irclog2html for #uclibc on 20100203

00:08.44*** join/#uclibc vapier (UserBah@wh0rd.org)
00:30.44*** join/#uclibc dileX (~sd@p5B2ECA41.dip.t-dialin.net)
01:15.07*** join/#uclibc austinf (~austinf@74.92.231.110)
01:16.35*** join/#uclibc austinf (~austinf@74.92.231.110)
02:05.53*** join/#uclibc DuratarskeyK (~wedge@95-86-249-203.pppoe.yaroslavl.ru)
02:46.56*** join/#uclibc dileX_ (~sd@p5B2EBD94.dip.t-dialin.net)
03:50.27dougmenckenhi, may I request md5sum --check (longoption for -c)?
03:55.16dougmenckenalso, I found a way to get some 'history' support: alias history="cat ${HOME}/.ash_history"
04:07.31dougmenckenbut how to make that alias to work on each login? ~/.ashrc doesn't help, it isn't executed
04:56.21opellomaybe .profile?
05:07.34dougmenckenno, .profile isn't executed too :( or maybe shall I +x it?
05:22.18opellolooks like it should be based on what is in git
05:23.16opellonot sure what else to think :/
06:10.35dougmenckenehm, to print line numbers, cat -n isn't supported, nl isn't in busybox
06:47.35*** join/#uclibc y_morin (~ymorin@ARennes-252-1-40-118.w83-195.abo.wanadoo.fr)
06:48.34*** join/#uclibc y_morin (~ymorin@ARennes-252-1-40-118.w83-195.abo.wanadoo.fr)
08:20.39*** join/#uclibc 18VAAASZ0 (~hw@p578b3905.dip0.t-ipconnect.de)
08:31.19*** join/#uclibc ncopa (~ncopa@245.39.189.109.customer.cdi.no)
09:27.02blindvt`dougmencken, egrep -B5 ^read_profile ../busybox/shell/ash.c
09:44.31*** part/#uclibc dougmencken (~Douglas@93.123.156.139)
09:44.38*** join/#uclibc dougmencken (~Douglas@93.123.156.139)
09:59.24*** join/#uclibc ncopa (~ncopa@245.39.189.109.customer.cdi.no)
10:03.40*** join/#uclibc dileX (~sd@vpn-eu1.unidsl.de)
10:07.11dougmenckenblindvt`, even with egrep -B10 -A40 ^read_profile ${BUSYBOX-GIT}/shell/ash.c I can't get it: http://pastebin.com/m31e0090d; can you please explain it to me
10:08.35dougmenckenalso, sed/awk profi can you help me to add line numbers to output?
10:27.54*** join/#uclibc Ulfalizer (~ulf@pat.se.opera.com)
10:38.13*** join/#uclibc matteo (~matteo@openwrt/developer/matteo)
10:39.15dougmenckenalias history='(n=0; while read x; do n=$((n+1)); echo "$n $x"; done) < ${HOME}/.ash_history'
10:41.03Ulfalizershould "generic x86 device support" be switched on whenever you're developing for an x86 system?
10:44.39CIA-4903vda.linux 07master * r80361b7c7421 10uClibc/libc/sysdeps/linux/powerpc/bits/sysdep.h: powerpc/bits/sysdep.h: move confusingly placed #undef
10:52.31Ulfalizera simple app that does write(1, "foo", 3) causes a segfault both on my dev. and target systems (both i686, 2.6.31 kernels). compiling with the native system compiler, it works fine on the dev. system.
10:56.54Ulfalizercompiled with -static. apps that don't call any uclibc functions seem to work fine.
11:06.52*** join/#uclibc dileX_ (~sd@p5B2EBD94.dip.t-dialin.net)
11:07.11Ulfalizeri tried compiling a version without a shared libgcc and cross-compiling without -static, but that just causes an immediate segfault, even for a program that does not call any uclibc functions
11:09.20dougmenckenUlfalizer, looks like broken cross-toolch
11:11.14Ulfalizerdougmencken: any ideas why? i used pretty default buildroot settings on a 2.6.31-17 ubuntu system, with gcc 4.4.1.
11:12.25blindvt`dougmencken, /etc/profile and ~/.profile are sourced, not executed. Think . /etc/profile
11:16.50blindvt`dougmencken, $ echo -e "one\ntwo\nthree" > /tmp/hist
11:16.58blindvt`$ ../busybox/busybox ash
11:16.58blindvt`$ history
11:16.58blindvt`ash: history: not found
11:16.58blindvt`$ alias history='i=0;cat /tmp/hist | while read l;do let i++; printf "%d %s\n" "
11:16.58blindvt`$i" "$l";done'
11:16.58blindvt`$ history
11:17.00blindvt`1 one
11:17.02blindvt`2 two
11:17.04blindvt`3 three
11:17.06blindvt`$
11:17.42CIA-4903vda.linux 07master * r557deb1014c1 10busybox/libbb/bb_askpass.c: libbb: fix bb_ask() to operate on correct fd
11:18.31blindvt`dougmencken, so just putting that alias into ~/.profile should make it available to every instance of ash that you run after having added it there.
11:21.48dougmenckenblindvt`, so why http://pastebin.com/m3c46195a << it doesn't work!
11:23.49dougmenckenthe same with "mv /root/.profile ./root/", "cat ./root/.profile", "./chroot-here.sh --noscreen --nomount", "history" --> /bin/sh: history: not found
11:24.47Ulfalizerweird.. using 'puts' from stdio works for outputing a string, but not write from unistd
11:25.14Ulfalizerlatter causes segfault
11:25.21Ulfalizerbut works fine with native compiler..
11:25.38dougmenckensegfault == linking with worng libs
11:25.45dougmenckentry to check with 'ldd'
11:29.11blindvt`Ulfalizer, which arch? which versions?
11:29.19Ulfalizerlibc.so.0 => not found (0x00000000)     /lib/ld-uClibc.so.0 => /lib/ld-uClibc.so.0 (0x00000000)    are the dependencies for both executables (had them statically compiled when i ran them)
11:31.09Ulfalizerblindvt`: i686, 2.6.31-17, bog-standard PC. the target is a virtual machine with similar specs.
11:31.27Ulfalizerversions coming..
11:32.12Ulfalizerbuildroot-2009.11, gcc 4.4.1 for building the toolchain
11:32.29Ulfalizerubuntu 9.10 system
11:32.59Ulfalizerstandard buildroot conf
11:33.15Ulfalizer*statically linked
11:36.10blindvt`Ulfalizer, works fine for me: $ file gethostid
11:36.10blindvt`gethostid: ELF 32-bit LSB executable, Intel 80386, version 1 (SYSV), dynamically linked (uses shared libs), not stripped
11:36.30blindvt`$ utils/ldd.host gethostid
11:36.30blindvt`ldd: can't open cache '/usr/i386-linux-uclibc/etc/ld.so.cache'
11:36.30blindvt`checking sub-depends for '/there/src/uClibc/lib/libc.so.0'
11:36.30blindvt`libc.so.0 => /there/src/uClibc/lib/libc.so.0 (0x00000000)
11:36.30blindvt`/usr/i386-linux-uclibc/lib/ld-uClibc.so.0 => /usr/i386-linux-uclibc/lib/ld-uClibc.so.0 (0x00000000)
11:36.37blindvt`$ ./gethostid
11:36.37blindvt`hostid=8323329
11:39.14Ulfalizerheh, if i put a call to 'puts' before the 'write', it works
11:39.33Ulfalizermaybe it could be related to this: http://lists.uclibc.org/pipermail/uclibc/2008-December/041623.html
11:43.11blindvt`Ulfalizer, interresting, please file a bug (if there isn't one already). TIA!
11:43.25UlfalizerTIA?
11:43.29Ulfalizerwill do :)
11:45.51blindvt`~TIA
11:45.52ibotconsole/xterm IDE for GNAT. URL: http://www.vaxxine.com/pegasoft/tia.html
11:45.57blindvt`ugh
11:46.01blindvt`~dict TIA
11:54.34*** join/#uclibc fabled (~fabled@letku109.adsl.netsonic.fi)
12:11.57*** join/#uclibc gustavoz (~gustavoz@host195.200-82-38.telecom.net.ar)
12:33.00*** join/#uclibc matteo` (~matteo@openwrt/developer/matteo)
13:30.24*** join/#uclibc curious (curious@bwv190.internetdsl.tpnet.pl)
13:36.15*** join/#uclibc gustavoz (~gustavoz@host195.200-82-38.telecom.net.ar)
14:15.43*** join/#uclibc matteo (~matteo@openwrt/developer/matteo)
14:23.18*** join/#uclibc wberrier (~wberrier@65-100-208-11.slkc.qwest.net)
15:18.39CIA-4903jacmet 07master * r9277cb760a4c 10buildroot/package/mtd/ (12 files in 3 dirs): mtd-utils: bump version, get rid of obsolete git version
15:18.41CIA-4903jacmet 07master * r6d34b1d0589c 10buildroot/package/mtd/mtd.mk: mtd: host support cleanup
15:18.44CIA-4903jacmet 07master * rf85ed72c7092 10buildroot/package/mtd/ (Config.in mtd.mk): mtd: zlib/lzo is only needed for mkfs.jffs2
15:18.45CIA-4903jacmet 07master * rd6e6a2e2e26c 10buildroot/package/mtd/Config.in: mtd: tweak default values
15:18.46CIA-4903jacmet 07master * rcb9da8512e94 10buildroot/target/jffs2/Config.in: target/jffs2: Don't select BR2_PACKAGE_MTD
15:18.48CIA-4903jacmet 07master * r653f7adb10ef 10buildroot/package/mtd/Config.in: mtd: don't put mtd options in a seperate submenu
15:32.05CIA-4903filippo.arcidiacono 07master * r23e2b7118e41 10uClibc/libc/stdlib/l64a.c: libc: Fix l64a to return the correct buffer pointer
15:32.05CIA-4903filippo.arcidiacono 07nptl * r8a32a3231790 10uClibc/libc/stdlib/l64a.c: libc: Fix l64a to return the correct buffer pointer
15:46.18*** join/#uclibc matteo` (~matteo@openwrt/developer/matteo)
17:22.23*** join/#uclibc ibot (ibot@rikers.org)
17:22.23*** topic/#uclibc is discussion of uClibc and Busybox | uClibc 0.9.30.2 was released 13 Jan 2010 | busybox 1.15.3 was released 12 Dec 2009 (unstable busybox 1.16.git) | buildroot 2009.11 was released 1 December 2009 | For general setup issues try #elinux or #edevreminds me that
18:05.18*** join/#uclibc y_morin (~ymorin@ARennes-252-1-40-118.w83-195.abo.wanadoo.fr)
18:45.25*** join/#uclibc matteo` (~matteo@openwrt/developer/matteo)
18:54.00busybot[New Bug] https://bugs.uclibc.org/1033 maj, P4, uClibc, denblomstertid@gmail.com->unassigned@uclibc.org, NEW, pending, i686 + static link + not calling any stdio functions = segfault
19:03.31gustavozJacmet: hi, around? got a preliminary new Config.in with some order in it if you wanna see
19:05.48*** join/#uclibc wbx (~wbx@e181230215.adsl.alicedsl.de)
19:05.59wbxhi.
19:07.03CIA-4903rep.dot.nop 07master * rf588c2d4bb2d 10uClibc/test/inet/gethostid.c: add gethostid() testcase
19:07.30wbxi am trying to compile uClibc 0.9.30.2 for mips64 (lemote notebook)
19:07.47wbxbut I get following error when linking:
19:07.49wbxreaddir64.c:(.text+0x40): undefined reference to `__getdents64'
19:07.59wbxmips64el-linux-uclibc/bin/ld: lib/libuClibc-0.9.30.2.so: hidden symbol `__getdents64' isn't defined
19:08.07*** join/#uclibc neale (~neale@proxyout.lanl.gov)
19:08.27nealeI'm having trouble getting buildroot to use my kernel config.  Is this the right channel?
19:09.23wbxuClibc 0.9.30.1 at least compiles fine
19:11.53khemwbx: getdents64 is not available for mips64 what kernel do you use
19:11.58khemfor kernel-headers
19:12.08wbx2.6.32
19:12.28khemhmm can you make sure they wired getdents64 syscall may be
19:14.07wbxwhat does that mean?
19:15.51khemwbx: never mind. Can you try this patch http://www.mail-archive.com/uclibc@uclibc.org/msg04960.html
19:15.56*** join/#uclibc PortaLu (~LucretiaL@5ac95b0c.bb.sky.com)
19:15.58khemand see if it helps
19:16.50wbxah, thanks. my search skills seems to be bad :}
19:17.15kheminfact try the redone one http://www.mail-archive.com/uclibc@uclibc.org/msg05005.html
19:17.23khemand report back
19:17.57khemneale: yes buildroot folks hang around here
19:18.06khemneale: I personally dont use buildroot
19:18.15nealeactually I think I just figured it out :)
19:18.28nealemy mistake was reading the wiki page instead of the makefiles.
19:19.08CIA-4903rep.dot.nop 07master * r8a227f31655d 10uClibc/libc/misc/internals/__uClibc_main.c: __uClibc_main: use __pagesize to protect against recursion
19:19.28blindvt`interresting bug, that 1033 :P
19:23.43wbxkhem: thanks. this fixes it for me.
19:23.50khemwbx: ok
19:24.24khemblindvt`: I think I have a hunch I know the problem re 1033
19:24.26wbxnow I will try to bootup the stuff, last time I had many issues with the pagesize (64kb). but i have seen a lot of changes which might fix it.
19:26.05khemwbx: I dont think uclibc will work for 64kb that easily
19:26.18khemblindvt:     if (likely(_stdio_init != NULL))
19:26.18khem<PROTECTED>
19:26.42blindvt`khem, shoot. My theory (which is essentially size-neutral) goes like this: uClibc.org/~aldot/uClibc/bug1033.patch
19:26.46khemin __uClibc_main.c
19:27.36blindvt`khem, the PITA is that __errno_location et al need the same crap, which isn't too pretty, so i'm all ears for better ideas :)
19:27.59blindvt`bbiab
19:28.44wbxkhem: so the changes to ld.so regarding pagesize are not enough?
19:31.14khemwbx: should be
19:31.43khemblindvt`: your patch is ok
19:34.15khemblindvt`: towards the end void (*foo)(void) = fini_table[0];
19:35.18khemI think ini and fini should be together
19:35.27khemfor comprehension
20:01.21*** join/#uclibc Dj-Death (~djdeath@potipota.net)
20:01.34CIA-4903jacmet 07master * r67a284d82040 10buildroot/ (3 files in 2 dirs): nbd: add server patches missing from 6aac003
20:02.16blindvtkhem, yes, i had moved it from _atexit to __uClibc_fini directly but then put it back. I don't see any reason why it should be in _atexit, do you?
20:04.56khemyeah
20:05.25khemI think __uClibc_fini sounds more related place
20:17.12blindvthttp://uClibc.org/~aldot/uClibc/0001-powerpc-Add-TLS-and-NPTL-support.patch
20:17.43CIA-4903austinf 07master * r5d5b6fe5f898 10uClibc/libc/sysdeps/linux/sparc/sigaction.c: Unbreak build for sparc on some config's
20:18.07blindvtif anyone with a real ppc box is bored, give it a whirl
20:18.28blindvtaustinf, please push it to the 0_9_30 branch too. TIA
20:23.07khemblindvt: I tried to get qemu working for me
20:23.15khembut I could not
20:23.27khemit seems I needed a ppc host to run ppc emu
20:24.45dougmenckenppc qemu almost useless at all (except for x86 targets)
20:29.49CIA-4903austinf 070_9_30 * r7b9641705369 10uClibc/libc/sysdeps/linux/sparc/sigaction.c: Unbreak build for sparc on some config's
20:31.34khemdougmencken: you mean qemu running on ppc ?
20:31.55khemblindvt: How did you run the qemu for ppc
20:38.57*** join/#uclibc trem (~trem@AToulouse-256-1-146-72.w90-45.abo.wanadoo.fr)
20:40.53*** join/#uclibc KaiForce (~chatzilla@adsl-70-228-89-235.dsl.akrnoh.ameritech.net)
21:50.41blindvtkhem, i seem to have used: qemu-system-ppc -kernel vmlinuxz -initrd initrd.xz -hda ../PPC.img -net nic,vlan=0 -net tap,vlan=0,ifname=tap0,script=/etc/qemu-brup -append "root=/dev/hdc4"
21:51.27blindvtdougmencken, worked fine for me, but ymmv
22:07.26blindvtkhem, austinf, i'd be interrested if ppc broke somebody else (i don't have another compiler at hand)
22:17.46austinfwell I applied your patch and i386 nptl still builds
22:17.55austinfso that's a good sign
22:20.39blindvtgreat.
22:21.14blindvtaustinf, i'll have to separate the testsuite/general/ppc-tls/ppc-nptl parts from each other
22:49.35nealeokay, can anyone help me set up a new board within buildroot?
22:49.46nealeI can't seem to get BR2_BOARD_NAME to stick
23:11.14CIA-4903dan 07master * rf111b67100d1 10busybox/editors/diff.c: diff: make code more portable

Generated by irclog2html.pl by Jeff Waugh - find it at freshmeat.net! Modified by Tim Riker to work with infobot logs, split per channel, etc.