| 00:43.31 | *** join/#uclibc rigidus (n=rigidus@host81-151-46-27.range81-151.btcentralplus.com) |
| 00:48.28 | *** join/#uclibc jaro (n=ja4aro@195.47.42.18) |
| 02:51.37 | *** join/#uclibc paravoid (i=paravoid@void.photonics.ece.ntua.gr) |
| 02:53.50 | *** join/#uclibc andersee (n=andersee@codepoet.org) |
| 02:55.07 | paravoid | andersee: finally! |
| 03:04.47 | CIA-10 | 03vapier * r11151 10uClibc/libc/sysdeps/linux/arm/bits/atomicity.h: import thumb support from jbowler in Bug 385 |
| 03:05.22 | paravoid | andersee: ping? |
| 03:06.20 | CIA-10 | 03vapier * r11152 10uClibc/libpthread/linuxthreads/sysdeps/arm/pt-machine.h: import thumb support from jbowler in Bug 385 |
| 03:14.24 | *** join/#uclibc __cow (n=cow@M866P030.adsl.highway.telekom.at) |
| 03:15.01 | CIA-10 | 03vapier * r11153 10uClibc/libc/sysdeps/linux/arm/ (Makefile ioperm.c iopl.c): rip out iopl from ioperm.c so it will override the common iopl.c |
| 03:15.57 | CIA-10 | 03vapier * r11154 10uClibc/libc/sysdeps/linux/common/iopl.c: arm has its own iopl.c now so we dont need ugly #ifdef check for arm |
| 03:18.04 | vapier | andersee: hey, for ldd.c with the EM_MACHINE and ELFCLASS, is it ok to rely on the linux headers setting those macros ? |
| 03:18.18 | vapier | it'd save us from having to include all those ugly #if defined(__arm__) ... stuff |
| 03:18.54 | vapier | asm/elf.h define ELF_ARCH and ELF_CLASS for us ... |
| 03:20.30 | vapier | linux-2.4 and linux-2.6 appear to define that properly |
| 03:31.08 | CIA-10 | 03vapier * r11155 10uClibc/libm/math_private.h: import thumb support from jbowler in Bug 385 |
| 03:39.47 | vapier | mjn3: any reason we need to make sure DT_DEBUG support is not compiled in for mips targets in ldso/dl-elf.h:__dl_parse_dynamic_info() ? |
| 03:39.58 | vapier | err ldso/include/dl-elf.h |
| 03:52.00 | CIA-10 | 03vapier * r11156 10uClibc/ldso/ (7 files in 7 dirs): move arch-specific checks out of shared dl-syscall.h and into arch-specific dl-syscalls.h |
| 04:58.45 | andersee | paravoid: howdy |
| 04:58.51 | andersee | paravoid: I'm here now... |
| 04:58.53 | paravoid | hey :) |
| 04:59.05 | paravoid | I've got some patches for you :) |
| 04:59.46 | paravoid | mail/DCC/copy-paste? |
| 05:00.09 | andersee | pick your poison |
| 05:00.31 | paravoid | heh |
| 05:00.57 | paravoid | first of all, Debian's libuclibc0 (ugh) maintainer has a patch in his .diff.gz |
| 05:01.06 | paravoid | it adds support for mipsel architecture |
| 05:01.45 | vapier | ha |
| 05:01.47 | andersee | uClibc runs on mips and mipsel just fine... |
| 05:02.00 | vapier | i run Gentoo/uClibc/mipsel |
| 05:02.11 | andersee | my amd alchemy board work great with both |
| 05:02.11 | paravoid | er |
| 05:02.43 | vapier | andersee: speaking of which, i was thinking of removing 'mipsel' from the arch list and moving it to a 'little/big endian' selection like arm has |
| 05:03.01 | paravoid | diff -Nur --exclude=debian ../native/uClibc/extra/Configs/Config.mips uClibc/extra/Configs/Config.mips |
| 05:03.01 | paravoid | --- ../native/uClibc/extra/Configs/Config.mips 2005-08-14 06:33:53.000000000 +0300 |
| 05:03.01 | paravoid | +++ uClibc/extra/Configs/Config.mips 2005-08-14 22:00:37.000000000 +0300 |
| 05:03.01 | paravoid | @@ -64,3 +64,7 @@ |
| 05:03.01 | paravoid | <PROTECTED> |
| 05:03.02 | paravoid | <PROTECTED> |
| 05:03.04 | paravoid | + |
| 05:03.06 | paravoid | +config ARCH_BIG_ENDIAN |
| 05:03.08 | paravoid | + bool |
| 05:03.10 | paravoid | + default y |
| 05:03.12 | paravoid | etc |
| 05:03.24 | paravoid | diffstat: |
| 05:03.24 | paravoid | <PROTECTED> |
| 05:03.24 | paravoid | <PROTECTED> |
| 05:03.24 | paravoid | <PROTECTED> |
| 05:03.24 | paravoid | <PROTECTED> |
| 05:03.26 | paravoid | <PROTECTED> |
| 05:03.28 | paravoid | <PROTECTED> |
| 05:03.30 | paravoid | <PROTECTED> |
| 05:03.32 | paravoid | <PROTECTED> |
| 05:03.36 | andersee | vapier: probably makes sense -- but not till after I release 0.9.28... |
| 05:03.52 | andersee | paravoid: what I said to vapier |
| 05:03.54 | andersee | :-) |
| 05:03.56 | paravoid | ok |
| 05:04.02 | paravoid | my patches now |
| 05:04.18 | paravoid | actually, one, for debian/ |
| 05:04.22 | paravoid | minor updates |
| 05:04.33 | vapier | andersee: pass it through mjn3 too ? |
| 05:04.39 | paravoid | diffstat: |
| 05:04.39 | paravoid | <PROTECTED> |
| 05:04.40 | paravoid | <PROTECTED> |
| 05:04.40 | paravoid | <PROTECTED> |
| 05:04.40 | paravoid | <PROTECTED> |
| 05:04.40 | paravoid | <PROTECTED> |
| 05:04.41 | paravoid | <PROTECTED> |
| 05:04.43 | paravoid | <PROTECTED> |
| 05:04.45 | paravoid | <PROTECTED> |
| 05:04.47 | paravoid | <PROTECTED> |
| 05:04.50 | mjn3 | vapier: i'm here |
| 05:04.55 | mjn3 | looking |
| 05:04.56 | andersee | paravoid: last I looked the debian stuff was still using the uClibc toolchain wrapper |
| 05:05.05 | mjn3 | ugh |
| 05:05.19 | paravoid | andersee: this is for your debian/, not Debian's |
| 05:05.25 | andersee | paravoid: ahh, ok |
| 05:05.55 | paravoid | updating shlibs 0.9.26->28, updating build-deps with kernel-source-2.4.27, removed pre-woody build-deps etc. |
| 05:06.18 | vapier | mjn3: ugh to what ? :) |
| 05:06.28 | paravoid | vapier: to the flooding maybe? :) |
| 05:06.39 | mjn3 | ugh that the debian package is still using the old toolchain wrapper |
| 05:06.46 | vapier | mjn3: yeah i saw that too :/ |
| 05:07.01 | vapier | mjn3: agree/disagree with my mipsel change to the arch list ? |
| 05:07.06 | mjn3 | sounds fine to me |
| 05:07.21 | paravoid | why is that so bad btw? |
| 05:07.44 | vapier | mjn3: see my ldso.c/DT_DEBUG question about mips ? |
| 05:08.34 | mjn3 | yeah. i honestly don't know and couldn't test anyway until tomorrow. in reality, i stay as far away from the ldso code as i can ;-) |
| 05:09.10 | vapier | heh, i'd like to get all the __ARCH__ stuff out of ldso source and keep em in ARCH/ subdirs |
| 05:09.32 | paravoid | andersee: so, interested? |
| 05:09.43 | vapier | andersee: see my ldd.c and asm/elf.h proposed changes ? |
| 05:09.54 | andersee | vapier: patch? |
| 05:10.03 | andersee | paravoid: send me a patch for review? |
| 05:10.14 | paravoid | a min |
| 05:11.05 | vapier | my box died ... but the basic idea is to drop all the arch checks in utils/ldd.c and replace it with #include asm/elf.h and just use like ELF_CLASS |
| 05:12.23 | vapier | since asm/elf.h defines ELF_ARCH to like EM_ARM |
| 05:16.12 | andersee | paravoid: why remove ldconfig from the postinst? |
| 05:16.32 | paravoid | it's uclibc-dev |
| 05:18.09 | andersee | paravoid: which contains /lib/libthread_db.so.1 and others |
| 05:18.20 | paravoid | hrm |
| 05:18.24 | andersee | paravoid: it really does want the ldconfig |
| 05:18.41 | paravoid | my bad |
| 05:18.45 | andersee | paravoid: the rest looks ok though |
| 05:18.48 | paravoid | want to resend? |
| 05:18.56 | andersee | nah |
| 05:19.06 | andersee | this is why god invented peer review |
| 05:19.08 | andersee | :-) |
| 05:19.15 | paravoid | oh I've added myself to the Uploaders field |
| 05:19.19 | paravoid | don't forget to remove it |
| 05:20.11 | andersee | k |
| 05:20.18 | paravoid | my name is on debian/changelog too |
| 05:20.58 | paravoid | (I added myself to the uploaders to make lintian shut up) |
| 05:22.31 | andersee | paravoid: applied |
| 05:22.42 | paravoid | thanks! :) |
| 05:22.51 | andersee | dunno why CIA-10 didn't tell us about it as usual |
| 05:23.00 | andersee | CIA-10: anybody home? |
| 05:23.10 | andersee | CIA-10: knock knock knock |
| 05:23.19 | andersee | ibot: zap CIA-10 |
| 05:23.19 | ibot | ACTION takes out a cattle prod and gives CIA-10 a good jolt. |
| 05:23.23 | paravoid | lol |
| 06:50.55 | *** join/#uclibc psaksa (n=pate@ip212-226-134-137.adsl.kpnqwest.fi) |
| 08:17.17 | *** join/#uclibc rigidus (n=rigidus@host81-151-41-146.range81-151.btcentralplus.com) |
| 08:30.41 | CIA-10 | 03jocke * r11158 10uClibc/ldso/libdl/libdl.c: Fix bug 387, missing _dl_progname for static apps. |
| 09:37.39 | *** join/#uclibc kergoth`zzz (n=kergoth@c-24-118-219-25.hsd1.mn.comcast.net) |
| 09:55.03 | *** join/#uclibc kergoth`1zz (n=kergoth@c-24-118-219-25.hsd1.mn.comcast.net) |
| 10:07.07 | Zta | Hm, I've inserted an USB flash memory dongle, and the usb-storage module seems to recognize it. |
| 10:07.28 | Zta | But where should I mount it from? |
| 10:08.03 | Zta | mount /de |
| 10:08.03 | Zta | v/sda1 -t vfat /tmp/usbmount/ |
| 10:08.04 | Zta | mount: Mounting /dev/sda1 on /tmp/usbmount failed: No such device or address |
| 10:15.28 | *** join/#uclibc rigidus (n=rigidus@host81-151-41-146.range81-151.btcentralplus.com) |
| 10:53.07 | andersee | JockeHome: around? |
| 11:02.15 | CIA-10 | 03andersen * r11159 10buildroot/package/coreutils/coreutils.mk: apply patches when requested to |
| 11:19.23 | *** join/#uclibc cow (n=cow@M1129P019.adsl.highway.telekom.at) |
| 11:40.34 | *** join/#uclibc GyrosGeier (n=geier@honey.kvedulv.de) |
| 11:47.10 | *** join/#uclibc Negative (n=negative@82.204.238.89) |
| 12:21.46 | Zta | Um, what scripts are called when an usb dongle is inserted? |
| 12:23.08 | Zta | I want to execute my own program/script, when an usb memory stick is inserted. |
| 12:24.26 | Zta | Specifically, I want to mount it and execute a program on the usb stick. |
| 12:40.06 | CIA-10 | 03andersen * r11160 10uClibc/include/stdlib.h: We do not supply a BSD style getloadavg() function |
| 12:52.47 | CIA-10 | 03andersen * r11161 10buildroot/package/coreutils/coreutils.mk: (log message trimmed) |
| 12:52.47 | CIA-10 | The coreutils configure script is a steaming <many explitives deleted |
| 12:52.47 | CIA-10 | in the interest of our younger readers>. This works around some of the |
| 13:03.34 | Zta | good night =) |
| 13:50.39 | cow | Zta, echo "/my/script" > /proc/sys/kernel/hotplug |
| 13:55.03 | *** join/#uclibc ambroseL (n=bjb@smtp2.symbium.com) |
| 13:57.04 | Zta | cow, I did that, but it doesn't get called =( Or in fact, I did mv /my/script /sbin/hotplug, which /proc/sys/kernel/hotplug is expecting. |
| 13:57.07 | Zta | chmod +x also |
| 13:57.28 | Zta | but it's never executed. |
| 13:57.39 | Zta | hub.c: new USB device <NULL>-1.2, assigned address 9 |
| 13:57.56 | Zta | This is what I get in the console. |
| 13:58.32 | Zta | I am trying to build hotplug for my system, but it's a pain in the ass |
| 13:58.58 | Zta | new module-init-tools to generate modules.usbmap... |
| 13:59.22 | Zta | But that shouldn't be necessary, for a plain stupid script to be run, when an usb dongle is inserted? |
| 14:00.31 | cow | Zta, iirc you primarily need CONFIG_HOTPLUG support in your kernel |
| 14:02.59 | Zta | I have. that gives me the proc. |
| 16:01.55 | *** join/#uclibc paravoid (i=paravoid@147.102.26.135) |
| 16:06.12 | *** join/#uclibc psm (n=mps@M705P021.adsl.highway.telekom.at) |
| 16:07.01 | psm | JockeHome: seen the bug about _dl_progname not defined in libdl for non_shared? |
| 16:08.27 | psm | bug 387 |
| 16:10.47 | psm | seen it ;) |
| 16:57.48 | *** join/#uclibc hanzzon (n=per@dialup-d2i-62-65-89-56.home.se) |
| 17:01.00 | hanzzon | hi, maybe I'm a complete moron, I just want tcpdump compiled for mipsel/bigendian but whenever I try to start the compile process I get a bunch of warnings from gcc "differ in signedness" am I not using the correct gcc? I do not add anything to my PATH but I though buildroot would take care of all that for me? |
| 17:03.51 | hanzzon | or would it be possible to find it precompiled somewhwre? (I have asked google) |
| 17:06.28 | psm | is it gcc-4? |
| 17:07.43 | hanzzon | yes 4.0.0 is what is installed on my Fedora Core 4 machine... |
| 17:11.11 | psm | gcc-4 emits many of these warnings, maybe you could try to apply the FC4 tcpdump patches (if any) |
| 17:12.40 | hanzzon | I have no idea how to do that... |
| 17:13.31 | *** join/#uclibc El-Lotso (n=Lotso@60.49.70.164) |
| 17:14.19 | psm | then live w/ those warnings ... |
| 17:14.50 | hanzzon | it finally fails at the end of the compile... I'm compiling it now again, will paste the error when it happens.. |
| 17:15.15 | *** join/#uclibc keturn (n=kevint@pdpc/supporter/sustaining/keturn) |
| 17:16.47 | hanzzon | smbutil.o(.text+0xc17):smbutil.c: undefined reference to `__fputc_unlocked' |
| 17:16.47 | hanzzon | collect2: ld returned 1 exit status |
| 17:16.47 | hanzzon | make[1]: *** [tcpdump] Error 1 |
| 17:29.37 | vapier | hanzzon: your build compiled wrong ... it used header files from your system (prob glibc) but compiled against the libc from uclibc |
| 17:31.35 | *** join/#uclibc psm (n=mps@M705P021.adsl.highway.telekom.at) |
| 17:35.14 | hanzzon | why did it do that then?/how to fix it? |
| 17:36.29 | vapier | because whatever you're trying to build added the wrong the -I paths, used the wrong compiler at some point, or you're doing it wrong |
| 17:36.54 | vapier | how are you building it ? |
| 17:37.17 | hanzzon | well, I am using buildroot to compile tcpdump... I thought buildroot would take care of everything for me? |
| 17:37.28 | vapier | tcpdump failed to compile ? |
| 17:37.51 | hanzzon | yes |
| 17:39.45 | vapier | in that case, scan bugs.uclibc.org to see if anyone else reported such a failure and maybe they posted a fix |
| 17:40.08 | vapier | and you're right, if you use buildroot, those details should be taken care of for you :) |
| 17:40.29 | hanzzon | ok, thnx |
| 17:41.04 | vapier | i think i should add a FAQ entry for the __fputc_unlocked error message ... took me a little digging the first time i hit that to find out what exactly was going on |
| 17:47.15 | hanzzon | not finding anything :( anyone wanting to do me the favor of compiling it for mipsel/big-endian? |
| 17:48.54 | vapier | i'll be able to look at in a few hours when i get home, so if you want to wait that long :p |
| 17:49.06 | vapier | also, when you say 'mipsel/big-endian', what exactly do you mean |
| 17:50.10 | hanzzon | a Netgear DG834GT router with a Broadcom bcm6348 CPU running Linux v2.4.17 and gcc v3.1 |
| 17:50.36 | vapier | isnt that a mipsel ? |
| 17:51.00 | hanzzon | yes, it is a mipsel big-endian (i.e. not little-endian ;-)) |
| 17:51.36 | vapier | ... |
| 17:52.05 | solar | ummm |
| 17:52.22 | vapier | i'm debating whether to keep trying |
| 17:52.50 | solar | unless I've missed something.. EL is never BE |
| 17:53.27 | vapier | yeah, i'm giving up |
| 17:53.31 | vapier | i'll just look at the tcpdump thing |
| 17:53.36 | vapier | since that looks legit ;) |
| 17:53.40 | hanzzon | ;-) |
| 18:06.28 | mjn3-work | tcpdump built fine here. just tested for mipsel with gcc 3.3.6 |
| 18:06.41 | mjn3-work | doesn't look like a buildroot issue to me |
| 18:06.47 | vapier | mjn3 trumps me |
| 18:06.59 | hanzzon | mjn3; could you DCC or e-mail the binary please? |
| 18:07.29 | mjn3-work | not from work... aka firewall hell |
| 18:07.36 | hanzzon | e-mail? |
| 18:08.15 | mjn3-work | and as i said, this was a little endian build anyway. you were wanting big endian |
| 18:08.44 | hanzzon | oh, little endian :( |
| 18:08.56 | hanzzon | yes, it needs to be big-endian |
| 18:09.49 | hanzzon | mjn3-work; can you send me your .config file, perhaps I made some glaring mistake? |
| 18:10.58 | psm | hanzzon, using mipsel means little endian for me, use mips instead |
| 18:11.25 | hanzzon | that might just be it... |
| 18:11.28 | psm | not only for me, also for all auto-tools |
| 18:24.11 | JockeHome | vapier: Mabe DT_DEBUG is readonly on MIPS? |
| 18:25.47 | vapier | maybe, but that seems kind of weird |
| 18:26.13 | JockeHome | vapier: BTW there are a few more non ldso ARM thumb assembler fixes in the thumb-interwork-asm.patch |
| 18:26.16 | vapier | why would mips alone exhibit that behavior ? |
| 18:26.40 | vapier | JockeHome: i dont know why jbowler filed all those patches together considering how unrelated they are ... |
| 18:26.52 | JockeHome | yup |
| 18:26.55 | vapier | i only went through and sniped the arm/thumb specific things |
| 18:27.30 | vapier | and the ones i was pretty sure were right/wouldnt screw up current arm code |
| 18:28.11 | vapier | i have to check with jbowler and figure out which ARM cpus we pretty lose support for by enabling thumb support all the time |
| 18:28.38 | vapier | for example, i dont want to lose ARM4 support seeing as how i use uClibc on my arm4 netwinders :) |
| 18:29.07 | JockeHome | I don't think you will enable thumb unless you compile with thumb option |
| 18:30.04 | JockeHome | I think he claimed that all ARM cpu can handle the thumb insns |
| 18:30.20 | vapier | i'm pretty sure not all ARM can handle it ... |
| 18:30.29 | vapier | but he hasnt come onto irc so i can poke him about it :) |
| 18:30.41 | JockeHome | :) |
| 18:31.53 | vapier | you seem to have been covering all the generic ldso changes though so i havent reviewed those ... anything you want some more eyes on ? |
| 18:33.06 | JockeHome | Nope, just look at the non ldso stuff in the above mentioned patch |
| 18:35.01 | vapier | yeah, that patch was what i wanted to talk to jbowler about and do some testing with before committing |
| 18:35.10 | vapier | the other patches were like no brainers ;) |
| 18:35.30 | JockeHome | what is does OE mean in jbowler last notes ? |
| 18:35.40 | vapier | he works on Open Embedded |
| 18:35.52 | vapier | and i dont really agree with his analysis ;) |
| 18:35.55 | JockeHome | ah |
| 18:36.13 | JockeHome | which one of them? |
| 18:36.16 | vapier | not everyone uses an automated build system for uClibc |
| 18:36.33 | vapier | the idea of making people override like CC or something |
| 18:37.28 | vapier | but the bug was getting crowded with your ldso chitchat so i was gonna wait for him to pop onto irc |
| 18:37.38 | vapier | maybe i'll just suck him into this channel |
| 18:38.35 | JockeHome | :) |
| 18:39.30 | JockeHome | I am a bit sensitive when people mess with dl_find_hash. Had major pain getting that one to work |
| 18:40.52 | vapier | yeah, that caused me some trouble on x86_64 |
| 18:40.54 | JockeHome | I think the chages to clone, mmap64 and vfork look like the ones you already applied |
| 18:41.42 | JockeHome | x86_64 OK now? |
| 18:42.35 | vapier | for dl_hash code ? yeah ... |
| 18:42.54 | vapier | it was using like unsigned long which is fine for 32bits because the hash code is stored in ELF's as 32bit values |
| 18:43.15 | vapier | but on x86_64, it was eating 2 hashes at a time since unsigned long -> 64bit but the stored hashes were still 32bit |
| 18:43.25 | vapier | ldso works on x86_64 now for the most part ... |
| 18:43.47 | vapier | i was able to test some basic packages like sed |
| 18:44.14 | vapier | but lazy binding is broken so all objects are relocated as if they were BIND_NOW |
| 18:44.48 | JockeHome | ouch, lazy bindning can be hard to get rigth |
| 18:46.23 | vapier | yeah, it looks like i have to invest more time reading the ELF supplement and banging my head :) |
| 18:46.44 | vapier | theres also the issue of pthreads doesnt want to work for me either |
| 18:46.54 | vapier | so those two features are missing from uClibc on x86_64 |
| 18:46.55 | JockeHome | look really hard at glibc |
| 18:48.56 | vapier | hmm, i wonder if theres some 32bit/64bit stuff thats screwing up pthreads ... |
| 18:49.02 | vapier | hadnt thought of that ... |
| 18:49.11 | JockeHome | thumb again, maybe add #ifdef __thumb__ to the ones that toches clon, mmap64 and vfork? |
| 18:49.22 | JockeHome | probably :) |
| 18:49.48 | JockeHome | good to have a suspect :) |
| 18:49.52 | vapier | blah i hate tracking that crap down heh |
| 18:50.30 | *** join/#uclibc GyrosGeier (n=geier@honey.kvedulv.de) |
| 18:51.04 | vapier | dont get fat |
| 18:51.08 | vapier | TV rotts your brain |
| 18:51.16 | JockeHome | :) already too late |
| 18:51.32 | JockeHome | atleat the brain part |
| 18:56.23 | *** join/#uclibc carlg (i=clrsrv@www.clearcore.com) |
| 19:15.23 | *** join/#uclibc cow (n=cow@M838P002.adsl.highway.telekom.at) |
| 19:24.31 | *** join/#uclibc CIA-10 (n=CIA@flapjack.navi.cx) |
| 19:50.09 | *** join/#uclibc SpanKY (n=UserBah@pool-151-203-204-157.bos.east.verizon.net) |
| 19:50.50 | *** join/#uclibc rigidus (n=rigidus@host86-134-240-50.range86-134.btcentralplus.com) |
| 21:24.48 | waiting4_0928 | Hi, how do I get argz.h to build gettext |
| 21:26.02 | solar | you really do not want gettext. But I would look at the gentoo or buildroot patches/build systems for that pkg |
| 21:26.32 | waiting4_0928 | solar, thx |
| 21:36.13 | andersee_sleep | bummer |
| 21:36.44 | andersee | last night's buildroot runs failed |
| 21:42.23 | nitinkg | solar, whats the story behind missing argz.h? |
| 21:50.14 | mjn3-work | what story? we simply never included (glibc) argz support |
| 21:51.59 | nitinkg | oh ok, is this something work-in-progress? |
| 21:52.48 | mjn3-work | should be easy enough to port. are you volunteering? |
| 21:54.25 | nitinkg | mjn3-work, I will take a look. I have to get a patch ready for gettext first |
| 21:56.07 | mjn3-work | what are you trying to do with gettext that requires a patch? |
| 21:56.31 | nitinkg | cleaner inclusion of argz.h |
| 21:57.15 | andersee | vapier: vapier is going to buy us all donuts! |
| 21:59.15 | andersee | vapier: contrary to popular opinion, I am trying to cut a release. Checking in random untested stuff is not helping... |
| 21:59.43 | andersee | vapier: buying donuts for everyone as penance would help however |
| 21:59.58 | andersee | vapier: as would not breaking the build |
| 22:07.44 | CIA-10 | 03vapier * r11162 10uClibc/ldso/include/dl-syscall.h: remove quote char which doesnt belong |
| 22:10.00 | CIA-10 | 03andersen * r11163 10uClibc/ldso/include/dl-syscall.h: Just quote the whole darn thing so we can print any error msg we want |
| 22:13.32 | CIA-10 | 03vapier * r11164 10/ (2 files in 2 dirs): Bernhard Fischer writes: fix devfsd usage |
| 22:14.30 | SpanKY | i was hoping to get it fixed before anyone noticed buy my devbox lost power so i couldnt do it remotely :/ |
| 22:18.35 | CIA-10 | 03andersen * r11165 10uClibc/libc/sysdeps/linux/common/getcwd.c: (log message trimmed) |
| 22:18.35 | CIA-10 | For current kernels, we just use the getcwd() syscall. For old 2.0 linux |
| 22:18.35 | CIA-10 | kernels, we need this bit to make the non-syscall version that recurses up the |
| 22:19.20 | andersee | SpanKY: was in the daily snapshot, so my buildroot runs for arm, ppc, x86, and mips all failed in the same spot |
| 22:27.16 | cow | anyone willing to reduce the size footprint of /me touching df while i get some rest? |
| 22:28.02 | andersee | cow: eh? |
| 22:28.08 | cow | vapier, .. and thanks for checking that in :) |
| 22:28.26 | cow | andersee, i sent a work in progress against df to the busybox list |
| 22:28.28 | andersee | cow: Subject: [BusyBox] df, add -P and -B? |
| 22:28.34 | cow | andersee, yes |
| 22:28.56 | andersee | cow: just saw that. That'll be very nice, esp since the ltp testsuite wants to use df -P |
| 22:29.23 | cow | g'night all |
| 22:29.28 | andersee | night |
| 22:29.53 | *** part/#uclibc nitinkg (n=ngupta@gateway-1237.mvista.com) |
| 22:41.31 | *** join/#uclibc paravoid (i=paravoid@147.102.26.135) |
| 22:45.53 | CIA-10 | 03vapier * r11166 10uClibc/docs/man/ (. ldconfig.8 ldd.1): write some good manpages |
| 22:46.18 | SpanKY | paravoid: that debian patch doesnt add mipsel support ... |
| 22:46.37 | SpanKY | paravoid: what it does is duplicate the mips config file so it can make the debian build easier |
| 22:46.43 | SpanKY | in other words, beyond debian, it's pointless |
| 22:46.45 | paravoid | SpanKY: so they told me |
| 22:47.00 | paravoid | SpanKY: it has some other changes too, but I'm too unfamiliar with uclibc to understand them |
| 22:47.17 | SpanKY | ive gone through the whole thing and taken the stuff that's worth it |
| 22:47.36 | SpanKY | i'm left with mips clone.S changes which mjn3 or andersee will have to review |
| 22:47.56 | paravoid | nice :) |
| 22:49.14 | SpanKY | and actually, most of the clone.S patch is just style changes |
| 22:49.20 | andersee | SpanKY: post em and I'll take a look -- after I get 0.9.28 out the door |
| 22:49.24 | SpanKY | http://rafb.net/paste/results/g6goKN99.html |
| 22:49.45 | SpanKY | that's the only thing left in the debian patchset ... no idea what it's for though, the changelog doesnt seem to indicate what it's forth |
| 22:51.00 | andersee | SpanKY: for mips, $31 is the return address register |
| 22:51.26 | andersee | SpanKY: so that code appears 100% equivalent to me... |
| 22:51.58 | SpanKY | same size when assembled i wonder |
| 22:53.33 | SpanKY | well other than that change, everything else in the debian clone.S patch doesnt do anything compile related |
| 22:54.55 | SpanKY | the two manpages were ugly/wrong so i just wrote two new ones |
| 22:55.09 | andersee | cool |
| 22:55.55 | andersee | you still owe us donuts, but that does work off needing to buy the jelly filled ones |
| 22:56.00 | andersee | :-) |
| 22:56.49 | SpanKY | there a donkin donuts near you ? ;) |
| 22:59.59 | SpanKY | andersee: do you have blackfin hardware ? |
| 23:00.55 | SpanKY | hmm MAINTAINERS says you do :) |
| 23:09.32 | *** join/#uclibc sjhill (n=sjhill@adsl-64-108-118-230.dsl.yntwoh.ameritech.net) |
| 23:12.33 | *** join/#uclibc mjn3-work (n=mjn3@h-66-134-116-110.dnvtco56.covad.net) |