| 02:19.57 | *** join/#edev mpk (n=mpk@aggr.com) |
| 02:27.09 | *** join/#edev Madsy (n=Madsy@fu/coder/madsy) |
| 02:54.06 | Mirell | Hmm... |
| 03:07.07 | Mirell | HoHum. |
| 03:07.15 | *** mode/#edev [+o Mirell] by ChanServ |
| 03:38.10 | *** join/#edev landley (n=landley@cpe-66-68-109-179.austin.res.rr.com) |
| 03:38.10 | *** mode/#edev [+o landley] by ChanServ |
| 04:11.40 | *** join/#edev landley (n=landley@66.68.109.179) |
| 04:11.40 | *** mode/#edev [+o landley] by ChanServ |
| 04:22.22 | *** join/#edev chiques (n=working_@pool-71-189-111-152.lsanca.fios.verizon.net) |
| 05:20.32 | monkey | Mirell: I was wrong about something. the board doesn't have uclinux, it's bare metal |
| 05:21.18 | Mirell | Ahh |
| 05:21.34 | monkey | that might help. haha. |
| 05:21.53 | Mirell | Just a bit! |
| 05:22.49 | monkey | that cross-compiler I downloaded had 64bit binaries |
| 05:23.19 | Mirell | Hmm... |
| 05:25.04 | monkey | uname is 686, and cross-compiler-armv4l/bin/armv4l-gcc: ELF 64-bit LSB executable, x86-64, version 1 (SYSV), dynamically linked (uses shared libs), for GNU/Linux 2.6.9, stripped |
| 05:25.55 | Mirell | monkey: Did you grab cross-static? |
| 05:26.56 | monkey | no, i got cross-compiler |
| 05:26.58 | Mirell | Because that's the i686 version |
| 05:26.59 | Mirell | Ah |
| 05:27.02 | monkey | ok |
| 05:27.16 | monkey | so, is that still the appropriate version given it's bare metal? |
| 05:27.19 | monkey | brb, dog needs out |
| 05:27.19 | landley | hmmm? |
| 05:27.30 | landley | monkey: download the -static versions. |
| 05:27.33 | landley | Those are 32 bit binaries. |
| 05:27.42 | landley | Next time, I'm just uploading the 32 bit ones for everything. |
| 05:27.43 | Mirell | landley: monkey is trying to work on bringing up an ARMv4L board |
| 05:27.53 | landley | armv4tl? |
| 05:27.55 | landley | With -eabi? |
| 05:28.04 | landley | Or armv4 without the thumb extensions? |
| 05:28.44 | landley | A bare metal compiler is actually a subset of the compilers we're building, so it should work fine. |
| 05:29.03 | Mirell | landley: ARM9TDMIS I believe. |
| 05:29.07 | landley | It means you won't need our C library, system headers, or ELF setup code stuff. |
| 05:29.12 | landley | That's armv4tl. |
| 05:29.40 | landley | If he's doing bare metal I'm not sure oabi vs eabi enters into it, but just in case it does he might want to grab the source code from the repository and build the new "armv4tl" target from source. |
| 05:30.02 | landley | We'll upload a prebuilt binary of that next release, but we hadn't added that target last release. |
| 05:30.21 | landley | I need to rename the old armv4l to armv4l-oabi. |
| 05:37.25 | monkey | landley: it's an arm7-tdmis |
| 05:37.32 | monkey | lpc2378 |
| 05:38.01 | monkey | landley: i wasn't sure about eabi or not |
| 05:38.34 | landley | monkey: eabi/oabi is purely a software thing. |
| 05:38.40 | landley | I dunno what your "bare metal" usage is. |
| 05:38.53 | landley | It specifies things like what registers arguments go in when you make a system call. |
| 05:39.22 | monkey | I can't answer that yet. I"m just getting started with this embedded program |
| 05:39.23 | landley | It's a bit like a.out vs elf executables, or solaris binaries vs linux binaries. Nothing to do with the actual hardware. |
| 05:39.26 | monkey | ok. thanks |
| 05:39.31 | landley | The exiting prebuilt armv4l compiler should be a decent start for you then. |
| 05:39.37 | landley | As long as it's consistent, it shouldn't matter. |
| 05:40.02 | monkey | the class is using is yagarto |
| 05:41.22 | *** join/#edev davidc__ (n=davidc__@96.49.165.44) |
| 05:42.06 | landley | http://www.arm.com/products/CPUs/ARM7TDMIS.html |
| 05:42.10 | landley | That's armv4t |
| 05:42.32 | monkey | will the armv4l work still? |
| 05:42.41 | landley | But t is an optional extension. |
| 05:42.45 | landley | You can ignore it. |
| 05:42.47 | monkey | ok |
| 05:42.53 | landley | It's a bit like having a processor with... |
| 05:42.56 | landley | I hate this cold. |
| 05:42.59 | landley | Those floating point extension thingies. |
| 05:43.01 | landley | mmx? |
| 05:43.08 | Mirell | MMX, SSE, SSE2... |
| 05:43.12 | monkey | ok. ah |
| 05:44.31 | landley | I should be online tomorrow if you have trouble with the toolchain. If it doesn't work out of the box for you, I want to know why so I can fix it. (rob @ landley.net). |
| 05:44.51 | monkey | ok. thanks. :) |
| 05:44.54 | landley | Actually, we should even have armv4tl snapshots up which _do_ use the thumb extensions, and eabi. |
| 05:45.24 | Mirell | landley: Symlink changes prevented nightly snapshots from uploading, which I realized during this conversation, and it's uploading currently... |
| 05:45.26 | monkey | w00t! |
| 05:45.27 | landley | I forgot we've got a nightly snapshot building cron ob. |
| 05:45.39 | landley | monkey: talk to Mirell, he's better at this part than I am. :) |
| 05:45.53 | monkey | mirell has been a bunch of help already |
| 05:48.12 | Mirell | Lovely. |
| 05:48.12 | Mirell | It looks like are core router lost its BGP session which took down our entire network (as far as we can tell). We have our network administrator heading to the data center now and should be there in 20 minutes or so. More updates to come. |
| 05:48.16 | Mirell | This is affecting all DreamHost services. |
| 05:48.40 | monkey | good thing I don't host with dreamhost |
| 05:48.53 | Mirell | monkey: Unfortunately I do |
| 05:48.58 | monkey | sad :( |
| 05:48.59 | monkey | ;) |
| 05:49.27 | Mirell | And it's where impactlinux.com is hosted. Grr. |
| 05:50.10 | monkey | *sigh* my dog wont stop whining |
| 05:50.27 | Mirell | Poor doggie |
| 05:50.58 | landley | monkey: there's a mirror on http://landley.net/code/firmware/downloads that you can download the -static thingies from, but it's slow and doesn't have the nightly snapshots. |
| 05:51.02 | landley | Just the last release version. |
| 05:51.23 | monkey | i already downloaded it, i probably wont get to it tonight. i need to finish other homework and turn it in within the hour |
| 05:52.02 | landley | No rush. :) |
| 05:52.38 | monkey | there is for me. i was unable to do homework since I couldn't compile anything. i amnaged to get it compiled on the mac right before class. but that's a less than ideal situation |
| 05:53.15 | Mirell | Hmm...I should look at Slicehost. http://www.slicehost.com/ |
| 05:54.27 | monkey | mmm, slices of pi |
| 05:55.43 | landley | I take it that the yatoolchain didn't work for you? |
| 05:56.03 | Mirell | landley: It's Windows-only |
| 05:56.27 | monkey | yagarto is osx/windows |
| 05:56.31 | monkey | i have an macbook pro |
| 05:57.02 | monkey | yagarto did compile it, but I did it as a last resort to turn something in |
| 05:57.12 | landley | ah. |
| 05:57.27 | landley | Ours is Linux, although Mirell was poking at getting it to work on a mac host. |
| 05:57.40 | landley | Why is osx not good for a macbook pro? |
| 05:57.47 | landley | is confused. |
| 05:57.54 | landley | has a cold. Easily confused at present. |
| 05:58.01 | landley | Even moreso than usual, which is an amazing thought. |
| 05:59.07 | monkey | I don't like the macbook. |
| 05:59.17 | landley | ah. |
| 05:59.30 | monkey | But, I just have it for work purposes. But I only use it when I have to |
| 05:59.45 | landley | Mirell: you've got a good bandwidth plan on dreamhost. The slicehost thing starts at 100 gigs, which we could probably blow through fast-ish. |
| 06:00.31 | Mirell | landley: Cheap, Lots of Bandwidth/Storage, Reliable, pick two. |
| 06:01.03 | monkey | who wants to add twos complements numbers for me? |
| 06:01.43 | landley | hmmm? |
| 06:02.15 | monkey | it's a signed hex numbers I'm supposed to do math with |
| 06:02.31 | landley | I know what they are, I'm just wondering the context of your question. |
| 06:02.53 | landley | Just add 'em together and drop the overflow. |
| 06:02.53 | monkey | oh, i just don't feel like doing this now, but it's due soon :) |
| 06:03.03 | landley | Python. |
| 06:03.31 | landley | print hex(0x413+0x722) |
| 06:03.48 | landley | Best desktop calculator ever. :) |
| 06:04.36 | monkey | w00t! |
| 06:08.54 | *** join/#edev lyakh (n=lyakh@p57BD1716.dip0.t-ipconnect.de) |
| 06:11.37 | monkey | how does something like the arm7 handle fp? |
| 06:19.05 | landley | soft-float, I expect. |
| 06:19.16 | landley | Fakes it with integer operations. |
| 06:21.11 | landley | Yeah, I don't see any floating point hardware mentioned at http://www.arm.com/products/CPUs/ARM7TDMIS.html |
| 06:21.37 | landley | And our toolchain is configured for soft-float anyway. (It replaces floating point operations with calls to a function library in libgcc.a) |
| 06:21.50 | monkey | ok |
| 06:22.05 | landley | what is it you're building and how are you expected to run it, anyway? |
| 06:22.16 | *** join/#edev SandGorgon (n=OmNomNom@122.160.41.129) |
| 06:22.25 | landley | Note that we tend to throw qemu at at things, which has the advantage that you can take it with you on your laptop. |
| 06:22.58 | monkey | it was just a general "hello world" that was pre-written, we just had to compile it and give some proof to the teacher |
| 06:23.27 | landley | I'm wondering how you run it. (How does it print hello world? Presumably using newlib, installed via the jtag...) |
| 06:24.04 | monkey | via jtag, i didn't get that far to actually transfer it to the board |
| 06:24.04 | monkey | there is an lcd on the board but we haven't touched that yet |
| 06:26.29 | monkey | are you an embedded dev as a profession? |
| 06:30.59 | landley | monkey: try to be. Underemployed at the moment, so it's being a hobby. |
| 06:31.09 | monkey | that sucks. |
| 06:31.16 | landley | shrugs. |
| 06:31.25 | monkey | or you can enjoy the time off |
| 06:31.35 | monkey | are you incharge of the cross-compiler thingie? |
| 06:31.37 | landley | I've been doing so, but it's getting a little old. |
| 06:31.51 | landley | Which cross compiler thingy, the firmware linux one? I'm the maintainer of that project. |
| 06:32.06 | monkey | cross-static-armv4l :) |
| 06:32.12 | landley | Yup. |
| 06:32.42 | monkey | is the embedded programming market competitive?(job wise) |
| 06:32.56 | landley | At the moment everything's competitive. |
| 06:33.04 | landley | There are 6 people looking for every job opening in the US. |
| 06:33.42 | Mirell | http://www.theonion.com/content/news/3_2_million_unemployed_americans |
| 06:35.30 | monkey | landley: I just tried to compile with cross-static, and got lots of `no symbol' missing from main hash table |
| 06:37.52 | landley | What did you try to compile? |
| 06:38.13 | monkey | that helloworld |
| 06:38.33 | landley | How did you try to compile it? "armv4l-gcc hello.c"? |
| 06:38.43 | monkey | makefile |
| 06:39.56 | monkey | <PROTECTED> |
| 06:40.02 | landley | Sure. |
| 06:41.58 | monkey | oh wait, i got a hex file, so it probably compiled. let me show you anyway |
| 06:42.52 | monkey | here http://pastebin.ca/1617374 |
| 06:43.48 | landley | What's in the linker script? |
| 06:43.58 | landley | prj/lpc2378_flash.ld |
| 06:44.42 | landley | Huh, actually that looks like it's defined but never used. |
| 06:44.44 | monkey | http://pastebin.ca/1617376 |
| 06:44.46 | landley | What error output are you getting? |
| 06:45.23 | landley | Ok, that looks vaguely reasonable... |
| 06:45.51 | landley | Your first pastebin isn't a complete makefile, it doesn't define any targets. Just a lot of variables. |
| 06:45.54 | monkey | http://pastebin.ca/1617378 |
| 06:46.06 | monkey | what's it msising? |
| 06:46.20 | monkey | oops, i see what i missed |
| 06:47.07 | landley | Ok, so the command line "armv4l-gcc ./src/crt.o ./src/main.o ./src/init.o ./src/print.o -mcpu=arm7tdmi -nostartfiles -T./prj/lpc2378_flash.ld -Wl,-Map=LPC2378HelloW.map,--cref,--no-warn-mismatch  -o LPC2378HelloW.elf" is what's producing the error output. |
| 06:47.12 | landley | Apparently the stuff before that worked. |
| 06:48.26 | landley | Cut and paste that line and run it from the command line to confirm it _is_ causing the problem, and then set try running it with WRAPPER_DEBUG=1 at the beginning (I.E. "WRAPPER_DEBUG=1 armv4l-gcc...") |
| 06:49.24 | landley | Also, try running "armv4l-rawgcc" instead of armv4l-gcc and see if that fixes it. |
| 06:50.06 | landley | I'm suspecting that my gcc wrapper has never encountered -nostartfiles before, and is misbehaving. (Easy enough to fix, but I'll probably have to do it in the morning. It's coming up on 2am here.) |
| 06:50.21 | monkey | yeah |
| 06:50.30 | monkey | ok, i'll give it a try |
| 06:50.32 | landley | You should be able to bypass that for the moment by using armv4l-rawgcc instead of armv4l-gcc, since you're doing bare metal builds anyway. |
| 06:51.08 | landley | There's also the possibility it's trying to access files out of newlib, which the other toolchain has and this doesn't. But I _think_ that's not the case, since that wouldn't really be a bare metal build either... |
| 06:51.09 | monkey | i tried it, got an error |
| 06:51.21 | landley | Tried which, the WRAPPER_DEBUG or the rawgcc? |
| 06:51.23 | landley | And was it the same error? |
| 06:51.25 | monkey | armv4l-rawgcc ./src/crt.o ./src/main.o ./src/init.o ./src/print.o -mcpu=arm7tdmi -nostartfiles -T./prj/lpc2378_flash.ld -Wl,-Map=LPC2378HelloW.map,--cref,--no-warn-mismatch -o LPC2378HelloW.elf |
| 06:51.29 | monkey | /home/monkey/Download/cross-static-armv4l/usr/bin/../libexec/gcc/armv4l-unknown-linux/4.2.1/ld: cannot find -lgcc |
| 06:51.36 | monkey | i changed both gcc things in the make file to rawgcc |
| 06:51.54 | landley | Sigh. Ok, it does need the wrapper so it can find its internal files. |
| 06:52.17 | landley | Could you do the WRAPPER_DEBUG run so I can see the dump of what it's rewriting the command line to? |
| 06:52.52 | monkey | where do I add that? |
| 06:53.24 | landley | It's an environment variable, add it to the start of the command line before the executable name. |
| 06:53.31 | landley | WRAPPER_DEBUG=1 armv4l-gcc hello.c |
| 06:53.45 | monkey | ok |
| 06:53.53 | landley | That's a special syntax to define environment variables for this command only. |
| 06:54.11 | monkey | should I switch it back to gcc? |
| 06:54.16 | landley | Yes please. |
| 06:54.29 | landley | The wrapper tells it where to find lgcc. |
| 06:55.21 | monkey | same garbage |
| 06:55.25 | monkey | as before |
| 06:55.33 | landley | It should print out more data at the beginning? |
| 06:56.14 | monkey | I didn't really see much. i just did WRAPPER_DEBUG=1 make |
| 06:56.22 | monkey | outgoing: armv4l-rawgcc -nostdlib -Wl,--dynamic-linker,/lib/ld-uClibc.so.0 -Wl,-rpath-link,/home/monkey/Download/cross-static-armv4l/bin/../lib -L/home/monkey/Download/cross-static-armv4l/bin/../lib -L/home/monkey/Download/cross-static-armv4l/bin/../gcc/lib -nostdinc -isystem /home/monkey/Download/cross-static-armv4l/bin/../include -isystem /home/monkey/Download/cross-static-armv4l/bin/../gcc/include -U__nptl__ ./src/crt.o ./src/main.o . |
| 06:56.30 | landley | Yeah, that. |
| 06:56.31 | monkey | that might have been different from last time, but the buffer is gone |
| 06:56.33 | monkey | oh ok. |
| 06:57.12 | landley | Yup, crt.o is still in there... |
| 06:57.38 | landley | I need to study what -nostartfiles is supposed to do. |
| 06:58.02 | monkey | ok. |
| 06:59.20 | landley | Ok, see the above big "outgoing" line? |
| 06:59.34 | landley | No, hang on a sec... |
| 06:59.43 | monkey | [monkey@orb LPC2378HelloW]$ WRAPPER_DEBUG=1 make |
| 06:59.43 | monkey | armv4l-gcc ./src/crt.o ./src/main.o ./src/init.o ./src/print.o -mcpu=arm7tdmi -nostartfiles -T./prj/lpc2378_flash.ld -Wl,-Map=LPC2378HelloW.map,--cref,--no-warn-mismatch -o LPC2378HelloW.elf |
| 06:59.46 | monkey | incoming: armv4l-gcc ./src/crt.o ./src/main.o ./src/init.o ./src/print.o -mcpu=arm7tdmi -nostartfiles -T./prj/lpc2378_flash.ld -Wl,-Map=LPC2378HelloW.map,--cref,--no-warn-mismatch -o LPC2378HelloW.elf |
| 07:00.05 | landley | The crt.o file is something this makefile is explicitly building. |
| 07:00.09 | landley | So that's probably ok then... |
| 07:01.30 | landley | That outgoing line is truncated, could you paste the second half of it? |
| 07:01.38 | landley | Or throw the output to pastebin? |
| 07:02.10 | monkey | that outgoing line is the whole thing |
| 07:02.32 | monkey | right after that starts ttht main hash table thing |
| 07:02.42 | landley | Here it makes it to "U__nptl__ ./src/crt.o ./src/main.o ." and stops. |
| 07:03.53 | monkey | brb, dog is whining still.... |
| 07:04.00 | landley | These things happen. :) |
| 07:05.22 | landley | Just confirmed, the wrapper script _is_ understanding -nostartfiles. |
| 07:05.23 | landley | So that's not it. |
| 07:10.24 | landley | monkey: add -v to the gcc invocation. |
| 07:10.37 | landley | It's going to be _very_ verbose, but it should tell you what it's sending to the actual linker. |
| 07:10.56 | landley | This "symbol 'no symbol'" thing is linker weirdness. |
| 07:14.04 | *** join/#edev bsdfox (n=h36sa@unaffiliated/bsdfox) |
| 07:14.41 | monkey | ha, ok |
| 07:15.34 | monkey | ok, hold on |
| 07:17.50 | monkey | http://pastebin.ca/1617416 |
| 07:21.34 | landley | monkey: I'm not seeing the error messages, were there more than 218 lines? |
| 07:21.59 | monkey | after that paste is the missing hash tables |
| 07:23.29 | landley | Why isn't it adding -static anywhere? |
| 07:23.42 | landley | This is a dynamic link, despite the -nostartfiles, which seems... odd. |
| 07:23.49 | landley | Try adding -static. |
| 07:24.06 | landley | To wherever -nostartfiles is specified, I mean. |
| 07:24.48 | monkey | clean build |
| 07:24.57 | landley | Probably a good idea, yes. |
| 07:25.08 | monkey | thanks |
| 07:25.19 | landley | Oh, you mean it _did_ build. :) |
| 07:25.27 | landley | Woot. (Does it run?) |
| 07:25.35 | monkey | well, it built before(at least had a hex file) i have no idea if it runs |
| 07:25.36 | monkey | lol |
| 07:25.38 | landley | Possibly -nostartfiles implies -static? |
| 07:25.43 | monkey | http://www.youtube.com/watch?v=RXJKdh1KZ0w&feature=related |
| 07:25.49 | monkey | probably not. |
| 07:25.54 | landley | I'm wondering if I should adjust my wrapper, or if gcc is supposed to already do that... |
| 07:25.57 | monkey | -static fixed it, so it's probably not -implied |
| 07:26.04 | landley | Oh, hang on, I think I understand. |
| 07:26.13 | landley | The other toolchain doesn't _have_ shared library support, so there's no need to disable it. :) |
| 07:26.17 | monkey | i mean, it compiled either way |
| 07:26.20 | landley | The other toolchain's static only. |
| 07:27.10 | landley | Do you have a test environment to run the result in? |
| 07:27.34 | monkey | i have the dev board it self ;) |
| 07:27.38 | landley | I'd like to make sure it actually works before going to bed. |
| 07:27.50 | monkey | lol |
| 07:28.02 | monkey | i haven't gotten as far as hooking it up. let me connect the baord/jtag thingie |
| 07:28.48 | monkey | now i need to find flash instructions |
| 07:29.02 | monkey | i have an openocd debugger, do I use openocd to flash it? |
| 07:29.13 | landley | Yup. |
| 07:29.17 | landley | You should have instructions for that already? |
| 07:29.27 | landley | If you wait until morning, prpplague on this channel can tell you all about openocd. |
| 07:29.41 | landley | I'm not an expert there, he is. |
| 07:30.12 | monkey | in the instructions for the class, they're using eclipse/yagarto/openocd drivers. i'd like to skip that |
| 07:30.23 | landley | Ah. |
| 07:30.42 | landley | I have a shell script here that flashed a nail board, but openocd needs a funky config file that tells it how to drive a given board. |
| 07:31.06 | monkey | there is an openocd.cfg file i believe |
| 07:31.28 | landley | It needs some black magic to initialize the dram controller or some such. |
| 07:31.29 | monkey | oh wait, i think they're using flash magic to flash the board |
| 07:32.14 | monkey | oh wait, flash magic is for using serial |
| 07:32.46 | monkey | shit, the files are batch, gross |
| 07:32.48 | landley | jtags often provide a serial interface. |
| 07:33.03 | landley | I think you really want to talk to prpplague about this. He generally logs on around 9am central time. |
| 07:33.06 | landley | What time zone are you in? |
| 07:33.14 | monkey | pst |
| 07:33.26 | landley | When you wake up in the morning, prpplague should be on. |
| 07:33.30 | monkey | thanks again, you've been a big help |
| 07:33.51 | landley | Glade I coudl help. I'll try to log on sometime in the morning to make sure the image you built worked for you. |
| 07:33.59 | landley | Just because it compiled, doesn't mean it'll run. :) |
| 07:34.09 | monkey | yeah. but this is lightyears ahead of where I was before |
| 07:35.04 | landley | By the way, I think you just run ./openocd -f thingy.cfg |
| 07:35.34 | landley | You may have to do "LD_LIBRARY_PATH=. ./openocd -f thingy.cfg" if your libflorist.sh is in the current directory instead of /lib. |
| 07:36.00 | landley | The LD_LIBRARY_PATH environment variable tells it extra library directories to look in, with . being the current one of course. :) |
| 07:36.02 | monkey | the openocd scripts are batch, not bash |
| 07:36.44 | landley | In my case, for the nail board prpplague whipped up "reset.cfg" which calls "reset.ocd" and similar. |
| 07:37.24 | landley | open obsessive compulsive disorder has its own data file format, which I do not particularly understand and prpplague does. |
| 07:37.41 | landley | (I'm fairly certain that's not what the actual program name stands for.) |
| 07:37.55 | landley | However, seeing the acronym... |
| 07:38.23 | monkey | haha |
| 07:39.49 | monkey | ok, it's bed time |
| 07:40.38 | landley | Same here. |
| 07:40.40 | landley | g'night. |
| 07:41.14 | *** part/#edev landley (n=landley@66.68.109.179) |
| 08:18.25 | *** join/#edev linac (n=lin@116.205.33.177) |
| 08:24.00 | *** join/#edev florian_kc (n=fuchs@port-217-146-132-69.static.qsc.de) |
| 09:01.06 | Mirell | "My cat has a wooden leg. OK, not a whole leg - just a faux paw." |
| 09:16.56 | monkey | lol |
| 09:17.18 | *** join/#edev SandGorgon (n=OmNomNom@122.160.41.129) |
| 09:56.01 | *** join/#edev bsdfox (n=h36sa@unaffiliated/bsdfox) |
| 10:15.20 | *** join/#edev pirho (i=pirho@gateway/gpg-tor/key-0x2CEEC9CB) |
| 11:57.16 | Redhatter | Has anyone done any bare metal development on i386? |
| 12:12.30 | chouimat | morning |
| 12:16.04 | *** join/#edev gustavoz (n=gustavoz@imhotep.toptech.com.ar) |
| 13:37.35 | *** join/#edev GPSFan (n=kenm@64.92.145.112) |
| 13:37.35 | *** mode/#edev [+o GPSFan] by ChanServ |
| 14:07.01 | *** join/#edev adyer (n=adyer@173-15-79-94-Illinois.hfc.comcastbusiness.net) |
| 14:18.48 | *** join/#edev eox425 (n=root@213.133.177.11) |
| 14:19.29 | *** part/#edev eox425 (n=root@213.133.177.11) |
| 14:31.45 | *** join/#edev eox425 (n=root@213.133.177.11) |
| 14:42.01 | eox425 | Hi guys |
| 14:42.24 | eox425 | Did someone use Nut/OS |
| 14:42.41 | eox425 | ??? Is it worth something ??? |
| 14:42.58 | eox425 | http://www.ethernut.de/en/firmware/index.html |
| 14:45.06 | florian | hi eox425 |
| 14:45.33 | florian | eox425: I didn't try it myself but I know people using it. |
| 14:57.06 | eox425 | florian: And how is it? is it worth it to use as an web server OS an Atmega128? |
| 14:57.53 | eox425 | <PROTECTED> |
| 14:58.19 | florian | eox425: People use it in this way iirc. Its not perfect but people like it because it is POSIX-like and quite easy to port. |
| 14:59.26 | eox425 | Is it GPL? |
| 15:01.59 | florian | no its BSD licensed |
| 15:13.40 | chouimat | hmmm look interesting |
| 15:14.54 | *** join/#edev mbuf (n=mbuf@124.123.179.93) |
| 15:58.32 | *** join/#edev kergoth (n=kergoth@206.112.117.35) |
| 16:56.57 | *** join/#edev eox425 (n=root@213.133.177.11) |
| 17:06.06 | *** join/#edev Redhatter (n=vk4msl@2001:388:f000:0:0:0:0:71) |
| 17:44.49 | *** join/#edev _dieter_ (n=quassel@p549B4894.dip.t-dialin.net) |
| 18:03.05 | *** join/#edev eox425 (n=root@213.133.177.11) |
| 18:14.28 | *** join/#edev dimm (i=gentux@unaffiliated/dimm) |
| 18:24.30 | *** join/#edev eox4251 (n=root@213.133.177.11) |
| 18:24.44 | *** join/#edev florian (n=fuchs@f048129147.adsl.alicedsl.de) |
| 18:28.08 | *** join/#edev pirho (i=pirho@gateway/gpg-tor/key-0x2CEEC9CB) |
| 18:58.40 | *** join/#edev brolin (n=brolin@200.24.16.217) |
| 19:03.21 | Mirell | Hmm.. |
| 19:08.40 | *** join/#edev eox425 (n=root@213.133.177.11) |
| 19:24.26 | *** join/#edev eox425 (n=root@213.133.177.11) |
| 20:00.01 | *** join/#edev dimm (i=gentux@unaffiliated/dimm) |
| 21:42.04 | *** join/#edev ccjoe (n=ccjoe@69.20.152.109) |
| 21:42.52 | *** join/#edev roxfan2 (n=dunno@91.182.210.233) |
| 21:50.34 | *** join/#edev ccjoe (n=ccjoe@69.20.152.109) |
| 22:10.43 | *** join/#edev brolin (n=brolin@201.232.107.129) |
| 22:34.27 | *** join/#edev ZeZu (n=null@c-98-227-56-17.hsd1.in.comcast.net) |
| 22:35.30 | ZeZu | what is a good chipset to use for SATA development ? |
| 22:39.19 | ZeZu | I'm not familiar with anything that has enough high speed serial ports to have a clue in this area ;| |
| 22:40.59 | roxfan | what kind of development? are you making sata controller, peripherals, or software? |
| 22:51.19 | *** join/#edev dieter_ (n=dieter@dslb-088-066-201-125.pools.arcor-ip.net) |
| 22:57.42 | *** join/#edev ccjoe (n=ccjoe@69.20.152.109) |
| 23:08.38 | ZeZu | roxfan: sorry, reading way too many docs for my liking ... raid controller |
| 23:09.45 | roxfan | hmm don't really know what you could use there |
| 23:11.19 | roxfan | there's marvell's sheeva cpu with pci-e and two sata channels but i think it's fit more for things like nas |
| 23:11.24 | roxfan | i hear some people use fpgas to make controllers/pci boards |
| 23:17.00 | ZeZu | yes, i was hoping to avoid that route, will take a look at sheeva though. thanks |
| 23:21.10 | roxfan | http://www.open-rd.org/ <- $150 board with all connections |
| 23:22.23 | roxfan | "The 88F6281 has two integrated 3 Gbps SATA ports. OpenRD-Base provides one |
| 23:22.23 | roxfan | internal SATA and one external SATA (eSATA) Port." |
| 23:23.13 | *** join/#edev archae0pteryx1 (n=snewman@207.47.42.130.static.nextweb.net) |
| 23:26.19 | *** join/#edev archae0pteryx2 (n=snewman@207.47.42.130.static.nextweb.net) |
| 23:29.09 | roxfan | googled up some chips |
| 23:29.10 | roxfan | http://www.pmc-sierra.com/sas-expanders-loopswitch-chips/ |