| 00:00.03 | landley | CosmicPenguin: I'm pretty happy to use an oldish version of x.org whenever possible. |
| 00:00.27 | CosmicPenguin | me too - it just never seems to work out that I can |
| 00:00.36 | landley | Kernel you have to update because jumping 3 versions blindsides you with strangeness. X less so. |
| 00:00.49 | landley | But then maybe I've been lucky. |
| 00:01.01 | landley | Half the systems I deal with are headless boxes, I can often _avoid_ having to deal with X... |
| 00:01.10 | CosmicPenguin | now that X is modular, breakage is way easier |
| 00:01.11 | landley | More like 80%, really... |
| 00:01.23 | CosmicPenguin | you can change the core and be blissfully unaware that 90% of the drivers broke |
| 00:01.40 | landley | CosmicPenguin: I'm 100% rooting for keith packard and co to render all memory of XFree86 development a distant historical curiosity. |
| 00:02.02 | CosmicPenguin | indeed |
| 00:02.09 | CosmicPenguin | but people have already forgotten |
| 00:02.18 | landley | And I've never gotten 3D to work _anywhere_ except as black magic. |
| 00:02.25 | CosmicPenguin | there was a thread on the X development list about how autotools sucks so bad and they should write their own configuration engine |
| 00:02.33 | landley | And yet the XFree86 website hasn't gone down yet. |
| 00:02.37 | *** join/#edev redb1ue (i=star@ppp131.108-253-207.mtl.mt.videotron.ca) |
| 00:02.47 | landley | And the one guy who drove everybody away was still checking stuff into it, last I checked. (Which was about a year ago.) |
| 00:03.08 | landley | CosmicPenguin: they should talk to ESR, he and I had a phone conversation about that a couple weeks ago. |
| 00:03.14 | landley | Autotools _does_ suck. |
| 00:03.34 | CosmicPenguin | yes |
| 00:03.39 | CosmicPenguin | but does nobody remember imake? |
| 00:03.57 | landley | Me? http://landley.net/writing/whynotmake.html |
| 00:04.20 | CosmicPenguin | exactly |
| 00:04.30 | CosmicPenguin | we finally got rid of that to something that at least I understand how to hack when it breaks |
| 00:04.36 | landley | Both ./configure and make have largely outlived their usefulness. |
| 00:04.47 | *** join/#edev prpplague (n=dave@70.249.150.251) |
| 00:04.50 | landley | But inertia's likely to keep them around for a few more years. |
| 00:05.26 | landley | What I'm looking forward to is a multi-threaded compiler that can do -j and take advantage of SMP internally, while compiling one big "cc one.c two.c three.c four.c -o blah" command line. |
| 00:05.50 | landley | But probably not this year. Too many other things to do... |
| 00:06.38 | landley | Huh... |
| 00:06.40 | landley | Interesting. |
| 00:06.47 | landley | I'd have to actually sit down and work out the numbers... |
| 00:07.04 | landley | But I think that the new quad core x86-64 chips from Intel are literally one _billion_ times faster than the original PDP-11. |
| 00:07.30 | landley | If not, it's probably only a few years away... |
| 00:09.03 | SpeedEvil | nowhere close |
| 00:09.21 | *** join/#edev g1powermac (n=g1powerm@unaffiliated/g1powermac) |
| 00:09.46 | SpeedEvil | call it 24Gops - a billion would take teh PDP-11 to 24 ops |
| 00:16.39 | ds2 | Mmmmm imake |
| 00:16.45 | ds2 | better then configure ;) |
| 00:16.51 | ds2 | and libtool |
| 00:17.16 | CosmicPenguin | you might consider checking yourself in for some professional care |
| 00:17.24 | CosmicPenguin | the nice guys will take you to your padded room now |
| 00:18.48 | ds2 | you actually like libtool/configure?! |
| 00:20.34 | CosmicPenguin | better then imake |
| 00:20.41 | CosmicPenguin | I hate libtool/configure - but I understand it |
| 00:20.58 | CosmicPenguin | so when I'm making build systems, I would much rather deal with that then trying to figure out N custom solutions |
| 00:21.03 | CosmicPenguin | which, by the way, all suck too |
| 00:21.36 | prpplague | hehe |
| 00:21.43 | Mirell | I has a laptop |
| 00:22.22 | prpplague | hehe the lawyers are after trolltracker now |
| 00:22.31 | prpplague | first it was PJ now its trolltracker |
| 00:32.11 | SpeedEvil | landley: http://freespace.virgin.net/roy.longbottom/whetstone.htm |
| 00:32.17 | SpeedEvil | says a million is probably sane |
| 00:32.32 | SpeedEvil | (times faster for current processors than the PDP/11-03 |
| 00:33.24 | RobotGuy | prpplague: I have a working toolchain! It's not standard, but it apparently works. :D |
| 00:33.40 | prpplague | RobotGuy: lovely |
| 00:33.47 | prpplague | RobotGuy: whats not standard about it? |
| 00:34.02 | RobotGuy | prpplague: I added binutils 2.18.* and fakeroot 1.9 from the current SVN. |
| 00:34.22 | prpplague | ahh |
| 00:34.37 | landley | SpeedEvil: quite possibly. |
| 00:34.38 | RobotGuy | I guess you could say I sort of hacked buildroot. |
| 00:35.00 | landley | I tried to look it up, but I keep forgetting how hard it is to actually measure _performance_. |
| 00:35.05 | landley | Other than by benchmarking applications. |
| 00:35.17 | landley | Nobody can even agree on _units_. :P |
| 00:35.19 | RobotGuy | All this while I am learning Squeak. :) |
| 00:36.00 | prpplague | it's time for "hammer's in spaaaaaccccceee" - http://www.pratthobbies.com/proddetail.asp?prod=CANSAT%2D1 |
| 00:36.11 | prpplague | ordered one of those to convert over to using the hammer |
| 00:36.35 | RobotGuy | prpplague: No, it's time to get some robots hammered.. |
| 00:36.51 | prpplague | RobotGuy: the bot board is done |
| 00:36.58 | prpplague | RobotGuy: just needs to be beta tested |
| 00:37.10 | RobotGuy | prpplague: I still don't know quite what to do about the /etc config file issue though. |
| 00:37.26 | prpplague | RobotGuy: what /etc config file issue? |
| 00:37.59 | RobotGuy | prpplague: Where it puts config files for external software I build, like Python, Squeak, other stuff I might want to add. |
| 00:38.01 | prpplague | RobotGuy: not till march or april on the bot board |
| 00:38.21 | RobotGuy | prpplague: I know, I know, I know. Curses. ;) |
| 00:38.34 | prpplague | RobotGuy: i told you how to do the configs |
| 00:38.51 | RobotGuy | prpplague: Copy them to where I want them? |
| 00:39.27 | prpplague | RobotGuy: put them on the thumb drive, in a shared directory, after the thumb drive is mount use a script in the inittab to copy them into place |
| 00:39.34 | RobotGuy | prpplague: Some configs belong in /etc, but I don't think others do. |
| 00:40.03 | prpplague | RobotGuy: on your thumb drive have a dir /usr/share/configs |
| 00:40.19 | prpplague | RobotGuy: with sub dirs like /usr/share/configs/etc |
| 00:40.51 | prpplague | RobotGuy: the other option is to generate symlinks |
| 00:41.44 | RobotGuy | prpplague: I'm still not quite sure I am following what you are suggesting. I copy the /etc files to /usr/share/configs ? |
| 00:42.15 | RobotGuy | I mean /usr/share/configs/etc |
| 00:42.21 | prpplague | RobotGuy: correct |
| 00:42.53 | prpplague | RobotGuy: then in your inittab after you mount your thumb drive, copy the contents of that dir over to /etc |
| 00:43.48 | RobotGuy | prpplague: That's a problem though because /etc is on the ramdisk which has a set size. I can try it though and see what happens. |
| 00:44.21 | prpplague | RobotGuy: when you generate the ramdisk it does give you some additional room |
| 00:44.47 | prpplague | RobotGuy: trust me it works, i've done it on half a dozen devices |
| 00:44.56 | RobotGuy | prpplague: I already have a script that creates my initrd, so I can easily add the copy of the etc to that so the configs get into my rootfs. |
| 00:45.18 | prpplague | ok |
| 00:45.28 | RobotGuy | prpplague: I understand what you are saying now, finally. :) |
| 00:45.55 | landley | prplague: is hammer at all radiation hardened? |
| 00:47.39 | prpplague | landley: base model, no |
| 00:48.23 | prpplague | landley: but for anything less than 80km, basic aluminum foil does the job |
| 00:48.28 | prpplague | landley: or case |
| 00:49.32 | landley | prpplague: or can. :) |
| 00:49.38 | RobotGuy | prpplague: Check this out: |
| 00:49.45 | landley | RobotGuy: initramfs still beats initrd. Scales dynamically... |
| 00:49.47 | RobotGuy | -rw-r--r-- 1 linuxguy staff 4.7M 2008-01-28 16:49 rootfs.arm.initrd.ext2 |
| 00:49.47 | RobotGuy | -rw-r--r-- 1 linuxguy staff 938K 2008-01-28 16:49 rootfs.arm.initrd.ext2.gz |
| 00:50.07 | prpplague | RobotGuy: and? |
| 00:50.10 | RobotGuy | landley: I don't know how to use initramfs |
| 00:50.15 | landley | Also more efficient use of space. An initrd will have filesystem formatted data and then a dentry/page cache formatted copy of the data. |
| 00:50.18 | RobotGuy | prpplague: Now to see if it boots. ;) |
| 00:50.19 | landley | RobotGuy: it's easy. |
| 00:50.32 | landley | How are you currently creating your initrd image? |
| 00:51.14 | landley | You have a directory full of files you package up somehow? |
| 00:51.14 | landley | During your build? |
| 00:51.21 | prpplague | landley: hehe, don't confuse him he just got his build work, hehe |
| 00:51.30 | prpplague | landley: yea he just has it configured in buildroot |
| 00:51.31 | landley | Ok. |
| 00:51.38 | landley | Ah, buildroot does it for you. |
| 00:51.42 | prpplague | landley: yea |
| 00:51.59 | RobotGuy | landley: Using buildroot |
| 00:51.59 | landley | Changing how buildroot works == black magic. Here there be dragons. |
| 00:52.03 | landley | Alas. |
| 00:52.14 | landley | It's easy to do, hard to make buildroot do. |
| 00:52.16 | landley | Many things are thus. |
| 00:52.22 | landley | Hence me not using buildroot. :) |
| 00:52.27 | RobotGuy | Yeah, don't cornfuzzle me! ;) |
| 00:52.49 | landley | After buildroot runs, do you have a directory of files that your initrd _was_ generated from? |
| 00:52.55 | landley | Still there, I mean? |
| 00:52.58 | prpplague | landley: still no coverage of the WIAV suit |
| 00:53.05 | prpplague | landley: yea |
| 00:53.09 | RobotGuy | landley: Yes. |
| 00:53.17 | landley | RobotGuy: cd into that directory, and run this: |
| 00:53.41 | landley | find . | cpio -o -H newc | gzip > ../initramfs.cpio.gz |
| 00:53.56 | landley | That creates an initramfs file from the contents of that directory. |
| 00:54.25 | prpplague | hehe or just turn on initramfs option in the config |
| 00:54.34 | landley | Then feed the kernel that initramfs file instead of your initrd file, and if you have an "/init" program in that filesystem, it'll run that and never fall through to the later initrd or root= code. |
| 00:55.04 | landley | prpplague: the build can do it for you, but this assumes that buildroot is populating your initrd contents directory _before_ it builds the kernel, and that you know the correct path to feed the build. |
| 00:55.30 | landley | RobotGuy: it's a fun thing to play with, but you can worry about it later. |
| 00:55.41 | prpplague | landley: yea, currently the kernel for the hammer is built outside of buildroot |
| 00:55.49 | landley | It _is_ a more efficient use of memory, and it scales dynamically. But initrd worked for years before they came up with something etter... |
| 00:56.20 | landley | prpplague: initramfs images can either be supplied externally (the way initrd is), or they can be compiled into the kernel binary. |
| 00:57.10 | prpplague | landley: right, was just note'n that the kernel was seperate |
| 00:57.16 | landley | Adding it to the kernel build would link it into the kernel binary. The above creates an external one. |
| 00:57.43 | landley | prpplague: I'll wait until robotguy runs out of problems with buildroot before trying to change things. |
| 00:57.49 | landley | I'm not sure I'll live that long, but... |
| 00:58.32 | prpplague | landley: hehe |
| 00:58.56 | prpplague | landley: actually i think thats fixed now |
| 01:00.18 | prpplague | landley: yea, if don't have a path to a initramfs image, it doesn't link it to the kernel |
| 01:02.05 | *** join/#edev ico2__ (n=ico2@modem-636.chimpanzee.dialup.pol.co.uk) |
| 01:06.26 | RobotGuy | HEY! |
| 01:08.09 | RobotGuy | landley: Just curious.. How would that affect the way I build stuff to flash into hammer? |
| 01:08.43 | prpplague | RobotGuy: it doesn't |
| 01:09.00 | prpplague | RobotGuy: just a different method of doing the initial root file system |
| 01:09.01 | landley | The least intrustive change is to just replace your current initrd image with an initramfs image. |
| 01:09.09 | landley | Keeping it external, and renaming /linuxrc to /init. |
| 01:09.23 | landley | (And if you're doing a pivot_root, using switch_root instead.) |
| 01:09.30 | landley | Beyond that, it's pretty much the same. |
| 01:09.50 | landley | Linking it into the kernel is a simplification you can do later. |
| 01:10.02 | prpplague | RobotGuy: i'd wait till you are comfortable with your current config |
| 01:10.07 | landley | However, it's not a pressing _need_. Just an improvement you could do. |
| 01:18.45 | ds2 | are distros so varied thesedays that they have to make dev. enviroments specific to one??!!? |
| 01:19.05 | ds2 | <-- irked at making dev enviroment specific to paths and distros |
| 01:24.41 | *** join/#edev cajaka (n=none@cpe-24-58-193-200.twcny.res.rr.com) |
| 01:24.53 | RobotGuy | prpplague: I did add the code to my script to create the initramfs and save it off to my project save location. Gosh, I have actually added comments to my script! |
| 01:25.06 | prpplague | hehe |
| 01:25.21 | prpplague | ds2: yep |
| 01:25.22 | RobotGuy | prpplague: Dang, this script may actually end up being useful to others. |
| 01:26.23 | RobotGuy | landley: So now that I have the code added to create the initramfs (thanks!), all I have to do later is link it to my kernel, right? |
| 01:27.00 | RobotGuy | landley: I'll wait to do that, like prpplague suggests, because I am not ready for that yet. |
| 01:27.33 | prpplague | RobotGuy: yea you can link it in,or just use it like you use the initrd |
| 01:27.59 | RobotGuy | prpplague: Ahh, so flash the initramfs instead of the initrd. |
| 01:28.20 | prpplague | RobotGuy: yea |
| 01:28.46 | RobotGuy | Here goes first test of new make initrd image script.. |
| 01:30.29 | RobotGuy | I scare myself sometimes.. It worked the FIRST TIME. |
| 01:30.36 | prpplague | hehe |
| 01:30.59 | RobotGuy | I added a new environment variable $HAMMER_PROJECT_ROOT |
| 01:31.30 | landley | RobotGuy: you can feed it in as an external image instead of initrd, or you can supply it as a file "usr/initramfs_data.cpio.gz" |
| 01:31.47 | landley | If that file's newer than the souce code, the build should just use it rather than recreating it, and it'll get linked into the kernel. |
| 01:31.57 | landley | That's linux-2.6.23/usr/initramfs_data.cpio.gz of course. :) |
| 01:32.37 | prpplague | RobotGuy: note if you link in the initramfs to the kernel, you'll need to change your bootloader settings |
| 01:32.48 | landley | prpplague: bootloaders, I don't know. :) |
| 01:33.24 | prpplague | landley: the bootloader is only copying 1024k of bytes for the kernel |
| 01:33.49 | landley | prpplague: yeah, linking in the initramfs might put that over. |
| 01:34.04 | prpplague | landley: so if the kernel image is bigger than 1024k , the default copy command will need to be altered |
| 01:34.17 | landley | The bootloader has no way to detect the size? |
| 01:34.45 | prpplague | landley: most bootloaders do not |
| 01:34.49 | landley | (Store a size in the first 4 bytes of the image, maybe?) |
| 01:34.50 | landley | Sad. |
| 01:34.54 | prpplague | landley: u-boot has a method for storing that |
| 01:35.07 | prpplague | landley: but it add some complexity to boot sequence |
| 01:35.27 | landley | The copy is more than a for loop from flash to dram? |
| 01:36.05 | landley | "First four bytes a are a length in CPU endianness, kernel image data starts at byte 4..." |
| 01:36.11 | landley | I'm not sure how this adds significant complexity... |
| 01:36.33 | landley | But I haven't looked too closely at how bootloaders work, because they change from platform to platform. |
| 01:36.38 | landley | With qemu, I bypass 'em completely. |
| 01:37.03 | landley | Oh speaking of which, prpplague: did you ever get those patches broken up to send to me? :) |
| 01:37.17 | prpplague | landley: indeed it sounds simple, but when you are running the mmu off, and alot of code done in assembly, it tends to get complicated quickly |
| 01:37.35 | prpplague | landley: the other piece is that the kernel has to be generated with the relevant infof |
| 01:37.40 | prpplague | landley: no not as of yet |
| 01:37.50 | prpplague | landley: i'll try to get them to you tomorrow for sure |
| 01:37.57 | prpplague | landley: just been busy with the "day job" |
| 01:38.48 | landley | Heh. |
| 01:39.00 | landley | "Failure is not an option" means failure is, in fact, mandatory. |
| 01:39.22 | ds2 | heh 1M kernel... that is almost amusing ;) |
| 01:39.25 | prpplague | landley: generally speaking, the copy routine is pretty fast , so most systems simply copy a max size |
| 01:39.31 | prpplague | ds2: why? |
| 01:39.50 | prpplague | ds2: my average kernel is less than 768k |
| 01:40.00 | landley | The kernel doesn't have to be generated with the relavant info. Any install script can read the length, write out 4 bytes, and cat the kernel data to that. |
| 01:40.01 | ds2 | prpplague: just don't seem many of them anymore... even the N800 zImages are 1.6M |
| 01:40.30 | landley | And doing a read from memory to get the length of a loop when the point of the loop is to copy data from memory... How is that hard? |
| 01:40.38 | landley | It's one extra asm instruction. |
| 01:40.42 | prpplague | landley: right, even though thats simple enough, its just another item that has to be done |
| 01:40.53 | ds2 | i would have a hard time even finding a uImage smaller then 1M |
| 01:40.58 | prpplague | landley: you'd be surprise how much trouble people have trying the procedure for u-boot |
| 01:41.04 | landley | Most of these bootloaders implement serial support and a command line. |
| 01:41.16 | landley | But this is hard? |
| 01:42.24 | prpplague | landley: hehe suprisingly yes, or rather its just not worth the addtional work |
| 01:42.31 | landley | You already have to be able to read from the location in order to copy the data... |
| 01:42.41 | landley | So all the stuff about MMU is a side issue. That setup must be done anyway. |
| 01:42.41 | prpplague | landley: gotta run take my daughter to the doctors, be back in a couple hours |
| 01:42.45 | landley | Ok. |
| 01:43.01 | prpplague | landley: i'll grab the relevant code from u-boot and we can have a look |
| 01:43.06 | prpplague | bbiab |
| 01:43.09 | landley | ok |
| 01:43.22 | ds2 | dejavu? |
| 01:43.38 | ds2 | I thought Ijust saw that same thing a few hours ago... |
| 01:44.06 | prpplague | ds2: hehe, it's one of those after hours clinic;s we were waiting for them to call |
| 01:44.14 | prpplague | gone for real this time :P |
| 01:45.00 | ds2 | oh okay thought there is some weird monday effect going on |
| 02:02.19 | RobotGuy | prpplague: Got that about the bootloader. :) |
| 02:04.41 | RobotGuy | Oh carp, my kernel is 1.1 MB now. Got to fix that. |
| 02:05.30 | RobotGuy | I can just make more stuff be modules. No problem. |
| 02:06.07 | MonMotha | RobotGuy: what about fish? |
| 02:06.53 | RobotGuy | MonMotha: Ha ha ha |
| 02:15.33 | *** join/#edev GPSFan (n=kenm@12.10.255.246) |
| 02:15.33 | *** mode/#edev [+o GPSFan] by ChanServ |
| 02:15.58 | RobotGuy | OK, got my kernel under 1 MB now. That was easy. |
| 02:16.21 | RobotGuy | notify landley |
| 02:19.21 | landley | hmmm? |
| 02:19.48 | landley | So did initramfs boot for you? |
| 02:20.07 | landley | (Alas, it gives no real debugging output when it doesn't boot.) |
| 02:20.17 | landley | But I've debugged it so many times I've got a checklist of "things that can go wrong"... |
| 02:20.19 | RobotGuy | landley: Have not tried the initramfs yet. |
| 02:21.25 | RobotGuy | landley: I am sticking to the initrd for right now. Got some other things I want to try before initramfs. :D |
| 02:21.41 | landley | ok |
| 02:21.57 | RobotGuy | landley: I am not done tinkering with buildroot yet. ;) ;) |
| 02:26.22 | RobotGuy | I have my second robot almost fully assembled now. :D |
| 02:32.27 | *** join/#edev ccjoe (n=ccjoe@75-169-14-207.slkc.qwest.net) |
| 02:38.01 | RobotGuy | I just have to connect the six servos to the MCU board. |
| 02:38.09 | RobotGuy | And power to the board. |
| 02:44.35 | *** join/#edev hd_chro (n=rhg_xu@59.108.23.235) |
| 02:46.15 | *** join/#edev ico2__ (n=ico2@modem-636.chimpanzee.dialup.pol.co.uk) |
| 02:49.19 | *** join/#edev chouimat|away (n=dieu@kde/developer/chouinard) |
| 02:53.00 | *** join/#edev T0mW (n=Tom@24.238.86.158.res-cmts.sth.ptd.net) |
| 02:59.22 | *** part/#edev hd_chro (n=rhg_xu@59.108.23.235) |
| 03:13.43 | RobotGuy | Uh oh, he's baaaa-aaaaack.. :) |
| 03:25.08 | *** join/#edev _lifeless (n=patrick@90.150.113.209) |
| 03:29.50 | landley | prpplague: so you said the nail samples showed up? |
| 03:30.20 | *** part/#edev T0mW (n=Tom@24.238.86.158.res-cmts.sth.ptd.net) |
| 03:31.23 | prpplague | landley: no, not yet, they got shipped from the board house today |
| 03:31.51 | landley | Ah, that was it. |
| 03:31.59 | RobotGuy | Durn, I lost some of my config stuff - having to recreate it. |
| 03:32.07 | prpplague | landley: they should arrive at the office wed. |
| 03:32.09 | landley | Something to do with learning PHP, I suspect. |
| 03:32.12 | landley | Cool. |
| 03:32.25 | prpplague | landley: then a day or two for assembly and testing |
| 03:32.58 | ds2 | this is ripe for out of context quotes ;) |
| 03:33.22 | landley | ds2: most things are. |
| 04:06.31 | landley | prpplague: so how much would differ between hammer in cansat and hammer in nail? |
| 04:06.49 | prpplague | landley: tech wise or cost wise? |
| 04:07.00 | landley | Developing software on nail and then deploying on cansat. |
| 04:07.16 | landley | I'm partly thinking off nail as "introduce people to embedded development". |
| 04:07.30 | landley | But I'm also thinking of it as "easy development platform". |
| 04:07.39 | landley | Which implies deploying in other things. |
| 04:07.53 | prpplague | landley: ahh, yea, no problem using the nail as a platform to do dev for something like the cansat |
| 04:07.55 | landley | The ability to plug in arbitrary USB crap to the downstream end (with a hub even) means you can have any devices you want in the thing. |
| 04:08.06 | landley | And wire them up differently and swap the drivers around later, that's just a different kernel .config... |
| 04:08.11 | prpplague | landley: yep |
| 04:08.23 | landley | I'm also thinking in terms of what would be involved _documenting_ it. |
| 04:08.35 | prpplague | landley: plus with the small expansion header adding things like a real time clock or temperature sensor is very very very easy |
| 04:08.47 | landley | Not just "can this be done" but "here are the issues you bump into trying to develop on nail and deploy on a differently wired variant". |
| 04:08.53 | landley | Yup. |
| 04:08.54 | prpplague | landley: documentation is the real hard part imho |
| 04:08.58 | landley | Expanding the nail. |
| 04:09.07 | landley | Documentation is the one I can really sink my teeth into... |
| 04:09.42 | landley | You may have noticed that my normal development method involves getting lost and taking a chainsaw to the wilderness. |
| 04:09.51 | landley | And then at the end I wanna draw a darn map so it's not so painful in future. |
| 04:09.59 | landley | 'cause otherwise I'll just get lost again. |
| 04:10.06 | landley | (You have no idea how often I refer to my own documentation.) |
| 04:10.16 | landley | Why do you think I keep such a detailed development blog? |
| 04:10.23 | landley | How did I make this work last time, again? |
| 04:11.04 | prpplague | hehe indeed i do the same, just with an internal text log |
| 04:11.38 | RobotGuy | prpplague: So do I. I even refer to what I've added to the Hammer Wiki frequently. :) |
| 04:12.20 | RobotGuy | My entire website is not much more than a log of my experiences. |
| 04:12.31 | RobotGuy | Well, one of the 10 sites I have is. |
| 04:12.36 | prpplague | hehe |
| 04:13.24 | RobotGuy | I'm waiting for my ISP to get .MOBI configured in their system. I just picked up a .MOBI domain for my robots. :D |
| 04:14.17 | landley | The hard part of documentation is getting the overall organization and sequencing right. |
| 04:14.25 | prpplague | landley: thats where we different, i like the free form of wiki's |
| 04:14.28 | RobotGuy | landley: I am still just learning how to write on the Wiki. |
| 04:14.28 | landley | Piling up lots of disconnected notes is easy, but not _useful_. |
| 04:14.41 | prpplague | landley: right |
| 04:14.51 | landley | Wikis are good for jotting stuff down. Not so good at coming up with an actual _plot_ to the thing. |
| 04:14.51 | prpplague | landley: the problem with alot of wiki work is that it isn't connected |
| 04:14.52 | RobotGuy | I have been trying to organize my stuff on the Wiki. |
| 04:14.59 | prpplague | landley: good wiki pages have lots of links |
| 04:15.13 | landley | Cross-linking is no substitute for organization. |
| 04:15.40 | prpplague | landley: hehe, indeed |
| 04:16.15 | landley | Until you've got an overall structure to the documentation, you can't even see where the holes are. |
| 04:16.24 | RobotGuy | prpplague hasn't slapped me for my Wiki work yet, so I must be doing OK. :) |
| 04:16.50 | landley | Wikis are a resource, but you mine them for raw materials from which to produce real documentation. |
| 04:17.08 | RobotGuy | Wikis can be organized though. |
| 04:17.47 | RobotGuy | I need to figure out how to title my pages rather than have the titles the Wiki software assigns. |
| 04:18.22 | RobotGuy | It's MediaWiki, right> |
| 04:22.53 | prpplague | landley: indeed |
| 04:23.28 | prpplague | landley: i'm hope'n that between you and me we can get some decent docs as well as some straight forward hardware for people to learn on |
| 04:23.48 | landley | prpplague: I'm waiting for a nail board to document. |
| 04:24.10 | landley | I'm also waiting for the kernel patches so I can not only push them upstream but get a reproducible kernel install from vanilla sources, so I can document that. |
| 04:24.26 | landley | I tend to prefer not to document "here's how you shouldn't do it". I prefer to _fix_ and then document it when it's working right. |
| 04:24.40 | landley | Personal idiosyncrasy. :) |
| 04:25.11 | prpplague | landley: hehe |
| 04:25.39 | landley | It does mean that I have the annoying tendency to go off on _enormous_ tangents... |
| 04:25.44 | prpplague | landley: yea i suspect that we'll probably be shipping you one a week from today |
| 04:25.49 | landley | Coolness. |
| 04:25.54 | prpplague | landley: you need an extra hammer module? |
| 04:26.07 | landley | Does the one in my current board pop off and go into this easily? |
| 04:26.34 | RobotGuy | landley: It comes out of the socket and goes back in very easily. |
| 04:27.01 | *** mode/#edev [+o prpplague] by ChanServ |
| 04:27.06 | RobotGuy | I take mine out of the socket when I want to solder to the carrier board. |
| 04:27.13 | landley | Gave it a yank and it didn't want to come off. Is there a lever or something? |
| 04:27.37 | landley | When hardware breaks, you can't restore from a backup. :( |
| 04:27.42 | landley | And I break _everything_. |
| 04:27.49 | RobotGuy | landley: Gently use the tip of a flat blade screwdriver on one end of the module. |
| 04:28.15 | RobotGuy | Pry gently upward and it will come right out. |
| 04:28.25 | landley | HA! |
| 04:28.27 | landley | Yes, that was it! |
| 04:28.51 | RobotGuy | If I can do it, IT IS EASY and safe. :D |
| 04:28.52 | landley | To fix mips, revert http://kernel.org/hg/linux-2.6/rev/74258 |
| 04:28.54 | landley | And then it boots. |
| 04:29.05 | prpplague | landley: yea its removable, but probably in your case, its best if we send you a spare, hehe |
| 04:29.51 | landley | I'm all for it. Thanks. :) |
| 04:29.59 | RobotGuy | I keep breaking the durned power terminals on my Lynxmotion bot board and SSC-32 though, but never have that problem with the terminals on my Sabertooth motor controller. |
| 04:30.26 | landley | The nail being a sealed piece of plastic is great for me. |
| 04:30.37 | landley | It's nice that I can open it up and look at the guts, but also nice that I don't _have_ to. |
| 04:31.09 | landley | Although presumably I can hook up the flyswatter to it to have two jtag and two serial port thingies? |
| 04:31.14 | landley | Or will the nail already be using that header? |
| 04:31.20 | RobotGuy | prpplague: You should check out what Lynxmotion is doing with the new SSC-NG board. They posted some pictures of it today and the designer has described its features. |
| 04:32.24 | prpplague | landley: no, you'll only be able to use the jtag on the nail board as well as the serial console, you can't use the flyswatter in conjuction with the nail board |
| 04:32.57 | prpplague | landley: just a note, the nail board we will send you won't have the plastic case yet, it will another 3-4 weeks for the plastics to arrive |
| 04:33.37 | landley | Er... |
| 04:33.51 | landley | So... how does one use it...? |
| 04:34.07 | landley | Wrap it in... paper towels? Duct tape? Plastic wrap? |
| 04:34.20 | RobotGuy | prpplague: http://www.lynxmotion.net/phpbb/viewtopic.php?t=3362 |
| 04:34.26 | landley | I have bubble wrap. |
| 04:34.28 | landley | Would bubble wrap help? |
| 04:35.30 | prpplague | landley: hehe, if you are worried, i'll find _something_ to put it in, hehe |
| 04:35.48 | landley | I'd like to be able to throw it in my backpack and take it along with me to the coffee shop. |
| 04:35.49 | prpplague | landley: lead time on plastics generally are in the range of 4-6 weeks |
| 04:35.56 | prpplague | landley: indeed |
| 04:36.03 | prpplague | landley: i'll dig up something |
| 04:36.05 | landley | I've done that to bare electronics. It was an unpleasant experience for both of us. |
| 04:36.11 | prpplague | landley: hehe, np |
| 04:36.15 | landley | Thanks. |
| 04:36.35 | landley | Possibly even power user. |
| 04:36.36 | prpplague | want me to paint some flames on it ? |
| 04:36.39 | landley | But developer? Not so much... |
| 04:36.52 | landley | prpplague: nah, it shouldnt' need to go faster. :) |
| 04:36.59 | landley | And we don't _want_ it to run hot... |
| 04:37.05 | RobotGuy | With a keyboard, wifi adapter, and nice LCD, you could have a small wifi terminal. |
| 04:37.22 | landley | robotguy: yup. |
| 04:37.36 | landley | You could even tape the electronics to the back of a flatscreen monitor, pretty easily. |
| 04:37.46 | RobotGuy | landley: Definitely |
| 04:38.02 | RobotGuy | Nail Board could even be used for robotics. |
| 04:38.24 | prpplague | RobotGuy: yea |
| 04:38.25 | landley | There should be a page on tin can tools listing uses that people have put hammer modules to. |
| 04:38.35 | RobotGuy | I bet one would fit on my Lynxmotion BRAT. |
| 04:38.39 | landley | Various dumb terminals. |
| 04:38.41 | landley | Robotics. |
| 04:38.45 | landley | Routers. |
| 04:38.52 | prpplague | landley: yea, we were still collecting users before we started that |
| 04:39.34 | RobotGuy | I've thought about turning a Hammer into a router. |
| 04:39.40 | landley | Ooh, there's a sick idea... |
| 04:39.47 | RobotGuy | Or at least a mini-firewall. |
| 04:39.48 | landley | You know how you can have a software defined USB device? |
| 04:40.05 | landley | To make it look like a USB disk? |
| 04:40.13 | landley | And you know how laptops can boot from a USB disk? :) |
| 04:40.14 | RobotGuy | Yep |
| 04:40.21 | landley | Now imagine working encryption into this... |
| 04:40.28 | landley | Using a nail board as more than just a boot dongle... |
| 04:40.47 | landley | The data is on the laptop's hard drive, but the encryption logic is in an initramfs supplied by the nail. |
| 04:40.57 | RobotGuy | You could fit a Hammer with an SD/MMC slot into a very small package. |
| 04:41.03 | landley | But you've got to log into the nail so it decrypts itself and lets you boot the rest of the way... :) |
| 04:41.19 | landley | And you can't just passively dump the USB image to do it... :) |
| 04:41.21 | prpplague | landley: nice idea |
| 04:41.41 | landley | I suspect there are some security companies that would like that. |
| 04:41.45 | RobotGuy | And, I have my eye on a 4 GB thumb drive, big brother to the 2 GB drive I have now. |
| 04:41.52 | prpplague | and some hackers |
| 04:41.56 | landley | Especially if the case had some anti-tampering logic that blanked the flash if it was opened... |
| 04:42.12 | prpplague | my bad |
| 04:42.16 | prpplague | s/hackers/crackers |
| 04:42.28 | prpplague | landley: so used to using that word around laypersons |
| 04:42.38 | RobotGuy | Yes, please don't call the bad guys hackers. |
| 04:42.48 | landley | I call the bad guys hackers. |
| 04:42.58 | landley | This is probably the biggest argument I have with Eric Raymond that doesn't involve libertarianism. |
| 04:43.05 | prpplague | hehe |
| 04:43.09 | landley | Nobody under 30 says "hacker" to mean anything other than "computer criminal". |
| 04:43.15 | prpplague | indeed |
| 04:43.19 | landley | Or at least people who break into systems. |
| 04:43.24 | landley | That battle's long lost. |
| 04:43.48 | RobotGuy | Well, if hackers are bad people, then I am not one. |
| 04:43.49 | landley | The younger generation took the word "geek" and made it our own. |
| 04:44.25 | landley | (I'm old enough to be in the "hacker" crowd, but as an old 8-bit kid I'm definitely on the "geek" rather than "hacker" side.) |
| 04:44.52 | RobotGuy | I remember before the term hacker was even used widely. |
| 04:44.53 | prpplague | hehe |
| 04:45.07 | prpplague | landley: i beat, heading to bed, got a long day tomorrow |
| 04:45.09 | landley | I'm happy to give up the word "hacker" as lost. The older generation will defend it to the death, but already know it won't outlive them. |
| 04:45.13 | landley | prpplague: good night. |
| 04:45.14 | RobotGuy | I have a milestone birthday this year. |
| 04:45.27 | landley | I turn... 36? What year is this? |
| 04:45.29 | landley | Yup, 36. |
| 04:45.32 | *** join/#edev DrZimmerman (n=theo@2001:5c0:8fff:fffe:0:0:0:6af1) |
| 04:45.37 | RobotGuy | G'Night prpplague |
| 04:45.42 | prpplague | landley: wow i thought you were younger than that |
| 04:45.50 | landley | Nah, just immature. |
| 04:45.59 | RobotGuy | I'm much older than prpplague |
| 04:46.04 | prpplague | hehe |
| 04:46.10 | prpplague | later folks |
| 04:48.34 | RobotGuy | I have to recreate all my startup stuff for my Hammer. |
| 04:49.13 | RobotGuy | I had it all setup to init the gadget ethernet device and everything. |
| 05:06.41 | Madsy | Hacker vs Cracker discussion again.. ? |
| 05:07.33 | *** join/#edev ico2__ (n=ico2@modem-636.chimpanzee.dialup.pol.co.uk) |
| 05:09.18 | Mirell | Mmm? |
| 05:09.25 | RobotGuy | landley: I feel old now. |
| 05:10.07 | Madsy | A hacker is obviously someone that is proficient in a certain field. Not that the media agrees on the term.. |
| 05:10.08 | Mirell | RobotGuy: I'm 22. |
| 05:10.12 | landley | Why do you feel old? |
| 05:10.13 | Madsy | s/that/who |
| 05:10.25 | RobotGuy | Mirell: Now I feel really old! :P |
| 05:10.44 | landley | Mirell was 18 when I met him... |
| 05:10.47 | RobotGuy | Because I seem to be the oldest here so far right now. |
| 05:10.53 | landley | He recently started his first real job. |
| 05:10.59 | RobotGuy | I know. |
| 05:11.02 | Madsy | Mirell: I think we are the last generation who knows what a BBS is. |
| 05:11.07 | Mirell | Madsy: Hehe |
| 05:11.09 | Madsy | Teenagers nowadays.. gah |
| 05:11.15 | Mirell | Madsy: I had my first e-mail at age 5. |
| 05:11.15 | landley | And is still surprised that ~3 weeks in he still has very little actual _work_ to do... :) |
| 05:11.17 | RobotGuy | He's a new wire, er I mean hire. |
| 05:11.23 | Mirell | Madsy: Which is nothing now, but that was in 1990. |
| 05:11.34 | landley | Madsy: no no, it's "kids theesh days..." |
| 05:11.37 | landley | Smack your gums while you do it. |
| 05:11.42 | landley | If you have a walking stick, shake it. |
| 05:11.43 | Madsy | haha |
| 05:11.48 | Madsy | Get of my lawn! |
| 05:11.53 | landley | Exactly! |
| 05:11.58 | Mirell | Madsy: Although actually, I never used BBSes. |
| 05:12.07 | Mirell | Madsy: Always had the Internet. |
| 05:12.08 | RobotGuy | Go fetch my robot, ya young whipper snapper! |
| 05:12.09 | landley | Mirell: I wrote three of 'em. |
| 05:12.22 | landley | What _is_ a whipper snapper? |
| 05:12.25 | RobotGuy | I have actually RUN a BBS. |
| 05:12.27 | Mirell | Madsy: Go Go University Systems Programmer for a dad. ;) |
| 05:12.28 | landley | What kind of whip, and how does one snap it? |
| 05:12.36 | Madsy | I did. My father had this blazingly fast 14400 baud modem back in 92. We used to download bad Doom clones from the states. |
| 05:12.49 | RobotGuy | I remember when 300 was fast. |
| 05:12.52 | landley | 300 baud modem plugged into a commodore 64. |
| 05:13.08 | Mirell | Pshaw. |
| 05:13.11 | Mirell | We had a VAX. |
| 05:13.28 | RobotGuy | My first online experience was with an ASR-33 TTY connected via a 300 BPS accoustic coupler to a local colledge computer. |
| 05:13.44 | landley | I used a horrible vax unix clone called "Eunice" via my father's dialin account at GE. |
| 05:13.51 | landley | It had amiga software on it, for download. |
| 05:13.55 | RobotGuy | This computer was an HP 3000E. |
| 05:13.58 | landley | Did I mention it was horrible? |
| 05:14.13 | RobotGuy | landley: Memory is the first to go.. :) |
| 05:14.43 | landley | RobotGuy: no it isn't, but when the memory _does_ go you forget the stuff that goes before that. |
| 05:15.03 | landley | My memory already kinda sucked when I was 7... |
| 05:15.20 | MonMotha | I cannot say I ever actually used an acoustic coupler |
| 05:15.28 | MonMotha | I did have a 2400bps modem, though |
| 05:15.30 | Madsy | On another note, LEGO is 50 years old now. |
| 05:15.32 | Madsy | http://cache.gizmodo.com/assets/resources/2008/01/lego-brick4-timeline.jpg |
| 05:15.34 | RobotGuy | I remember programming in FORTRAN IV when I was barely out of High School. |
| 05:16.07 | RobotGuy | I want Lego Mindstorms, about 5 sets should work good. |
| 05:16.22 | Madsy | I started out in QBasic. At that time, I thought it was the only way to make (awful) applications. |
| 05:16.32 | RobotGuy | Do you remember CP/M? |
| 05:16.37 | Madsy | It took four years before I even learned about C. |
| 05:17.35 | RobotGuy | I programmed in Applesoft and Apple Pascal |
| 05:18.04 | RobotGuy | In fact, I converted a bunch of Applesoft BASIC programs to Apple Pascal for a book. |
| 05:18.32 | RobotGuy | Geesh, I am almost as old as LEGO. |
| 05:18.43 | Madsy | RobotGuy: What a comparsion :-D |
| 05:21.34 | Madsy | I recently found one of the old games my dad used to play, while I was hidden behind the chair, scared shitless. |
| 05:21.38 | Madsy | Catacomb Abyss 3D |
| 05:22.42 | Madsy | He got a GUS very early, and hooked up his band equipment to the PC. Two badarse speakers which were taller than me. |
| 05:23.13 | Madsy | I think I can still hear the growls from the Doom monsters in my head. |
| 05:34.18 | RobotGuy | I remember ZORK, best text adventure I think there ever was. |
| 05:49.08 | Mirell | Oh, that reminds me... |
| 05:53.34 | RobotGuy | What's a Frotz? |
| 06:00.09 | Mirell | RobotGuy: Infocom/whatnot emulator |
| 06:06.15 | RobotGuy | I should have known, with a name like Frotz. |
| 06:06.59 | RobotGuy | Well, I think I will exit for awhile. Bye. |
| 06:23.40 | Mirell | Hmm... |
| 06:40.14 | *** join/#edev davidc__ (n=davidc__@S01060002b360aacd.vc.shawcable.net) |
| 06:49.34 | *** join/#edev steliosk (n=Stelios@85.72.110.56) |
| 07:09.15 | *** join/#edev ico2__ (n=ico2@modem-1776.cheetah.dialup.pol.co.uk) |
| 07:09.34 | *** join/#edev chouimat|busy (n=dieu@r2351064.cidc.net) |
| 08:02.50 | *** join/#edev harpal (n=harpal@124.125.76.8) |
| 09:09.08 | *** join/#edev ico2__ (n=ico2@modem-35.cougar.dialup.pol.co.uk) |
| 09:29.28 | *** join/#edev lyakh (n=lyakh@p57BD27D6.dip0.t-ipconnect.de) |
| 09:58.58 | *** join/#edev hw_ (n=hw@p578b3905.dip0.t-ipconnect.de) |
| 10:03.32 | *** join/#edev Madsy (n=mads@48.80-202-23.nextgentel.com) |
| 10:17.04 | *** join/#edev steliosk_ (n=Stelios@athedsl-288139.home.otenet.gr) |
| 11:09.41 | *** join/#edev ico2_ (n=ico2@modem-3268.chimpanzee.dialup.pol.co.uk) |
| 11:14.00 | *** join/#edev steliosk (n=Stelios@athedsl-288139.home.otenet.gr) |
| 11:37.08 | harpal | Hey I have made one system and in that I am just synchronize my time with ntpdate but its taking UTC I want it to be IST, What should i do? |
| 11:39.43 | SpeedEvil | unix systems generally should always run on UTC |
| 11:40.00 | SpeedEvil | You set teh timezone so that it displays your local time |
| 11:42.51 | *** join/#edev steliosk_ (n=Stelios@athedsl-105085.home.otenet.gr) |
| 11:53.03 | *** join/#edev howtoo (n=howtoo@195.159.62.66) |
| 11:56.11 | *** join/#edev dijenerate (n=dijenera@66.54.118.16) |
| 12:02.04 | harpal | SpeedEvil: I ahve set my time zone in /etc/timezone file |
| 12:02.22 | harpal | but still its not dispaying my time |
| 12:20.40 | *** join/#edev steliosk (n=Stelios@athedsl-105085.home.otenet.gr) |
| 12:24.04 | *** join/#edev emil|k (n=emil@85.179.239.4) |
| 12:30.31 | Mirell | Mmm. |
| 12:54.57 | harpal | SpeedEvil: Thanks a lot i ahve solved that problem |
| 12:56.37 | SpeedEvil | Great - sorry I wasn't of more help - I realised after your comment that I haven't actually set mine up as I'm on GMT for a few months yet |
| 13:10.25 | *** join/#edev ico2_ (n=ico2@modem-3268.chimpanzee.dialup.pol.co.uk) |
| 13:23.30 | SpeedEvil | anyone gutted any tomtom's? |
| 13:29.51 | *** join/#edev bbradley (n=bbradley@82-32-35-42.cable.ubr05.azte.blueyonder.co.uk) |
| 13:35.02 | Mirell | Sweet, got suspend-to-ram and suspend-to-disk working. |
| 13:44.20 | *** join/#edev sjhill (n=sjhill@real.realitydiluted.com) |
| 13:46.06 | *** join/#edev prpplague (n=dave@mail.americanmicrosystems.com) |
| 13:46.19 | *** join/#edev dijenerate (n=dijenera@69.73.221.13) |
| 13:52.58 | prpplague | ho diddly ho ho |
| 14:12.51 | *** join/#edev jayna (n=chenyun-@125-238-109-42.broadband-telecom.global-gateway.net.nz) |
| 14:14.46 | *** join/#edev GPSFan (n=kenm@12.10.255.246) |
| 14:14.46 | *** mode/#edev [+o GPSFan] by ChanServ |
| 14:24.37 | *** join/#edev chouimat|work (n=mchouina@209.217.106.98) |
| 14:26.54 | chouimat|work | morning |
| 14:44.48 | prpplague | chouimat|work: greetings |
| 14:44.53 | prpplague | GPSFan: http://www.elinux.org/Hammer_LED_Driver |
| 14:44.57 | prpplague | GPSFan: hows that look? |
| 14:46.13 | GPSFan | prpplague: looksok, is there a /proc interface aslo? |
| 14:47.00 | prpplague | GPSFan: no proc entry, trying not to encourage people to use deprecated methods |
| 14:47.57 | GPSFan | prpplague: k, then that's also a good way to demonstrate adding new platform entries. |
| 14:48.36 | GPSFan | prpplague: have you looked at the s3c2410 support in kernel 2.6.24? lots of stuff seems to have changed. |
| 14:48.50 | *** join/#edev sunrunner20 (n=jamesdoe@pool-96-226-38-56.dllstx.fios.verizon.net) |
| 14:49.04 | SpeedEvil | Variable clock rate support is also being worked on |
| 14:49.14 | prpplague | GPSFan: yea, i have a copy that i'm running on a s3c2440 |
| 14:49.19 | SpeedEvil | which should help a fair bit with power. |
| 14:50.30 | GPSFan | prpplague: I tried porting the hammer patches to 2.6.24, and got most applied, a few were not needed, but when I compiled the kernel, all sorts of .h file locations have changed as well as their contents. |
| 14:51.53 | prpplague | GPSFan: yea the patch i have includes a bunch of fixes that were done in 2.6.24 |
| 14:52.00 | prpplague | GPSFan: most of the patch can be ignored |
| 14:52.14 | prpplague | GPSFan: i'll generate a 2.6.24 in a few days |
| 14:53.11 | GPSFan | prpplague: cook, there were just too many undefined references for me to tackle at once. but it looks like hammer specific stuff will be a snap. |
| 14:53.18 | GPSFan | cool even. |
| 14:53.26 | *** join/#edev RobotGuy (n=robotguy@dsl093-038-072.pdx1.dsl.speakeasy.net) |
| 14:53.46 | prpplague | GPSFan: yea, most of the stuff that probably failed are the items i fixed in 2.6.22 that are now fixed in 2.6.24 |
| 14:53.53 | RobotGuy | Good Morning! |
| 14:54.06 | RobotGuy | prpplague: I am having trouble getting usbfs mounted. I can |
| 14:54.13 | RobotGuy | not find the correct way. |
| 14:55.15 | RobotGuy | I lost all my setup I had for Hammer. :( Have to recreate it all. |
| 14:55.16 | prpplague | RobotGuy: mount -t usbfs usbfs /proc/bus/usb |
| 14:55.45 | RobotGuy | prpplague: Hmm, that's what I have in my inittab but it is not getting mounted. Maybe I am missing something in my kernel. |
| 14:56.04 | RobotGuy | I have it right after the proc mount in /etc/inittab |
| 14:56.12 | sjhill | RobotGuy: did you actually configure usbfs into your kernel |
| 14:56.28 | RobotGuy | prpplague: I believe so, but I have to check to be sure. |
| 14:56.30 | prpplague | RobotGuy: is proc mounted? |
| 14:56.40 | SpeedEvil | can you mount it manually RobotGuy ? |
| 14:57.04 | prpplague | RobotGuy: check to make sure proc is mounted first |
| 14:57.09 | sjhill | or cat /proc/filesystems to see if usbfs is in there |
| 14:57.11 | sjhill | sheesh |
| 14:57.29 | RobotGuy | prpplague: proc is mounted |
| 14:57.48 | RobotGuy | prpplague: usbfs is configured in my kernel |
| 14:58.22 | RobotGuy | sjhill: What is the problem? |
| 14:58.33 | prpplague | RobotGuy: so if you issue the command manually what happens? |
| 14:58.43 | RobotGuy | prpplague: It fails |
| 14:58.51 | prpplague | RobotGuy: with what error? |
| 14:59.00 | SpeedEvil | Does the usb drive show up in lsusb? |
| 14:59.11 | SpeedEvil | and what's the error message on failure |
| 14:59.24 | SpeedEvil | you need scsi and sd_mod installed usually |
| 14:59.33 | RobotGuy | SpeedEvil: I know that. |
| 14:59.45 | *** part/#edev harpal (n=harpal@124.125.76.8) |
| 14:59.46 | RobotGuy | SpeedEvil: That has nothing to do with mounting usbfs though. |
| 14:59.48 | SpeedEvil | Just covering the stupid stuff. |
| 14:59.56 | SpeedEvil | oh - sorry, my fault |
| 15:00.00 | RobotGuy | SpeedEvil: I understand, and thanks. :) |
| 15:01.13 | prpplague | RobotGuy: what error does respond with? |
| 15:02.31 | RobotGuy | prpplague: http://www.pastebin.ca/878773 |
| 15:03.01 | RobotGuy | SpeedEvil: No need to be sorry. It is easy to forget the obvious stuff. |
| 15:03.14 | prpplague | RobotGuy: ok, so does /proc/bus/usb exist? |
| 15:03.25 | RobotGuy | prpplague: The dir exists |
| 15:03.44 | SpeedEvil | IIRC that's the behaviour if you have no working root hub |
| 15:03.52 | SpeedEvil | ignore that |
| 15:03.55 | SpeedEvil | no it's not |
| 15:04.14 | prpplague | RobotGuy: ok and what is the output of cat /proc/filesystems ? |
| 15:05.00 | SpeedEvil | mount -t usbfs /proc/bus/usb/ /proc/bus/usb |
| 15:05.22 | SpeedEvil | works for me to go from an empty /proc/bus/usb directiory on my neo to a populated one |
| 15:05.34 | GPSFan | RobotGuy: is / mounted rw? |
| 15:05.55 | RobotGuy | prpplague: There is no /proc/bus dir! I guess I have to create that, so added it to my devices file. |
| 15:06.17 | RobotGuy | prpplague: I will also add /proc/bus/usb |
| 15:06.31 | RobotGuy | prpplague: I don't remember having to create those. |
| 15:07.03 | GPSFan | RobotGuy: you don't, they are created by mounting /proc |
| 15:07.52 | prpplague | doh |
| 15:07.52 | SpeedEvil | or by loading a usb module with usb device filesystem enabled in (/proc/bus/usb) |
| 15:07.54 | prpplague | <prpplague> RobotGuy: ok, so does /proc/bus/usb exist? |
| 15:07.54 | prpplague | <RobotGuy> prpplague: The dir exists |
| 15:07.56 | *** join/#edev RobotGuy (n=robotguy@dsl093-038-072.pdx1.dsl.speakeasy.net) |
| 15:08.02 | prpplague | ~lart RobotGuy |
| 15:08.02 | purl | whips out his power stapler and staples RobotGuy's foot to the floor |
| 15:08.03 | GPSFan | prpplague: I had a problem similar to this, because / was not being remounted rw correctly. I had to move the order around in inittab. |
| 15:08.31 | prpplague | GPSFan: correct, which is why i asked him if proc was mounted |
| 15:08.39 | prpplague | GPSFan: didn't i ask if that directory existed? |
| 15:08.42 | RobotGuy | prpplague: proc IS mounted |
| 15:09.07 | prpplague | RobotGuy: show me a paste of the results of the mount command with no args |
| 15:09.58 | GPSFan | if ls /proc isn't full os stuff this is probably the problem. |
| 15:10.10 | *** join/#edev ico2_ (n=ico2@modem-2174.cougar.dialup.pol.co.uk) |
| 15:10.18 | RobotGuy | prpplague: http://www.pastebin.ca/878814 |
| 15:10.30 | RobotGuy | GPSFan: proc has lots of stuff |
| 15:11.55 | prpplague | RobotGuy: sounds like you are missing something in your kernel if you don't have a /proc/bus/usb entry |
| 15:12.02 | GPSFan | RobotGuy: good, thsi is called diagnostics. you may know what's going on on your system, but we don't, ans often have to ask some very stupid/obvious sounding questions to help educate ourselves. |
| 15:12.20 | RobotGuy | GPSFan: I understand. |
| 15:12.56 | GPSFan | car /proc/filesystems |
| 15:12.58 | RobotGuy | GPSFan: I don't have any problems with that and give the info that is asked for. :) |
| 15:13.10 | GPSFan | cat even |
| 15:14.31 | RobotGuy | prpplague and GPSFan: http://www.pastebin.ca/878838 |
| 15:15.17 | prpplague | RobotGuy: no usbfs listed |
| 15:15.21 | RobotGuy | I do have usbfs configured in my kernel. |
| 15:15.24 | RobotGuy | prpplague: I know. |
| 15:16.30 | prpplague | GPSFan: http://www.elinux.org/Hammer_Buzzer_Driver |
| 15:16.32 | SpeedEvil | Is it modular?The directroy is not created till the USB module is loaded, if it's modular |
| 15:16.57 | RobotGuy | prpplague: Here's my inittab - http://www.pastebin.ca/878851 |
| 15:17.27 | prpplague | RobotGuy: i'd rather see your kernel config |
| 15:17.36 | RobotGuy | SpeedEvil: I have a modular kernel, but try to have the main stuff right in the kernel. |
| 15:19.24 | RobotGuy | prpplague: Kernel Config - http://www.pastebin.ca/878864 |
| 15:20.31 | prpplague | RobotGuy: you have the host controller as a module |
| 15:20.37 | prpplague | RobotGuy: is the module loaded? |
| 15:20.46 | RobotGuy | This time I am going to backup my stuff. |
| 15:21.06 | RobotGuy | prpplague: No modules are being loaded. I have kernel auto load modules enabled |
| 15:21.17 | SpeedEvil | ah |
| 15:21.28 | prpplague | RobotGuy: no host controller, so why would you expect the usbfs to be supported? |
| 15:23.53 | RobotGuy | prpplague: I usually have that IN the kernel, and don't know why I had it as a module this time. |
| 15:24.22 | *** join/#edev CosmicPenguin (n=nobody@163.181.251.103) |
| 15:24.22 | *** mode/#edev [+o CosmicPenguin] by ChanServ |
| 15:25.05 | prpplague | RobotGuy: hehe, and as usually you have a thousand modules you don't need, hehe |
| 15:25.09 | RobotGuy | I think I will just table all this until Friday. I ran out of one of my medicines. :( |
| 15:25.21 | RobotGuy | prpplague: That is according to you. :) |
| 15:25.42 | prpplague | RobotGuy: thats according to the guy who makes a living doing embedded dev, hehe |
| 15:26.13 | RobotGuy | prpplague: I understand and respect that. However, I have the stuff in my kernel I want to be there now and that I will be tinkering with. |
| 15:26.56 | RobotGuy | prpplague: I don't want to have to rebuild my kernel everytime I want to tinker with something different. ;) |
| 15:27.57 | prpplague | RobotGuy: don't have to, just build the modules you need, but anyway, doing so many options just invited problems when trying to debug |
| 15:29.01 | RobotGuy | prpplague: I think I will not be here for a few days. I won't have any more of the medicine I ran out of until Friday. |
| 15:29.34 | prpplague | RobotGuy: doh |
| 15:29.45 | prpplague | RobotGuy: can't call it in to a Rx ? |
| 15:30.00 | RobotGuy | prpplague: I am broke right now. :( |
| 15:31.12 | RobotGuy | I'll be back Friday or Saturday. Bye for now. |
| 15:31.20 | *** join/#edev redb1ue (i=star@ppp227.108-253-207.mtl.mt.videotron.ca) |
| 15:32.41 | prpplague | RobotGuy: okie dokie |
| 15:35.31 | *** join/#edev T0mW (n=Tom@24.238.86.158.res-cmts.sth.ptd.net) |
| 15:35.57 | T0mW | CosmicPenguin: you there? |
| 15:36.00 | CosmicPenguin | yes |
| 15:36.17 | T0mW | CosmicPenguin: should I buy a Phenom 9600 or wait? |
| 15:40.32 | prpplague | GPSFan: http://www.elinux.org/Hammer_Board_Example_Kernel_Drivers |
| 15:42.51 | *** join/#edev roxfan (n=dunno@246.213-200-80.adsl-dyn.isp.belgacom.be) |
| 15:44.45 | *** join/#edev dijenerate (n=dijenera@66.54.118.16) |
| 15:59.02 | GPSFan | prpplague: that looks nice, you don't always get that kind of HW/SW support from board vendors, it's just "use our configs and pay for what you don't understand" |
| 15:59.19 | GPSFan | s/pay/pay more |
| 15:59.52 | prpplague | GPSFan: indeed, thats what we want to avoid |
| 16:00.51 | prpplague | GPSFan: i need to add some info on the buzzer config |
| 16:01.08 | prpplague | GPSFan: and also i need to add the configuration for doing audio via pwm |
| 16:01.12 | GPSFan | prpplague: I think Robotguy's main problem is trying to do too much at once, and expecting everything to "just work" while ignoring many details that usually bite people in the a$$ |
| 16:01.22 | prpplague | GPSFan: indeed |
| 16:01.31 | prpplague | GPSFan: i've tried to talk to him about that |
| 16:01.48 | prpplague | GPSFan: any other items you can think of? |
| 16:01.59 | prpplague | GPSFan: i need to add a section on the lcd stuff, and the enc28j60 |
| 16:05.53 | GPSFan | prpplague: Stuff I think the wiki should touch on: led, switch, buzzer, spi, i2c, keyboard (matrix using gpios or an external pic via spi/i2c), lcd (simple char & graphics), sound (via usb), ethernet, wireless (via usb), usb (using client port as a host, or usb-networking) |
| 16:06.38 | GPSFan | bbl.. |
| 16:07.31 | SpeedEvil | GPSFan: bluetooth (over usb) |
| 16:07.38 | SpeedEvil | Though that's a whole new can of worms. |
| 16:09.16 | prpplague | GPSFan: okie dokie |
| 16:10.38 | SpeedEvil | Has anyone pulled apart a tomtom one? |
| 16:13.45 | *** join/#edev lyakh (n=lyakh@p57BD27D6.dip0.t-ipconnect.de) |
| 16:18.26 | landley | And lo, I hath shipped Firmware Linux 0.3.1. |
| 16:18.27 | landley | So there. |
| 16:19.22 | prpplague | jeje |
| 16:19.25 | prpplague | hehe |
| 16:27.05 | *** join/#edev dijenerate (n=dijenera@72.51.91.199) |
| 16:30.42 | sunrunner20 | just make sure |
| 16:30.47 | sunrunner20 | there are no time bombs in it >_> |
| 16:38.54 | *** join/#edev FireEgl (i=FireEgl@Sebastian.FireEgl.Com) |
| 16:47.51 | *** join/#edev ico2_ (n=ico2@modem-2174.cougar.dialup.pol.co.uk) |
| 17:10.33 | *** mode/#edev [+o prpplague] by ChanServ |
| 17:15.32 | *** join/#edev dijenerate (n=dijenera@72.51.91.199) |
| 17:28.21 | *** join/#edev DrZimmerman (n=theo@2001:5c0:8fff:fffe:0:0:0:6af1) |
| 17:30.30 | prpplague | Crofton: ping |
| 17:31.54 | landley | sunrunner20: time bombs? |
| 17:32.10 | prpplague | landley: unaddressed issues |
| 17:32.18 | landley | prpplague: ping lwn@lwn.net as well. |
| 17:32.32 | landley | Any specific unaddressed issues he was referring to? (In FWL?) |
| 17:32.35 | *** join/#edev steliosk_ (n=Stelios@athedsl-28944.home.otenet.gr) |
| 17:32.41 | prpplague | landley: yea to the fwl release |
| 17:33.07 | landley | Haven't done the 32-bit hosted build yet, so the cross compilers for the x86 host are still from the previous version... |
| 17:33.12 | prpplague | landley: i'll give lwn a ping, but the general response has been, if it aint' involving linux or OSS , we aren't interested |
| 17:33.23 | landley | And I hit a darn "internal compiler error" trying to build uClibc for m68k. :P |
| 17:33.32 | landley | Time to upgrade to gcc 4.2.2 I guess... |
| 17:43.04 | prpplague | doh |
| 17:54.49 | prpplague | Crofton: ping |
| 17:57.24 | landley | prpplague: I mentioned I break stuff. |
| 17:57.49 | prpplague | landley: indeed |
| 18:11.10 | Crofton | prpplague, pong |
| 18:11.42 | prpplague | Crofton: rtty |
| 18:11.50 | prpplague | Crofton: i assume you are familiar with that |
| 18:12.12 | prpplague | Crofton: correct? |
| 18:13.19 | Crofton | I know what it sonds like :) |
| 18:13.37 | Crofton | not super familiare with the demod though |
| 18:13.54 | Crofton | although I suspect brute force methods are easy enough .... |
| 18:14.34 | prpplague | Crofton: yea i was curious if you thought the hammer would be capable of decoding rtty |
| 18:14.53 | prpplague | Crofton: if we use the adc to sample the audio level |
| 18:14.56 | Crofton | as long as you feed the data in at audio rate |
| 18:15.12 | Crofton | or low data radte |
| 18:15.20 | prpplague | Crofton: rtty is 45 baud |
| 18:15.30 | prpplague | Crofton: some are up to 75 baud |
| 18:15.35 | Crofton | there are lots of people who feed audio from a receiver into a PC via sound card and run demod |
| 18:15.41 | Crofton | yeayh |
| 18:15.55 | prpplague | Crofton: yea, i was hope'n to do without the audio codec |
| 18:16.01 | prpplague | Crofton: just use the adc |
| 18:16.03 | Crofton | there are like two tones a few hundrend hertz apart |
| 18:16.12 | Crofton | I suspect that should work |
| 18:16.18 | Crofton | brb |
| 18:19.05 | Crofton | can you control the sample rate? |
| 18:19.27 | prpplague | yea |
| 18:19.43 | Crofton | where are you getting the input from? |
| 18:19.55 | prpplague | let me check what the best grainularity is |
| 18:20.13 | prpplague | Crofton: don't know yet, just researching |
| 18:20.43 | Crofton | ok |
| 18:21.13 | prpplague | Crofton: 10 bit resolution at about 500KSPS |
| 18:21.44 | prpplague | Crofton: 500KSPS is max of coruse |
| 18:21.49 | Crofton | yeah |
| 18:22.25 | Crofton | the basic principle is use the slowest sample rate you can without loss of information |
| 18:22.36 | Crofton | you want like 10ksps for rtty |
| 18:22.45 | *** join/#edev animesh (n=animesh@61.8.138.196) |
| 18:23.05 | prpplague | Crofton: thats pretty much what i figured, just wanted to confirm |
| 18:23.14 | Crofton | figuring 3khZ BW, 6kHz nyquist sampling, add 4 khz because nothing is ideal :) |
| 18:23.20 | animesh | hi |
| 18:23.28 | prpplague | animesh: greetings |
| 18:23.46 | prpplague | Crofton: so you think its doable? |
| 18:23.50 | Crofton | yeah |
| 18:24.07 | Crofton | if the hammer is as fast as you say |
| 18:24.14 | prpplague | Crofton: okie dokie thanks |
| 18:24.23 | Crofton | are there any special dsp instructions? |
| 18:24.24 | prpplague | Crofton: question now is the radio output |
| 18:24.28 | Crofton | yep |
| 18:24.32 | Crofton | that is the hard part |
| 18:24.34 | prpplague | Crofton: no nothing like that for the s3c2410 |
| 18:24.59 | prpplague | Crofton: these guys already have the radio stuff working for interfacing to a sound card |
| 18:25.00 | Crofton | you may have to code some section in asm |
| 18:25.05 | Crofton | ok |
| 18:25.29 | prpplague | Crofton: so i think if they have it working to a sound card, it should be fairly easy to tweak it for standard adc |
| 18:25.35 | Crofton | yep |
| 18:26.01 | prpplague | Crofton: you still thinking of a project using the hammer for SDR? |
| 18:26.16 | Crofton | thiking |
| 18:26.25 | prpplague | okie dokie |
| 18:26.26 | Crofton | but I have a big back log of stuff I want to do :( |
| 18:26.31 | prpplague | Crofton: indeed |
| 18:26.46 | prpplague | Crofton: i might have to pester you for a few questions, that ok? |
| 18:26.57 | Crofton | ok |
| 18:27.05 | Crofton | http://linpsk.sourceforge.net/ |
| 18:27.10 | Crofton | may have some smaple code |
| 18:27.40 | prpplague | thanks |
| 18:27.55 | *** join/#edev RobotGuy (n=robotguy@dsl093-038-072.pdx1.dsl.speakeasy.net) |
| 18:28.01 | Crofton | most of the psk31 sw also does rtty |
| 18:29.31 | prpplague | thats what is was noticing |
| 18:30.28 | prpplague | Crofton: hmm, they have a sample rtty audio file, but they don't tell you what the text is suppose to be |
| 18:31.01 | RobotGuy | It's the ohci-hcd module I need for host USB, right? Yeah, I am back, and better now. :) |
| 18:31.48 | RobotGuy | I am unable to load any modules. :( |
| 18:31.53 | RobotGuy | # modprobe ohci-hcd |
| 18:31.53 | RobotGuy | modprobe: failed to load module ohci-hcd |
| 18:33.54 | prpplague | RobotGuy: re |
| 18:34.27 | prpplague | RobotGuy: hmm several things could be a problem |
| 18:34.56 | prpplague | RobotGuy: trying doing an insmod on it instead |
| 18:52.15 | RobotGuy | # insmod /lib/modules/2.6.22/kernel/drivers/usb/host/ohci-hcd.ko |
| 18:52.15 | RobotGuy | Illegal instruction |
| 18:52.54 | RobotGuy | prpplague: As far as I can tell, everything is setup for modules correctly. |
| 18:53.06 | CosmicPenguin | doh |
| 18:53.15 | CosmicPenguin | stop breakin' the law! |
| 18:53.17 | RobotGuy | CosmicPenguin: ?? |
| 18:53.54 | RobotGuy | Oh, wait, I got it now. :D:D Silly Penguin! |
| 18:58.01 | RobotGuy | Anyone know about http://www.siliconmotion.com.tw/en/en2/products_sm502.html ?? It looks like it would be a great companion for Hammer. |
| 18:58.34 | *** part/#edev T0mW (n=Tom@24.238.86.158.res-cmts.sth.ptd.net) |
| 19:03.05 | RobotGuy | There is already a kernel driver for the SM501. |
| 19:11.28 | prpplague | RobotGuy: the min qty for the chip is very very very high, plus it requires a 32-bit bus interface ,which the hammer doesn't have |
| 19:21.59 | *** join/#edev steliosK__ (n=Stelios@athedsl-105805.home.otenet.gr) |
| 19:22.09 | sunrunner20 | prpplague: man, this wind is a bitch |
| 19:22.27 | prpplague | sunrunner20: indeed, its the windiest i've ever seen |
| 19:23.14 | sunrunner20 | windiest as far as I remember |
| 19:23.46 | sunrunner20 | execpt for that tornado |
| 19:23.48 | sunrunner20 | but that doesn't count |
| 19:28.50 | RobotGuy | prpplague: Durn, OK, well it was an idea. |
| 19:38.46 | RobotGuy | prpplague: Could it be a problem with Busybox? |
| 19:39.12 | RobotGuy | prpplague: I get some messages about fakeroot everytime I build an initrd. |
| 19:40.45 | RobotGuy | prpplague: http://www.pastebin.ca/880253 |
| 19:54.41 | prpplague | RobotGuy: definetly a problem somewhere |
| 19:58.05 | *** part/#edev dbtid (i=qwi0tu5z@cpe-71-72-252-171.columbus.res.rr.com) |
| 20:02.48 | RobotGuy | prpplague: I don't know where though. |
| 20:05.08 | prpplague | RobotGuy: usually when you have illegal instruction, its one of the following issues |
| 20:05.30 | prpplague | RobotGuy: wrong library, i.e. it was compiled with a different set of options or different arch |
| 20:06.03 | prpplague | RobotGuy: or the instruction involved a floating point and you don't have fpe enabled |
| 20:06.56 | RobotGuy | prpplague: I don't set any compile options. How could I have a wrong library? Everything is default settings for buildroot. |
| 20:07.08 | *** join/#edev redb1ue (i=star@ppp227.108-253-207.mtl.mt.videotron.ca) |
| 20:07.56 | prpplague | RobotGuy: don't know |
| 20:07.59 | RobotGuy | prpplague: I will make a new default buildroot and use my current config file and see what happens. |
| 20:08.06 | prpplague | RobotGuy: you have fpe enable in your kernel? |
| 20:09.34 | RobotGuy | prpplague: NWFPE emulation is enabled in the kernel. |
| 20:10.22 | prpplague | RobotGuy: ok, well i'd start back with the default config for the hammer, |
| 20:13.29 | RobotGuy | prpplague: I am doing that now. |
| 20:17.27 | chouimat|work | ARRRGGGG |
| 20:17.54 | *** join/#edev steliosk (n=Stelios@athedsl-105805.home.otenet.gr) |
| 20:18.25 | chouimat|work | I really start to hate this openspace concept office ... one of my coworker listens to crappy pop music and he sings it too |
| 20:22.22 | Crofton | awesome |
| 20:23.24 | chouimat|work | Crofton: seeing him doing this is worse ... he looks like a teenage girl in front of a mirror |
| 20:24.11 | Crofton | even more awesome |
| 20:24.16 | Crofton | this is why I work alone |
| 20:24.49 | chouimat|work | the usb missile launcher is usefull in this case\ |
| 20:26.07 | steliosk | chuimat|work : The upgraded version ? with the extra hardened head ? :) |
| 20:27.16 | chouimat|work | steliosk: yes |
| 20:36.43 | prpplague | ho ho hum |
| 20:48.17 | prpplague | jeeze - http://www.setexasrecord.com/news/205753-texas-company-seeks-judgement-to-stop-potential-patent-suit |
| 20:48.19 | prpplague | oops |
| 20:48.22 | prpplague | wrong paste |
| 20:48.46 | prpplague | jeeze - http://www.vscom.de/6801.htm#b |
| 20:48.56 | prpplague | these guys are shipping a new product with a 2.4 kernel |
| 20:50.36 | chouimat|work | prpplague: I know but some use "prepackaged kernel" made by somone else |
| 20:51.24 | prpplague | chouimat|work: yep, just amazes me |
| 20:51.57 | chouimat|work | prpplague: and since debian is stale errr mean stable for a long time it's ok for lab devices |
| 20:52.03 | prpplague | ok well introductory pricing for the nail board with hammer module is to be $199 plus shipping |
| 20:52.25 | chouimat|work | nice |
| 20:53.54 | CosmicPenguin | prpplague: heh - wern't you violently anti 2.6 not so long ago? :) |
| 20:54.39 | prpplague | CosmicPenguin: not opposed, my argument was for legacy devices |
| 20:55.00 | prpplague | CosmicPenguin: new products should use current kernel, imho |
| 20:55.40 | prpplague | CosmicPenguin: i can understand maybe with it being two years ago or even one year ago |
| 20:55.40 | chouimat|work | 2.4.32 is fairly recent |
| 20:56.04 | prpplague | CosmicPenguin: but now there isn't any reason to move forward to 2.6 |
| 20:57.08 | prpplague | bbiaf |
| 21:11.19 | *** join/#edev RobotGuy (n=robotguy@dsl093-038-072.pdx1.dsl.speakeasy.net) |
| 21:34.17 | sunrunner20 | fsdafdas |
| 21:34.22 | sunrunner20 | check check 123? |
| 21:36.37 | prpplague | sunrunner20: no one home, go away |
| 21:36.47 | prpplague | sunrunner20: you ever go to DPRG meetings? |
| 21:37.14 | prpplague | RobotGuy: you find your module issue? |
| 21:46.18 | sunrunner20 | dprg? |
| 21:46.29 | sunrunner20 | dallas ? robitics group? |
| 21:52.13 | prpplague | yea |
| 21:53.38 | prpplague | GPSFan: ping |
| 21:59.08 | *** join/#edev landley (n=landley@rrcs-67-78-38-61.sw.biz.rr.com) |
| 21:59.15 | prpplague | landley: hey hey |
| 22:01.04 | prpplague | landley: question for ya |
| 22:01.14 | GPSFan | prpplague: ping |
| 22:01.30 | prpplague | GPSFan: hey, this was for you too |
| 22:01.49 | prpplague | GPSFan / landley thoughts on this size of a case for the a version of the nail board - http://www.pactecenclosures.com/Plastic-Enclosures/PP.html |
| 22:02.06 | prpplague | that has basically the same dimension for x/y as a pcmcia card |
| 22:02.22 | *** join/#edev landley_ (n=landley@rrcs-67-78-38-61.sw.biz.rr.com) |
| 22:02.27 | prpplague | landley_: re |
| 22:03.04 | GPSFan | prpplague: it's only 3.7" long will the nail fit in that? |
| 22:03.16 | prpplague | GPSFan: no, not in current form |
| 22:03.24 | SpeedEvil | That looks reasonable. |
| 22:03.43 | GPSFan | oh, for a future formfactor rev, then yes that looks ok. |
| 22:03.45 | prpplague | GPSFan: the pcb is wider, so we'd have to shorten the pcb and widen it a little |
| 22:04.17 | prpplague | GPSFan: yea, i was curious as to whether it took away from the coolness factor |
| 22:04.41 | prpplague | i can get that case in custom colors too - http://www.pactecenclosures.com/customcolors.html |
| 22:05.37 | *** join/#edev landley__ (n=landley@rrcs-67-78-38-61.sw.biz.rr.com) |
| 22:06.00 | prpplague | arm-linux-uclibc-gcc -o float-test float-test.c |
| 22:06.00 | prpplague | /tmp/ccaVnpFX.o: In function `main': |
| 22:06.00 | prpplague | float-test.c:(.text+0xc4): undefined reference to `cos' |
| 22:06.00 | prpplague | collect2: ld returned 1 exit status |
| 22:06.10 | agaffney | prpplague: I agree |
| 22:06.20 | prpplague | agaffney: you agree what? |
| 22:06.32 | agaffney | with banging your head on your desk |
| 22:06.35 | agaffney | :P |
| 22:06.48 | agaffney | I've had the urge to do so a few times today :) |
| 22:06.49 | prpplague | ahh |
| 22:07.28 | prpplague | i can't ffigure out why i'm not getting a good softfloat build |
| 22:09.05 | RobotGuy | prpplague: No. I started a new buildroot, default config file, dir make, and it bombed out. |
| 22:09.24 | prpplague | RobotGuy: on what? |
| 22:10.05 | SpeedEvil | prpplague: I can't quite make that out - is there a seperate insert panel in the top cover?> |
| 22:10.13 | RobotGuy | prpplague: http://www.pastebin.ca/881095 |
| 22:10.17 | prpplague | SpeedEvil: yea |
| 22:10.47 | RobotGuy | prpplague: All I changed was the gcc version to 4.2.1 |
| 22:13.32 | prpplague | RobotGuy: well, i've not tested any of 4.2.1 |
| 22:14.30 | RobotGuy | prpplague: I thought you used 4.2.1 |
| 22:14.52 | prpplague | 4.1.2 |
| 22:16.20 | RobotGuy | prpplague: Switched back to 4.1.2 and rebuilding default buildroot |
| 22:17.13 | RobotGuy | I zapped all the host and build dirs before starting the rebuild to force everything to be rebuilt. |
| 22:18.40 | *** join/#edev landley_ (n=landley@rrcs-67-78-38-61.sw.biz.rr.com) |
| 22:20.54 | prpplague | SpeedEvil: think that the size of the case takes away from the coolness of the nail board? |
| 22:21.48 | SpeedEvil | prpplague: Don't know. |
| 22:21.54 | RobotGuy | prpplague: It looks like a mini-coffin. |
| 22:21.57 | SpeedEvil | I happened to have http://www.analog.com/en/prod/0,,760_1077_AD7142,00.html open at the same time |
| 22:22.10 | prpplague | RobotGuy: we are discussing a different case |
| 22:22.13 | SpeedEvil | which is a $1.5 touch-button controller. |
| 22:22.15 | *** join/#edev landley__ (n=landley@rrcs-67-78-38-61.sw.biz.rr.com) |
| 22:22.25 | RobotGuy | Oh |
| 22:22.27 | SpeedEvil | that can do all sorts of capacitive keypads over I2C |
| 22:22.44 | SpeedEvil | Are you intending to put a battery in? |
| 22:22.50 | RobotGuy | SpeedEvil: They would be very useful for robotics also. |
| 22:22.57 | SpeedEvil | Or is it USB only? |
| 22:23.59 | prpplague | SpeedEvil: no |
| 22:24.05 | prpplague | SpeedEvil: this is usb powered only |
| 22:24.13 | SpeedEvil | Thought so. |
| 22:25.15 | SpeedEvil | I was just wondering about the possibility of making it a generic thing. I do think the case makes more sense - if only for mounting. |
| 22:26.14 | prpplague | SpeedEvil: yea, we are doing a specific target for this |
| 22:35.14 | SpeedEvil | That is an interesting case vendor. Free samples, cheap international shipping, and reasonable small quantity pricing. |
| 22:35.17 | *** join/#edev landley_ (n=landley@rrcs-67-78-38-61.sw.biz.rr.com) |
| 22:35.31 | prpplague | SpeedEvil: pactec? |
| 22:36.25 | SpeedEvil | yes |
| 22:37.28 | RobotGuy | Does Pactec have a site? |
| 22:37.36 | SpeedEvil | http://www.pactecenclosures.com/Clear-Utility-Cases.php |
| 22:37.46 | SpeedEvil | or the url prpplague posted a while back |
| 22:37.46 | prpplague | http://www.pactecenclosures.com/Plastic-Enclosures/PP.html |
| 22:38.06 | prpplague | SpeedEvil: yea they are kind of the industry standard for generic cases |
| 22:38.22 | prpplague | SpeedEvil: pactec is what the rat shack sold for yers |
| 22:38.23 | prpplague | years |
| 22:39.22 | prpplague | i personally like the size of a pcmcia card, but i think landley would still prefer the thumb drive size |
| 22:40.38 | prpplague | serpac has a nice case but it has a screw post right where the hammer module should be |
| 22:41.16 | prpplague | http://serpac.com/products_c-28.htm |
| 22:43.48 | prpplague | the h series would work as well and i can get it in colors too - http://serpac.com/products__h-home.htm |
| 22:46.02 | prpplague | that c series also has a pocket clip option |
| 22:47.38 | SpeedEvil | Pocket clips are _horrid_ to get right |
| 22:47.56 | SpeedEvil | they need to pull off at just the right degree of force |
| 22:48.06 | SpeedEvil | otherwise they just snap the case or the clop |
| 22:48.23 | SpeedEvil | clip |
| 22:52.35 | prpplague | hmm |
| 22:59.05 | *** join/#edev redb1ue (i=star@ppp053.109-253-207.mtl.mt.videotron.ca) |
| 23:01.56 | *** join/#edev CosmicPe1guin (n=nobody@163.181.251.103) |
| 23:02.19 | *** mode/#edev [+o CosmicPenguin] by ChanServ |
| 23:08.29 | RobotGuy | prpplague: http://www.pastebin.ca/881414 |
| 23:09.47 | prpplague | RobotGuy: lordy i don't understand why you are having sooooo much trouble |
| 23:10.03 | RobotGuy | prpplague: It's getting quite frustrating for me, to say the least. |
| 23:10.37 | prpplague | RobotGuy: so you used the official buildroot with the official config |
| 23:10.37 | RobotGuy | I'm not even running Fedora Core 8 anymore and things don't seem any better. |
| 23:10.50 | prpplague | RobotGuy: and you are getting that error? |
| 23:11.05 | RobotGuy | prpplague: I changed the stock hammer config back to gcc 4.1.2. That's the ONLY thing changed from last time. |
| 23:11.30 | prpplague | RobotGuy: sounds like you still have some stamp files left over |
| 23:11.34 | RobotGuy | prpplague: Stock Hammer Buildroot and stock config. |
| 23:11.54 | prpplague | RobotGuy: interesting |
| 23:12.00 | RobotGuy | prpplague: Hmmmm. I zapped all the -build and -host dirs before rebuilding. |
| 23:12.21 | RobotGuy | prpplague: I did not zap the uCLibc_dev dir though. |
| 23:12.22 | prpplague | RobotGuy: i'd just delete buildroot dir and start over |
| 23:13.05 | RobotGuy | prpplague: I already did that when I changed back to stock buildroot. I'll just zap the toolchain_build_arm dir this time. |
| 23:13.19 | prpplague | RobotGuy: okie dokie |
| 23:13.29 | prpplague | RobotGuy: i've built 12 working toolchains today |
| 23:14.15 | RobotGuy | That should insure that none of those dang stamp files exist. |
| 23:15.16 | RobotGuy | prpplague: I also copied the hammer config to .config to make absolutely sure I am building the stock buildroot. |
| 23:15.35 | prpplague | RobotGuy: then do a make oldconfig |
| 23:15.58 | RobotGuy | prpplague: OK, did that. |
| 23:16.21 | RobotGuy | Is make oldconfig required everytime you replace the .config? |
| 23:27.13 | *** join/#edev lyakh (n=lyakh@p57BD27D6.dip0.t-ipconnect.de) |
| 23:27.28 | prpplague | RobotGuy: its precautionary |
| 23:31.21 | RobotGuy | That didn't answer my question entirely. |
