only other chip now is that ana one |
| 22:26.56 | prpplague | ds2: LAK |
| 22:27.29 | ds2 | prpplague: see the mail about the different ARM trees and stuff? sounds like they are trying to address your complain (to an extent) |
| 22:27.56 | prpplague | ds2: sorry no, i've not seen anything, i'll have a look at the archives |
| 22:28.42 | g1powermac | prpplague, and for that LPC915, I wonder how we can confirm it is actually an 8051. . . |
| 22:28.58 | ds2 | the second hand copy shows a sent date of Mon 28 APR around 4:29 |
| 22:29.02 | prpplague | g1powermac: any manufacturer markings on the ANA chipset |
| 22:29.51 | g1powermac | checking. . . |
| 22:30.41 | g1powermac | no, you can see it clearly with this shot: http://www.drtune.com/zipit2/bigboard1.jpg |
| 22:31.25 | roxfan | there's a little spot in the corner |
| 22:31.33 | roxfan | but it could be pin1 |
| 22:31.46 | g1powermac | thats what I'd suspect |
| 22:32.38 | Lars_G | g1powermac: still hacking the... what was it called? |
| 22:32.44 | g1powermac | Lars_G, zipit |
| 22:32.46 | Lars_G | that |
| 22:33.34 | g1powermac | yupper |
| 22:34.04 | Lars_G | is it comming along nicely? |
| 22:34.31 | g1powermac | ya, was playing around with a shell running on the zipit :-) |
| 22:34.53 | g1powermac | and we definitely got some more info on these other chips thanks to prpplague and GPSFan |
| 22:36.19 | Lars_G | Well it's a cheap nice device for a linux capable thingy with built in lcd and kbd |
| 22:36.32 | g1powermac | yup, thats what drew me to it |
| 22:39.06 | prpplague | g1powermac: gotta run, but it appears that the ANA chip is this - http://focus.ti.com/docs/prod/folders/print/bq24035.html |
| 22:39.21 | GPSFan | g1powermac: you still need a closeup of that other chip? |
| 22:39.48 | g1powermac | GPSFan, well, part numbers work too |
| 22:39.55 | g1powermac | prpplague, wow, you're good :-) |
| 22:41.11 | prpplague | heads home |
| 22:41.46 | roxfan | packaging doesn't seem to match |
| 22:42.36 | roxfan | the part at url is flat quadpack no-lead |
| 22:43.02 | roxfan | and 20 pin only... |
| 22:43.07 | g1powermac | its also 20 pins. . .the ana chip is 24 |
| 22:43.44 | g1powermac | ok, can't be it then |
| 22:46.06 | g1powermac | ok, tonight I'm cracking open the zipit again and probing the aer chip |
| 22:46.27 | SpeedEvil | measuring which are VCC/gnd can help |
| 22:46.33 | g1powermac | with datasheet of the LPC915 in hand, I'll compare pinouts |
| 22:46.51 | g1powermac | obviously it could be different since this chip isn't made by philips |
| 23:02.47 | *** join/#edev prpplague (n=dave123_@ppp-70-244-164-200.dsl.rcsntx.swbell.net) |
| 23:02.55 | g1powermac | wb prpplague |
| 23:03.19 | g1powermac | prpplague, the ana chip can't be the one you found, its not the right packaging |
| 23:03.36 | prpplague | thanks |
| 23:03.42 | prpplague | you got your page updated? |
| 23:03.50 | g1powermac | yup |
| 23:03.57 | prpplague | hmm |
| 23:03.59 | prpplague | looks |
| 23:04.12 | g1powermac | but did not update the ana chip listing |
| 23:04.33 | prpplague | double checking |
| 23:04.39 | prpplague | can you repost your main page url |
| 23:04.53 | g1powermac | http://zipit2system.sourceforge.net/board.html |
| 23:07.44 | prpplague | yea doesn't look right |
| 23:09.26 | Lars_G | prpplague: btw this is the other choice I've found so far: http://www.kwikbyte.com/KB9260.html |
| 23:10.59 | ds2 | Lars_G: not sure if you care but their payment system is via paypal |
| 23:11.10 | prpplague | lordy lordy |
| 23:11.14 | Lars_G | ds2: it's not really a hinder right now. |
| 23:12.03 | Lars_G | prpplague: ? |
| 23:12.40 | Lars_G | g1powermac: you know that thing does look nice |
| 23:15.39 | Lars_G | sigh |
| 23:15.50 | Lars_G | the pxa is clocked higher than my palm TX's |
| 23:16.02 | g1powermac | :-) |
| 23:16.10 | ds2 | isn't the TX also a PXA? |
| 23:17.06 | roxfan | wikipedia says 312 too |
| 23:17.44 | roxfan | for TX |
| 23:21.00 | Lars_G | yes it's a PXA |
| 23:29.46 | Lars_G | I was mistaken about the clocking I see |
| 23:30.29 | g1powermac | hmm, definitely need to work on the keyboard driver on this sucker, debouncing isn't happening too well |
| 23:31.14 | Lars_G | Omg that thing has 802.11? |
| 23:31.32 | g1powermac | yupper, 802.11b/g |
| 23:31.35 | g1powermac | minisd |
| 23:31.41 | g1powermac | too |
| 23:31.45 | Lars_G | damnit. very nice indeed |
| 23:32.01 | g1powermac | and after I'm done with it, it'll have a camera, accelerometer, and a gsm module :-) |
| 23:32.12 | Lars_G | sniffs |
| 23:32.15 | Lars_G | they grow so fast |
| 23:33.17 | g1powermac | hehe |
| 23:37.25 | g1powermac | hmm, this P89LPC915 does _alot_ of stuff |
| 23:37.59 | g1powermac | could be tough to figure out if the aer chip is a P89LPC915 by assuming some pin function alone |
| 23:39.04 | g1powermac | definitely a nice 8051 to use for some microcontroller projects |
| 23:45.00 | MonMotha | boo 8051. Yea AVR |
| 23:45.10 | MonMotha | though I will admit there are times when the 8051 is actually better |
| 23:45.14 | ds2 | those SiLab 8051's are cool |
| 23:45.16 | g1powermac | bah avr, yay 8051 :-) |
| 23:45.21 | MonMotha | especially the fancy "fast" (single cycle on most instructions) 8051s |
| 23:45.26 | g1powermac | is an 8051 guy |
| 23:45.33 | g1powermac | yup, love the ds89c450s |
| 23:45.36 | MonMotha | I used to be an 8051 guy, then I programmed AVR |
| 23:45.47 | ds2 | does anyone make a toolstick equiv for the AVR's? |
| 23:46.13 | MonMotha | at90usbkey |
| 23:46.20 | MonMotha | it's a ~$35 dev/eval board put out by atmel |
| 23:46.21 | *** join/#edev Sienna (n=Sienna@pool-71-125-184-233.bltmmd.east.verizon.net) |
| 23:46.24 | MonMotha | with their USB uC on it |
| 23:46.52 | MonMotha | though that's actually a USB enabled uC, not a debugging interface |
| 23:47.09 | ds2 | that is a one peice board |
| 23:47.12 | MonMotha | that is one thing I don't like so much about AVR: Atmel charges for the JTAG pod and doesn't document the JTAG debug interface |
| 23:47.17 | *** join/#edev jarv (n=jarv@96.245.16.205) |
| 23:47.32 | ds2 | I like the toolstick where you can use the little dev boards in a project |
| 23:47.37 | MonMotha | though their JTAGICE works really well. Much better than, say, the USB OCD for ARM stuff |
| 23:48.13 | MonMotha | I'm thinking of the Olimex POS |
| 23:48.35 | g1powermac | atmel has some issues with some stuff |
| 23:48.39 | g1powermac | like their fpgas |
| 23:48.47 | MonMotha | yeah, I'm not a big fan of their FPGAs |
| 23:48.56 | g1powermac | they make awesome fpgas, but their software isn't free |
| 23:49.11 | MonMotha | well, most FPGA dev software isn't free |
| 23:49.15 | MonMotha | and in fact most of it is quite expensive |
| 23:49.31 | MonMotha | I do like what Xilinx does with the free stripped-down version for their smaller FPGAs |
| 23:49.44 | g1powermac | true, but xilinx is nice |
| 23:49.52 | g1powermac | in regards to their software |
| 23:50.03 | MonMotha | their hardware is pretty good, too |
| 23:50.14 | MonMotha | though they tend to go for the big and bad market |
| 23:50.20 | g1powermac | yea, not bad, still like atmel's fpgas in theory |
| 23:50.21 | MonMotha | not the tiny/low power market |
| 23:50.48 | g1powermac | I actually have a batch of atmel fpga/cpld samples floating around |
| 23:50.49 | MonMotha | I would love to tinker with the FPSLIC stuff |
| 23:50.57 | g1powermac | got some of them too |
| 23:50.59 | MonMotha | the AVR hard core married with FPGA |
| 23:51.21 | g1powermac | has so many free samples, its almost sickening ;-) |
| 23:52.26 | g1powermac | holy crap, Hans Reiser was found guilty of first degree murder |
| 23:52.34 | SpeedEvil | Indeed :( |
| 23:52.43 | SpeedEvil | has been reading the liveblogging. |
| 23:52.50 | jarv | link? |
| 23:52.58 | g1powermac | http://yro.slashdot.org/yro/08/04/28/2243232.shtml |
| 23:53.20 | jarv | thx |
| 23:53.23 | SpeedEvil | My major problem with it is http://yro.slashdot.org/yro/08/04/28/2243232.shtml |
| 23:53.48 | SpeedEvil | my major problem with it is that if it was first degree - premeditated murder - he did a really, really sucky job. |
| 23:54.15 | g1powermac | yea |
| 23:54.35 | jarv | pretty dumb to take the stand |
| 23:54.40 | SpeedEvil | And appearing dysfunctional on the stand. |
| 23:54.43 | jarv | yeah |
| 23:54.53 | SpeedEvil | (reading the liveblogging - which is pretty close to transcripts) |
| 23:54.58 | g1powermac | well, I expect he'll appeal |
| 23:55.22 | SpeedEvil | Sure. |
| 23:56.10 | *** join/#edev prpplague (n=dave123_@ppp-70-244-164-200.dsl.rcsntx.swbell.net) |
| 23:56.35 | prpplague | returns |
| 23:56.57 | g1powermac | prpplague, hans reiser has just been convicted of murder |
| 23:57.15 | Sienna | and he is special why? |
| 23:57.22 | SpeedEvil | reiserfs |
| 23:57.32 | SpeedEvil | active in the open-source community. |
| 23:57.37 | SpeedEvil | (until his arrest) |
| 23:57.41 | Sienna | but if he was convicted, he should be treated the same as everyone else |
| 23:58.40 | SpeedEvil | As to why he is special - reading hte liveblogging - which is quite close to transcript level detail - much of the juries impression of him will be wierdness that is not atypical of CS geeks. |
| 23:59.03 | Sienna | actually, just ignore me... i think i might be having a really cynical phase right now |
| 23:59.03 | SpeedEvil | This does not in itself mean he murdered her, simply that he's wierd. |
| 23:59.17 | g1powermac | well, time to go home, bbiab |
