01:14.01 | *** join/#uclibc ChanServ (ChanServ@services.) |
01:14.01 | *** join/#uclibc andersee (~andersee@codepoet.org) |
01:14.01 | *** join/#uclibc waldi (~waldi@bblank.thinkmo.de) |
01:14.01 | *** join/#uclibc kergoth`bbl (~kergoth@dsl081-228-056.chi1.dsl.speakeasy.net) |
01:14.01 | *** join/#uclibc mjn3 (~mjn3@cdm-68-226-136-114.laft.cox-internet.com) |
01:14.01 | *** mode/#uclibc [+o ChanServ] by zelazny.freenode.net |
01:44.47 | *** join/#uclibc Flimzy (~jonhall@ip68-102-11-171.ks.ok.cox.net) |
01:45.37 | Flimzy | When compiling uclibc 0.2.21, I get the following error: + ln -fs /home/jonhall/ob/uClibc-0.9.21/kernel-source-2.4.20/include/asm-i386 include/asm |
01:45.37 | Flimzy | + unset -x |
01:45.37 | Flimzy | ./extra/scripts/fix_includes.sh: line 154: unset: -x: invalid option |
01:46.01 | Flimzy | I'm building using the Debian facilities... |
01:46.06 | Flimzy | dpkg-buildpackage -rfakeroot -b -us -uc |
01:46.24 | Flimzy | Any ideas what the specific problem is? |
01:50.14 | Flimzy | what is 'unset -x' supposed to do? |
01:50.24 | Flimzy | -x does not appear to be supported in my shell... |
02:10.48 | mjn3 | Flimzy: i think that was fixed in cvs last week |
02:12.49 | andersee | Flimzy: as mjn3 said... fixed in the latest from cvs |
02:14.58 | Flimzy | ah, okay. Thanks |
02:16.22 | Flimzy | Are there CVS snapshot tarballs on the FTP site? |
02:16.55 | Flimzy | n/m I answered my own question |
05:39.21 | *** join/#uclibc ChanServ (ChanServ@services.) |
05:39.21 | *** mode/#uclibc [+o ChanServ] by zelazny.freenode.net |
07:13.14 | *** join/#uclibc tahorg (~loic@perax.net1.nerim.net) |
13:24.13 | *** join/#uclibc dsmith (~dsmith@mail.actron.com) |
14:28.41 | *** join/#uclibc rfs (~ralphs@mail.blueairnetworks.com) |
18:27.17 | *** join/#uclibc dsmith (~dsmith@mail.actron.com) |
19:18.54 | *** join/#uclibc andersee (~andersee@codepoet.org) |
19:18.59 | mjn3 | andersee: morning |
19:19.11 | andersee | mjn3: morning |
19:19.29 | kergoth | hey mjn3, andersee |
19:19.41 | mjn3 | kergoth: hey |
19:19.57 | kergoth | whats goin on? |
19:20.18 | andersee | kergoth: just waking up |
19:20.40 | andersee | kergoth: was up late hacking together yet-more .debs |
19:20.51 | kergoth | ah, cool |
19:21.15 | kergoth | playing around with zeroconf and multicast dns a bit myself |
19:22.19 | mjn3 | i'm still working on the new stdio, and sending the dietlibc list a link to a publicly available draft of the C99 standard (since they could really use it) |
19:22.29 | mjn3 | :-) |
19:22.52 | kergoth | hehe. how subtle of you |
19:22.52 | kergoth | :) |
19:25.27 | mjn3 | kergoth: subtlety is lost on many... |
19:26.50 | kergoth | true that. |
19:27.46 | mjn3 | kergoth: felix got on my case for not reporting some bugs a while back. so i've been trying to be helpful while still avoiding doing their work for them |
19:28.33 | mjn3 | kergoth: part of that process is elevating their awareness of how buggy their code is, and emphasizing the need for testing and code review |
19:29.23 | kergoth | makes sense. good luck |
19:29.23 | kergoth | heh |
19:30.45 | andersee | mjn3: maybe they'll instead decide to give up |
19:31.06 | mjn3 | andersee: either that or ban me from the mailing list :-) |
19:31.14 | andersee | mjn3: or (heaven forbid) decide to copy working code |
19:31.27 | mjn3 | i doubt that will ever happen |
19:32.46 | andersee | agreed |
19:39.23 | dsmith | andersee: What kinda .debs? |
19:40.30 | andersee | dsmith: the kind you get when you have a uClibc only system, and you run 'apt-get source -b <package>' to compile a package from debian woody and then run 'dpkg -i <package>' to install it |
19:41.09 | andersee | dsmith: I've been working on bootstrapping such a system for a couple of weeks now (I think) |
19:41.49 | andersee | dsmith: just as an fyi -- bootstrapping such a system is a _lot_ of work |
19:43.38 | dsmith | Yes! |
19:44.08 | dsmith | Bringing up a woody nftroot system was relativly painless. |
19:44.21 | dsmith | andersee: What arch? |
19:44.54 | andersee | dsmith: x86 thus far |
19:45.19 | andersee | dsmith: I'm not sure if I will do mipsel |
19:45.21 | dsmith | Oh well. |
19:45.34 | andersee | dsmith: the bootstrapping process really needs to be done natively |
19:46.04 | andersee | dsmith: and my mips box is so slow, and has so little memory, it would probably take me a year and half |
19:46.06 | andersee | :-( |
19:46.19 | dsmith | Yes. |
19:48.05 | andersee | dsmith: of course, if someone were to, say, send me a really fast mips box with plenty of ram |
19:55.24 | andersee | I wonder if a company like broadcom would be willing to sponsor such a thing? |
19:55.50 | andersee | they seem to be selling a lot of mips cpus to the various wireless ap folks |
19:56.23 | andersee | Of course, they do so under a "Broadcom owns all theis open source so you can sell it all you want" illegal license |
19:56.30 | andersee | ;-) |
20:17.06 | *** join/#uclibc rfs (~ralphs@mail.blueairnetworks.com) |
20:18.45 | andersee | rfs: pong |
20:19.11 | rfs | andersee: long time no hear :0 |
20:19.18 | andersee | rfs: :-) |
20:19.28 | andersee | rfs: I've been busy I suppose |
20:19.35 | rfs | andersee: i was going to report a bug in 0.9.1 but I see it was already fixed in CVS :( |
20:20.00 | andersee | rfs: and there has not been much netwinder traffic that seemed worthy of response |
20:20.02 | andersee | :-) |
20:20.14 | rfs | true, things are pretty dead right now... |
20:20.25 | andersee | rfs: yeah, there were several little gotchas in the last release |
20:20.38 | andersee | rfs: we really should push out another release |
20:20.48 | rfs | andersee: or at least a new tarball :) |
20:20.56 | andersee | heh |
20:21.47 | rfs | I got caught on the $(subst -g,,$(CFLAGS)) |
20:21.55 | rfs | took a while to spot it :) |
20:22.57 | andersee | ugh |
20:22.59 | andersee | yeah |
20:23.07 | andersee | That one was sortof nasty |
20:23.47 | rfs | yeah... i fixed it using patsubst instead of subst, but your way is probably more portable ;) |
20:24.15 | rfs | I did manage to build a working uclibc and busybox for my winder |
20:24.30 | rfs | static exe was 270kB though :( |
20:24.48 | rfs | still, got my kernel down to 570kB, so the two fit into 1M flash :0 |
20:26.14 | andersee | busybox can get a bit large, depending on what you enable |
20:26.19 | andersee | busybox version? |
20:26.33 | rfs | latest released |
20:26.40 | rfs | -pre3 or somesuch |
20:26.45 | andersee | rfs: hmm, so you are replacing the netwinder firmware? |
20:26.50 | andersee | rfs: ok cool |
20:26.57 | andersee | neeto |
20:27.06 | andersee | indeed |
20:27.44 | andersee | And honestly, without tab completion and w/o even a choice of kernels to load, the firmware does leave a bit to be desired |
20:27.58 | rfs | yeah, multiple profiles would be nice |
20:29.19 | dsmith | I always wanted to get a netwinder. |
20:29.30 | andersee | I suppose I have gotten a bit spoiled from using grub |
20:29.53 | andersee | rfs: so something grub like would be awfully cool |
20:29.54 | rfs | andersee: i'm not entirely happy with grub either... |
20:30.14 | rfs | some features are cool, but its not intuitive enough |
20:30.22 | andersee | rfs: oh it has its warts (what were they thinking with the (hdX,Y) scheme?) |
20:30.23 | rfs | and did they have to invent yet another way of naming devices? |
20:30.54 | rfs | i think it came from sun bootloader or somesuch. |
20:31.11 | andersee | rfs: but I do really like reading files direct from your partitions with tab completion |
20:31.48 | rfs | andersee: yes, and its easy to do as well - for some subset of filesystems anyways. |
20:31.52 | andersee | rfs: and having a menu of kernels to boot proved awfully handy when you hose up your kernel |
20:32.10 | rfs | yup, or have moved disk drives between units and dont' remember what is on it ;) |
20:32.22 | andersee | rfs: with the netwinder, when I build a broken kernel it is quite painful |
20:33.05 | rfs | ah, i move drives around a lot, and some of them have different partition setups, etc... |
20:38.33 | andersee | back later |
20:47.03 | rfs | cu Erik... |
20:57.32 | *** part/#uclibc rfs (~ralphs@mail.blueairnetworks.com) |