| 00:04.43 | *** join/#edev bsdfox (n=h36sa@c-67-161-174-21.hsd1.ca.comcast.net) |
| 00:20.41 | wmat | prpplague_afk: pong |
| 00:21.24 | *** join/#edev htns (n=htns@61.6.64.6) |
| 00:24.53 | *** join/#edev htns (n=htns@61.6.64.6) |
| 00:38.25 | Mirell | I have reduced the build time from 1.5 hours to .5 hours. |
| 00:38.33 | Mirell | That's a decent trade-off for three days of work... |
| 00:38.36 | Mirell | Now for sleep... |
| 00:42.07 | *** join/#edev MonMotha (n=monmotha@cpe-65-29-71-209.indy.res.rr.com) |
| 01:34.42 | *** join/#edev kergoth (n=kergoth@ip24-251-170-95.ph.ph.cox.net) |
| 01:51.40 | *** join/#edev linac (n=lin@122.90.120.98) |
| 01:54.32 | *** join/#edev landley (n=landley@cpe-66-68-109-179.austin.res.rr.com) |
| 01:54.33 | *** mode/#edev [+o landley] by ChanServ |
| 01:56.36 | *** join/#edev SpeedEvil (i=1000@tor/regular/SpeedEvil) |
| 02:12.33 | *** join/#edev kergoth (n=kergoth@ip24-251-170-95.ph.ph.cox.net) |
| 02:34.46 | SpeedEvil | landley: on swap to flash - a couple of days ago. Yes swap to flash has inherent problems - but 100k writes * nblocks is actually a very large number. The primary problem ATM with swap to most flash is that the swap allocator does not understand that it should write linearly. |
| 02:35.51 | SpeedEvil | landley: Which would benefit write speed generally - as linear write speed is lots better than others - as it involves less copying of blocks and does erases with optimal efficiency. Barring the OS having deep knowledge of the flash - which would be the better way |
| 03:25.21 | landley | Sigh. Eric is upgrading the server from Ubuntu 8.10 to 9.10. My website and email may be down for a while... |
| 03:25.56 | landley | SpeedEvil: you'd think the elevator code would deal with at least some of this. |
| 03:26.01 | landley | That is what it's for... |
| 03:26.08 | *** join/#edev SandGorgon (n=OmNomNom@122.173.249.152) |
| 03:34.00 | solar | landley: you still do @contract work? |
| 03:35.38 | landley | solar: yup. |
| 03:35.40 | solar | landley: if so. This seems an easy one for you that you already worked with in the past. http://elinux.org/CELF_Open_Project_Proposal_2010 |
| 03:35.56 | solar | see Tim Bird's proposal |
| 03:36.15 | solar | it says sample.. But maybe he wants that :) |
| 03:36.43 | landley | solar: in my sleep. :) |
| 03:36.46 | *** join/#edev morthog (i=551998b9@gateway/web/freenode/x-qityapudwbwxsezb) |
| 03:37.02 | solar | that's why you came to mind first :p |
| 03:38.22 | solar | Ahh yep. His is 1 of 3 proposals |
| 03:38.33 | landley | And actually, I can think of several other proposals I could wave at them to see if they'd bite. |
| 03:38.52 | solar | it's a wiki. Edit it |
| 03:39.03 | landley | Adding a full m68k system emulation to qemu, with mmu. |
| 03:39.07 | landley | Alas, I don't do wikis much. |
| 03:39.15 | landley | I thought the proposals should be sent to the mailing list? |
| 03:39.44 | landley | Upgrading tinycc so it can build the linux kernel and a minimal busybox+uClibc root filesystem (about 6 months full time, I expect, they probably wouldn't bite). |
| 03:39.47 | solar | yeah true |
| 03:40.02 | monkey_ | hey landley how is your asembly? |
| 03:40.09 | landley | monkey_ depends on the platform? |
| 03:40.17 | monkey_ | arm7tdmi |
| 03:40.20 | monkey_ | or armv4 |
| 03:40.29 | landley | I can pull up a reference and hunt-and-peck my way through it. |
| 03:40.35 | landley | What's the question? |
| 03:40.51 | monkey_ | i need to optimize a memcopy in assembly |
| 03:41.55 | landley | solar: do you think it would be gauche to attach a completed patch to the proposal? December 20th is along way away... :) |
| 03:42.06 | solar | basic arm http://www.embedded.com/columns/technicalinsights/19205567?_requestid=368268 |
| 03:42.09 | landley | Eh, full schedule for the month already. :) |
| 03:42.37 | solar | landley: You wont get paid for it |
| 03:42.38 | landley | monkey_ in theory your libc should include a reasonable inlineable memcpy already. |
| 03:42.47 | landley | solar: yeah, true. |
| 03:42.51 | solar | Or rather I would not pay you if I controled the $$ |
| 03:43.01 | solar | haha thanks for the pathc |
| 03:43.07 | solar | patch |
| 03:43.18 | *** part/#edev morthog (i=551998b9@gateway/web/freenode/x-qityapudwbwxsezb) |
| 03:43.20 | solar | not really. Tim is a good guy. |
| 03:43.29 | landley | solar: yup, and I'd smile and put you in my spam filter. |
| 03:43.52 | solar | just it might make life harder for him to justify the resource when a free patch exists |
| 03:44.02 | landley | I haven't actually downloaded the package yet, maybe the shell script is so hideous it really is 2 weeks worth of work... |
| 03:44.15 | landley | solar: I wasn't that serious. :) |
| 03:44.15 | solar | it's umm easy. |
| 03:44.29 | SpeedEvil | landley: sort-of, initially. But block 0-1000 are written to swap, Then some time later, blocks 100, 134, 180-193 are freed. The swapper may then think it's quite OK to use block 100,134,... - but writing block 100 may in fact mean that blocks 96-104 need read, written to a new location, the block erased. You'd much rather it simply wrote to 1000-2000 |
| 03:44.39 | solar | really it could all be done in rcS alone. But to do it semi proper. fine c it is |
| 03:44.53 | landley | Alas, I can't subscribe to celinux-dev and send them an email until my mail server goes back up. |
| 03:45.04 | landley | Eric is perpetrating system adminsitration. |
| 03:45.07 | landley | Be afraid. Be very afraid. |
| 03:45.19 | solar | you can move your MX record to me if you want |
| 03:45.54 | landley | SpeedEvil: ah, got it. You need a weighting or ordering algorithm or some such. |
| 03:46.37 | SpeedEvil | landley: What you really want is some way to tell the flash 'I don't want this block, reclaim it at your leisure' - but that's not there yet. |
| 03:46.38 | landley | SpeedEvil: is round robin on the free space best? Or would some attempt at packing help? |
| 03:46.46 | SpeedEvil | It's icky. |
| 03:46.50 | solar | that would get you back up. Looks like you have a timeout of 300 seconds |
| 03:47.13 | landley | solar: the server isn't currently running a mail or web server. |
| 03:47.17 | SpeedEvil | You're trying to guess what the wear leveling and internal algorithms of the card are doing |
| 03:47.22 | landley | The dns is on the same box that hosts the actual services. |
| 03:48.04 | landley | SpeedEvil: no, wasn't thinking of wear levelling, I was thinking about write speed optimization. |
| 03:48.19 | solar | landley: ahh then you would have to change it at the domain registar |
| 03:48.36 | solar | landley: how long do you expect Eric to be down? |
| 03:48.42 | SpeedEvil | landley: That's the fun part - you don't know where the blocks are mapped due to the wear leveling algorithm. In practice, linear writes seem to be daster for all hardware. |
| 03:48.50 | landley | Namely, if I have 5 pages outstanding to write, do I want to write to this stripe which is entirely free, or do I want to write to the stripe that has the closest number of free blocks? |
| 03:49.03 | landley | solar; the machine's grelber, Eric Raymond is the guy whose basement it lives in. |
| 03:49.12 | solar | if over 8hrs then it's probably worth the effort. |
| 03:49.14 | landley | He expects the upgrade to be finished in an hour or so. |
| 03:49.29 | solar | ahh. You will be fine |
| 03:49.37 | landley | He expects upgrading from Ubuntu 8.10 to Ubuntu 9.10 not to actually cause significant collateral damage. |
| 03:49.50 | landley | I am dubious about that last point. |
| 03:49.51 | SpeedEvil | landley: you probably want the free one - the erase block size is typically 128K - so any hole smaller than 128K will probably cause performance to plummet |
| 03:50.16 | landley | SpeedEvil: you don't have to overwrite swap entries that aren't being used anymore, you just have to track 'em. |
| 03:50.34 | landley | So you have an in-memory bitmap or whatever it is that tracks which ones have been "freed", not the same as overwritten. |
| 03:50.48 | landley | You only overwrite when you have fresh data to put there. |
| 03:50.54 | SpeedEvil | landley: yes. |
| 03:51.04 | solar | the 9.10 upgrade breaks quite a bit from what I've heard from co-workers |
| 03:51.13 | landley | The erase block size is the same whether you're rewriting 90% of the data in the erase block or whether your're rewriting none of it and it's all fresh. |
| 03:51.13 | SpeedEvil | falls asleep |
| 03:51.23 | landley | solar: I break everything. |
| 03:51.42 | landley | This is a server I use. I'm not doing the upgrade, but there's still quantum entanglement, as it were. |
| 03:51.49 | landley | I'm trying not to think about it. |
| 03:51.52 | solar | heh |
| 03:52.17 | solar | tiz my day job |
| 03:52.42 | landley | He says he's doing it because he's putting out a gpsd and needs to give someone else on that project a login so they can reproduce errors. (They're a mac developer, I think.) |
| 03:52.52 | landley | I'm a bit unclear on why he's doing it this way, but oh well... |
| 03:53.28 | landley | His email is down as long as mine is, it's unlikely to go unresolved for too long... |
| 03:53.48 | landley | And tomorrow, I get on a plane to california. |
| 03:54.17 | *** join/#edev morthog (i=551998b9@gateway/web/freenode/x-qityapudwbwxsezb) |
| 03:54.28 | morthog | hello |
| 03:54.58 | solar | landley: where in cali are you going? |
| 04:02.25 | Mirell | solar: Mountain View, I would assume. |
| 04:03.33 | Mirell | landley: 9.10 is really, really broken. |
| 04:03.42 | Mirell | landley: You'll probably lose everything. |
| 04:05.34 | morthog | has anybody come across an error like this when building an embedded system? http://pastebin.com/d41d099fe |
| 04:14.10 | morthog | init points to /bin/busybox |
| 04:17.55 | landley | Yay, email server back up. |
| 04:17.59 | landley | solar: The Googleplex. |
| 04:18.11 | landley | San Jose airport, then "somewhere". |
| 04:19.13 | landley | morthog: that means it couldn't run init. |
| 04:19.31 | landley | Put a statically linked "hello world" program as init and see if that outputs hello world. |
| 04:19.40 | landley | Then once you've got that working, make it dynamically linked. |
| 04:19.52 | landley | Then try busybox only _after_ those two steps are working first. |
| 04:20.14 | landley | Right now there are literally a couple dozen things that could be wrong. |
| 04:20.20 | landley | Your kernel .config might not have ELF binary support in it. |
| 04:20.28 | landley | Your shared libraries might have leaked references to libc6. |
| 04:20.38 | landley | Your dynamic linker might be at a different path than the ELF header says it is. |
| 04:21.20 | landley | At least it says it mounted root, so your block device and filesystem drivers are compiled in correctly. That's something. |
| 04:21.26 | morthog | ok. just a sec |
| 04:21.54 | landley | Loopback mount your ext2 image and make sure that A) everything's where you think it is and not in a subdirectory or something, B) your init has the executable bit set. |
| 04:25.19 | morthog | landley: busybox is in /bin and init is a symbolic link to it located at /sbin/init... loader passes init=/sbin/init to kernel at boot |
| 04:27.27 | morthog | does that sound right? |
| 04:28.16 | landley | morthog reasonably. |
| 04:28.46 | landley | Add a static hello program and set init=/hello and see what it does. |
| 04:28.51 | monkey_ | landley: i have to optimize this: http://pastebin.ca/1699604 |
| 04:29.00 | morthog | right. forgot about that. just a sec |
| 04:29.38 | landley | monkey_: so they gave you an assignment and you're asking us to do it for you? :) |
| 04:29.52 | monkey_ | nooooooooooooooooooooooooooooooooooo :) |
| 04:30.30 | morthog | landley: the hello program goes inside the ramdisk right? |
| 04:30.50 | landley | morthog: if it's a ramdisk then you use rdinit= not init= |
| 04:31.33 | morthog | ... |
| 04:32.00 | landley | But I think /sbin/init is one of the kernel's fallback paths anyway. |
| 04:32.46 | landley | Yup, in the linux kernel source, init/main.c line 839. |
| 04:33.01 | landley | <PROTECTED> |
| 04:33.01 | landley | <PROTECTED> |
| 04:33.01 | landley | <PROTECTED> |
| 04:33.11 | landley | <PROTECTED> |
| 04:33.22 | landley | <PROTECTED> |
| 04:33.56 | morthog | yeah. that's why i put it there... |
| 04:34.27 | landley | morthog: depends on what you're trying to do. It can be. |
| 04:34.38 | landley | Sorry, responding to backscroll, my bad. |
| 04:34.51 | landley | Had the scroll bar at the wrong bit, thought something had been reposted. |
| 04:35.00 | morthog | ...but what you said about rdinit. that could be it. lemme check |
| 04:35.05 | landley | ok |
| 04:35.09 | morthog | i've been using init and this is all taking place on a ramdisk |
| 04:35.20 | landley | yeah, that won't work. |
| 04:35.48 | landley | The hello world program is probably still a good idea. |
| 04:41.57 | solar | landley: cough.. If you come to my town. Atleast let me buy you you drinks in the downtown sjc area durring the trip |
| 04:43.00 | landley | solar: woot. |
| 04:43.25 | morthog | yeah. the hello world program did not run... |
| 04:43.44 | landley | My plane departs at 3 and the travel time is 3 hours 40 minuts, minus 2 hours of time zone change, carry the zero... |
| 04:43.49 | landley | I think I arrive yesterday. |
| 04:44.06 | landley | morthog: ok, that's good to know. |
| 04:44.13 | landley | It's not a shared library issue, not a busybox issue. |
| 04:44.23 | landley | It's a "can I run a simple elf binary" issue. |
| 04:44.40 | landley | It was statically linked? |
| 04:44.59 | morthog | i'm using a precompiled suse kernel for 8.1 |
| 04:45.05 | landley | solar: my hotel is 860 east el camino bignum. |
| 04:45.07 | morthog | i didn't compile it myself |
| 04:45.24 | landley | morthog: so this is an x86 or x86_64 embedded system? |
| 04:45.28 | morthog | x86 |
| 04:45.42 | landley | Run "file" on the hello world and tell me what it says. |
| 04:46.01 | morthog | hello_world: ELF 32-bit LSB executable, Intel 80386, version 1 (SYSV), for GNU/Linux 2.4.1, dynamically linked (uses shared libs), not stripped |
| 04:46.09 | monkey_ | landley: sub sp, sp, #4 //how does this allocate 4 bytes on the stack? |
| 04:46.17 | landley | morthog: dynamically linked? |
| 04:46.30 | landley | monkey_: moves the stack poiner, I'd guess? |
| 04:46.33 | landley | pointer |
| 04:46.40 | morthog | landley: sorry? |
| 04:46.51 | landley | On a lot of platforms, the stack grows downwards. |
| 04:46.57 | monkey_ | ok. right. |
| 04:47.04 | landley | This means you can dereference sp+offset to get your struct of local variables. |
| 04:47.25 | landley | morthog: the above dump says that your hello world is dynamically linked. |
| 04:47.39 | morthog | ok. what does that mean? |
| 04:47.53 | landley | gets a drink. |
| 04:48.14 | landley | morthog: ok, do you know what shared libraries are? |
| 04:48.25 | morthog | and when busybox compiled i was left with 'busybox' and busybox_stripped' |
| 04:48.43 | landley | morthog: that has to do with stripping off debug information, that's a separate issue. |
| 04:48.43 | morthog | libraries that the app requires to run? |
| 04:49.06 | landley | Ok, ELF files. |
| 04:49.10 | morthog | sorry. dont mean to make you go into a lesson here |
| 04:50.11 | landley | morthog: work your way up to understanding http://www.muppetlabs.com/~breadbox/software/tiny/teensy.html someday, it's really fun. |
| 04:50.38 | landley | Ok, the when the kernel tries to exec() a file, the first thing it does is determine its file type. |
| 04:50.53 | landley | There are a number of options, but the two most interesting ones are scripts and ELF binaries. |
| 04:51.04 | landley | A script is anything that starts with the two bytes #! |
| 04:51.13 | Mirell | The amount of times I see long option parsing redone in scripts I just wonder why bash doesn't provide a sane way of doing it. |
| 04:51.48 | morthog | ok |
| 04:51.51 | landley | The kernel takes the first whitespace character as an absolute path to a file to run. |
| 04:52.01 | landley | Mirell: bash? Sane? |
| 04:52.15 | landley | Sorry, the kernel takes everything up to the first whitespace character... |
| 04:52.53 | landley | If there's whitespace in the line, everything after the whitespace is fed into the new program as its first argument. |
| 04:53.06 | morthog | got it |
| 04:53.13 | landley | (This means that if you go #!/echo "this is a test" it will print out "This is a test" including the quotes. |
| 04:53.22 | morthog | yes |
| 04:53.25 | landley | It only splits out _one_ argument, and that argument is optional. |
| 04:53.37 | landley | The line goes to a newline (ascii 10). |
| 04:53.54 | morthog | so #!/echo "this is a" "test" would only print out "this is a" |
| 04:54.01 | landley | The _second_ argument (or the first if there was no whitespace in the line) is the name of the file they actually tried to execute. |
| 04:54.17 | landley | morthog: no, it would print "this is a" "test" |
| 04:54.18 | landley | Try it. |
| 04:54.31 | landley | Create an empty file, start it with #!/bin/echo "blah blah blah" |
| 04:54.36 | landley | And set the executable bit on it with chmod +x |
| 04:54.44 | morthog | yeah. ive worked with scripts |
| 04:55.01 | morthog | have a pdf on shell scripting here somewhree |
| 04:55.02 | landley | Ok. If it's not a shell script, the other type the kernel might detect is an ELF file. |
| 04:55.08 | morthog | ok |
| 04:55.11 | landley | Thos start with "ELF\0x7a" I think. |
| 04:55.15 | landley | Different binary signature. |
| 04:55.38 | landley | Those files start with a structure called an "ELF header", which has pointers to a linked list of "elf segments" of various types. |
| 04:56.15 | morthog | ok |
| 04:57.45 | landley | The actual spec is here, defining all the structures, if you care. http://www.skyfree.org/linux/references/ELF_Format.pdf |
| 04:57.56 | Mirell | ~/src/firmware/build/root-filesystem-i686/lib/ld-uClibc-0.9.30.1.so |
| 04:57.57 | Mirell | Standalone execution is not supported yet |
| 04:57.58 | Mirell | Aww. |
| 04:58.20 | landley | You can also look in /usr/include/elf.h although it won't explain what most of the structs are for. |
| 04:58.31 | landley | So you have a specially formatted file full of chunks of data. |
| 04:58.45 | landley | The "readelf" command will dump a directory listing of the elf stuff as test. |
| 04:58.50 | landley | Try "readelf -a /bin/ls" |
| 04:58.55 | landley | Might want to pipe it to less. |
| 04:58.59 | morthog | yeah. i'm looking at the pic of the structure on wikipedia |
| 04:59.29 | landley | The kernel has to follow all the pointers (actually file offsets) to find the next structures and interpret them and load everything into memory and link it together at runtime, and that's a serious pain. |
| 05:01.08 | landley | That would be the kernel's "elf loader". |
| 05:01.14 | morthog | ok |
| 05:01.15 | landley | In fs/binfmt_elf.c |
| 05:01.24 | landley | (Don't ask me why it's in the filesystem directory, I don't know.) |
| 05:01.33 | landley | The kernel only looks at _one_file_. |
| 05:01.57 | landley | With shared libraries, you load more than one file at runtime, and mmap them into the appropriate places and link all the pointers together and so on. |
| 05:02.15 | landley | This is done by a program called a "runtime library loader". |
| 05:02.27 | landley | It's basically another linker, like ld, except it works in memory rather than on disk. |
| 05:02.56 | landley | Do a "readelf -a /bin/ls | grep ld-" |
| 05:03.12 | landley | It should say "Requesting program interpreter" followed by an absolute path to a filename. |
| 05:03.12 | morthog | i'm still scrolling through the first one... |
| 05:03.23 | landley | search for "program interpreter" |
| 05:03.31 | landley | less shoudl let you search if you hit / |
| 05:03.35 | morthog | # readelf -a /bin/ls | grep ld- [Requesting program interpreter: /lib/ld-linux.so.2] |
| 05:03.39 | landley | That. |
| 05:03.40 | morthog | that's all it says |
| 05:03.44 | landley | That's your runtime linker. |
| 05:03.45 | morthog | didn't need grep |
| 05:03.50 | morthog | there was only one line of output |
| 05:04.02 | morthog | oh... nm |
| 05:04.04 | landley | There's a special field in the ELF header that's either empty or points to that string. |
| 05:04.04 | morthog | ignore that |
| 05:04.25 | landley | If it points to that string, then instead of loading this file, it loads that _other_ file and feeds this file in as a command line argument. |
| 05:04.32 | landley | Just like the #!/blah stuff earlier. :) |
| 05:04.48 | landley | This linker's job is then to load your program and find all the shared libraries it needs. |
| 05:05.00 | morthog | ok |
| 05:05.05 | landley | If you type "ldd /bin/ls" it should list all the shared libraries it needs to run. |
| 05:05.26 | landley | (You can get the same info from readelf, you just have to dig through a lot of unrelated stuff. The ldd command gives you _just_ shared libraries.) |
| 05:05.58 | morthog | so are you saying that libraries that are needed to run are not present when hello or init are being run? |
| 05:06.01 | landley | The dynamic linker produces one big blob with all the data loaded at once, it just hunts through lots of shared libraries at runtime to find all the data. |
| 05:06.23 | landley | When you "statically link" program, you tell ld to hunt through at _compile_ time, and find all this data, and put it in one big executable. |
| 05:06.33 | landley | That big executable doesn't depend on any external libraries. |
| 05:06.41 | morthog | got it |
| 05:06.51 | landley | Its library loader field is blank, and thus the kernel loads the file directly and runs it without bouncing off a second program. |
| 05:06.57 | landley | The gcc option is "--static" |
| 05:07.35 | landley | We'll save pic and pie for another day, I think. :) |
| 05:07.43 | landley | But do read the muppet labs link, it's marvelous. |
| 05:07.59 | morthog | i have it bookmarked |
| 05:08.00 | morthog | thx |
| 05:08.06 | landley | You're welcome. |
| 05:08.21 | morthog | but what does this have to do with what is happening when i try to boot the system? |
| 05:08.23 | landley | Long ago I gave classes on some of this stuff at the local community college. |
| 05:08.33 | landley | My idea of an "introduction to unix" class might have been a _touch_ unorthodox. :) |
| 05:08.43 | morthog | heh. |
| 05:08.50 | landley | morthog: the error message you get says that exec() failed on all the filenames it tried. |
| 05:08.53 | landley | It doesn't say _why_ it failed. |
| 05:08.55 | morthog | what? are you waiting at an airport |
| 05:08.57 | landley | It just said it couldn't run them. |
| 05:09.03 | landley | morthog: that's tomorrow. |
| 05:09.09 | landley | 3pm |
| 05:09.35 | morthog | is this a problem with the kernel? |
| 05:10.14 | landley | morthog: the kernel has notoriously bad error reporting during early boot. |
| 05:10.30 | landley | It hasn't got a lot to work with yet. Mostly it delegates this sort of thing to userspace. |
| 05:10.39 | landley | When it's trying to get userspace up and running, it can't delegate to it. |
| 05:10.56 | morthog | all of my efforts on this project have come to a halt right here |
| 05:10.59 | landley | It couldn't run your init program. This might be because the dynamic loader the ELF header pointed to wasn't there. |
| 05:11.03 | morthog | it's driving me nuts |
| 05:11.33 | landley | If one of the shared libraries the dynamic loader needed to load wasn't there, the loader should spit out a better error message. (Well, modern ones will. Circa red hat 6, not so much.) |
| 05:11.49 | landley | A statically linked hello world program has no external dependencies. No reason not to run. |
| 05:11.55 | morthog | this is a OLD kernel |
| 05:11.58 | landley | If it can find it in the filesystem, and it has the executable bit, it should run. |
| 05:12.10 | landley | morthog: how old are we talking? 1998? |
| 05:12.16 | morthog | slackware 8.1 |
| 05:12.24 | landley | meaningless to me. |
| 05:12.38 | morthog | me either. someone else called it ancient though |
| 05:12.41 | MonMotha | I've managed to get binaries to just immediately spit out "No such file or directory" when you try to exec them. It's annoying, but usually means a missing ld-linux |
| 05:12.52 | landley | But not to google. 2003. |
| 05:12.55 | landley | Not that old, really. |
| 05:13.15 | MonMotha | also annoyingly, ldd will claim "not a dynamic executable" if you run it on the binary to try to figure out the missing library |
| 05:13.24 | landley | "No such file or directory" is the kernel's way of saying "I can't find your interpreter". |
| 05:13.32 | landley | It'll do the same for #!/does/not/exist |
| 05:13.45 | MonMotha | I thought it was a little more specific in that instance |
| 05:13.52 | landley | MonMotha: use the uClibc ldd, the glibc one is crap. |
| 05:13.53 | morthog | wow. the filesize is much larger than the one that's not statically linked |
| 05:14.02 | landley | The uclibc ld is essentially a modified readelf. |
| 05:14.03 | MonMotha | landley: the glibc one is a shell script :) |
| 05:14.20 | landley | The glibc one runs the dynamic linker with an extra environment variable so it prints out data instead of running the file. |
| 05:14.42 | landley | (Yes literally. You want to really hos somebody's system, set that environment variable in their /etc/profile and let them TRY to figure out what's wrong.) |
| 05:14.49 | landley | s/hos/hose |
| 05:14.56 | MonMotha | hah |
| 05:15.02 | landley | morthog: yes, that's the downside of static linking. |
| 05:15.48 | landley | The uClibc ldd should work on any file (possibly even ones for different architectures as long as the 32/64 bit thing is right. Structures are a different size otherwise.) |
| 05:15.52 | MonMotha | anyone who isn't idle done much with Xilinx FPGAs? I'm doing something that I've not done before (attempting to use Selectmap) and I'm curious if anyone's ever messed with it |
| 05:16.03 | landley | The glibc one only works on _some_ glibc binaries. |
| 05:16.41 | landley | MonMotha: I almost got a contract to work on a xylinx bsp a year ago. |
| 05:16.58 | landley | I said "I have december free", and they got back to me in January, after I was already hip-deep in another contract. |
| 05:17.01 | landley | Alas. |
| 05:17.05 | MonMotha | nice |
| 05:17.11 | landley | (Tried to find 'em another engineer to do the work, but it wasn't a good fit.) |
| 05:17.16 | MonMotha | I've done FPGA stuff before, but I've always just chucked a platform flash device on the board |
| 05:17.42 | MonMotha | in this case, I want the CPU to download the bitstream to the FPGA, and I want the FPGA to sit on the CPU's bus anyway, so selectmap seems a good fit |
| 05:17.50 | landley | I dunno what selectmap is. The stuff I looked at was kernel and toolchain with the FPGA part already loaded. |
| 05:17.52 | MonMotha | unfortunately, it uses a motorola-ish bus handshake, not intel |
| 05:18.06 | landley | morthog: how goes it? |
| 05:18.08 | morthog | landley: what environment do you use to test your systems? |
| 05:18.15 | landley | morthog: qemu. |
| 05:18.18 | morthog | me too |
| 05:18.19 | morthog | ... |
| 05:18.20 | landley | http://impactlinux.com/fwl |
| 05:18.52 | landley | That builds uClibc+busybox systems for every QEMU target architecture I've managed to get working so far, and boots 'em under qemu. |
| 05:19.04 | landley | With a native toolchain so you can compile stuff inside the emulated environment. :) |
| 05:19.31 | landley | http://impactlinux.com/fwl/screenshots/ might help. |
| 05:19.53 | landley | I'm redoing that page to be less ugly as we speak, although it probably wont' be up until I get back from california. |
| 05:19.55 | morthog | aparrently the initrd is too big now and it's requesting too much memory |
| 05:20.08 | landley | morthog: run strip on the static hello world. |
| 05:20.19 | landley | It can be statically linked _and_ have debug information in it. |
| 05:20.34 | landley | You need the static linking, but not the debug info just now. We're not running gdb on this. |
| 05:24.11 | morthog | sorry. so you create a hard drive image and then create your embedd system on that? |
| 05:26.24 | landley | morthog: I boot it under an emulator, qemu. |
| 05:27.13 | landley | If I want to run stuff on real hardware, I either use the "hw-" targets to feed in a different kernel .config or just repack the build/root-filesystem-$ARCH directory by hand and build my own kernel elsewhere. |
| 05:27.42 | landley | I'm in the process of rewriting the documentation on three different fronts. :) |
| 05:27.48 | landley | One of which is for Mirell, and needs to be done soonish. |
| 05:28.07 | landley | Creating the hard drive image comes last. |
| 05:28.19 | landley | I populate a root filesystem directory first. (You can chroot into it if you like.) |
| 05:28.31 | landley | Then the system-image.sh script creates a hard drive image from a root-filesystem directory. |
| 05:28.40 | landley | As well as building a kernel. |
| 05:29.02 | morthog | sorry. you boot what under an emulator? |
| 05:29.14 | landley | morthog: ever heard of qemu? |
| 05:29.19 | morthog | i'm working with a floppy disk image and tell qemu to boot that |
| 05:29.36 | landley | http://www.qemu.org/download.html |
| 05:29.39 | landley | Right. |
| 05:29.41 | morthog | i'm using it :) |
| 05:30.05 | landley | If you look at http://impactlinux.com/fwl/downloads/binaries/ you'll see "system-image" tarballs for various architectures. |
| 05:30.15 | landley | Grab one, extract it, cd into it, and ./run-emulator.sh |
| 05:30.26 | landley | (Not you need the full qemu package installed, not just kvm.) |
| 05:30.43 | morthog | i have qemu and kqemu |
| 05:30.48 | landley | That would be http://impactlinux.com/fwl/FAQ.html#ubuntu_mispackaged_qemu |
| 05:30.56 | landley | (I have a FAQ! It needs more questions.) |
| 05:31.04 | landley | morthog: yay. |
| 05:31.29 | landley | So those prebuilt binary system images are uClibc+busybox+gcc hard drive images and appropriate kernel, plus a run-emulator.sh script that invokes qemu against 'em. |
| 05:31.49 | landley | I've got it working for mips, arm, powerpc, sh4, x86, and x86_64. |
| 05:31.58 | landley | Partially working (not really) for sparc. |
| 05:32.18 | morthog | my final goal is something that will fit on a floppy |
| 05:32.25 | morthog | i just started with this a few days ago |
| 05:32.26 | landley | Not working at all for m68k, although if you download the root-filesystem tarball you can chroot into it with qemu-mips. (The application emulation version, not the system emulation version.) |
| 05:32.38 | *** join/#edev mbuf (n=mbuf@118.102.130.6) |
| 05:32.49 | landley | And I have stubs for alpha and blackfin and such lying around, not merged yet. They don't do anything interesting at all yet. |
| 05:35.24 | landley | morthog: if you set the environment variable NATIVE_TOOLCHAIN=none then my build scripts will create a root filesystem with no native compiler. That should fit on a floppy. |
| 05:35.31 | landley | Or you could just use tomsrtbt |
| 05:35.54 | landley | The most linux ever fit on a floppy: http://www.toms.net/rb/ |
| 05:35.57 | morthog | i'd really like to build my own |
| 05:35.59 | landley | You still have a floppy drive? |
| 05:36.15 | morthog | yeah. i've read about his work a few years ago |
| 05:36.42 | morthog | first on i ever heard of was LOAF |
| 05:36.45 | morthog | *one |
| 05:37.02 | morthog | that was back in the 90s |
| 05:48.08 | monkey_ | landley: not to do the assignment, but maybe a tip/pointer... i have till monday to finish 4 assignments ;( |
| 05:48.12 | monkey_ | and a final |
| 05:57.36 | landley | monkey_: I'd have to do the same thing you would, read the documentation. |
| 05:58.06 | monkey_ | ok |
| 05:58.58 | landley | monkey_: beat memcpy in what sense? Runtime? |
| 05:59.01 | monkey_ | yes |
| 05:59.34 | landley | http://en.wikipedia.org/wiki/Duff%27s_device |
| 06:01.07 | monkey_ | thanks |
| 06:05.26 | monkey_ | now to write that in asm ;) |
| 06:06.54 | landley | Um, ldr is a 32-bit operation, right? |
| 06:07.02 | monkey_ | could be? |
| 06:07.38 | monkey_ | judging by the asm alreayd, i'd say yes |
| 06:08.03 | monkey_ | does ldr just mean load register? |
| 06:08.10 | landley | Yes. |
| 06:08.20 | landley | And the address is in bytes, not in words? |
| 06:08.36 | monkey_ | that i couldn't tell you |
| 06:08.43 | landley | So it's doing a bunch of overlapping save/store operations? |
| 06:08.47 | monkey_ | you stll have that pastebin? |
| 06:08.51 | landley | Seems darn inefficient if you ask me. |
| 06:09.04 | monkey_ | so, i could load/store in one? |
| 06:09.06 | landley | And I have no idea what happens if you ask it to copy _less_ than 4 bytes. |
| 06:09.17 | landley | you could load/store once and then increase the counter by 4, not by 1. |
| 06:09.19 | landley | I think. |
| 06:09.26 | landley | Assuming that ldr is using 4 byte values... |
| 06:09.43 | monkey_ | <PROTECTED> |
| 06:09.43 | monkey_ | <PROTECTED> |
| 06:10.12 | landley | Ah, it's adding 4 to each value there. Nevermind. |
| 06:10.42 | landley | Yeah, it's number of words. Ok. |
| 06:11.21 | *** join/#edev PureSine (n=PureSine@87.247.173.124) |
| 06:11.38 | monkey_ | i can't remember, on arm, a word is 2 or 4 bytes/ |
| 06:11.44 | landley | 4 |
| 06:11.47 | landley | 32 bit processor |
| 06:11.52 | monkey_ | yeah |
| 06:12.02 | monkey_ | but a word isn't always 4 bytes right? |
| 06:12.04 | landley | The #4 i also adding 4 to r1, I think? |
| 06:12.12 | landley | monkey: there are 16 bit processors, sure. |
| 06:12.33 | monkey_ | 4 to the address? |
| 06:12.39 | landley | Something called a "nybble" used to be important. |
| 06:12.44 | monkey_ | http://pastebin.ca/1699604 |
| 06:12.48 | monkey_ | nibbles? |
| 06:12.49 | monkey_ | right |
| 06:12.53 | landley | The #4 constant, which I vagule recall gets added to r1. |
| 06:12.54 | monkey_ | i have heard of that. don't really remember |
| 06:13.03 | landley | monkey_: you haven't needed to care since about 1980. |
| 06:13.15 | monkey_ | i wasn't even born then |
| 06:13.19 | landley | There you go. |
| 06:14.16 | landley | Unless there's a memory/memory copy without going through a register (which I believe arm doesn't have), and unless there's something clever you can do with the loop increment/test itself (again, 43,000 transistors total for the core logic), I believe loop unrolling is where they want you to go with this. |
| 06:14.43 | monkey_ | with duff's device? |
| 06:14.48 | landley | Now on x86 there was all sorts of optimization black magic. |
| 06:14.53 | monkey_ | i suppose i can just write it in C and dissabmle it |
| 06:14.55 | monkey_ | lol |
| 06:14.55 | landley | monkey_: an instance of loop unrolling. |
| 06:15.23 | monkey_ | ok |
| 06:15.26 | landley | http://groups.google.com/group/net.lang.c/msg/66008138e07aa94c |
| 06:16.20 | monkey_ | lol, 1984 |
| 06:16.56 | landley | monkey_: as long as the porn keeps flowing, the internet brings us many things and remembers them all. |
| 06:17.33 | landley | Although the youtube entry where they made a music video for "the internet is for porn" out of sesame street muppet clips might qualify as incest, I'm not sure. |
| 06:17.38 | monkey_ | i got in trouble for porn on the computer when i was younger. then last year my dad told me he was jealous because he didn't know where to find it. |
| 06:17.45 | landley | It's darn close to recursive, at any rate. |
| 06:17.51 | monkey_ | ha |
| 06:17.59 | landley | google? |
| 06:18.32 | landley | Yup, I just typed "porn" into google, and responded real fast. |
| 06:18.47 | monkey_ | lol, this was pre-google |
| 06:18.56 | monkey_ | i'm older than that |
| 06:19.09 | landley | I pondered going to bing and typing "google" into it to see what it did, but I don't really want to get any of that site on me. |
| 06:19.38 | monkey_ | i wasborn in 81, and started using computers from a very young age |
| 06:20.34 | monkey_ | Results 1 - 10 of about 27,500,000 for porn [definition]. (0.11 seconds) |
| 06:20.55 | monkey_ | that, is mighty impressive |
| 06:20.56 | landley | Strange furry-on-robot illustrated slash fiction with sound effects and full motion video, you don't even see it after a while in a college compute rlab. Microsoft's attempt at killing google: that's just wrong and I won't have it in the house. |
| 06:21.13 | monkey_ | did you find that shit on bing? i might have to switch! |
| 06:21.13 | landley | You must have safe search on. |
| 06:21.30 | landley | I've never voluntarily brought up microsoft's bling site. |
| 06:21.36 | monkey_ | hat, is mighty impressive |
| 06:22.08 | landley | Keep in mind, I had the internet in 1993. |
| 06:22.12 | monkey_ | http://www.google.com/search?hl=en&q=Strange+furry-on-robot+illustrated+slash+fiction+with+sound+effects+and+full+motion+video&btnG=Search&cts=1259822239233&aq=f&oq=&aqi= |
| 06:22.15 | landley | This isn't exactly new to me. |
| 06:22.25 | landley | monkey_ I don't even want to know what it came up with. |
| 06:22.25 | monkey_ | so did i! |
| 06:22.28 | monkey_ | i think |
| 06:22.31 | monkey_ | lol |
| 06:22.33 | landley | I'm not asking. |
| 06:22.34 | monkey_ | not what you'd think |
| 06:22.37 | landley | I don't care. |
| 06:22.48 | landley | I plead Rule 34. |
| 06:23.06 | monkey_ | nothing on the first page is porn or anything. |
| 06:23.45 | landley | A friend of mine sent me _enzyme_porn_ last month, which she found on 4chan. Just to prove that rule 34 really is the exception to "every rule has an exception, including this one" |
| 06:24.10 | monkey_ | enzyme porn? hot! |
| 06:24.15 | landley | http://xkcd.com/305/ |
| 06:24.42 | monkey_ | lol, i remember that |
| 06:25.03 | landley | Ok, doing a google image search for "rule 34" is just _wrong_. |
| 06:26.47 | monkey_ | jesus christ |
| 06:30.10 | monkey_ | ok, back to homework |
| 07:03.26 | solar | landley: sorry. looking up that addr now |
| 07:05.09 | solar | the street addr you said fails to let good compute |
| 07:05.16 | solar | google |
| 07:06.45 | solar | anyway. Give me a half a days head notice anywhere in the bay area and we can probably find each other |
| 07:09.50 | morthog | landley: ...what about the "Warning: unable to open an initial console." part? |
| 07:15.18 | solar | landley: sure it's not bingham ? |
| 07:16.13 | solar | which puts you in palo alto next to calculus |
| 07:35.26 | morthog | so i can just use uClibc if the libraries aren't being found when init is being loaded, right? |
| 07:38.41 | solar | well no |
| 07:38.54 | solar | static vs dynamic linking |
| 07:39.16 | solar | pick one or the other. sounds like you need static |
| 07:39.25 | *** join/#edev hw_ (n=hw@p578b3905.dip0.t-ipconnect.de) |
| 07:42.11 | morthog | ... |
| 07:42.38 | morthog | solar: i'm running 'make menuconfig' right now. should i go for the 'static flat'? |
| 07:44.28 | morthog | ...under target file format |
| 07:46.33 | *** join/#edev remotesine3 (n=chatzill@87.247.173.124) |
| 07:51.01 | *** join/#edev PureSine (n=PureSine@87.247.173.124) |
| 08:01.28 | *** join/#edev PureSine (n=PureSine@87.247.173.124) |
| 08:05.43 | landley | solar: sorry, el camino real. |
| 08:05.58 | landley | Very very old computer science joke I've never had the chance to use before. |
| 08:06.39 | *** join/#edev PureSine (n=PureSine@87.247.173.124) |
| 08:06.48 | landley | morthog: that means /dev/console doesn't provide a controlling tty. If you build toybox and use "oneit" that'll fix it. |
| 08:07.03 | morthog | i'm using busybox |
| 08:07.14 | landley | Just ignore it for the moment. It means ctrl-c and ctrl-z won't work. |
| 08:07.41 | morthog | okie |
| 08:07.59 | morthog | oh, and i'm compiling uclibc, what target should i go for? |
| 08:08.35 | landley | for x86? Probably i686. |
| 08:09.13 | landley | morthog: so the static hello world ran for you? |
| 08:09.25 | landley | And presumably you made progress since then? I was off doing other things for a bit... |
| 08:09.57 | morthog | no. it didn't |
| 08:10.03 | morthog | ... |
| 08:10.07 | landley | Alas. |
| 08:10.11 | landley | Presumably it should have... |
| 08:10.26 | landley | You did eventually get a statically linked version built, right? |
| 08:10.43 | morthog | built, stripped, zipped into the initrd and failed |
| 08:10.56 | landley | It ran on the host? |
| 08:10.56 | morthog | *gzipped |
| 08:11.06 | morthog | yes |
| 08:11.16 | landley | How did you put it into the initrd? Loopback mounted ext2 image? |
| 08:11.29 | morthog | i'm going to try again just in case i was temporarily stupid and missed something |
| 08:11.36 | morthog | yeah. loopback |
| 08:11.49 | landley | You unmounted the loopback image before gzipping it, right? |
| 08:12.08 | morthog | yeah. been doing this for the past two days. hundreds of times by now. it's automatic |
| 08:12.17 | landley | Did it ever _work_? |
| 08:12.22 | morthog | and ... fail |
| 08:12.30 | morthog | no |
| 08:12.46 | landley | So you have an automatic process that may or may not be producing a useful result, you can't tell. |
| 08:13.03 | landley | How are you booting it, and how are you feeding it to the kernel? |
| 08:13.16 | morthog | qemu -fda <imagename.img> -m 02 |
| 08:13.20 | morthog | *-m 20 |
| 08:13.34 | landley | That feeds it in as a floppy image. |
| 08:13.39 | morthog | exactly |
| 08:13.42 | landley | You said you were using a ramdisk. |
| 08:13.49 | landley | You're putting the kernel on the fda? |
| 08:14.06 | morthog | the ramdisk is inside the floppy image with the kernel, and bootloader |
| 08:14.15 | landley | syslinux? lilo? |
| 08:14.30 | morthog | syslinux, and required syslinux.cfg file too |
| 08:14.39 | morthog | it's all there |
| 08:14.52 | morthog | brb |
| 08:15.13 | landley | So you loopback mount an ext2 image, copy the init into it, unmount it, gzip it, loopback mount a floppy image, copy the initrd.gz into that, and the unmount the floppy. |
| 08:15.41 | *** join/#edev PureSine (n=PureSine@87.247.173.124) |
| 08:16.13 | landley | morthog: did anybody ever show you http://tldp.org/HOWTO/Bootdisk-HOWTO/ |
| 08:16.29 | landley | It may be useful. |
| 08:19.12 | morthog | no |
| 08:19.34 | landley | Sorry I didn't think of it earlier. That's where I learned some of the early stuff from. |
| 08:19.42 | landley | (An earlier version of that document, anyway.) |
| 08:20.31 | morthog | i loopback mount the uncompressed initrd, copy busybox and my filesystem onto it, unmount. gzip it to initrd.gz. then put that on the floppy with the bootloader/syslinux.cfg and the kernel |
| 08:20.46 | landley | So you have an actual floppy lying around? |
| 08:22.14 | *** join/#edev PureSine (n=PureSine@87.247.173.124) |
| 08:22.48 | morthog | what? yeah. of course |
| 08:22.58 | morthog | wait. you might be interested in this |
| 08:23.06 | morthog | have you heard of coyote linux? |
| 08:23.27 | morthog | it's another floppy distro. when i try to boot qemu with that, i get the same message |
| 08:23.27 | landley | Vaguely. |
| 08:23.42 | landley | morthog: that's unusual. Does it boot from the actual floppy? |
| 08:23.45 | morthog | but if i boot my computer from it it works |
| 08:23.48 | morthog | yeah. it does..... |
| 08:23.58 | morthog | maybe i should try booting mine from floppy for real |
| 08:24.01 | morthog | duh |
| 08:24.03 | *** join/#edev JoeLlama (n=snork@206.169.63.10) |
| 08:24.10 | landley | Ok, try "cp /dev/fd0 tempfile.img" and boot qemu from the .img. |
| 08:24.40 | landley | Might be a strange permissions or driver issue with the floppy driver, which really hasn't been properly tested in a long time because nobody has them anymore. |
| 08:25.04 | morthog | that's what i was doing. i did a 'dd if=/dev/fd0 of=/coyote.img' and booted qemu with 'qemu -fda coyote.img -m 30' to try it out, and it failed miserably |
| 08:25.12 | morthog | lemmy do it again and get the exact message |
| 08:25.50 | landley | It could also be that qemu's floppy device emulation hasn't been properly tested in a while, for similar reasons. Try tomsrtbt. |
| 08:26.14 | morthog | everyting ok up until "Kernel panic: VFS: Unable to mount root fs on 01:00" |
| 08:26.32 | morthog | i better reboot with a real floppy of my project. this is just silly |
| 08:27.00 | morthog | is qemu's floppy emulation buggy? |
| 08:31.23 | landley | qemu over here just booted tomsrtb just fine. |
| 08:31.40 | landley | qemu 0.11.0 I think. |
| 08:32.00 | landley | 0.11.50 it says. |
| 08:34.38 | morthog | how do i check the version? |
| 08:34.58 | morthog | wow 0.9.1 |
| 08:35.11 | morthog | that might be it |
| 08:51.08 | *** join/#edev PureSine (n=PureSine@87.247.173.124) |
| 08:52.45 | *** join/#edev mogwart (i=551998b9@gateway/web/freenode/x-mdgxfxoialhkaydi) |
| 08:52.58 | mogwart | back. had some troubles there |
| 08:53.20 | *** join/#edev S_A (i=d2766cfe@gateway/web/freenode/x-dsscsjeeirrhipci) |
| 08:54.33 | S_A | Hi! I want to start on Linux porting and distro making but currently I have no hardware. so i thought of using qemu. Is there any step by step tutorial ? |
| 09:00.48 | *** join/#edev PureSine (n=PureSine@87.247.173.124) |
| 09:24.28 | landley | S_A: you might try http://impactlinux.com/fwl/downloads/presentation.pdf |
| 09:24.39 | landley | There are a few other things scattered around the fwl project's website. |
| 09:24.50 | landley | But alas, it is after 3am here and I go fall over now. |
| 09:24.55 | landley | g'night. |
| 09:40.53 | *** join/#edev PureSine (n=PureSine@87.247.173.124) |
| 09:44.05 | *** join/#edev PureSine (n=PureSine@87.247.173.124) |
| 09:58.15 | *** join/#edev florian (n=fuchs@Maemo/community/contributor/florian) |
| 10:00.22 | florian | good morning |
| 10:13.31 | *** join/#edev mogwart (i=8d4c2d22@gateway/web/freenode/x-dzcwjopaeviwgbxn) |
| 10:42.34 | *** join/#edev JoeLlama (n=snork@206.169.63.10) |
| 10:43.17 | *** join/#edev PureSine (n=chatzill@87.247.172.135) |
| 11:10.33 | *** join/#edev PureSine_ (n=chatzill@87.247.172.147) |
| 11:25.15 | *** join/#edev pirho (i=pirho@gateway/gpg-tor/key-0x2CEEC9CB) |
| 11:42.23 | *** join/#edev pirho (i=pirho@gateway/gpg-tor/key-0x2CEEC9CB) |
| 11:52.20 | *** join/#edev mpk (n=mpk@aggr.com) |
| 11:54.18 | *** join/#edev ChanServ (ChanServ@services.) |
| 11:54.18 | *** join/#edev florian (n=fuchs@Maemo/community/contributor/florian) |
| 11:54.18 | *** join/#edev S_A (i=d2766cfe@gateway/web/freenode/x-dsscsjeeirrhipci) [NETSPLIT VICTIM] |
| 11:54.18 | *** join/#edev hw_ (n=hw@p578b3905.dip0.t-ipconnect.de) [NETSPLIT VICTIM] |
| 11:54.18 | *** join/#edev mbuf (n=mbuf@118.102.130.6) [NETSPLIT VICTIM] |
| 11:54.18 | *** join/#edev SpeedEvil (i=1000@tor/regular/SpeedEvil) [NETSPLIT VICTIM] |
| 11:54.18 | *** join/#edev linac (n=lin@122.90.120.98) [NETSPLIT VICTIM] |
| 11:54.18 | *** join/#edev MonMotha (n=monmotha@cpe-65-29-71-209.indy.res.rr.com) [NETSPLIT VICTIM] |
| 11:54.18 | *** join/#edev bsdfox (n=h36sa@unaffiliated/bsdfox) [NETSPLIT VICTIM] |
| 11:54.18 | *** join/#edev archae0pteryx (n=snewman@207.47.42.130.static.nextweb.net) [NETSPLIT VICTIM] |
| 11:54.18 | *** join/#edev MuscleNerd (i=eric@adsl-75-27-189-151.dsl.irvnca.sbcglobal.net) [NETSPLIT VICTIM] |
| 11:54.18 | *** join/#edev CosmicPenguin (n=nobody@66-162-99-237.static.twtelecom.net) [NETSPLIT VICTIM] |
| 11:54.18 | *** join/#edev rektide (i=rektide@voodoowarez.com) [NETSPLIT VICTIM] |
| 11:54.18 | *** join/#edev eox425 (n=root@213.133.177.11) [NETSPLIT VICTIM] |
| 11:54.18 | *** join/#edev olivdt (n=x@unaffiliated/olivdt) [NETSPLIT VICTIM] |
| 11:54.18 | *** join/#edev keesj (n=keesj@ip49-193-210-87.adsl2.static.versatel.nl) [NETSPLIT VICTIM] |
| 11:54.18 | *** join/#edev monkey_ (n=monkey@216-243-6-35.users.condointernet.net) [NETSPLIT VICTIM] |
| 11:54.18 | *** join/#edev lilalinux (n=lilalinu@ist.deswahnsinns.de) [NETSPLIT VICTIM] |
| 11:54.18 | *** join/#edev vapier (i=UserBah@wh0rd.org) [NETSPLIT VICTIM] |
| 11:54.18 | *** join/#edev Madsy (n=Madsy@fu/coder/madsy) [NETSPLIT VICTIM] |
| 11:54.18 | *** join/#edev Redhatter (n=vk4msl@gentoo/developer/redhatter) [NETSPLIT VICTIM] |
| 11:54.18 | *** join/#edev dfarnsworth (n=dfarnswo@nat/montavista/x-lawzzmnatmzfsksb) [NETSPLIT VICTIM] |
| 11:54.18 | *** join/#edev Crofton|work (n=balister@pool-71-171-31-89.nwrknj.east.verizon.net) [NETSPLIT VICTIM] |
| 11:54.18 | *** join/#edev calculus (n=calculus@gentoo/user/calculus) [NETSPLIT VICTIM] |
| 11:54.18 | *** join/#edev ccole (n=cole@coledd.com) [NETSPLIT VICTIM] |
| 11:54.18 | *** join/#edev Ian_Daniher (n=it@nat-pool-128-88.olin.edu) [NETSPLIT VICTIM] |
| 11:54.19 | *** join/#edev dijenerate (n=dijenera@64.210.44.37) [NETSPLIT VICTIM] |
| 11:54.20 | *** join/#edev mnemoc (n=amery@shell.opensde.net) [NETSPLIT VICTIM] |
| 11:54.20 | *** join/#edev cjameshuff (n=cjameshu@pool-71-97-195-26.aubnin.dsl-w.verizon.net) [NETSPLIT VICTIM] |
| 11:54.20 | *** join/#edev chouimat (n=quassel@CPE002129b5a060-CM0011e6c40c15.cpe.net.cable.rogers.com) [NETSPLIT VICTIM] |
| 11:54.20 | *** join/#edev ohsix (i=ohsix@66.220.111.99) [NETSPLIT VICTIM] |
| 11:54.20 | *** join/#edev biberao (i=mapd@unaffiliated/biberao) [NETSPLIT VICTIM] |
| 11:54.20 | *** join/#edev armin76 (n=armin@gentoo/developer/armin76) [NETSPLIT VICTIM] |
| 11:54.20 | *** join/#edev HcE (n=egtvedt@egtvedt.no) [NETSPLIT VICTIM] |
| 11:54.20 | *** mode/#edev [+o ChanServ] by irc.freenode.net |
| 11:54.22 | *** join/#edev mpk (n=mpk@aggr.com) |
| 11:54.26 | *** join/#edev pirho (i=pirho@gateway/gpg-tor/key-0x2CEEC9CB) |
| 11:54.26 | *** join/#edev solar_ (n=solar@smtp.gentoo.org) |
| 11:54.27 | *** join/#edev PureSine (n=chatzill@87.247.172.147) |
| 11:54.27 | *** join/#edev SandGorgon (n=OmNomNom@122.173.249.152) |
| 11:54.41 | *** join/#edev PureSine_ (n=chatzill@87.247.172.147) [NETSPLIT VICTIM] |
| 11:54.41 | *** join/#edev roxfan (i=dunno@189.218-200-80.adsl-dyn.isp.belgacom.be) |
| 11:54.41 | *** join/#edev ccjoe (n=ccjoe@69.20.152.109) [NETSPLIT VICTIM] |
| 11:55.16 | *** join/#edev Spida (n=timo@ns1.spinnennetz.org) [NETSPLIT VICTIM] |
| 11:55.57 | *** join/#edev ccjoe (n=ccjoe@69.20.152.109) |
| 12:12.55 | *** join/#edev gustavoz (n=gustavoz@host248.201-252-31.telecom.net.ar) |
| 12:43.32 | *** join/#edev Crofton (n=balister@c-69-255-207-92.hsd1.va.comcast.net) |
| 13:17.31 | *** join/#edev PureSine (n=chatzill@87.247.172.147) |
| 13:19.17 | *** join/#edev PureSine (n=chatzill@87.247.172.147) |
| 13:21.07 | *** join/#edev PureSine (n=chatzill@87.247.172.147) |
| 13:23.03 | *** join/#edev RemoteSine (n=chatzill@87.247.172.147) |
| 13:33.38 | *** join/#edev Zoiah (i=Zoiah@matryoshka.zoiah.net) |
| 13:40.43 | *** join/#edev Crofton (n=balister@12.165.222.78) |
| 13:48.09 | *** join/#edev kergoth (n=kergoth@ip24-251-170-95.ph.ph.cox.net) |
| 14:09.44 | *** join/#edev GPSFan (n=kenm@64.92.145.112) |
| 14:09.44 | *** mode/#edev [+o GPSFan] by ChanServ |
| 14:14.01 | *** join/#edev likewise (n=chatzill@82-171-51-231.ip.telfort.nl) |
| 14:55.06 | *** part/#edev remotesine8 (n=chatzill@87.247.172.147) |
| 15:28.50 | *** join/#edev davidc__ (n=davidc__@user-387g3od.cable.mindspring.com) |
| 15:46.59 | *** join/#edev mbuf (n=mbuf@122.175.76.14) |
| 15:50.32 | chouimat | morning |
| 15:50.46 | florian | hi chouimat |
| 15:56.28 | *** join/#edev kergoth (n=kergoth@ip24-251-170-95.ph.ph.cox.net) |
| 15:59.51 | chouimat | is reading the 2.6.32 changelog page at kernelnewbies |
| 16:20.12 | mnemoc | does the dance of joy because of udevtmpfs |
| 16:20.23 | kergoth | udevtmpfs? |
| 16:20.27 | kergoth | hmm |
| 16:21.33 | chouimat | kergoth: some people call it devfs 2.0 ... |
| 16:21.43 | mnemoc | :) |
| 16:22.08 | florian | I tried the first board with it some hours ago... works like expected :-) |
| 16:22.21 | chouimat | mnemoc: this will make my life and probably of a few others a lot easier |
| 16:22.29 | mnemoc | die coldplug die |
| 16:22.43 | mnemoc | chouimat: yeah |
| 16:23.12 | *** join/#edev JoeLlama (n=snork@206.169.63.10) |
| 16:24.09 | chouimat | removes the cat from his desk ... development boards don't like cats |
| 16:24.40 | mnemoc | open windows like cats |
| 16:24.55 | mnemoc | and cats like to fly |
| 16:25.17 | chouimat | mnemoc: not sure he will like it much |
| 16:25.28 | mnemoc | try it :) |
| 16:26.11 | chouimat | mnemoc: it's ok he now on my sparc server and he will sleep there for a while |
| 16:27.09 | chouimat | mnemoc: and beside that, he's a lot less annoying than my former coworkers ... |
| 16:27.46 | mnemoc | :) |
| 16:41.39 | *** join/#edev cbx (n=cbx@unaffiliated/cbx) |
| 16:56.38 | chouimat | prpplague: good morning |
| 16:56.45 | prpplague | greetings |
| 16:58.07 | florian | hi prpplague |
| 16:58.23 | chouimat | hates sql |
| 16:58.24 | prpplague | florian: hey bud |
| 17:04.15 | *** join/#edev SandGorgon (n=OmNomNom@122.173.249.152) |
| 17:09.54 | prpplague | finds it interesting how people outside texas view texas and it's residents |
| 17:15.33 | solar | thinks of beef when thinking of texas |
| 17:16.49 | chouimat | only thinks of trigger happines when thinking of texas ;) |
| 17:18.02 | CosmicPenguin | Yeah, its not a popular state.... |
| 17:20.09 | prpplague | CosmicPenguin: yea seems so |
| 17:20.36 | prpplague | CosmicPenguin: the thing is i don't understand the hostility about it |
| 17:20.48 | solar | really? |
| 17:21.17 | solar | lots of texans are pretty closed minded about topics that the rest of the councty has been ok with for the past 20 years. |
| 17:21.23 | kergoth | didn't really think about texas at all until he moved there, but thought it was a great place to live, liked DFW |
| 17:21.25 | solar | country |
| 17:21.28 | prpplague | solar: such as? |
| 17:21.46 | SpeedEvil | is never going to texas. All the chainsaw massacres. |
| 17:22.21 | solar | prpplague: immigration. there seems to be some hostility towards the hispanic cultures there |
| 17:23.13 | CosmicPenguin | Honestly, I think that Texas's biggest problem is the pride that Texans have for Texas.... |
| 17:23.27 | kergoth | eh, there's no shortage of people that are annoyed at illegal immigrants anywhere |
| 17:23.30 | kergoth | texas is no exception |
| 17:24.20 | solar | sure. Just seems I see more ppl foaming at the mouth about that topic there. |
| 17:24.24 | CosmicPenguin | Texas is still just insular enough that the ideal of the The Republic of Texas persists |
| 17:24.35 | solar | which seems logical based on where it's umm located |
| 17:25.03 | CosmicPenguin | See also - Irish sentiment in New York, circa late 1800s |
| 17:25.11 | kergoth | solar: http://www.npr.org/templates/story/story.php?storyId=88002493 |
| 17:25.53 | prpplague | solar: hmm, see i find that interested, it has nothing to do with hispanic culture it has to do with illegal, not legal, immigration |
| 17:26.08 | SpeedEvil | many people conflate the two. |
| 17:26.38 | SpeedEvil | Personally I think a native language test would be useful. |
| 17:26.38 | prpplague | answers phone, brb |
| 17:26.53 | SpeedEvil | but, that's obviously controversial. |
| 17:27.35 | chouimat | SpeedEvil: you have to be able to speak English or French if you want to immigrate to Canada |
| 17:27.49 | solar | still the first things that comes to mind is beef, ranches and cowgirls |
| 17:28.04 | SpeedEvil | ponders ranches of cowgirls. |
| 17:28.20 | chouimat | I think this is called a Harem |
| 17:29.46 | prpplague | returns |
| 17:30.06 | prpplague | CosmicPenguin: i've heard that comment before about the pride of texans |
| 17:32.20 | kergoth | was reading stephen fry's in america the other day, english guy going state to state through the entire US, to get a feel for the similarities and differences. goes crab fishing in maine, etc. pretty amusing, always interesting to see the opinion of the country from an outsider, and also interesting for most of us, since we haven't even been to every state in the us ourselves |
| 17:32.25 | prpplague | like i said, it is the hostility that i don't get, i can understand disagreement, just not the hostility |
| 17:34.33 | prpplague | i mean, i don't assume that everyone in new york or california have the exact same opinions or attitudes, why is texas different? |
| 17:35.22 | prpplague | goes back to coding |
| 17:36.37 | chouimat | prpplague: I guess it's probably because texas is more vocal about stuff and some people use it as an example of what is wrong in your country ... just a guess |
| 17:37.50 | prpplague | chouimat: do you think texas is an example of what is wrong with the USA? |
| 17:38.09 | chouimat | prpplague: no |
| 17:38.20 | chouimat | prpplague: just the whole country ;) |
| 17:38.30 | prpplague | chouimat: hehe |
| 17:39.09 | CosmicPenguin | The Texas first mentality is a turn off for most americans, I think |
| 17:39.46 | prpplague | CosmicPenguin: yea i have heard that complaint before as well |
| 17:40.15 | CosmicPenguin | Its not that Texas is part of the United States, it is that Texas allows the United States to include it |
| 17:42.04 | prpplague | my interest is on the personal level, i find stereotypes used for texans very degrading |
| 17:42.57 | CosmicPenguin | You will have to join the club on that one |
| 17:43.14 | prpplague | funny thing is, i keep hearing how america is suppose to becoming more tolerant, but i find less and less so |
| 17:43.23 | CosmicPenguin | Like the stereotypes for Californians or Midwesterners or New Yorkers are any different |
| 17:44.04 | prpplague | CosmicPenguin: agreed, just never seen people who are hostile towards those stereotypes |
| 17:44.33 | CosmicPenguin | You've never seen anybody be hostile toward a Californian stereotype? |
| 17:44.47 | CosmicPenguin | Come hang about in Colorado for a while |
| 17:45.17 | prpplague | CosmicPenguin: why are they hostile to the californian stereotypes? |
| 17:45.25 | Mirell | God, like, |
| 17:45.33 | Mirell | For real, like, I'm in LA, like, it's totally awesome. |
| 17:45.34 | CosmicPenguin | Because the californians are invading and we're not amused |
| 17:45.35 | Mirell | Like yah. |
| 17:45.49 | prpplague | Mirell: ?? |
| 17:45.54 | Mirell | prpplague: LA girl |
| 17:46.17 | prpplague | not sure what Mirell is refering to |
| 17:46.40 | Mirell | You've never seen anybody be hostile toward a Californian stereotype? |
| 17:47.07 | prpplague | Mirell: i haven't |
| 17:47.20 | Mirell | Like, oh say, airheaded LA girls? |
| 17:47.58 | CosmicPenguin | Or wine country types? |
| 17:48.01 | CosmicPenguin | Or surfer dudes |
| 17:48.05 | prpplague | Mirell: maybe make fun of them, but not hostile |
| 17:48.19 | CosmicPenguin | or silicon valley tech heads? |
| 17:48.37 | prpplague | i think we are talking two different things |
| 17:48.38 | solar | CosmicPenguin: oregon seem to feel the same way about the california invasion |
| 17:48.50 | CosmicPenguin | Oregon, Idaho, Utah, Colorado |
| 17:48.59 | prpplague | i talking like "i hate that person and going to beat the @#$ out of them" |
| 17:49.17 | Mirell | San Francisco gay dudes? |
| 17:50.04 | CosmicPenguin | prpplague: most of us dislike Texas, but I think it is a rather broad brush to say that we violently hate texas - we just choose not to visit |
| 17:50.23 | CosmicPenguin | That said, there are violent people everywhere |
| 17:50.43 | prpplague | Mirell: yes i understand that, what i don't understand is a judgement based soley on what state you are from |
| 17:51.13 | Mirell | prpplague: Which I agree to, especially being in Austin. |
| 17:51.15 | solar | it's a sign of a shallow minded person. Best to avoid them types |
| 17:51.36 | CosmicPenguin | I'm sure everybody here has made an impression about somebody based on where they are from |
| 17:51.45 | CosmicPenguin | Its nearly unavoidable human nature |
| 17:52.35 | prpplague | indeed |
| 17:53.00 | prpplague | Mirell: indeed i bet you get it pretty heavy as well |
| 17:53.10 | CosmicPenguin | I'm from Wyoming - its nearly impossible not to come up with a stereotype about me |
| 17:53.15 | CosmicPenguin | Unfortunately, they are all mostly true |
| 17:53.20 | CosmicPenguin | Except for the sheep thing... |
| 17:54.53 | solar | a blank comes to mind when saying Wyoming. Maybe horseback riding comes to mind |
| 17:55.03 | prpplague | solar: i was about to say the same |
| 17:55.09 | SpeedEvil | Is it in the middle somewhere? |
| 17:55.36 | chouimat | SpeedEvil: the word you're looking is nowhere |
| 17:55.38 | prpplague | <CosmicPenguin> prpplague: most of us dislike Texas, but I think it is a rather broad brush to say that we violently hate texas - we just choose not to visit |
| 17:55.43 | kergoth | i seem to remember wyoming and nebraska both being fairly boring to drive through, thats about all i've got there |
| 17:56.01 | prpplague | CosmicPenguin: thats how i feel about new york and california |
| 17:56.18 | CosmicPenguin | Sounds like we're all on the same page |
| 17:56.37 | CosmicPenguin | And please feel free not to visit Wyoming |
| 17:56.52 | CosmicPenguin | We will go ahead and keep it to ourselvse |
| 17:56.53 | prpplague | hmm, looks like someone is trying to "Slam" my business longdistance service |
| 17:58.30 | chouimat | hmmm I think I'm sick ... I have craving for KFC today |
| 17:59.19 | CosmicPenguin | stops putting off debuging a performance problem |
| 17:59.21 | Mirell | prpplague: They're still doing the long distance slamming? Wow. |
| 17:59.42 | *** join/#edev JoeLlama (n=snork@206.169.63.10) |
| 18:00.00 | prpplague | Mirell: yea apparently they target businesses now |
| 18:01.16 | chouimat | loves this job posting ... http://capital-ridge.com/site/candidates/current-searches/mobile-os-developer/ |
| 18:04.16 | Mirell | chouimat: HowSo? |
| 18:04.46 | chouimat | Mirell: they ask for a java developer at the beginning ... and only ask for 5 years of C++ with linux :) |
| 18:05.21 | Mirell | Ah, heh. |
| 18:05.52 | chouimat | Mirell: and not sure I want to apply since I have the feeling it's for my former employer |
| 18:06.13 | Mirell | Always awkward.. |
| 18:07.04 | chouimat | ok I think i will go grab something to eat ... I'm tired of doing sql |
| 18:09.04 | Mirell | select food from things where calories < 2000 |
| 18:09.51 | chouimat | Mirell: I'm just writing a small application that will give me the next available name from the naming scheme and the next free ip ... |
| 18:23.42 | *** join/#edev dirk2 (n=dirk@p5B04194C.dip0.t-ipconnect.de) |
| 18:27.07 | *** join/#edev likewise (n=chatzill@82-171-51-231.ip.telfort.nl) |
| 18:38.01 | *** join/#edev florian (n=fuchs@Maemo/community/contributor/florian) |
| 18:42.51 | prpplague | ho ho hum |
| 18:42.54 | prpplague | dirk2: hey bud |
| 18:43.37 | prpplague | dirk2: what where you working on for this http://www.elinux.org/NaviEngine ? |
| 18:43.40 | Mirell | HoHum indeed. |
| 18:46.24 | prpplague | is doing battle with seasonal depression |
| 18:46.39 | prpplague | something about the holidays that always gets me depressed |
| 18:53.14 | Mirell | prpplague: You have SAD |
| 18:53.52 | prpplague | Mirell: Seasonally Aggravated Depression ? |
| 18:55.44 | Mirell | Heh |
| 18:55.53 | Mirell | is staring at Tcl Exception handling |
| 18:55.56 | Mirell | It ain't pretty. |
| 18:56.48 | CosmicPenguin | I am hungry |
| 18:56.55 | MonMotha | Mirell: that's OK. Everybody just does: try { ... } catch(Ex) {} anyway |
| 18:59.52 | Mirell | MonMotha: Tcl doesn't have that. |
| 19:00.15 | Mirell | MonMotha: They're still dickering around with it potentially: http://www.tcl.tk/cgi-bin/tct/tip/329 |
| 19:13.44 | calculus | solar: looks like landley's hotel is in sunnyvale |
| 19:28.17 | Mirell | Mmm, Sunnyvale. |
| 19:31.59 | Mirell | Wonder if Buffy is still hanging out there |
| 19:34.53 | wmat | prpplague: I haven't fixed the mass of Wanted Pages btw, I've just changed the MW setting to only show pages with 2 or more links ;) |
| 19:35.05 | wmat | prpplague: once I fix it, I'll set it back to 1 or more |
| 19:35.17 | wmat | stupid language settings munged everything |
| 19:35.29 | prpplague | wmat: yea i noticed the language items |
| 19:51.20 | *** topic/#edev by prpplague -> A channel for the discussion of embedded systems (Hardware/Software/Linux/WinCE/Commercial Device Repurposing) : Firmware Linux Discussion : Busybox and Buildroot Discussion : Gold 1218.00 : Silver 18.75 |
| 20:28.05 | prpplague | wmat: http://www.elinux.org/Special:LonelyPages |
| 20:43.40 | *** join/#edev cbx (n=cbx@unaffiliated/cbx) |
| 21:06.28 | *** join/#edev Inopia_ (n=niels@145.94.205.172) |
| 21:19.25 | *** join/#edev MonMotha (n=monmotha@cpe-65-29-71-209.indy.res.rr.com) |
| 21:53.57 | Mirell | Woah. |
| 21:54.00 | Mirell | They have free soup here |
| 22:14.51 | solar | gold has gone down ten bucks from what's listed in the /topic |
| 22:16.44 | *** join/#edev Crofton (n=balister@12.165.222.65) |
| 22:17.07 | prpplague | solar: where do you see that? |
| 22:17.08 | MonMotha | solar: it's still over $1200/oz, which is nvts |
| 22:17.30 | prpplague | solar: http://www.marketwatch.com/investing/future/future-us-gold |
| 22:20.21 | prpplague | solar: futures were up as high as 1226.00 - http://www.marketwatch.com/story/gold-taps-more-records-nears-1225-2009-12-02 |
| 22:29.23 | MonMotha | I wouldn't be surprised if the scrap price of my old class ring is now more than I paid for it |
| 22:29.44 | MonMotha | I have some old processors I've been meaning to ebay "for gold recovery" |
| 22:32.41 | prpplague | hehe |
| 22:33.02 | prpplague | well i expect to see $1300 by end of the year and $1800 by end of feb |
| 22:33.45 | prpplague | i'm curious to see how high it goes tomorrow after the job numbers are released |
| 22:35.00 | prpplague | alot of people are expecting the unemployement rate to hit 10.5 on tomorrows numbers |
| 22:35.29 | Spida | prpplague: in which country? |
| 22:35.40 | Spida | usa? |
| 22:35.41 | prpplague | Spida: USA |
| 22:36.34 | MonMotha | Gold is looking like a bubble at this point |
| 22:36.43 | MonMotha | I don't think those kind of numbers are sustainable |
| 22:36.44 | Mirell | All those sub-prime gold loans |
| 22:36.53 | MonMotha | it's like oil at $160/bbl |
| 22:37.09 | prpplague | MonMotha: normally i would agree with you |
| 22:37.33 | MonMotha | The question is when it will crash (or otherwise fall). That I cannot say. There's a lot of forces bidding it up. |
| 22:37.49 | prpplague | MonMotha: problem is USD, if the fed keeps printing money people are going to keep putting their money in something they can be sure of |
| 22:37.54 | MonMotha | true |
| 22:38.07 | MonMotha | though I'm not sure the other international currencies are holding up as well as people seem to think, Yen in particular |
| 22:38.52 | prpplague | MonMotha: alot better than the USD that is for sure |
| 22:38.56 | MonMotha | they said the same thing about oil being so high |
| 22:39.06 | MonMotha | well yeah, USD is getting hammered, but I think some of that is fear |
| 22:39.37 | wmat | prpplague: those orphans are likely a result of a broken link; i've fixed a few in the past where the name of the original link was slightly different from the name of the new page |
| 22:39.40 | MonMotha | domestic inflation has thus-far been fairly low, and the Fed has indicated that they will stop printing money if inflation does become a concern |
| 22:39.52 | MonMotha | hopefully that won't be too sudden for the fed to react |
| 22:40.19 | MonMotha | dinner... |
| 22:40.50 | prpplague | wmat: yea, i was just showing that we've gotten it down from 153 to less than 30 |
| 22:41.12 | Mirell | I evidently need to learn FPGA programming... |
| 22:41.14 | prpplague | wmat: i found alot of the pages were linked but linked using a full url rather than a wiki link |
| 22:41.22 | prpplague | Mirell: fun fun |
| 22:43.12 | prpplague | ho ho ho |
| 22:43.19 | SpeedEvil | china wants to support hte dollar - for a while |
| 22:43.27 | prpplague | has too much work to do and doesn't feel like it |
| 22:43.36 | SpeedEvil | Oil however is no longer 30/bbl |
| 22:44.40 | prpplague | SpeedEvil: indeed, which is also due to demand |
| 22:44.45 | solar | prpplague: looking at the spot price. |
| 22:44.55 | SpeedEvil | Bouncing around the high 70s |
| 22:44.57 | prpplague | solar: url? |
| 22:45.08 | solar | prpplague: http://www.kitco.com/market 24hgold.com etc. but I do most of my trading on bulliondirect.com |
| 22:46.07 | solar | gold and oil usually follow each other in the market place |
| 22:46.12 | prpplague | solar: right |
| 22:46.31 | prpplague | solar: and gold(along with other metals) is inversely related to dollar |
| 22:46.59 | solar | pretty much. better gold is going usually means the worse amercians are doing |
| 22:47.09 | prpplague | solar: indeed |
| 22:57.32 | *** join/#edev SpeedEvil (n=user@tor/regular/SpeedEvil) |
| 23:00.11 | Mirell | I wish I could expensive a nice Aeron chair... |
| 23:00.22 | Mirell | I've expensed a $12K server and $15K switch today already... |
| 23:00.26 | Mirell | Expense rather |
| 23:01.59 | SpeedEvil | It's not a chair, it's a network controller support device. |
| 23:02.52 | Mirell | Nice. |
| 23:17.29 | solar | dislikes Aeron chairs. the plastic frame makes the sitting experience less fun. I'm like the only guy in my office that took his and put it into the execve room. Then snagged a good old leather chair |
| 23:18.29 | Mirell | solar: There's three different sizes. |
| 23:18.55 | Mirell | solar: You have to get one where the frame doesn't bit into your thighs |
| 23:19.01 | Mirell | solar: Bite, rather. |
| 23:21.02 | solar | Mirell: yeah we have a few diff sizes around. |
| 23:21.11 | solar | it's just not the chair for my back. shrug |
| 23:23.23 | Mirell | Ahh |
| 23:23.41 | Mirell | I'm just the opposite, I don't like the leather/foam things, just because I get too warm |
| 23:24.04 | Mirell | I also like to keep my apartment at 68F..so. |
| 23:27.53 | *** join/#edev bsdfox (n=h36sa@unaffiliated/bsdfox) |
| 23:37.34 | chouimat | Mirell: I miss the 1.5K chair I had at my former workplace |
| 23:43.36 | solar | 1.5k on a chair #$% |
| 23:43.57 | solar | my sofa at home was less |
| 23:44.17 | *** join/#edev bsdfox__ (n=h36sa@c-67-161-174-21.hsd1.ca.comcast.net) |
| 23:45.41 | Mirell | Is your sofa at home sat in 8 hours a day for years? |
| 23:46.20 | solar | not yet. But it's part of my plan |
| 23:53.17 | Mirell | Heh |