| 00:00.11 | prpplague | landley hows the ear? |
| 00:01.07 | Dr{Afk} | i have never seen an odd # release considered "stable" on kernel.org |
| 00:02.51 | RobotGuy | prpplague: I don't have a clue what you are talking about, sorry. All I want is good drivers for Ralink stuff. |
| 00:03.30 | prpplague | RobotGuy hehe we do seem to have commo problem sometimes |
| 00:03.45 | Dr{Afk} | is this the project? http://rt2x00.serialmonkey.com/wiki/index.php?title=Main_Page |
| 00:03.55 | RobotGuy | prpplague: This kernel does not look right to me. I did not see anything to do with Samsung processors. |
| 00:04.08 | RobotGuy | prpplague: Probably all my fault. |
| 00:04.09 | prpplague | RobotGuy the patch that is posted on the wiki was designed to apply to the kernel to set everything, including the cross compiler and arch |
| 00:04.23 | prpplague | RobotGuy the patch you just applied only adds the core support |
| 00:04.30 | RobotGuy | prpplague: OK, so I have to set that stuff manually for rc8 then. |
| 00:04.30 | prpplague | RobotGuy you have to turn on the other options |
| 00:04.43 | prpplague | RobotGuy correct |
| 00:05.00 | landley | prpplague: http://www.tvsquad.com/2008/03/28/star-trek-the-next-generation-angel-one/ |
| 00:05.19 | prpplague | RobotGuy now , the linux arm kernel maintainers are not going to accept the patch on the wiki , because it does specific things for the hammer |
| 00:05.25 | landley | prpplague: the ear is much better, thanks. |
| 00:05.38 | prpplague | RobotGuy so i have to make a "generic" patch that can be accepted by LAK |
| 00:05.51 | RobotGuy | prpplague: Isn't that the whole idea of getting patches for hardware into the dist kernel tree?? To support more hardware? |
| 00:06.04 | prpplague | RobotGuy so as long as we "try" to submit patches upstream we will have two patches |
| 00:06.20 | prpplague | RobotGuy yes, that is the goal |
| 00:06.24 | RobotGuy | prpplague: That really isn't how it should be and it sucks that it is that way. |
| 00:06.34 | prpplague | RobotGuy thats reality |
| 00:06.41 | RobotGuy | Reality sucks |
| 00:07.05 | ce_geek | prpplague: New release for hammer soon? |
| 00:07.15 | prpplague | ce_geek getting close |
| 00:07.18 | RobotGuy | I don't know what to set CROSS_COMPILE to. Is it the prefix for the compiler? |
| 00:07.23 | landley | RobotGuy: PATH=/path/to/cross/compiler:$PATH make ARCH=arm CROSS_COMPILE=armv4l- |
| 00:07.26 | ce_geek | prpplague: cool cool |
| 00:07.41 | landley | RobotGuy: yes. |
| 00:07.46 | landley | The compilers I make, it's armv4l- |
| 00:07.50 | prpplague | RobotGuy set them in the Makefile |
| 00:07.54 | landley | The ones you're using it's whatever comes before "gcc". |
| 00:08.00 | landley | You don't have to set them in the Makefile, it works just fine from the command line. |
| 00:08.18 | prpplague | landley agreed, but most people don't like typing it over and over |
| 00:08.27 | landley | This is why the shell script was invented. |
| 00:08.28 | prpplague | RobotGuy have a look at the makefile for your old kernel |
| 00:08.33 | prpplague | landley hehe indeed |
| 00:09.38 | *** mode/#edev [+o prpplague] by ChanServ |
| 00:10.28 | wmat | prpplague: no elinux.org still |
| 00:10.37 | prpplague | wmat and no update? |
| 00:10.52 | prpplague | is the server dead? or just a dns problem? |
| 00:11.36 | wmat | prpplague: it's not dead, it just won't allow ssh access. No update from godaddy yet or Tim R. who's looking at the DNS. |
| 00:11.54 | prpplague | wmat ahh ok |
| 00:11.59 | Crofton | who hosts elinux.org? |
| 00:12.00 | prpplague | wmat atleast thats good news |
| 00:12.08 | wmat | Crofton: godaddy |
| 00:12.18 | wmat | we just moved to them actually |
| 00:12.19 | Crofton | should have picked geekisp.com :) |
| 00:13.01 | RobotGuy | prpplague: I think I figgered it out. I set CROSS_COMPILE to "arm-linux-: |
| 00:13.01 | wmat | Crofton: don't think I didn't try |
| 00:13.19 | prpplague | RobotGuy that should work, but "arm-linux-uclibc-" would be better |
| 00:13.29 | prpplague | RobotGuy and make sure you set the subarch |
| 00:13.29 | Crofton | I guess someone like the semi-nudie ads |
| 00:13.50 | wmat | prpplague: btw, our spring release is next week, so new Lite toolchains will be out if you're interested |
| 00:14.10 | prpplague | yea want to grab one and get familiar with it |
| 00:14.18 | prpplague | wmat for several reasons |
| 00:15.18 | RobotGuy | prpplague: SUBARCH is the part that is all screwy. It bases the selection on the host system. |
| 00:15.49 | prpplague | RobotGuy look at the same section in your old kernel |
| 00:16.08 | prpplague | RobotGuy just go the blank line below and do: SUBARCH := arm |
| 00:16.35 | RobotGuy | Shouldn't "SUBARCH := $(shell uname -m" be SUBARCH := $(ARCH) ??? Or something similar?? |
| 00:17.09 | RobotGuy | prpplague: ARCH=arm, so why would SUBARCH=arm also? |
| 00:17.19 | Dr{Afk} | uname -m will pull in your dom0 boxes data. |
| 00:17.35 | RobotGuy | Dr{Afk}: I know. |
| 00:17.42 | Dr{Afk} | k |
| 00:17.57 | RobotGuy | Dr{Afk}: That's why I am saying it is wrong in the Makefile. |
| 00:18.03 | Dr{Afk} | ahh |
| 00:18.12 | RobotGuy | It should be based on ARCH |
| 00:18.14 | Dr{Afk} | i see what your saying ya should not be uname -m |
| 00:18.31 | prpplague | RobotGuy sorry my bad long day, i said SUBARCH and meant ARCH |
| 00:18.44 | RobotGuy | prpplague: I have ARCH=arm |
| 00:19.18 | prpplague | okie dokie, should be able to copy the hammer-config over to .config and then do a make oldconfig |
| 00:19.21 | RobotGuy | The selection for SUBARCH needs to be based on what ARCH is, NOT the host info as in uname -m |
| 00:19.44 | RobotGuy | prpplague: I have a good config from rc7. Can't I use that? |
| 00:20.01 | prpplague | RobotGuy yea you can, make sure you do a make oldconfig |
| 00:20.09 | RobotGuy | prpplague: I did. :) |
| 00:21.01 | RobotGuy | If I replace this: $(shell uname -m with $(ARCH) wouldn't that fix it? |
| 00:21.25 | RobotGuy | I mean $(ARCH |
| 00:22.04 | prpplague | RobotGuy you asking me? |
| 00:22.09 | RobotGuy | This line: SUBARCH := $(shell uname -m | sed -e s/i.86/i386/ -e s/sun4u/sparc64/ \ |
| 00:22.26 | RobotGuy | Would become: SUBARCH := $(ARCH | sed -e s/i.86/i386/ -e s/sun4u/sparc64/ \ |
| 00:22.36 | RobotGuy | prpplague: If you know the answer, then yes. :) |
| 00:22.37 | prpplague | RobotGuy give me a second to pull it up |
| 00:22.58 | prpplague | RobotGuy what is it you are trying to do? |
| 00:23.28 | landley | Heh. solar@gentoo may wind up going to CELF. |
| 00:23.33 | RobotGuy | prpplague: I am trying to fix that SUBARCH selection to be correct. It is wrong now. |
| 00:23.38 | landley | Sitting in on my cross compiling tutorial. |
| 00:23.52 | landley | Having _him_ watch _me_ give a tutorial on cross compiling is... |
| 00:23.59 | RobotGuy | prpplague: That uname -m is wrong |
| 00:23.59 | landley | Analogies fail me, but it's ironic. |
| 00:24.13 | prpplague | RobotGuy see you have me all confused, hehe |
| 00:24.46 | RobotGuy | prpplague: The selection of SUBARCH should NOT be based on the host info as in uname -m |
| 00:24.47 | prpplague | RobotGuy below the SUBARCH section on the blank line just do: SUBARCH := arm |
| 00:25.04 | RobotGuy | prpplague: I want to fix it RIGHT, not patch it. :) |
| 00:25.25 | RobotGuy | prpplague: Why does SUBARCH = ARCH ?? |
| 00:25.37 | RobotGuy | prpplague: That does not seem right. |
| 00:25.48 | Dr{Afk} | a good read prpplague.. i finaly spent the time to diff the kernel on the iso to the stock 2.6.22 dam thats a lot of work. |
| 00:26.33 | prpplague | RobotGuy read the text right about the subarch section |
| 00:27.03 | Dr{Afk} | have not had more time to play with the gadget driver yet. you have any time to verify it works? |
| 00:27.25 | prpplague | Dr{Afk} depends on which diff you are looking at, some of that diff is some tweak patches that i didn't do, just collected |
| 00:27.53 | prpplague | Dr{Afk} sorry no, been swamped with a new board bringup |
| 00:27.57 | Dr{Afk} | i just took the 2.6.22 you provided and diffed it to stock 2.6.22 to see all changes.. |
| 00:28.00 | RobotGuy | prpplague: The way they do this is sooooooooooooooooooooo stupid. When you see SUB anything, it should be a lower level selection, not equal to the same as the higher level (ARCH). |
| 00:28.34 | prpplague | Dr{Afk} yea thats a mix of alot of collected patches |
| 00:28.46 | RobotGuy | For instance, if ARCH=arm, then SUBARCH should be like arm4vl or whatever is below arm. |
| 00:28.49 | prpplague | Dr{Afk} 2.6.22 wasn't completely stable for the s3c2410 |
| 00:29.22 | prpplague | RobotGuy you are welcome to submit patches to the kernel mailing lists :) |
| 00:29.46 | Dr{Afk} | np.. i hope to find more time this week as well.. have to head to get my teen from school. some teen idol thing she stayed later for.. ill be back on tomarrow. |
| 00:29.51 | RobotGuy | prpplague: And undergo all that crap? I don't think so and would end up not being nice to somebody in the process. |
| 00:30.01 | prpplague | hehe exactly |
| 00:30.07 | prpplague | hence why its still a mess |
| 00:30.28 | prpplague | and yet another reason why so many companies maintain their own trees |
| 00:30.55 | Dr{Afk} | ohh ya.. |
| 00:30.57 | prpplague | RobotGuy oh and btw, i'm a buildroot and busybox maintainer, i can accept patches for either |
| 00:31.07 | RobotGuy | prpplague: I don't think that's the real reason it's a mess. It's that some people are just doing things in stupid ways and they don't care. |
| 00:31.29 | Dr{Afk} | hmm prp.. i didnt know that i will have to send you a few. |
| 00:31.38 | RobotGuy | prpplague: Cool. I've already given you several things that need fixing, but that is to our 'official' buildroot. |
| 00:32.17 | RobotGuy | prpplague: However, I have fixed a couple things in the svn, but unfortunately can't remember what they are. :( |
| 00:32.23 | prpplague | RobotGuy yea, i meant in regards to the post the forum |
| 00:32.35 | prpplague | RobotGuy what svn? |
| 00:32.47 | landley | wmat: i point out that solar@gentoo runs an ISP as his day job... |
| 00:32.57 | RobotGuy | prpplague: Oh, you mean what dunk and I are discussing about adding wlan-ng? |
| 00:32.58 | Dr{Afk} | bash is broken and the bash maintainers are not responding to my find of a memory exposure security bug that is exposed in buildroot's bash-011 patch |
| 00:33.04 | agaffney | what about solar? |
| 00:33.07 | prpplague | RobotGuy yea |
| 00:33.15 | RobotGuy | prpplague: Oh, OK, cool. |
| 00:33.29 | RobotGuy | prpplague: I've been toying with a port or two myself. |
| 00:34.07 | RobotGuy | prpplague: Would it be appropriate to add modules for Python such as pyserial to buildroot? |
| 00:34.19 | prpplague | RobotGuy submit a patch |
| 00:34.27 | RobotGuy | prpplague: OK. :) |
| 00:34.57 | landley | prpplague: send _me_ patches and I can forward them to the list. |
| 00:35.04 | landley | has some experience doing this. :) |
| 00:35.31 | landley | never tries to fix FSF packages. |
| 00:35.41 | landley | tries to avoid using them, and carries patches when that's not feasible. |
| 00:35.55 | prpplague | hehe |
| 00:35.56 | landley | I'll happily navigate the kerenel development process, but not the FSF. |
| 00:36.01 | landley | It's one of those "all they can do is kill you" things... |
| 00:36.12 | landley | There are fates worse than Al Viro. |
| 00:36.35 | Dr{Afk} | bash32-011 enables bash32's built in getcwd() function that has big problem copying memory out of bounds on the stack into a local var.. this is fine from userland but as run by process 0 it will cause a kernel panic.. and as userland it "can" expose memory and "may" be a vulnerability. |
| 00:36.38 | landley | agaffney: solar's day job is at an ISP. |
| 00:37.00 | landley | Or do you mean him coming to the Consumer Electronic Linux Forum next week? |
| 00:37.15 | landley | He says he lives 30 minutes south of it. I'm trying to get them to comp him a badge. |
| 00:37.30 | prpplague | landley / RobotGuy i generally don't fight the stuff unless it is pretty bad, i'm mostly concerned about getting my product out the door |
| 00:38.00 | agaffney | landley: I know solar :P he's an admin at GNi |
| 00:38.09 | agaffney | I was just wondering why you were talking about solar |
| 00:39.01 | landley | agaffney: because elinux.org just switched to godaddy, and has been down for over 8 hours now. |
| 00:39.27 | prpplague | which is pretty darn bad |
| 00:40.00 | landley | And because he offered to sit in on the cross compiling tutorial I'm giving at CELF next week, if the high price of the admission doesn't prevent him from attending. |
| 00:40.21 | landley | (I remember when they had a big PowerPC Linux conference in Austin, and I didn't go because it was $500 to attend. That was _insane_.) |
| 00:40.42 | ds2 | landley: isn't it just $50? |
| 00:41.52 | landley | ds2: CELF? Dunno, I'm presenting. |
| 00:42.06 | *** join/#edev ccjoe (n=ccjoe@75-169-37-101.slkc.qwest.net) |
| 00:42.07 | prpplague | i thought it was alot more than 50 |
| 00:42.14 | ds2 | landley: think so. for companies it is $300 but for hobbiest and individuals it should be 50 |
| 00:42.17 | ce_geek | Anyone played with boa? Any pointers? |
| 00:42.28 | ce_geek | As soon as I get it to show me a web page, I get to go home. |
| 00:42.32 | ce_geek | That was the goal. |
| 00:42.47 | ce_geek | I got it installed, it exists. Look like I need to add group "nobody" |
| 00:42.57 | landley | According to http://embeddedlinuxconference.com/elc2008/index.html "ticket at door: $350". |
| 00:43.05 | RobotGuy | prpplague: I do not see the OHCI usb driver anywhere now. |
| 00:43.07 | prpplague | Price: $300 Professional, $50 Hobbyist |
| 00:43.07 | prpplague | Ticket at Door: $350 |
| 00:43.42 | RobotGuy | I would be in way over my head at a conference like that |
| 00:44.06 | ds2 | 450 if he prereg |
| 00:44.22 | ds2 | $50 |
| 00:44.36 | ds2 | most of the celinuxforum pages are dead... I assume cohosted with elinux.org? |
| 00:44.47 | *** join/#edev dijenerate (n=dijenera@66.54.118.16) |
| 00:45.08 | landley | Quite possibly. |
| 00:45.13 | prpplague | RobotGuy thats because its not in 2.6.25-rc8 |
| 00:45.18 | RobotGuy | prpplague: I don't see most of the s3c2410 stuff at all. |
| 00:45.24 | prpplague | RobotGuy thats because its not in 2.6.25-rc8 |
| 00:45.31 | RobotGuy | prpplague: Well how am I going to do USB without the OHCI driver?? |
| 00:45.46 | landley | RobotGuy: tweak the .config? |
| 00:45.56 | landley | It should still be in the kernel tree. |
| 00:45.56 | prpplague | RobotGuy wait for a patch that is more than "core" support |
| 00:46.01 | landley | I think he means it's not in the defconfig? |
| 00:46.08 | RobotGuy | prpplague: Oh good. :( |
| 00:46.20 | RobotGuy | landley: No, it's disabled or just not there right now. |
| 00:46.44 | ce_geek | Hmmn. Boa is complaining: gethostbyname:: No such file or directory |
| 00:47.20 | prpplague | RobotGuy actually i think the 2.6.25-rc8 does have ohci support |
| 00:47.26 | prpplague | checks the git repo |
| 00:48.03 | RobotGuy | It's telling me I have to do make mrproper now to fix links to the asm stuff. |
| 00:49.28 | landley | There's USB_ARCH_HAS_OCHI... |
| 00:49.30 | ds2 | bit bang USB host driver! ;) |
| 00:49.37 | prpplague | RobotGuy ?? sounds like you've got something misconfigured in your makefile |
| 00:50.51 | landley | OHCI HCD support. |
| 00:51.00 | prpplague | RobotGuy yea looks like the s3c2410 ohci is there, just hasn't been added to the makefile or kconfig |
| 00:51.05 | landley | CONFIG_USB_OHCI_HCD |
| 00:51.37 | prpplague | landley makefile doesn't build the source |
| 00:51.41 | landley | device drivers->usb support->Support for Host-side USB |
| 00:52.04 | landley | last one I built was -rc7, lemme try... |
| 00:52.52 | landley | PATH=~/firmware/firmware/build/cross-compiler-armv4l/bin:$PATH make ARCH=arm CROSS_COMPILE=armv4l- -j 2 |
| 00:53.01 | landley | Building fine here so far... |
| 00:54.05 | prpplague | landley it actually builds the ohci-s3c2410.c ? |
| 00:55.15 | landley | It built with the OCHI config enabled. I didn't check for specific .c files... |
| 00:55.56 | prpplague | landley yea, you'll need to see if you have a ohci-s3c2410.o |
| 00:56.34 | landley | Only mention of "ohci" in the compiler output was |
| 00:56.35 | landley | <PROTECTED> |
| 00:56.48 | ce_geek | Sweet, Boa works. |
| 00:57.32 | prpplague | landley yea thas what i thought |
| 00:57.38 | prpplague | ce_geek cool |
| 00:57.44 | landley | I dont' see an ohci-s3c2410.c file in the source. |
| 00:57.44 | prpplague | ce_geek what are testing with |
| 00:57.57 | ce_geek | prpplague: Oh I just wanted "Hello World." |
| 00:57.57 | prpplague | ahh |
| 00:58.09 | ce_geek | prpplague: On hammer. |
| 00:58.09 | prpplague | ce_geek sorry i mean with what hardware |
| 00:58.24 | landley | Ah, hang on: ./drivers/usb/host/ohci-s3c2410.c |
| 00:58.53 | prpplague | landley yea thats it |
| 00:59.09 | landley | #ifdef CONFIG_ARCH_S3C2410 |
| 00:59.09 | landley | #include "ohci-s3c2410.c" |
| 00:59.14 | landley | Look in ohci-hcd.c |
| 00:59.17 | landley | Ye mighty and despair. |
| 00:59.22 | prpplague | hehe |
| 00:59.32 | prpplague | ahh ok |
| 00:59.32 | landley | It's not a seprate .o file, the .c file is #included from another .c file. |
| 00:59.44 | prpplague | thats new iirc |
| 00:59.46 | landley | Keep in mind I built -rc7 and it worked onna nail. |
| 01:00.08 | prpplague | RobotGuy there ya, go, just select the ohci and your good |
| 01:00.12 | landley | just built -rc8, lemme try it on the nail. |
| 01:00.17 | landley | Your defconfig selects ohci. |
| 01:00.22 | ce_geek | I want a sledge. :( |
| 01:00.28 | prpplague | hehe |
| 01:01.16 | prpplague | ce_geek what are you using for the tcp/ip connection? |
| 01:02.12 | landley | The nail is re-flashing. |
| 01:02.22 | prpplague | landley okie dokie |
| 01:02.24 | landley | Be nice if it could flash the LED or something when it does this... |
| 01:02.33 | prpplague | thats doable |
| 01:02.37 | prpplague | makes a note |
| 01:03.54 | prpplague | ho ho hum |
| 01:04.02 | prpplague | landley still likeing the nail? |
| 01:04.03 | landley | And lo, it is enbootenating. |
| 01:04.09 | prpplague | hehe |
| 01:04.17 | landley | Kernel panic, no init found. |
| 01:04.25 | landley | But that's actually to be expected if I haven't reset my kernel command line yet. :) |
| 01:04.25 | prpplague | landley doh |
| 01:04.30 | landley | Reflashing apex resets it... |
| 01:04.36 | prpplague | yep |
| 01:05.23 | landley | Hey, I got a command prompt. |
| 01:05.26 | prpplague | landley you don't have to reflash the bootloader every time |
| 01:05.27 | RobotGuy | landley: I have rc7 working on Hammer also. |
| 01:05.34 | landley | 2.6.25-rc8 works just fine with your latest patch, with my armv4l toolchain. |
| 01:05.39 | RobotGuy | prpplague: Found the OHCI |
| 01:05.44 | landley | prpplague: I made a script. |
| 01:05.46 | prpplague | good deal |
| 01:05.46 | landley | I run the script. |
| 01:05.59 | landley | Using -rc8 here. |
| 01:06.21 | landley | So rc7->rc8 didn't break anything obvious. :) |
| 01:06.36 | landley | prpplague: yes, I like the nail. |
| 01:06.45 | landley | I just have a shortage of time to do all open source development. |
| 01:06.58 | landley | Got pinged by another google recruiter today (via email, no reply to my reply yet). |
| 01:07.02 | prpplague | there is a recursion problem with rc8 but it only applies to very specific sdram configuration |
| 01:07.05 | landley | If that pans out, possibly I'd have more time to work on stuff... |
| 01:07.15 | ds2 | landley: they are still recruiting? |
| 01:07.16 | landley | prpplague: which doesn't impact hammer/nail. |
| 01:07.25 | landley | ds2: this is for the Austin office, which is just starting up. |
| 01:07.27 | prpplague | landley correct |
| 01:07.40 | prpplague | landley spent 1/2 of today debugging it |
| 01:07.43 | ds2 | landley: wow. and this is right after shedding a bunch of people here |
| 01:07.45 | landley | I remember. :) |
| 01:07.56 | landley | Where' "here"? |
| 01:08.20 | prpplague | landley google has a "lot" of the hammer kits |
| 01:08.28 | prpplague | no clue what they are doing with them |
| 01:08.36 | landley | prpplague: it all depends on what the _austin_ office is doing. |
| 01:08.38 | ds2 | here == Silly valley |
| 01:08.55 | prpplague | landley our 1st and 2nd biggest orders have been from google |
| 01:09.06 | landley | The reason I didn't go to work for Google back in 2000 was it required a move to CA. |
| 01:09.25 | landley | (I recently dug up the thank you packet they sent me after I gave them their first financial press coverage, by the way.) |
| 01:09.27 | prpplague | landley yea that would be a big draw back |
| 01:09.31 | landley | Yay unpacking old boxes... |
| 01:09.43 | landley | Austin office == not silicon valley. |
| 01:09.57 | prpplague | indeed |
| 01:10.35 | landley | Heh: http://www.fool.com/portfolios/rulemaker/1999/rulemaker990604.htm |
| 01:10.53 | landley | That was the column they responded to with stickers and a t-shirt. |
| 01:11.02 | landley | Couldn't keep the t-shirt. Value was probably over $10. :( |
| 01:11.10 | prpplague | doh |
| 01:11.12 | prpplague | hehe |
| 01:11.13 | landley | The stickers, I kept. Put one on my car. |
| 01:11.19 | landley | That was 2 cars ago... |
| 01:11.35 | prpplague | i wish our company had a similiar policy |
| 01:11.51 | prpplague | that dick head EE is getting a boat load of freebies and meals from vendors |
| 01:12.06 | prpplague | 1/2 the time he is meeting with vendors we have 0 chance of using |
| 01:12.27 | landley | Similar policy? |
| 01:12.33 | prpplague | nothing over $10 |
| 01:12.37 | landley | Ah yes. |
| 01:12.47 | landley | The thing is, The Motley Fool is a financial industry site. |
| 01:12.56 | landley | There are all sorts of strange rules covering financial analysts. |
| 01:13.05 | landley | Even when you have no _idea_ what you're doing, like I didn't. |
| 01:13.25 | RobotGuy | prpplague: I want some freebies from the EE. :) |
| 01:13.27 | prpplague | hehe yea i've heard that before |
| 01:13.37 | prpplague | RobotGuy uh what? |
| 01:13.47 | landley | Who was it who said "I'm pretty dangerous when I don't know what I'm doing". |
| 01:13.53 | landley | Sounds like The Doctor. |
| 01:14.05 | prpplague | landley yea that does sound like something he would say |
| 01:14.08 | RobotGuy | Now kernel won't compile. I give up for now. Maybe RC9 will be good. |
| 01:14.20 | RobotGuy | NVM. |
| 01:14.28 | prpplague | RobotGuy i think you misunderstood |
| 01:14.38 | prpplague | RobotGuy vendors are trying to bribe our EE into using their products |
| 01:14.53 | RobotGuy | prpplague: I'd be happy to let them bribe me. :) |
| 01:15.08 | ds2 | it isn't unusual for vendors to show up with donuts and a coffee |
| 01:15.15 | landley | RobotGuy: yeah, so is the EE it sounds like... |
| 01:15.22 | prpplague | RobotGuy well, it doesn't sit well with the rest of engineering |
| 01:15.23 | landley | wants donuts now. |
| 01:15.34 | RobotGuy | prpplague: I understand. I wasn |
| 01:15.46 | RobotGuy | wasn't being serious. |
| 01:15.54 | prpplague | RobotGuy ahh ok |
| 01:16.13 | landley | prpplague: so the upshot of telling the owner that the EE was bad juju was... nothing? |
| 01:19.06 | landley | boggles at the freebsd developers. |
| 01:19.18 | landley | Their "bootonly" cd boots you into an installer, not a shell prompt. |
| 01:19.21 | bsdfox_ | ? |
| 01:19.22 | landley | There's no way to _get_ a shell prompt. |
| 01:19.44 | landley | Their livefs iso did the same thing. |
| 01:19.54 | bsdfox_ | uh |
| 01:19.59 | bsdfox_ | does it boot you into sysinstall? |
| 01:20.02 | landley | I tried installing from the livefs, but it wants to talk to an ftp server (not http), and that won't go out through the firewall. |
| 01:20.10 | landley | bsdfox_ dunno what it's called. Blue background, curses thingy. |
| 01:20.13 | bsdfox_ | yeah |
| 01:20.19 | landley | Looks like Turbo Linux circa 1992. |
| 01:20.23 | bsdfox_ | that's sysinstall.. did you try just exiting out of it? |
| 01:20.33 | landley | It told me it would reboot the system if I did that. |
| 01:20.47 | ds2 | do they have virtual consoles? |
| 01:20.52 | landley | Under qemu? |
| 01:20.56 | landley | alt-1 didn't do much. |
| 01:21.01 | ds2 | hmm |
| 01:21.26 | landley | alt-f2 did something, but not something useful. |
| 01:21.34 | landley | it says: |
| 01:21.43 | landley | DEBUG: ioctl(3, TIOCCONS,NULL) = 0 (success) |
| 01:21.50 | landley | DEBUG: MADT: Found CPU APIC ID 0 enabled |
| 01:21.56 | landley | DEBUG: Add mapping for /dev/cuad0 to sl0 |
| 01:22.00 | landley | That's it. |
| 01:22.09 | landley | Alt f3-F8 or so didn't change it. |
| 01:22.16 | landley | Alt-F1 gives me back the installer. |
| 01:22.21 | ds2 | fun |
| 01:22.25 | landley | (You can change virtual consoles if you let qemu do a mouse grab. I'd forgotten that.) |
| 01:22.36 | bsdfox_ | landley: 6.3 or 7.0? |
| 01:22.43 | landley | 7.0 |
| 01:22.47 | landley | Trying to exit, and... |
| 01:22.58 | landley | Um, it seems to be hung. |
| 01:23.06 | bsdfox_ | I haven't done a clean install in a couple years but the cd used to boot into a shell then you'd manually run /sbin/sysinstall |
| 01:23.12 | bsdfox_ | err /stand/sysinstall actually |
| 01:23.31 | landley | It rebooted. |
| 01:23.31 | landley | So yes, if you exit the installer, it reboots. |
| 01:23.35 | bsdfox_ | lame |
| 01:23.51 | bsdfox_ | landley: try booting into single user mode? |
| 01:23.52 | landley | I don't think anybody's done a clean install of any BSD in about 10 years. |
| 01:23.56 | prpplague | landley sorry back phone |
| 01:23.58 | landley | bsdfox_: how? |
| 01:24.02 | prpplague | landley so far nothing |
| 01:24.12 | bsdfox_ | does it pop up a text picture of beastie before it boots the kernel? |
| 01:24.19 | prpplague | landley but i did notice yesterday on monster.com a job posting |
| 01:24.40 | landley | Ok, hit space at the bootloader prompt, then hit 4. |
| 01:24.45 | prpplague | landley doesn't specifically say for AML, but it does match the area and job description for the EE job |
| 01:24.47 | landley | And it boots me into the same stupid installer. |
| 01:24.53 | landley | prpplague: heh. :) |
| 01:25.05 | landley | You could always apply for it... :) |
| 01:25.07 | prpplague | landley so its a wait and see |
| 01:25.37 | prpplague | landley so far i've emailed the boss(not the owner) with 43 mistakes in the current design |
| 01:25.51 | prpplague | landley i'm not EE |
| 01:25.56 | landley | Ah. |
| 01:26.00 | landley | can't tell the difference. |
| 01:26.04 | landley | My father is an electrical engineer. |
| 01:26.10 | landley | Has been one since before I was born. |
| 01:26.23 | landley | Me, I do software. |
| 01:26.42 | prpplague | landley i'm more like a "liason" |
| 01:26.51 | prpplague | landley middle man for the EE and software |
| 01:26.57 | landley | bsdfox: not that I noticed. |
| 01:27.13 | landley | It pops up a big ascii art picture of the words "Free BSD", but that's the bootloader. |
| 01:27.20 | bsdfox_ | landley: yeah |
| 01:27.26 | bsdfox_ | is there a single user mode option there? |
| 01:27.41 | landley | bsdfox_: yes, that was the "hit 4" part above. |
| 01:27.45 | landley | It still booted me into the installer. |
| 01:28.15 | prpplague | RobotGuy its very tense at the office right now about the EE, so knowing if you were joking was kind of hard |
| 01:28.26 | landley | FreeBSD isn't really seeing a lot of new users trying it out for the first time these days, is it? |
| 01:28.27 | RobotGuy | prpplague: Sorry |
| 01:28.36 | bsdfox_ | doubt it |
| 01:28.40 | RobotGuy | landley: I like FreeBSD |
| 01:28.58 | landley | RobotGuy: I'm trying to install it because somebody wanted to build tinycc on it, and couldn't get it to work. |
| 01:29.04 | ds2 | prpplague: what's his background? Dervy? |
| 01:29.06 | bsdfox_ | I've run a pair of freebsd servers for almost 10 years now |
| 01:29.11 | landley | I need a test environment. I was hoping to find a livecd with development tools. |
| 01:29.31 | landley | bsdfox_: I know several people who have been using freebsd for a long time. |
| 01:29.33 | bsdfox_ | landley: pc-bsd would work for you I think |
| 01:29.42 | landley | I don't know anybody new who has adopted it in the past 5 years. |
| 01:29.45 | bsdfox_ | pretty sure that boots as a livecd |
| 01:29.45 | prpplague | landley hows the run time option for tinycc working? |
| 01:29.55 | landley | prpplague: it works fine on a 32 bit host. |
| 01:30.01 | landley | It hasn't got a 64-bit code generator. |
| 01:30.08 | RobotGuy | prpplague: Do you know where I could get an S3C2440 based board with 128MB/128MB? |
| 01:30.15 | landley | So if you use -run on a 64 bit host, it tries to call a 32 bit function in 64 bit mode and goes "boing". |
| 01:30.27 | landley | Unless you build for a 32 bit target and have /lib32 on your host... |
| 01:30.44 | landley | needs to write a 64 bit code generator. |
| 01:31.08 | prpplague | landley one thing AML would be interested in would be to use tinycc in run time for some small db apps |
| 01:32.20 | landley | prpplague: ONe of my short term todo items is testing arm code generator on arm host. |
| 01:32.39 | landley | I know that the x86 code generator won't work on an arm host because it does stupid things with unaligned access to ints. |
| 01:32.45 | landley | But I _think_ that the arm code generator might work. |
| 01:32.56 | landley | plans to fix the x86 one soonish... |
| 01:33.04 | prpplague | landley what we'd want to do is generate C on a windows pc and then send it to an arm device and run it there |
| 01:33.20 | landley | Um, ok? |
| 01:33.38 | landley | hasn't touched the windows host code, but has a test environment set up for it... |
| 01:33.50 | landley | It's been a bit far down on my todo list, but interested users tend to bubble things up... |
| 01:34.01 | landley | Ah, generating _c_ on a windows host. |
| 01:34.03 | landley | Ok, that makes more sense. |
| 01:34.11 | prpplague | landley sorry what i mean is we'd have an app on the windows side that would generate a C text file on the windows pc |
| 01:34.17 | prpplague | landley right |
| 01:34.21 | landley | Sure, I should be able to make that work. |
| 01:34.48 | landley | I meant to test the tinycc arm hosted under qemu last night, but wound up falling asleep at 6pm and sleeping through until morning. |
| 01:34.58 | landley | And I meant to do it tonight but here I am still at work at 8:30. :) |
| 01:35.07 | landley | should go get dinner and head to Starbucks. |
| 01:35.16 | prpplague | landley you in houston? |
| 01:35.21 | landley | Alas, yes. |
| 01:35.46 | landley | Why do you ask? |
| 01:35.47 | prpplague | landley ahh |
| 01:35.51 | prpplague | landley just curious |
| 01:36.00 | prpplague | landley i know being away from home sucks, hehe |
| 01:36.46 | prpplague | landley for data collection devices, a common item is to be able to configure |
| 01:36.53 | prpplague | landley a program on the windows side |
| 01:37.11 | prpplague | landley i.e. ask the upc code, enter qty and then shelf location |
| 01:37.22 | prpplague | landley then transfer it to the handheld |
| 01:37.55 | MonMotha | prpplague: you still work for that company that makes handheld barcode scanner stuff |
| 01:38.00 | MonMotha | ? |
| 01:38.11 | prpplague | landley problem is you either have to use an interpretive language, which is slow and cumbersome, or compile the code on windows for arm which means you have to include a cross compiler under windows |
| 01:38.15 | prpplague | MonMotha yep |
| 01:38.21 | prpplague | MonMotha atleast for now |
| 01:38.35 | MonMotha | any I could buy in low quantity (maybe 3) that are somewhat OSS friendly? |
| 01:39.00 | prpplague | depends on what price range |
| 01:39.08 | prpplague | MonMotha got a particular project? |
| 01:39.14 | MonMotha | setting up something at work that does inventory tracking, and a "smart" wireless barcode scanner platform would be handy |
| 01:39.16 | landley | prpplague: I'll try to test tinycc on the arm host after dinner, and let you know what I find out tomorrow. |
| 01:39.27 | MonMotha | so I could scan, view quantity of a part in the box, "pull"/"place", tec. |
| 01:39.27 | landley | If it breaks, I'm pretty sure I can fix it. :) |
| 01:39.42 | prpplague | landley np, not a high priority item, but it is on the radar for not just us |
| 01:39.44 | MonMotha | as for price range, I don't know what the budget would be. It's just an idea at this point |
| 01:40.01 | prpplague | landley i got some info thru the grape vine that some others are watching your tinycc dev too |
| 01:40.10 | MonMotha | the system is in development, but any wireless terminal functionality would be additional |
| 01:40.28 | prpplague | MonMotha of course the big thing is when you add an intergrate bc scanner |
| 01:41.06 | MonMotha | well, that would be the point in this case |
| 01:41.13 | MonMotha | so I can scan the shelf and box right on the spot |
| 01:41.14 | ds2 | hammer + LCD + cuecat? |
| 01:41.39 | MonMotha | there's no reason I can't pull the box then take it to a terminal, but scanning it on the spot minimizes error and misplacements |
| 01:41.53 | MonMotha | the cuecat is hardly the idea scanner for this applicaion |
| 01:42.03 | prpplague | MonMotha the M7220 is relatively new and has a 802.11g radio |
| 01:42.11 | ds2 | cheap and well hacked |
| 01:42.19 | prpplague | MonMotha with everything opensource |
| 01:42.54 | prpplague | http://www.amltd.com/m7220.asp |
| 01:43.30 | prpplague | the m7220 is has a basically a hammer as the core |
| 01:43.43 | MonMotha | Linux on ARM9 |
| 01:43.47 | MonMotha | sounds hackable enough |
| 01:44.02 | landley | prpplague: cool. |
| 01:44.08 | landley | _I'm_ watching it, for one thing. :) |
| 01:44.16 | MonMotha | prpplague: "ROM" is flash, I presume |
| 01:44.23 | prpplague | MonMotha yea market speak |
| 01:44.58 | MonMotha | ah. 16MB NOR Flash |
| 01:45.24 | RobotGuy | prpplague: Check this out.. http://www.pastebin.ca/977772 |
| 01:45.37 | MonMotha | prpplague: what's ballpark pricing on that? |
| 01:45.52 | MonMotha | it looks like it would be more than sufficient, but not total overkill |
| 01:46.16 | prpplague | MonMotha are sitting down? |
| 01:46.35 | MonMotha | that bad, eh? |
| 01:46.39 | prpplague | MonMotha ballpark is around $1200 |
| 01:46.41 | RobotGuy | I think their chipset detection is screwed up. |
| 01:46.46 | MonMotha | well, that's likely a deal breaker, then |
| 01:47.05 | prpplague | RobotGuy indeed |
| 01:47.08 | MonMotha | to be fair, there IS a lot of "stuff" in there |
| 01:47.42 | prpplague | MonMotha check with me in the morning, let me see if can get a more exact value |
| 01:47.58 | prpplague | MonMotha is the software going to just used inhouse or resold? |
| 01:48.22 | wmat | landley: which ISP would that be? |
| 01:48.42 | MonMotha | prpplague: this is in-house only |
| 01:48.58 | MonMotha | there's the remote possibility it might eventually get sold, but it's being developed to fulfill in-house needs |
| 01:49.53 | landley | wmat: the one solar runs? |
| 01:49.57 | landley | Or works for? |
| 01:50.17 | wmat | landley: both, i guess |
| 01:50.26 | landley | wmat: gni.com |
| 01:51.09 | wmat | landley: ah, neat |
| 01:51.20 | prpplague | MonMotha ahh ok, any chance that if you do develop it, we might be able to do a press release or some such market stuff? |
| 01:51.41 | MonMotha | probably |
| 01:52.01 | landley | wmat: that said, according to his comments on #gentoo-embedded earlier today, the ISP he works for isn't at the cheap end of the pool. :) |
| 01:52.13 | MonMotha | don't count on it being developed into a commercial product, but if that were to happen I'm guessing we'd be able to let you guys do that |
| 01:52.35 | wmat | landley: as of right now, given the state of elinux.org, cost would likely be our last concern |
| 01:52.51 | landley | wmat: /msg him. Or log onto #gentoo-embedded and ask him about it there. |
| 01:52.52 | MonMotha | this is a bill of materials management system coupled with parts inventory tracking |
| 01:52.54 | landley | He's online now... |
| 01:53.04 | MonMotha | basically, it lets us manage alternatives for parts and track our inventory |
| 01:53.16 | *** join/#edev dijenerate (n=dijenera@65.48.147.188) |
| 01:53.37 | landley | (Admittedly for a variant of "now" that was an hour ago...) |
| 01:54.11 | prpplague | MonMotha i know the marketing is looking for more success stories for folks using the opensource aspect of the handhelds |
| 01:54.30 | MonMotha | (the biggest reason for going custom is that the goal is to get it completely integrated into my ECAD software) |
| 01:54.42 | MonMotha | well, I can look into that |
| 01:54.49 | MonMotha | right now, I've got nothing more than a backend |
| 01:54.55 | wmat | landley: i'll note the contact and ping him if we move hosts. We just moved though, so I'd rather not go through it again anytime soon :) |
| 01:54.57 | MonMotha | I've got a database and an upload page |
| 01:55.11 | prpplague | MonMotha give me shout in the morning |
| 01:55.41 | MonMotha | ok |
| 01:55.45 | prpplague | MonMotha what we've been looking for are more ISV's that can take advantage of open source |
| 01:55.56 | MonMotha | well, that is exactly what our company does |
| 01:56.16 | MonMotha | we do HW/SW product development, but obviously there are times when it's best to just spec somebody else's gear as a component of the total solution |
| 01:56.40 | prpplague | indeed we do the same |
| 01:56.49 | MonMotha | we have 4 software guys and only one hardware guy (me) |
| 01:57.00 | MonMotha | so that should give you an idea of the balance of time spent on our projects |
| 01:57.03 | MonMotha | heck, I even do software |
| 01:57.27 | landley | wmat: having your first experience at the new host be over 8 hours of downtime can't be good. |
| 01:57.27 | MonMotha | I'll talk with you in the morning if you're busy now. I'd be at work then, too |
| 01:57.58 | MonMotha | we currently have all the work we can handle, but I know the people up top are always interested in good contacts |
| 01:58.32 | prpplague | MonMotha naw, not busy its just i can't get with the sales mutants to get firm pricing |
| 01:59.04 | MonMotha | ah, well, firm pricing is not needed at this point |
| 01:59.21 | MonMotha | the handheld aspect is one of those "ooooh. wouldn't it be cool if..." type things |
| 01:59.26 | prpplague | MonMotha i probably could get you a loaner to eval |
| 01:59.27 | MonMotha | so a ballpark number is sufficient |
| 01:59.46 | MonMotha | well, I'll keep in tough wrt an eval unit. Like I said, just not to that stage yet |
| 02:00.31 | prpplague | MonMotha np, i still have a set of lcd's for you |
| 02:00.40 | prpplague | MonMotha just never gotten clearance to ship |
| 02:00.57 | MonMotha | ah, cool |
| 02:01.16 | MonMotha | I'm not sure what the status is on that project. It was a back-burner project to begin with |
| 02:01.28 | MonMotha | I'm sure I can find other uses, though :) |
| 02:01.55 | prpplague | MonMotha np |
| 02:02.35 | MonMotha | and oh wow, I need to brush up on my SQL |
| 02:02.38 | MonMotha | it's been so long.. |
| 02:03.59 | prpplague | MonMotha imho thats kind of like saying you'd like to brush up on your LSD usage |
| 02:05.01 | MonMotha | well, SQL can have some similar effects |
| 02:05.11 | MonMotha | the S decidedly does NOT mean "Simple" |
| 02:05.32 | MonMotha | (either officially or unofficially) |
| 02:13.06 | prpplague | heads to bed |
| 02:13.11 | prpplague | later folks |
| 02:13.21 | prpplague | wmat please give me an update asap |
| 02:13.27 | prpplague | wmat via email |
| 02:16.40 | *** join/#edev RobotGuy (i=robotguy@dsl093-038-072.pdx1.dsl.speakeasy.net) |
| 02:17.12 | RobotGuy | returns |
| 02:52.28 | *** join/#edev CP|home (n=nobody@www.crouse-house.com) |
| 02:59.56 | *** join/#edev agaffney (n=agaffney@gentoo/developer/pdpc.active.agaffney) |
| 04:01.43 | *** join/#edev DrZimmerman (n=theo@tbaumgartner.broker.freenet6.net) |
| 04:44.14 | *** join/#edev RobotGuy (n=robotguy@dsl093-038-072.pdx1.dsl.speakeasy.net) |
| 05:02.44 | *** join/#edev harpal (n=harpal@124.125.79.150) |
| 05:23.09 | *** join/#edev harpal_ (n=harpal@121.246.75.165) |
| 05:59.20 | *** join/#edev drath_ (i=vmaster@p5B07C793.dip.t-dialin.net) |
| 07:37.28 | *** join/#edev Appleman1234 (n=Appleman@131.181.47.39) |
| 07:44.11 | *** join/#edev linac (n=lin@59.61.124.100) |
| 08:02.57 | *** join/#edev [AD]Turbo (n=turbo@host131-128-dynamic.57-82-r.retail.telecomitalia.it) |
| 08:03.48 | [AD]Turbo | yo all |
| 08:17.44 | RobotGuy | Alloo |
| 08:34.37 | *** join/#edev Appleman1234 (n=Appleman@CPE-124-186-229-83.qld.bigpond.net.au) |
| 09:02.23 | *** join/#edev harpal (n=harpal@124.125.79.150) |
| 09:13.04 | *** join/#edev Madsy (n=madsy@ti500710a342-1985.bb.online.no) |
| 09:13.39 | *** join/#edev harpal_ (n=harpal@121.246.75.165) |
| 09:50.22 | *** join/#edev hw_ (n=hw@p578b3905.dip0.t-ipconnect.de) |
| 11:02.35 | *** join/#edev lyakh (n=lyakh@p57BD2C6D.dip0.t-ipconnect.de) |
| 11:23.21 | RobotGuy | Pssst, anyone awake?? |
| 11:23.54 | RobotGuy | tries not to wake SpeedEvil up. :) |
| 12:13.47 | *** join/#edev RobotGuy (n=robotguy@dsl093-038-072.pdx1.dsl.speakeasy.net) |
| 12:26.25 | *** join/#edev [AD]Turbo (n=turbo@82.57.128.131) |
| 12:35.44 | *** join/#edev linac (n=lin@59.61.124.100) |
| 12:38.55 | *** join/#edev prpplague (n=dave123_@ppp-70-244-80-235.dsl.rcsntx.swbell.net) |
| 12:39.01 | prpplague | wmat ping |
| 12:39.19 | prpplague | wmat any word on elinux.org? |
| 12:40.11 | prpplague | heads to work |
| 13:04.24 | *** join/#edev bobf (i=bob@unaffiliated/bobf) |
| 13:05.02 | RobotGuy | Greetings bobf! |
| 13:05.41 | bobf | Hello. :) |
| 13:07.28 | RobotGuy | This is where the heavy hitter called prpplague hangs out. He works for TinCanTool. |
| 13:07.42 | SpeedEvil | passes RobotGuy an s. |
| 13:08.12 | bobf | Cool, let's hope it's what I'm looking for. |
| 13:09.12 | RobotGuy | bobf: From what you were describing to me on #python, it looks like Hammer could be a good fit. |
| 13:09.32 | RobotGuy | Yeah yeah, SpeedEvil. :P |
| 13:09.38 | bobf | Yeah, this looks very close to what I want. |
| 13:09.44 | bobf | It'll certainly work for a prototype. |
| 13:10.24 | RobotGuy | SpeedEvil: I have WiFi configured at boot now, with Hammer. :D |
| 13:10.53 | RobotGuy | bobf: The price of the whole dev kit is pretty reasonable. |
| 13:11.08 | SpeedEvil | RobotGuy: boot over nfs over wifi? |
| 13:11.21 | RobotGuy | SpeedEvil: From Flash |
| 13:11.42 | SpeedEvil | Boring. |
| 13:11.56 | RobotGuy | SpeedEvil: Oh yeah? I think it's rather exciting. |
| 13:12.06 | RobotGuy | I am also working on the Bluetooth link. |
| 13:12.15 | bobf | I just need a board that has a USB interface, flash card slot, can preferably run Python, and that will let me stick a dtmf processor chip on too. |
| 13:13.00 | RobotGuy | bobf: The Hammer Carrier Board has a pretty generous prototype area, TWO USB Host/Slave ports. |
| 13:13.11 | bobf | Yeah, it looks perfect. |
| 13:13.24 | RobotGuy | It does have an MMC/SD slot that is accessed via SPI. |
| 13:13.42 | bobf | It'll be a device that sits between the a phone and a phone line socket and processes numbers and does stuff. |
| 13:13.50 | RobotGuy | I need to get an MMC or SD card so I can tinker with that. |
| 13:14.00 | bobf | Yeah, as long as it's got some storage on there it's great. |
| 13:14.07 | bobf | I only need a little. |
| 13:14.36 | RobotGuy | bobf: You can also use a USB Thumb Drive. They work great! I have a 2 GB PNY unit I got at Best Buy. |
| 13:14.40 | bobf | I'm glad to finally see something that actually has everything I need on it - it's been a nightmare trying to research this, having no previous knowledge at all. :) |
| 13:14.50 | bobf | Ah yeah, neat. |
| 13:14.57 | RobotGuy | bobf: It's great we connected up then. |
| 13:15.06 | bobf | Yeah, thanks a lot. :) |
| 13:15.20 | RobotGuy | I'm glad to help anytime. I spend a lot of time on here. |
| 13:15.35 | bobf | I'll tell my boss to buy me one and I can start playing. |
| 13:15.45 | *** join/#edev dijenerate_ (n=dijenera@66.54.118.16) |
| 13:15.55 | RobotGuy | Wonderful! |
| 13:16.28 | RobotGuy | I don't think you'll find a smaller full Linux system than Hammer (40 pin DIP). |
| 13:16.45 | RobotGuy | There is also a Hammer-RDP board in development now. |
| 13:18.11 | *** join/#edev prpplague (n=dave@mail.americanmicrosystems.com) |
| 13:18.20 | bobf | So the hammer kit comes with the cpu? |
| 13:18.42 | SpeedEvil | yes |
| 13:18.53 | SpeedEvil | it's a little board 1*2" or so, with 40 pins |
| 13:19.02 | SpeedEvil | It's got CPU, flash, RAM onboard. |
| 13:19.03 | prpplague | morning all |
| 13:19.09 | prpplague | wmat: any update? |
| 13:20.22 | bobf | Hm, I think just the Hammer and the Hammer Carrier should do what I want. |
| 13:20.56 | prpplague | bobf: unless you have a good jtag device already, i highly suggest getting one |
| 13:21.07 | RobotGuy | bobf: Very probably. If you need smaller, the Hammer-RDB will be about the size of a Lynxmotion Bot Board II. |
| 13:21.16 | RobotGuy | RDP that is |
| 13:21.42 | bobf | prpplague: This device is something that I only need one of though, right? It's not something that would be included on a production model? |
| 13:21.44 | RobotGuy | prpplague: Greetings! I have good news!! |
| 13:21.53 | prpplague | bobf: that is correct |
| 13:22.06 | bobf | Okay cool, I'll tell the boss to fork out a bit more cash then. :) |
| 13:22.08 | prpplague | RobotGuy: you won the lottery and your girl friend is pregnant? |
| 13:22.34 | RobotGuy | prpplague: 1) I didn't win the lottery. 2) I don't have a girlfriend. |
| 13:22.39 | prpplague | RobotGuy: doh |
| 13:22.42 | RobotGuy | :P |
| 13:22.57 | prpplague | RobotGuy: do tell |
| 13:23.08 | RobotGuy | However, I -do- have kernel 2.6.25-rc8-hammer working AND Wifi comes up at boot now. :D:D |
| 13:23.38 | RobotGuy | I have not slept yet. ;) |
| 13:24.45 | RobotGuy | I can't test the WiFi any more though, because I don't have a working WiFi router at present (got two that don't do WiFi or much of anything else anymore though). |
| 13:26.08 | RobotGuy | All the configuration short of associating with a router/AP is correct though. Well except for the broadcast address and netmask. I'm not sure if that will make a difference or not yet. |
| 13:27.37 | prpplague | RobotGuy: good deal |
| 13:27.55 | prpplague | begins his lcd fun |
| 13:28.25 | *** join/#edev chouimat|work (n=dieu@209.217.106.98) |
| 13:28.44 | prpplague | wish i could get an update on the wiki |
| 13:30.47 | RobotGuy | prpplague: For the LCD? |
| 13:31.07 | RobotGuy | Why can't you get an update? |
| 13:31.57 | prpplague | RobotGuy: wmat hasn't gotten any info from godaddy yet about the domain |
| 13:32.09 | prpplague | RobotGuy: lcd on the new board |
| 13:32.20 | RobotGuy | prpplague: Ah, OK. New domain? |
| 13:32.29 | prpplague | ~lart RobotGuy |
| 13:32.29 | purl | whips out a shotgun, trudges over to RobotGuy, and goes postal |
| 13:32.32 | RobotGuy | I have 11 domains now. |
| 13:32.39 | prpplague | RobotGuy: elinux.org has been down since yesterday morning |
| 13:32.45 | RobotGuy | ACK. |
| 13:33.14 | RobotGuy | prpplague: Eeeeeeeeeeek, that's not good. I have not needed to check there for at least that long. |
| 13:33.33 | RobotGuy | I've been so busy working on Hammer and WiFi and such. |
| 13:34.42 | RobotGuy | prpplague: I found the box that has my Honeywell compass module, RTC, and 3-axis accelerometer. :D |
| 13:35.01 | chouimat|work | morning |
| 13:35.06 | RobotGuy | The compass is 3.3V so I can connect it direct to Hammer. |
| 13:35.55 | RobotGuy | chouimat|work: Greetings |
| 13:47.24 | RobotGuy | Cool, my RTC module (Maxim DS1307) is also 3.3V. |
| 13:48.20 | RobotGuy | That gives me two I2C devices I can tinker with right away. Sienna will be happy to hear that. |
| 13:51.40 | RobotGuy | Cool, even my 3 axis accelerometer is 3.3V. :D:D |
| 13:56.51 | *** join/#edev LyosNoreze1 (n=Ralph@75-93-179-25.day.clearwire-dns.net) |
| 14:30.29 | *** join/#edev Mirell_ (n=mirell@otomi.ling.utexas.edu) |
| 14:35.48 | RobotGuy | prpplague: I need to find out about this inactive memory. Now I have 2708 inactive. :( |
| 14:49.36 | *** join/#edev harpal__ (n=harpal@124.125.78.84) |
| 14:51.43 | *** join/#edev CosmicPenguin (n=nobody@163.181.251.103) |
| 14:51.43 | *** mode/#edev [+o CosmicPenguin] by ChanServ |
| 15:19.04 | *** join/#edev landley (n=landley@adsl-68-93-84-18.dsl.hstntx.swbell.net) |
| 15:19.04 | *** mode/#edev [+o landley] by ChanServ |
| 15:21.26 | landley | Morning. |
| 15:21.40 | landley | prpplague: arm tinycc works just fine on an arm host once the config is worked out. |
| 15:22.03 | prpplague | landley: cool cool |
| 15:22.13 | *** mode/#edev [+o prpplague] by ChanServ |
| 15:22.28 | prpplague | CosmicPenguin: still got a job? |
| 15:23.21 | landley | ds2: http://www.fiwix.org/ |
| 15:23.57 | prpplague | RobotGuy: yea, when i checked mine i only had around 256k inactive |
| 15:24.20 | RobotGuy | prpplague: So why would I have so much inactive memory? |
| 15:24.34 | prpplague | RobotGuy: not sure |
| 15:24.56 | prpplague | RobotGuy: i'll be happy to look when i have time, but i have a major deadline to deal with |
| 15:25.09 | RobotGuy | Ok. |
| 15:27.51 | *** part/#edev LyosNorezel (n=Ralph@unaffiliated/lyosnorezel) |
| 15:30.21 | CosmicPenguin | prpplague: my badge worked |
| 15:30.24 | CosmicPenguin | thats all I know |
| 15:35.29 | *** join/#edev LyosNorezel (n=Ralph@unaffiliated/lyosnorezel) |
| 15:37.56 | *** join/#edev chouimat|work (n=dieu@209.217.106.98) |
| 15:47.24 | prpplague | CosmicPenguin: probably won;t let you know till friday at 4:30 |
| 15:49.41 | CosmicPenguin | nah - thats not their style |
| 15:50.05 | prpplague | look like the celinuxforum.org server was hacked |
| 15:50.19 | prpplague | which means that the wiki was probably compromised as well |
| 15:51.00 | prpplague | wmat: any news on elinux.org? |
| 15:53.19 | *** join/#edev magnetron (n=magnetro@unaffiliated/magnetron) |
| 16:11.49 | landley | CosmicPenguin: was there a likelihood of you losing your job? |
| 16:11.55 | landley | (Did you look under the sofa?) |
| 16:12.50 | CosmicPenguin | its always in the last place you look |
| 16:14.50 | *** join/#edev lyakh (n=lyakh@p57BD2C6D.dip0.t-ipconnect.de) |
| 16:17.29 | *** join/#edev dirk3 (n=dirk@F3239.f.strato-dslnet.de) |
| 16:34.07 | dirk3 | Sorry if this a faq, but any news about elinux.org status? |
| 16:35.09 | dirk3 | Seems that these days e.g. http://etrunko.blogspot.com/2008/04/android-running-on-n810.html is very popular, but links to elinux are dead... |
| 16:35.56 | dirk3 | I already saw a page about above topic linking to the Google cache of elinux |
| 16:36.17 | wmat | prpplague: no news, i'm going to call GoDaddy now |
| 16:37.29 | prpplague | wmat: i was told that celinuxforum and several other sites related to celinux had been hacked |
| 16:37.50 | wmat | prpplague: i knew celoinuxforum was hacked last weeked |
| 16:37.57 | prpplague | dirk3: hopefully hear some news soon |
| 16:38.05 | prpplague | wmat: is it hosted on the same server? |
| 16:38.34 | wmat | prpplague: no, only elinux.org is at godaddy |
| 16:38.43 | prpplague | ahh ok |
| 16:41.41 | LyosNorezel | my website is also at godaddy... and mine's working fine |
| 16:42.09 | LyosNorezel | so whatever it is... it's not a global godaddy issue |
| 16:44.19 | prpplague | is rikers pages up? |
| 16:44.25 | prpplague | ~nslookup www.rikers.org |
| 16:44.52 | RobotGuy | I have Python and SQLite in Flash now. :D\ |
| 16:45.29 | SpeedEvil | ~nslookup qkwv.com |
| 16:47.08 | prpplague | ok sooo, i normally like a good hardware puzzle, but this board is starting to really piss me off |
| 16:47.54 | wmat | prpplague: yes, Tim R's pages are fine |
| 16:47.55 | LyosNorezel | ~nslookup www.elinux.org |
| 16:48.31 | SpeedEvil | prpplague: is there a second board - in case it's a one-off? |
| 16:48.51 | prpplague | SpeedEvil: these are design issues |
| 16:51.14 | SpeedEvil | Sure - however assembly errors that are assumed to be design errors can be anoying. |
| 16:51.53 | prpplague | SpeedEvil: these are pure design errors |
| 16:52.17 | prpplague | points towards to EE's office |
| 16:52.20 | wmat | prpplague: godaddy is clueless |
| 16:52.34 | prpplague | wmat: lovely |
| 16:52.44 | prpplague | wmat: so what are we going to do? |
| 16:52.58 | Crofton | switch to geekisp? |
| 16:53.09 | wmat | heh |
| 16:53.28 | prpplague | thats fine, but we need to get the latest data from the server |
| 16:53.33 | Crofton | godaddy hosts elinux and can't find it? |
| 16:53.50 | wmat | well, first we need Tim R. to somehow recover the site, as part of my condition of switching to a debian based VM was that he admin it. |
| 16:54.10 | prpplague | wmat: can we even get into the box? |
| 16:54.13 | wmat | i'm awaiting his email or his presence in #elinux |
| 16:54.19 | prpplague | wmat: as am i |
| 16:54.33 | wmat | prpplague: not presently. |
| 16:54.35 | Dr{Afk} | prp.. what exactly is the issue? what is godaddy '''not''' doing.. updating the dns records? |
| 16:54.44 | Crofton | it would be just our luck he is on vaction :( |
| 16:54.52 | prpplague | Dr{Afk}: no clue |
| 16:55.08 | prpplague | Crofton: he's on holiday? |
| 16:55.12 | wmat | Dr{Afk}: I actually dobt this is godaddy's fault |
| 16:55.29 | Crofton | prpplague, jusst speculation |
| 16:55.37 | prpplague | wmat: is godaddy able to access the server? |
| 16:55.56 | Dr{Afk} | k ill have my cisco network admin look see if he can find anyting |
| 16:55.57 | prpplague | Crofton: ahh |
| 16:56.47 | prpplague | the server is responding to pings |
| 16:56.58 | Dr{Afk} | ok the data all comes up for dns correct so if its pointed to the correct ip or the ip its pointed to is net setup with apache or a web server |
| 16:57.08 | Dr{Afk} | i would contact the noc. |
| 16:57.14 | wmat | prpplague: yes, but i want Tim R. to call them and tell them what to fix |
| 16:57.33 | prpplague | wmat: hmm, no phone contact for tim? |
| 16:57.34 | Crofton | ip-208-109-232-74.ip.secureserver.net |
| 16:57.37 | Crofton | from tracepath |
| 16:58.13 | Dr{Afk} | noc@godaddy.com +1-480-505-8809 |
| 16:58.51 | Crofton | whois elinux.org has some numbers |
| 16:59.35 | Crofton | google suggests it will reach Tim :) |
| 16:59.43 | Crofton | loves stalking people |
| 16:59.44 | wmat | i've emailed the Tims again and now must get back to the dayjob |
| 16:59.48 | wmat | prpplague: i'll keep you posted |
| 17:00.10 | prpplague | wmat: ok |
| 17:01.31 | Dr{Afk} | btw.. prpplague.. if you do need hosting im sure others here have too but i have over commited my bandwith. i have 20mbits and only use a little so if you need a mirror or antything we are discussing how we can use this resource and at least get something for our money. |
| 17:02.20 | prpplague | Dr{Afk}: CELF normally handles that, i'm mainly just nag/editor of elinux stuff |
| 17:02.23 | Dr{Afk} | or you know someone.. we have talked about putting up a game server or mirror site shrug.. |
| 17:03.51 | Dr{Afk} | need to find a way to burn 15mbits : / or just let it rot. |
| 17:04.31 | prpplague | pr0n server? |
| 17:04.39 | Dr{Afk} | hehe.. |
| 17:08.42 | Dr{Who} | interesting though i have never used godaddy's myself ever time i have seen a virtual setup on godaddy the server has a blank landing page by default.. so it does seem odd that it would be a configuration error on the customers end.. |
| 17:08.50 | RobotGuy | Dr{Who}: Greetings |
| 17:08.50 | Dr{Who} | is it just a standard godaddy virtual? |
| 17:08.56 | Dr{Who} | hay robotguy. |
| 17:09.38 | RobotGuy | I have space on my hosting account at webfaction.com |
| 17:10.43 | RobotGuy | Dr{Who}: I have Python and SQLlite on flash now. :) |
| 17:10.44 | Dr{Who} | hmm that server is running an irc server and looks like a windows box??????? |
| 17:10.52 | Dr{Who} | wow.. |
| 17:10.56 | Dr{Who} | how big did python get? |
| 17:11.26 | RobotGuy | Well, I have some other dev stuff also, but the whole JFFS2 image is 9 - 10 MB. |
| 17:11.27 | Dr{Who} | sqllite i have done its nice and small. but all my testing of python make me think its going to be too big. |
| 17:11.41 | wmat | Dr{Who}: it was a standard godaddy vm until we reprovisioned it with a debian installation |
| 17:11.41 | RobotGuy | Dr{Who}: It fits fine. |
| 17:11.52 | Dr{Who} | not bad. did you buildroot python or custom cross build it? |
| 17:12.13 | RobotGuy | Dr{Who}: /dev/mtdblock1 14.8M 10.2M 4.5M 69% / |
| 17:12.40 | RobotGuy | Dr{Who}: It's all buildroot |
| 17:13.15 | Dr{Who} | cool. |
| 17:13.18 | RobotGuy | Not bad for a fully self contained Hammered robot. :) |
| 17:13.24 | Dr{Who} | not at all. |
| 17:13.43 | RobotGuy | And I still have plenty of room for Python code. |
| 17:17.35 | Dr{Who} | i helped guido back in 97 resolve the thread safety issues on python. he didnt go with my method but at least i helped push him fix the core problem in python back then. its much better today thanks to that mejor design change. |
| 17:18.34 | RobotGuy | I also have all the utils to firewall W.A.L.T.E.R. so he's protected as well as he can be. |
| 17:20.40 | RobotGuy | It's kind of scary.. I am really getting where I can manipulate buildroot. |
| 17:20.42 | Dr{Who} | i basicly refactored all of python to be thread safe using a state variable and found all of the globals and wrapped them. his method was to use tls. I still think the state object is a better method. |
| 17:20.49 | RobotGuy | Cool! |
| 17:21.37 | RobotGuy | I am planning to put either DJango or Pylons (Web framework) and Phenny (IRC Bot) on W.A.L.T.E.R. also. |
| 17:21.51 | landley | CosmicPenguin: the reason it's in the last place you look is that when you find it, you stop looking. |
| 17:21.52 | RobotGuy | It's all written in Python. |
| 17:23.42 | Dr{Who} | nice. python is fun. |
| 17:26.00 | RobotGuy | Well, I wanted everything to be in a single language - much easier to glue things together that way, and I like Python. Oh yeah, I will hopefully be using PyRo (Python Robotics) as the control framework for W.A.L.T.E.R. |
| 17:26.08 | landley | RobotGuy: how much space did python take up on hammer? |
| 17:26.44 | RobotGuy | landley: Well, I'm not sure how much space just Python took up. I'd have to do a build with and without it to see. |
| 17:27.54 | RobotGuy | I just know I have Python and SQLlite on Flash, and still have about 4.5 MB of space left. |
| 17:28.06 | RobotGuy | There is a bunch of other stuff on the flash to though. |
| 17:32.56 | RobotGuy | Now I just need a good WiFi router (not a gateway) to test W.A.L.T.E.R.'s WiFi with. |
| 17:51.56 | *** join/#edev NishanthMenon (n=gnat@nat/ti/x-6b3a85b693bce8b4) |
| 17:57.48 | landley | RobotGuy: this is jffs2, which is compressed, right? |
| 17:57.57 | landley | prpplague: what would be involved in getting uboot up on hammer? |
| 17:57.58 | RobotGuy | landley: Right |
| 17:58.19 | landley | I realize it has no obvious benefits, but there seem to be a lot of people familiar with it and wanting it as a security blanket. |
| 17:58.46 | prpplague | landley: uboot already supports the s3c2410, so it would of little effort to do it |
| 17:59.08 | prpplague | landley: but speaking for tct we will only support apex |
| 17:59.20 | landley | shrugs. |
| 17:59.25 | landley | It seems to be an emacs/vi thing. |
| 17:59.30 | prpplague | landley: indeed |
| 17:59.32 | landley | Matter of what people are familiar with. |
| 17:59.52 | prpplague | ye |
| 17:59.53 | prpplague | yea |
| 18:01.24 | RobotGuy | Could APEX theoretically boot up a full distribution of Linux provide there were adaquate resources? |
| 18:01.26 | landley | I'll put it on my hammer todo list. |
| 18:01.31 | landley | RobotGuy: sure. |
| 18:01.45 | landley | The bootloader doesn't care what the kernel it hands control off to does. |
| 18:02.18 | landley | You used to be able to boot a full linux distro via cat bzImage > /dev/fd0 |
| 18:02.38 | landley | Back before they yanked the built-in floppy bootsector as totally irrelevant to modern anything and something they didn't wanna support no more. :) |
| 18:04.27 | RobotGuy | I guess they call that progress. |
| 18:07.02 | landley | RobotGuy: more that loading sectors off a floppy drive into high memory (to avoid the 640k kernel size limit, despite the bios being 16 bit code) is something modern bios writers really don't debug much anymore. |
| 18:07.09 | landley | How many laptops still have a floppy drive? |
| 18:07.17 | landley | And if your floppy drive is on USB, will that bios code work? |
| 18:07.26 | landley | It never worked for non-floppy drives, by the way. |
| 18:07.31 | RobotGuy | My PCs both still have floppy drives. |
| 18:07.59 | landley | How old is your newest PC? |
| 18:11.07 | RobotGuy | 3 years probably, going on 4. |
| 18:12.26 | *** part/#edev NishanthMenon (n=gnat@nat/ti/x-6b3a85b693bce8b4) |
| 18:19.48 | *** join/#edev T0mW (n=Tom@24.115.218.135.res-cmts.sth.ptd.net) |
| 18:27.28 | *** join/#edev dijenerate (n=dijenera@65.48.147.188) |
| 18:32.14 | *** join/#edev emil|k (n=emil@e179235114.adsl.alicedsl.de) |
| 18:37.13 | landley | The iMac was really strange when it discarded the floppy drive back around 1999. |
| 18:37.23 | landley | By about 2005, everybody else had caught up. |
| 18:37.29 | *** part/#edev dirk3 (n=dirk@F3239.f.strato-dslnet.de) |
| 18:37.41 | landley | These days, if you want to copy files from computer to computer, you use a network connection. |
| 18:37.51 | landley | If you need boot media, there are burnable CDs and USB keys. |
| 18:38.08 | landley | (Which you can also use for "sneakernet" if you really want to, or for backup purposes.) |
| 18:38.26 | landley | Although really, they're too small for serious backups. Even burnable DVDs are on the small side, you need to burn a stack to copy off a whole 100+gig hard drive. |
| 18:38.41 | landley | I bought a 500 gig USB hard drive for backups, and rsync to severs. |
| 18:38.44 | chouimat|work | landley: or you can do and hexdump to the printer ... |
| 18:38.56 | landley | My USB key just backs up my email and a portion of my website, and even then 4 gigs is getting tight. |
| 18:39.15 | landley | chouimat|work: would you like me to do the math on how many centuries it would take the printer to finish printing a hex dump of my laptop? :) |
| 18:39.33 | chouimat|work | landley: :) |
| 18:40.27 | prpplague | ho ho hum |
| 18:40.41 | prpplague | tries to calm down and get this pwm backlight working |
| 18:41.38 | landley | A rough estimate is that my /dev/sda2 (sata shows up as scsi, long story) is 542 thousand floppy images. |
| 18:41.49 | landley | prpplague: the EE again? |
| 18:42.04 | prpplague | landley: yea, really pissing me off |
| 18:42.15 | chouimat|work | landley: nice :) |
| 18:43.45 | landley | Did he do something new, or just his continuing existence? |
| 18:44.06 | chouimat|work | landley: he started breathing :) |
| 18:56.21 | Crofton | curses Wall Street |
| 18:56.32 | chouimat|work | Crofton: ? |
| 18:56.38 | Crofton | market is down |
| 19:00.19 | RobotGuy | ncurses stupid people |
| 19:07.01 | *** join/#edev ce_geek (n=ce_geek@roger-blanchette-4.umelst.maine.edu) |
| 19:14.45 | *** join/#edev ce_geek (n=ce_geek@roger-blanchette-4.umelst.maine.edu) |
| 19:15.04 | ce_geek | So I noticed that when I have the flyswatter connected to my PC and the hammer, if I reboot my system the hammer board reboots. |
| 19:17.41 | RobotGuy | That's normal. Most boards do the same. |
| 19:18.25 | ce_geek | What causes this? Is it the JTAG interface because the USB device is getting reset by my PC? |
| 19:18.36 | ce_geek | It must not be the serial connection... |
| 19:18.51 | ce_geek | (I guess I could test these theories...) |
| 19:20.02 | ali_as | If flyswatter connects to nRST that would cause that. |
| 19:20.58 | ali_as | Hmm, maybe. |
| 19:23.13 | ce_geek | Fair enough. |
| 19:23.25 | landley | Crofton: the market often goes down. |
| 19:23.55 | Crofton | Yes, and I should not check all the time :) |
| 19:24.38 | ce_geek | So how does the filesystem on the hammer work (as shipped) ? Seems like everything new I create disappears when I reboot, I imagine because when it boots it just unpacks the compressed root filesystem. What about logs and such that might be nice to have after a reboot? |
| 19:25.12 | landley | The filesystem as shipped is initrd. It's a ramdisk. |
| 19:25.22 | landley | What you write goes into the ramdisk, and isn't saved to the flash. |
| 19:25.37 | landley | If you want a persistent fiilesystem, you need jffs2. |
| 19:26.00 | ce_geek | gotcha. I saw discussion about that yesterday, that was not mentioned as one of the great benefits. |
| 19:26.33 | ce_geek | So far it seems that aside from some funky initial setup, JFFS2 is _the_ way to go. Still trying to dig up other negative effects though, aside from the initial setup. Are there any? |
| 19:27.19 | RobotGuy | ce_geek: I'm booting my Hammer from flash now. |
| 19:27.48 | RobotGuy | I have Python and SQLlite on there, with other dev and important stuff. |
| 19:28.11 | ce_geek | Cool. |
| 19:29.21 | ce_geek | Okay so here's a practical exercise for me. I want the IP address to be automagically configured when the hammer boots. Until I implement JFFS2, such a change will require me to modify the initrd. |
| 19:29.21 | RobotGuy | ce_geek: All the software I'll be using for my robot is written in Python. |
| 19:29.41 | RobotGuy | ce_geek: That can be done in your inittab. |
| 19:29.57 | ce_geek | Right. so I need to edit inittab in the initrd. |
| 19:30.30 | ce_geek | And one way I might do this is to uncompress the initrd and mount it on my local machine, change the file, recompress and jtag that over to the hammer. |
| 19:30.31 | ce_geek | right? |
| 19:30.31 | RobotGuy | ce_geek: Yes. It's buildroot/target/device/Samsung/Hammer/tar*/etc/inittab |
| 19:31.10 | ce_geek | Ah okay. So if I were to edit that file, what do I need to do within buildroot to get it to make that change to the initrd it spits out in binaries/Hammer/? Will "make" do it? |
| 19:31.20 | RobotGuy | ce_geek: You could do that, but it's so easy to just recreate your initrd. Everything you put in your initrd takes away from memory for applications. |
| 19:31.58 | RobotGuy | The inittab and such are automatically a part of anything built from buildroot. |
| 19:32.18 | ce_geek | I'm talking about adding one line to inittab... seems like that shouldn't be a big space hog. |
| 19:32.37 | RobotGuy | Just be sure you do a make rootclean after you have modified anything in the Samsung/Hammer/targ* |
| 19:32.40 | RobotGuy | dir |
| 19:33.13 | ce_geek | ehhhhhhhhhhhhh I've been down this rootclean path before. yuk. I'll have to go review the logs. Hopefully the next software release for hammer will have the problems fixed. |
| 19:33.19 | RobotGuy | ce_geek: Buildroot ignores the stuff there if there is not a clean root |
| 19:33.30 | ce_geek | That's good to know. |
| 19:33.39 | RobotGuy | ce_geek: Which problem? |
| 19:34.44 | RobotGuy | My hammer has about 23 MB RAM available now, but for some reason has 2708 KB inactive. |
| 19:35.13 | ce_geek | There's some stuff that doesn't quite completely get taken care of by simply doing a "make rootclean". There are directories and things that need to vaporize to make things work right. One problem I had was that boa was messed up, it doesn't get added to the initrd the second time around. I have to remove the folder for it in binaries and then it will work. |
| 19:35.26 | ce_geek | There's something else too but I'll have to go hunt for that in the log. |
| 19:36.56 | RobotGuy | My current jffs2 image is around 10 MB. |
| 19:37.39 | RobotGuy | I even have WiFi configuring at bootup now. :) |
| 19:37.44 | ce_geek | Cool |
| 19:38.26 | ce_geek | Alright I have to scram. I'll be talkin' to ya later. Things are starting to get rolling with my hammer project(s) |
| 19:38.32 | RobotGuy | Well, I have not slept yet, so I am going to go take a nap. |
| 19:38.37 | RobotGuy | Bye for now. |
| 19:38.39 | ce_geek | later |
| 19:40.06 | *** join/#edev roxfan (n=dunno@126.205-240-81.adsl-dyn.isp.belgacom.be) |
| 21:03.02 | *** join/#edev calamari (n=calamari@ip24-255-58-177.tc.ph.cox.net) |
| 21:03.10 | *** part/#edev calamari (n=calamari@ip24-255-58-177.tc.ph.cox.net) |
| 21:07.46 | *** part/#edev LyosNorezel (n=Ralph@unaffiliated/lyosnorezel) |
| 21:24.41 | *** join/#edev dijenerate_ (n=dijenera@69.73.223.242) |
| 21:28.23 | *** join/#edev dijenerate__ (n=dijenera@72.22.133.170) |
| 21:36.09 | landley | Heh. |
| 21:36.30 | landley | The penguicon panel list is up. (http://penguicon.dabbledb.com) |
| 21:38.17 | landley | Where else are you going to see Vernor Vinge, Eric Raymond, and Peter Salus on the same panel? :) |
| 21:38.22 | *** join/#edev GPSFan (n=kenm@12.10.255.246) |
| 21:38.22 | *** mode/#edev [+o GPSFan] by ChanServ |
| 21:42.50 | ds2 | hmmm elinux down is unrelated to celinuxforum down Hmmm |
| 21:44.27 | landley | I'm told that the celinuxforum stuff involved being hacked, and the elinux stuff involved moving to godaddy. |
| 21:44.30 | landley | Not sure which is worse. |
| 21:45.48 | ds2 | basically Tim's mail right? |
| 21:51.11 | *** join/#edev CosmicPenguin (n=nobody@163.181.251.103) |
| 21:51.11 | *** mode/#edev [+o CosmicPenguin] by ChanServ |
| 22:07.47 | wmat | prpplague: we have to reprovision elinux.org |
| 22:07.56 | prpplague | wmat: yea i was just reading |
| 22:07.59 | prpplague | wmat: that bites |
| 22:08.04 | wmat | prpplague: no doubt |
| 22:08.10 | prpplague | wmat: i'd not done a backup since the move |
| 22:08.21 | CosmicPenguin | ah - the default ubuntu background for hardy is cute |
| 22:08.37 | wmat | prpplague: so you did a backup right after the move? |
| 22:08.51 | prpplague | wmat: before the move |
| 22:09.09 | prpplague | wmat: we might get timr to put the g.elinux.org back in place for now |
| 22:32.33 | prpplague | heads home |
| 22:36.49 | *** join/#edev Sienna (n=Sienna@pool-71-125-188-214.bltmmd.east.verizon.net) |
| 22:38.04 | Sienna | Landley! I am sure you would know... Which way does the JTAG Interface plug into the Hammer? |
| 22:38.20 | Sienna | I didn't see it in any docs, and the wiki was down |
| 22:39.48 | landley | The wiki is sort of back up. |
| 22:39.55 | landley | http://movial.elinux.org/Main_Page |
| 22:40.06 | landley | As for how does it plug into the hammer: which jtag interface, and which carrier board? |
| 22:40.10 | landley | pulls out my nail board... |
| 22:40.27 | landley | Ok, on the end of the hammer there are some sticky-up pins? On the top? |
| 22:40.50 | landley | (Sienna: That's a technical term, sticky-up pins there. It means "I have no idea what that's called".) |
| 22:41.03 | landley | A little ribbon cable plugs into that. |
| 22:41.37 | landley | Sienna: http://movial.elinux.org/Category:TCT-Hammer |
| 22:41.42 | Sienna | I know the ribbon cable plugs into the headers (thats what they are called) |
| 22:41.53 | Sienna | i just don't know which side is pin 1 |
| 22:41.57 | landley | #include <ribbon-cable.h> |
| 22:42.04 | landley | ah, that. |
| 22:42.07 | landley | The red stripe is on... |
| 22:42.20 | landley | if the headers are at the top, it's on the left. |
| 22:42.51 | landley | That help? |
| 22:42.53 | Sienna | thats ambigous to me... is the red stripe on the pin 1 side or the pin 40 side of the hammer |
| 22:43.37 | landley | I thought the red stripe was always pin 1. |
| 22:43.51 | CosmicPenguin | sticky up pins... I must add that to my regular rotation |
| 22:43.52 | Sienna | it is... but to me, 'top' is the physical top of the board |
| 22:43.53 | landley | And in this case, the "1" on the nail board is on the red stripe side. |
| 22:44.06 | Sienna | ok |
| 22:44.15 | landley | If you hold the hammer board so that the bottom pins are facing away from you, and the headers are at the top. |
| 22:44.24 | Sienna | so, in its natural rest position, the cable comes over the hammer, right? |
| 22:44.28 | landley | The red stripe is on the left, and according to the writing on the nail board that's where pin 1 plugs in. |
| 22:44.39 | landley | Ah, you have a flyswatter. |
| 22:44.45 | Sienna | yes, sorry |
| 22:44.47 | landley | In its natural position, the flyswatter wants to turn upside down. |
| 22:44.51 | landley | No problem, I had one. |
| 22:44.56 | landley | I mailed it to DrWho here, I think. |
| 22:45.11 | landley | Pin 1 on the nail carrier board should be the same as pin 1 on the flyswatter. |
| 22:45.13 | landley | (I'd hope.) |
| 22:45.31 | landley | http://movial.elinux.org/Hammer_Board may help. |
| 22:45.32 | Sienna | me too! |
| 22:45.33 | landley | That's up now. |
| 22:45.41 | landley | Notice "movial" instead of "www". |
| 22:45.48 | landley | That's their old box from the previous ISP, I think. |
| 22:46.01 | landley | It's not quite up to date, but the godaddy box appears to be toast and they haven't got a newer backup. |
| 22:46.08 | landley | (TimRiker was discussing this over on #elinux.) |
| 22:46.31 | landley | The drawing on the page identifies pin1, but I'm not that great reading hardware schematics. |
| 22:47.01 | landley | Actually, that's not schematics, that's a spec sheet. |
| 22:47.02 | Sienna | thats pin 1 of the Hammer itself, I just needed to identify on the 2x7 JTAG header which pin is pin 1, because those were not labelled |
| 22:47.07 | landley | But I'm only medium at those either. |
| 22:47.47 | landley | Over here, it's on the left side when you're holding the hammer with the headers at the top and the headers facing towards you. |
| 22:48.18 | Sienna | ok, got it |
| 22:48.20 | landley | I take it your evaluation kit arrived. :) |
| 22:48.28 | Sienna | yes it did, just today! |
| 22:48.48 | landley | My response to the evaluation kit was to complain enough they designed a new carrier board. |
| 22:48.53 | Sienna | so, now its my turn to pester everyone with biuldroot stuff |
| 22:48.56 | landley | This is a _slight_ oversimplification of what actually happened. |
| 22:49.02 | landley | doesn't use buildroot. |
| 22:49.09 | Sienna | yeah yeah :D |
| 22:49.09 | landley | uses http://landley.net/hg/firmware |
| 22:49.15 | landley | Which I got to boot on the hammer. |
| 22:49.17 | landley | is proud of that. |
| 22:49.24 | landley | is also testing tinycc on hammer hardware. |
| 22:49.32 | ds2 | doesn't use build systems ;) |
| 22:49.33 | landley | Worked on qemu-arm last night... |
| 22:49.34 | Sienna | so you compiling natively now? |
| 22:49.44 | landley | Well, I have a native compiler. |
| 22:49.48 | landley | There's a difference. |
| 22:49.54 | landley | tinycc won't build an unmodified linux kernel yet. |
| 22:50.14 | landley | It's almost a complete c99 compiler, and supports many gcc extensions. |
| 22:50.23 | ds2 | landley: can the modifications to the kernel be done automatically with a sed or perl script? |
| 22:50.30 | landley | But "almost" involves not supporting variable length array declarations until I finish implementing them. |
| 22:50.33 | Sienna | can't you just use 'notsotinyastinycc'cc? :P |
| 22:50.57 | landley | Sienna: alas, I have to go with something that's actually been implemented. |
| 22:51.03 | landley | tinycc is actually a quite functional compiler. |
| 22:51.11 | landley | In the same way busybox is actually a quite functional set of command line tools. |
| 22:51.30 | Sienna | busybox? I thought that was a window manager |
| 22:51.38 | landley | But when you actually try to use a new implementation with something that's always used some other implementation, there's a lot of debugging and implementing of strange and obscure missing pieces. |
| 22:51.40 | ds2 | you are thinking of matchbox |
| 22:51.42 | landley | Sienna: busybox.net |
| 22:51.49 | landley | I maintained that sucker for a while. |
| 22:51.57 | ds2 | busybox is a multicall binary like what AIX did |
| 22:52.15 | landley | Busybox is an implementation of a hundred or so SUSv3 command line utilities in a single executable that's < 1 megabyte in size. |
| 22:52.28 | landley | It does the "swiss army knife" thing, it figures out what command to act like based on the name it was called under. |
| 22:52.31 | Sienna | ah, cool |
| 22:52.37 | landley | So you create a bunch of symlinks to busybox: cat, ls, mv, and so on... |
| 22:52.42 | Sienna | (and actually ds2, i had used fluxbox in the past...) |
| 22:52.55 | landley | I'm reimplementing the whole thing from scratch, of course. http://landley.net/code/toybox |
| 22:53.02 | Sienna | toybox :D |
| 22:53.06 | ds2 | Sienna: ah... matchbox seems to be used in a lot of products thesedays |
| 22:53.09 | landley | Because now that I know how they did it, I think I can do a better job. :) |
| 22:53.19 | landley | Getting X11 up and running is a whole 'nother step. |
| 22:53.23 | landley | I've done that before, but not recently. |
| 22:53.32 | ds2 | toybox with internal Kdrive? :) |
| 22:53.40 | landley | What's kdrive again? |
| 22:53.43 | Sienna | please... no X on the Hammer! |
| 22:53.44 | ds2 | tinyx |
| 22:53.45 | landley | (Sounds familiar, but I'm not placing it.) |
| 22:53.47 | landley | Ah. |
| 22:53.50 | landley | That's obsolete. |
| 22:54.09 | landley | Back under XFree86, Keith Packard and co were writing a new small not-insanely-bloated X11 engine. |
| 22:54.27 | landley | Then the maintainer of XFree86 grew a pencil-thin mustache and learned to do the whole evil laugh thing... |
| 22:54.31 | Sienna | Ok, the other piece of advice i need... how in the heck do you get this hammer out of its socket? :P |
| 22:54.38 | CosmicPenguin | kdrive still lives |
| 22:54.40 | landley | Kicked keith out of the project, and keith did x.org. |
| 22:54.44 | landley | And everybody and their dog moved over. |
| 22:54.50 | landley | x.org is based on what used to be tinyx. |
| 22:54.53 | ds2 | I have wonder if Kdrive/TinyX is worthwhile - could one be better off running Xvncserver + a directfb vncviewer? |
| 22:55.09 | landley | CosmicPenguin: as far as I know, _all_ of x.org is now tinyx/kdrive. |
| 22:55.23 | ds2 | landley: there is X server going around that doesn't use a config file; everything is configured on the command line |
| 22:55.31 | landley | Cool. |
| 22:55.35 | CosmicPenguin | no - there still xfree86ish design there too |
| 22:55.39 | landley | That's actually a remarkably sane way to do it. |
| 22:55.44 | CosmicPenguin | its all an all in one package of fun and games |
| 22:55.47 | landley | CosmicPenguin: they're still cleaning out old crap, yes. |
| 22:56.01 | landley | I'm not sure anybody's ever going to reimplement support for some of the ancient and obscure graphics cards under a new framework. |
| 22:56.14 | CosmicPenguin | its not all kdrive |
| 22:56.15 | landley | Sienna: vnc on the hammer, maybe. |
| 22:56.20 | CosmicPenguin | kdrive device management is very different |
| 22:56.30 | ds2 | Sun CGtwelve display support! |
| 22:56.31 | ds2 | ;) |
| 22:56.32 | landley | But how big of a bitmap can you really support in 32 megs? |
| 22:56.38 | CosmicPenguin | but much of the kdrive rendering engine has made it into the core code in some fashion |
| 22:56.50 | landley | Keep in mind no graphics card means no graphics memory, so you have to eat some of your 32megs. |
| 22:57.01 | ds2 | CosmicPenguin: what does the OLPC use? the full X server? |
| 22:57.04 | CosmicPenguin | yep |
| 22:57.06 | landley | 1024x768x24 bit color is 3 megs. |
| 22:57.29 | CosmicPenguin | It really is not your father's X server - not with the modular packages and all |
| 22:57.29 | landley | My understanding is that kdrive is to x.org what gecko was to mozilla. |
| 22:57.34 | landley | You can still bloat around it all you want. :) |
| 22:58.13 | landley | Sienna: did you get the hammer out of its socket? |
| 22:58.25 | CosmicPenguin | kdrive still makes more sense on really embedded platforms - for for your nominal x86 platform you would be better off going fully leaded |
| 22:58.41 | Sienna | landley, nope, not yet... I am so scared of breaking it |
| 22:59.10 | ds2 | landley: if Sienna did, I would recommend some better wedges ;) |
| 22:59.13 | ds2 | ducks |
| 22:59.29 | landley | Sienna: that's my response to most hardware. |
| 22:59.44 | landley | prpplague would know, and RobotGuy probably would too. |
| 23:00.06 | landley | In theory you lift from both edges at once. There's a chip puller tool for this sort of thing, but I don't own one. |
| 23:00.11 | Sienna | Ah, i am working it out... using a tiny flathead screwdriver |
| 23:00.37 | landley | Out of morbid curiosity, _why_ do you want to lift it out of its carrier? |
| 23:01.05 | Sienna | Cause I have a morbid obsession of taking working things apart to see what i can see :D |
| 23:01.13 | Sienna | and success by the way~! |
| 23:01.36 | landley | does that with software. |
| 23:01.45 | landley | But hardware is difficult to restore from a backup once broken. |
| 23:01.53 | Sienna | My lord... those are microscopic vias! and whoever soldered those tiny little resistors is mad! |
| 23:02.07 | landley | That would probably be prpplague. :) |
| 23:02.19 | landley | Actually I suspect they have a machine for it... |
| 23:02.25 | MonMotha | those don't get hand soldered Sienna |
| 23:02.43 | landley | MonMotha: speak for yourself. |
| 23:02.44 | MonMotha | and are they really microvias? or are they just small mechanically drilled vias? |
| 23:02.49 | landley | I have a nail prototype meaning it has patch wires. |
| 23:02.56 | landley | They warned me about those when they sent it to me. |
| 23:03.02 | MonMotha | but was the WHOLE THING hand assembled? |
| 23:03.05 | landley | And when it arrived, I initially couldn't _find_ them, because they were so small. |
| 23:03.15 | MonMotha | I mean, yes, I *can* older 0402s, but I don't do it if I don't have to |
| 23:03.18 | MonMotha | *solder |
| 23:03.22 | landley | MonMotha: hammer itself is now presumably somewhat mass produced. |
| 23:03.24 | MonMotha | but no, I cannot solder 0201s :) |
| 23:03.47 | Crofton | anyone done a bga in a toaster oven? |
| 23:03.55 | ds2 | do you even have a tweezer to pickup and the optics to look at 0201's? ;) |
| 23:04.02 | Sienna | the vias are half the diameter of the pin that sticks out of a .1" socket |
| 23:04.02 | MonMotha | Crofton: it wouldn't surprise me if somebody has. The issue is placement |
| 23:04.11 | Sienna | (at least thats how big they look to me) |
| 23:04.12 | MonMotha | Sienna: that a BIG via these days... |
| 23:04.27 | MonMotha | 8mil finished mechanically drilled hole size is rather common |
| 23:04.37 | Sienna | you people are crazy! :tug |
| 23:04.56 | MonMotha | I tend to stick with 15mil if I can, because it's just about the right size to solder 30awg wire into |
| 23:04.59 | ds2 | in the whole scheme of things 8mils is pretty big |
| 23:05.17 | Sienna | (and its back in its socket now) |
| 23:05.19 | MonMotha | you can get 3 or even 4 mil if you go laser drilled |
| 23:05.31 | MonMotha | er "4 or even 3" |
| 23:05.39 | ds2 | think they make mechanical drills in 3-4 mil range |
| 23:05.55 | ds2 | just donno how many would one snap drilling one hole ;) |
| 23:06.00 | MonMotha | ds2: they may, but I can't imagine they last very long |
| 23:06.01 | Sienna | remind me not to get involved with a drill bit the size of my hair... |
| 23:06.12 | MonMotha | um, that's likely smaller than your hair |
| 23:06.28 | Sienna | (truth to tell, i have no idea how big hair is...) |
| 23:06.49 | MonMotha | some smaller PCB features are rivaling the larger IC features of a few years back |
| 23:07.08 | Sienna | ok, now all I need is for my new soldering station to arrive |
| 23:07.23 | Sienna | (cause I don't feel like touching my new toy with my RS firestarter) |
| 23:07.45 | ds2 | heheh |
| 23:07.51 | MonMotha | I don't know how people use Ratshack irons on small stuff these days |
| 23:07.58 | MonMotha | I just have a weller, nothing overly fancy |
| 23:08.07 | MonMotha | but I couldn't live without temp control and a decent tip |
| 23:08.24 | landley | ds2: carbon fibers. |
| 23:08.44 | landley | The problem isn't snapping 'em, the problem is keeping 'em _straight_ rather than letting them bend. |
| 23:09.31 | Sienna | what is the appropriate temp? |
| 23:10.00 | MonMotha | depends on what you're doing |
| 23:10.03 | Sienna | (and I ordered a hakko 936, nothing overly fancy) |
| 23:10.05 | MonMotha | my "standard" temp is 325c |
| 23:10.35 | MonMotha | that's enough to readily work on Pb-Free if I have to (and boy do I hate that crap) |
| 23:10.40 | MonMotha | but it isn't overly hot |
| 23:11.12 | ds2 | with the rat shack irons, painting on flux helps |
| 23:16.09 | Sienna | that reminds me... flux, how do you clean it off boards? |
| 23:16.56 | ds2 | use no-clean flux? :) |
| 23:17.17 | Sienna | I read that still leaves a residue? |
| 23:21.36 | MonMotha | isopropyl alcohol works well when used in moderation |
| 23:21.45 | MonMotha | though some components may not like it |
| 23:22.14 | Crofton | prefers freon |
| 23:22.18 | Sienna | cool. Thanks! |
| 23:22.25 | Crofton | and hates Pb free stuff :) |
| 23:22.44 | Crofton | Sienna, look for the highest alcohol content you can find |
| 23:23.26 | MonMotha | indeed |
| 23:23.28 | Sienna | that would be the vodka in the freezer.... |
| 23:23.30 | Sienna | ? |
| 23:23.31 | Sienna | :D |
| 23:23.43 | CosmicPenguin | everclear - tasty and pratical |
| 23:23.52 | MonMotha | well, it's at least practical |
| 23:24.02 | Sienna | vodka makes white russians |
| 23:24.20 | ds2 | anyone tried using ISO-HEET to clean boards? from what i can tell, it should basically be anhydrous isopropyl in a small bottle |
| 23:25.52 | Sienna | And on a practical note... is there any real danger from leaving flux on boards for hobbyists? |
| 23:26.05 | ds2 | some flux are potentially conductive |
| 23:26.15 | ds2 | think no clean is nonconductive, IIRC |
| 23:26.38 | Sienna | i had ordered some rma cored solder, is rma potentially conductive? |
| 23:27.23 | Sienna | (please keep in mind, i was a software person in college, I just feel my way through all this electronics stuff!) |
| 23:27.46 | ds2 | donno what RMA cored is... I usually use rosin core |
| 23:28.01 | Sienna | rosin mildly activated |
| 23:30.12 | ds2 | probally not |
| 23:30.13 | ds2 | I have never cleaned it and haven't had a problem traceable to flux residue |
| 23:48.18 | Crofton | hmm, the old elinux.org is back |
| 23:48.31 | Crofton | this is a start I guess |
| 23:55.22 | landley | remains a software person. |
| 23:58.57 | ds2 | hardware is more fun, you get to make things go bang |