IRC log for #uclibc on 20110113

00:18.48*** join/#uclibc bkuhn (~bkuhn@fsf/director/conservancy.president.bkuhn)
00:42.31*** join/#uclibc tsukasa (~tsukasa@unaffiliated/tsukasa)
01:37.30*** join/#uclibc ynezz (ynezz@ibawizard.net)
02:23.48*** join/#uclibc khem (~khem@99-57-141-118.lightspeed.sntcca.sbcglobal.net)
02:48.44*** join/#uclibc xiangfu (~xiangfu@124.205.34.122)
03:07.15*** join/#uclibc xiangfu (~xiangfu@fidelio.qi-hardware.com)
03:13.33*** join/#uclibc Whoopie_ (Whoopie@unaffiliated/whoopie)
04:09.23*** join/#uclibc CapnKernel (~mjd@ppp59-167-188-93.static.internode.on.net)
04:21.13daliascan anyone recommend an alternative for gnu m4? busybox doesn't seem to have one and gnu m4 is one of the most horribly nonportable gnu packages i've run into :/
04:28.13*** join/#uclibc neli (micha@244-233.ftth.onsbrabantnet.nl)
06:59.32*** join/#uclibc Artemys (~quassel@stp25-1-82-225-18-227.fbx.proxad.net)
07:01.01*** join/#uclibc neli (micha@77-232.ftth.onsbrabantnet.nl)
07:12.19*** join/#uclibc neli (micha@77-232.ftth.onsbrabantnet.nl)
07:25.48*** join/#uclibc kos_tom (~thomas@humanoidz.org)
07:27.07kos_tomhello
07:34.29*** join/#uclibc xiangfu_ (~xiangfu@124.205.34.122)
07:43.22*** join/#uclibc xiangfu_ (~xiangfu@124.205.34.122)
07:44.42speakmankos_tom: morning
07:45.45*** join/#uclibc neli (micha@39-234.ftth.onsbrabantnet.nl)
08:05.55*** join/#uclibc xiangfu_ (~xiangfu@124.205.34.122)
08:06.30*** join/#uclibc neli (micha@11-235.ftth.onsbrabantnet.nl)
08:23.12*** join/#uclibc xiangfu__ (~xiangfu@fidelio.qi-hardware.com)
08:35.09*** join/#uclibc neli (micha@77-232.ftth.onsbrabantnet.nl)
09:22.54ncopamorning
09:23.00ncopawhere is __UCLIBC__ defined?
09:23.58ncopafeatures.h
09:49.38CIA-5303jacmet 07master * r877044a73823 10buildroot/ (Makefile docs/copyright.txt docs/footer.html): Update copyright info for 2011
09:59.40*** join/#uclibc CapnKernel1 (~mjd@204.37.233.220.static.exetel.com.au)
10:35.11*** join/#uclibc pt3asw (~pt3asw@net-93-65-45-116.cust.dsl.vodafone.it)
10:41.06*** join/#uclibc doc2 (~Dieter@139.12.1.252)
11:16.11*** join/#uclibc roadies (~chrislrob@stun.cmf.nrl.navy.mil)
11:37.20*** join/#uclibc gustavoz (~gustavoz@201.252.45.98)
12:01.45*** join/#uclibc neli (micha@41-239.ftth.onsbrabantnet.nl)
12:15.36*** join/#uclibc doc2 (~Dieter@139.12.1.252)
12:17.21*** join/#uclibc luneff (~yury@84.51.195.188)
12:43.40*** join/#uclibc landley (~landley@64.134.149.230)
12:59.58*** join/#uclibc xiangfu__ (~xiangfu@123.119.66.73)
13:02.07*** join/#uclibc felipec (~felipec@nat/nokia/x-porcqpdnwilkbxkf)
13:18.29*** join/#uclibc xiangfu__ (~xiangfu@fidelio.qi-hardware.com)
13:20.29*** join/#uclibc landley (~landley@64.134.149.230)
13:23.13*** join/#uclibc bkuhn (~bkuhn@fsf/director/conservancy.president.bkuhn)
14:00.40*** join/#uclibc WildPikahcu_ (~nkukard@dsl-247-41-48.telkomadsl.co.za)
14:19.10*** join/#uclibc WildPikahcu_ (~nkukard@41-133-198-70.dsl.mweb.co.za)
15:08.23CIA-5303dvlasenk 07master * r068444629fbc 10busybox/ (include/usage.src.h procps/top.c): top: add keyboard commands to --help
15:11.14*** join/#uclibc curious (curious@bwv190.internetdsl.tpnet.pl)
15:22.50*** join/#uclibc curious (curious@bwv190.internetdsl.tpnet.pl)
16:00.41gustavozJacmet: we aren't doing anything with powerpc ABIs in target/Config.in.arch, know why?
16:10.50*** join/#uclibc georgem (~georgem@mail.novatech-llc.com)
16:14.06georgemI was using uClibc 0.9.27 on a system running linux 2.6.26.8 ARM nommu and everything was working file. I upgraded linux to 2.6.33.7 and now after my kernel prints "Freeing init memory: 72K" fget_light causes a data abort exception. Anyone think this could be because my uClibc is too old?
16:15.13Jacmetgustavoz: no, not really - It seems to have bitrotten
16:15.43gustavozJacmet: hmm i'll give it a shot of re-enabling it
16:15.51Jacmetgustavoz: ok, thanks
16:16.12gustavozJacmet: otherwise we don't do SPE and that would be good for the thing i've got here :)
16:16.25Jacmetgustavoz: I haven't noticed any issue with not defining it, but don't have altivec/spe/.. hw
16:16.41gustavozJacmet: i lack softfloat ppc to test the changes i do though
16:16.56Jacmetgustavoz: I have softfloat here
16:17.07gustavozJacmet: for now i've got e300c3 (classic) & e500v2 (spe or classic) and both are hard
16:17.29gustavozJacmet: e500* is by definition hard IIRC
16:17.59Jacmetgustavoz: I have some old 405 hw somewhere I could dig up
16:18.22gustavozJacmet: cool, i can do qemu but i think it's somewhat lax with respect to emulation in the cpu department
16:18.41gustavoz(allows things that wouldn't work on real hardware)
16:18.43Jacmetgustavoz: and otherwise I have e300 hw as well
16:19.08gustavozJacmet: wrt hard fp we could do some extra work too since some are single and others double precision FP
16:19.48Jacmetgustavoz: how is that supported in gcc? do you fall back to kernel emulation for double stuff?
16:19.49gustavozJacmet: for arm i'm somewhat tempted to go for ISA rather than core
16:20.36gustavozwe've got a lot of redundancy there
16:22.16Jacmetgustavoz: true
16:46.22gustavozkhem: hi, around? i'm seeing this initially with git http://pastebin.com/gsHhDTVt and when i patch-fix it i get to this http://pastebin.com/M3SAhuan
16:47.07georgemIs anyone here running uClibc on an ARM with NOMMU Linux kernel 2.6.27.x or new?
16:56.09pt3aswi get the following error message when installing uclibc headers: http://pastebin.com/TgMRkNyi
16:56.40pt3aswi don't know if it is because of gcc or uclibc
16:57.41pt3aswthe target is arm cortex-a8
17:10.40*** join/#uclibc mnt_real (~mnt_real@174.91.218.153)
17:25.15*** join/#uclibc calculus (~calculus@gentoo/user/calculus)
17:35.14*** join/#uclibc y_morin (~ymorin@ARennes-256-1-53-25.w90-32.abo.wanadoo.fr)
17:35.15speakmanqt 4.7.1 build fails... hmm
17:36.57speakmantrying again with sourcery 2010q1
17:43.11*** join/#uclibc mnt_real (~mnt_real@174.93.215.254)
17:57.38speakmanyes, 2010q1 did the trick
18:09.53Jacmetspeakman: if what config does it fail for you?
18:22.41speakmanJacmet: not sure I understand. What in Qt caused the error? I'm still running 2010.11 btw.
18:22.54gustavozspeakman: which toolchain for instance?
18:24.26speakmanoh I forgot to tell, it's CodeSourcery ARM 2010.09
18:24.29speakmanthat fails
18:27.42Jacmetspeakman: sorry, meant 'with what config did it fail?'
18:28.04Jacmetspeakman: ok, I built qt with the internal toolchain recently without problems
18:28.16speakmanJacmet: oh, it's a home rolled with neon enabled
18:28.44speakmanJacmet: CS 2010q1 didn't fail though
18:30.22pt3aswthe cc1: error: unrecognized command line option "-mbig-endian" is still persecuting me..
18:30.38pt3aswi'm new to the arm world
18:32.57pt3aswtarget triplet is arm-qemu-linux-gnueabi, but i'm not sure it is right
18:33.59pt3aswthis is because i believe "gnu" is for glibc, while "uclibc" is for... uclibc
18:34.12speakmanpt3asw: which defconfig?
18:34.44pt3aswi haven't used defconfig. i used oldconfig
18:34.46y_morinpt3asw: *-gnueabi is for glibc. And *-uclibcgnueabi is for uClibc.
18:34.57y_morinI mean, for EABI.
18:35.14y_morinpt3asw: and for OABI, resp *-gnu and *-uclibc.
18:35.35pt3aswy_morin, where did you find this info?
18:37.15pt3asw(by the way, in the meantime i also tried with ct-ng, and it seems to work - arm-unknown-linux-uclibcgnueabi)
18:39.28pt3aswalso, CFLAGS_FOR_TARGET="-Os -mcpu=cortex-a8 -pipe"
18:42.28y_morinpt3asw: look in a config.sub from /recent/ autstuff.
18:42.45y_morins/autstuff/autostuff/
18:43.30y_morinpt3asw: and I learnt it the hard way from my beginings at building toolchains.
18:45.51pt3aswanyway, same error, even with arm-qemu-linux-uclibcgnueabi
18:47.37y_morinpt3asw: what are you doing that gives you this 'cc1 error' ?
18:48.27pt3aswi have built the gcc cross compiler without the headers. i installed the linux headers. now i'd like to install the uclibc headers, so i can then use them to build a better gcc
18:49.21y_morinpt3asw: what uClibc version?
18:49.32pt3aswlatest 0.9.32-rc1
18:50.01y_morinpt3asw: NPTL or LinuxThreads?
18:50.17speakmanany particular reason you're aiming for big endian?
18:50.23pt3aswLINUXTHREADS_NEW=y
18:50.58pt3aswspeakman, i haven't choose bigendian by myself
18:51.20y_morinpt3asw: there's been a fix recently committed to git wrt respect headers generation. Lemme dig the archives to get you the pointer...
18:51.27pt3aswCONFIG_ARM_EABI=y
18:52.01y_morinpt3asw: http://git.buildroot.org/uClibc/commit/?id=8764f2e75c875757529aa81e5cd96e17e6f6e5e4
18:53.31pt3aswok, so i either patch 0.9.32-rc1 that way or i use git..
18:53.51y_morinpt3asw: try patching first.
18:55.07y_morinpt3asw: are you building out-of-tree?
18:55.30pt3aswby "tree" you mean the sources tree?
18:55.48y_morinpt3asw: yes, building in a directory that is not the source tree.
18:56.19pt3aswyes with gmp, mpfr, mpc, binutils and gcc; no for linux, uclibc and busybox
18:57.02y_morinpt3asw: I meant for uClibc. So you will not need: http://git.buildroot.org/uClibc/commit/?id=fc91a4155d265a189ae6aa50b63640d8da7b3609
18:59.50pt3aswi just patched uclibc the way you said before. it works now! even in the tree
19:00.15y_morinpt3asw: good! :-)
19:00.17pt3aswthank you y_morin!
19:01.20pt3aswfor the future, i don't know if to use the latest released tarballs for the toolchain packages, or the various git/hg/svn versions
19:01.51pt3aswthe latter requires more knowledge by my side, but i'm happy to learn
19:02.38y_morinpt3asw: depends on what you want. If you want production-ready toolchain, go for the released tarballs. If you want a bleeding-edge, ready to break toolchain and/or test various components, then go for the repos heads.
19:04.35pt3aswwell. i cross compile just for hobby. also, i'd like to help when i'll be more knowledgeable. so i think i'll try the repository versions :)
19:06.57y_morinpt3asw: I would recommend that you look at the way crosstool-NG does its thing. There are a hell-lot of corner cases and weird stuff goin' on in toolchain land...
19:07.33y_morinG2G, BBL.
19:07.35y_morinis going for dinner! :-)
19:17.14*** join/#uclibc PortaLu (~LucretiaL@5ac2a2b2.bb.sky.com)
19:34.50*** join/#uclibc Montreal1 (~goran@modemcable134.206-131-66.mc.videotron.ca)
19:35.05Montreal1hi folks
19:37.20Montreal1i don't know if this is the right place to ask...but i have some weird prob with my Motorola IPTV STB box
19:37.33Montreal1this box running on busybox
19:38.18Montreal1and even if the original stream is 4:3 ( 720x576 ) i still get wide pic...16:9 or so
19:38.28Montreal1i have access via terminal only
19:38.33Montreal1what can i do?
19:38.40Montreal1to fix this?
19:42.22Montreal1can i access somehow in stings?
20:07.43xMffthis is not related to busybox at all
20:09.42*** join/#uclibc WildPikachu (~nkukard@about/linux/staff/wildpikachu)
20:12.28Montreal1but same livestreams i can see it straith on my WDTV Live
20:12.48Montreal1wdtv live show them right
20:13.37xMffbusbox is a shell for the os, video streaming happens, display management and conversion happens elsewhere, most likely in some proprietary vendor software
20:13.55*** join/#uclibc trem (~trem@mol92-1-81-57-136-23.fbx.proxad.net)
20:14.55Montreal1i seee, but coz wdtv show'em strait, so i thought if someth is wrong...will be on both box-es
20:15.11Montreal1not just on Motorola
20:38.32y_morinELC `11 CFP is on-line: http://events.linuxfoundation.org/events/embedded-linux-conference/cfp
20:46.49*** join/#uclibc yorttsed (52e20d9c@gateway/web/freenode/ip.82.226.13.156)
20:58.32CIA-5303rep.dot.nop 07master * r049262dd838e 10uClibc-website/developing.html: document personal repo
21:13.29*** join/#uclibc tiny (~ivob@BSN-77-54-208.dsl.siol.net)
21:13.33*** join/#uclibc tiny (~ivob@unaffiliated/tiny)
21:29.07CIA-5303rep.dot.nop 07master * r0d91b7a45322 10uClibc-website/developing.html: tweak misplaced closing paragraph
22:21.37*** join/#uclibc calculus (~calculus@gentoo/user/calculus)
23:21.30*** join/#uclibc bkuhn (~bkuhn@fsf/director/conservancy.president.bkuhn)
23:33.26*** join/#uclibc calculus (~calculus@gentoo/user/calculus)
23:46.24*** join/#uclibc KaiForce (~chatzilla@adsl-70-228-82-202.dsl.akrnoh.ameritech.net)
23:57.54*** join/#uclibc calculus (~calculus@gentoo/user/calculus)

Generated by irclog2html.pl Modified by Tim Riker to work with infobot.