IRC log for #uclibc on 20070406

00:29.00*** join/#uclibc Dj_Andy (i=Dj_Andy@e180234238.adsl.alicedsl.de)
00:29.47Dj_Andyis here anybody on??? i need help :(
00:32.21Dj_Andyis here anybody on??? i need help :(
00:58.17*** join/#uclibc wrobbie (n=rob@cm74.kappa84.maxonline.com.sg)
01:59.37*** join/#uclibc Skapare (n=phil@sedna.ipal.net)
02:15.23*** join/#uclibc kasc_ (n=kasc@dslb-084-060-107-123.pools.arcor-ip.net)
03:17.11*** join/#uclibc flatronf701B (n=flatronf@202.75.186.154)
06:06.31*** join/#uclibc blindvt__ (n=bf@M796P011.adsl.highway.telekom.at)
06:54.08*** join/#uclibc blindvt` (n=blindvt`@chello213047219050.surfer.at)
07:21.36*** join/#uclibc vrm (n=vrm@210.23.98-84.rev.gaoland.net)
07:22.35vrmhi
08:53.23blindvt`wget -s http://msn.com/index.py
08:53.23blindvt`Connecting to msn.com (207.68.172.246:80)
08:53.23blindvt`wget: bad header line: status_CODE: NotFound
08:53.33blindvt`M$ sucks
09:11.28blindvt`heh, that one is even better (throws a runtime-error):
09:11.48blindvt`www.msn.com and then get /index.py HTTP/1.0
09:12.48blindvt`sounds like a nice opportunity to stress that machine a bit ;)
10:41.07CIA-903vda * r18346 10busybox/networking/nc_bloaty.c: nc: code shrinkage, bugfixes. -50 bytes code size
11:55.38CIA-903aldot * r18347 10buildroot/package/bash/bash.mk: - using target's LDFLAGS as LDFLAGS_FOR_BUILD is b0rked
12:00.35*** join/#uclibc vrm (n=vrm@210.23.98-84.rev.gaoland.net)
12:51.45CIA-903aldot * r18348 10buildroot/package/libelf/libelf.mk: - bump version. memmove works so let configure know about this fact
14:06.53*** join/#uclibc blindvt_ (n=bf@62.47.144.144)
14:09.53*** join/#uclibc dalias (i=dalias@brightrain.aerifal.cx) [NETSPLIT VICTIM]
14:09.53*** join/#uclibc triel (i=triel@pandora.stupidisco.net) [NETSPLIT VICTIM]
14:09.53*** join/#uclibc r0ute (n=r0ute@ina044000004.lancs.ac.uk) [NETSPLIT VICTIM]
14:14.34*** join/#uclibc quitte (n=quitte@stgt-d9beaa25.pool.mediaWays.net)
14:21.02CIA-903aldot * r18349 10buildroot/package/libelf/ (Config.in libelf.mk): (log message trimmed)
14:21.02CIA-9- someone disabled shared lib support without fixing the libelf rule leading to
14:21.02CIA-9<PROTECTED>
14:21.50*** join/#uclibc wrobbie (n=rob@cm74.kappa84.maxonline.com.sg)
14:24.24*** join/#uclibc blindvt_ (n=bf@M815P028.adsl.highway.telekom.at)
15:01.34CIA-903aldot * r18350 10buildroot/package/openssl/openssl.mk: - hit awk on steroids with a clue bait
15:05.06CIA-903aldot * r18351 10buildroot/package/ltrace/Config.in: - doesn't compile if LFS is off
16:31.07*** join/#uclibc mjn3-work (n=mjn3@h-66-134-116-110.dnvtco56.covad.net)
16:32.06CIA-903aldot * r18352 10buildroot/package/readline/readline.mk: - commentary typo fix
16:36.49CIA-903aldot * r18353 10buildroot/ (10 files in 9 dirs): - make sure to download fakeroot if needed
18:01.49*** join/#uclibc KhemHome (i=KhemHome@nat/montavista/x-2750dafa0a4f824f)
22:06.54*** join/#uclibc flyback (n=sankyo@c-67-186-27-63.hsd1.pa.comcast.net)
22:07.04flybackcanucks!
22:07.04flyback:P
22:07.08flybackno seriously
22:07.26flybackis there any secvurity trade offs with uclibc vs glibc
22:07.32flybackon a platform with a mmu
22:08.01solarthere are no known security problems with uClibc.
22:08.26solarnow as far as proactive security measures some arches have more advantages than others
22:08.34flybackI know that much
22:08.44flybackoh hey you are the dude that did that exec patch yrs ago right?
22:08.44solarppc for example recently got (thanks to Jocke) the addition of SecurePLT
22:09.04solarwhich properly splits up the PT_LOAD segments
22:09.22solarsparc,alpha still suffer from the problem (but we don't care about those arches)
22:09.48solarI think you are thinking of Solar Designer from the OWL project
22:10.44flybackoh
22:10.47flybackthen you suck then
22:10.47flybackdie
22:10.50flybackj/k
22:10.51flyback:P
22:10.54flybackthx for the info
22:10.58flybackneat
22:11.07flybackI was thinking about using this mostly x86
22:11.13flybackfor a bunch of old pc's I have lying around
22:11.46solarwell the best security wise you can do is opt to use a hardened-uclibc-x86 toolchain
22:12.03flybackbut I do have some juiceboxes (arm no mmu), a webpal (arm mmu), and palm t5, 2 old palm (m68k no mmu), and a ppc 405gp I will be messing with in the future
22:12.14flybackthe sparctation 2 is going in the trash unless someone needs a board
22:12.17flybackworthless
22:12.22flybackpc beats it in every respect
22:12.42flybackit's almost as worthless as my ibm pcjr
22:12.52flybackthx
22:13.08flybackI just wondered if in general if security measures had to be sacrified to get the size down
22:13.09flybackoverall
22:13.21solarhttp://tinderbox.dev.gentoo.org/uclibc/i386/ (prebuilt uclibc stable security based bins)
22:13.29flybackneato thx
22:14.25flybackvmware makes this more possible too :)
22:14.36flybackwhen I don't have time to setup a lot of native linux boxes
22:14.40flybackor extra ones for cross compiling
22:14.41flyback:)
22:17.27flybackyeah this virtual pc and vmware stuff has completely changed my world
22:17.40flybackI can test pretty much anything I want in minutes instead of hrs
22:27.34ashesCore was generated by `./a.out'.
22:27.34ashesProgram terminated with signal 11, Segmentation fault.
22:27.34ashes#0  0x080483dd in __stack_chk_fail () at libc/sysdeps/linux/common/ssp.c:107
22:27.34ashes107     {
22:28.32asheswhen i build a main(){} with a i686-pc-linux-uclibc-gcc and -fstack-protector-all, it segfaults
22:29.05*** join/#uclibc blindvt_ (n=bf@M874P014.adsl.highway.telekom.at)
22:29.09asheswhen i build a main(){} with a i686-pc-linux-gnu-gcc and -fstack-protector-all, linked to libuClibc.so, it does an infinite "stack smashing detected" loop
22:29.37ashesif i build with i686-pc-linux-uclibc-gcc and -fstack-protector (without the -all), then its fine
22:30.27ashesssp.c:107 is actually ssp.c:99, i messed with it a bit so the line numbers are off
22:32.39ashesif i remove the "while(1)" from __stack_chk_fail, then the ssp-test.c built with i686-pc-linux-gnu-gcc does not do an infinite loop, and aborts properly. but removing the while(1) causes a compiler warning that a noreturn function has a return
22:33.22ashesremoving the while(1) will still cause segfaults with i686-pc-linux-uclibc-gcc's a.out's
22:35.50ashesa.out = main(){}
22:42.14solarssp + gcc-4.x = blah;
22:42.32solarso many patches to have clean upgrade paths. it sickens us.
22:42.39solarpsm has it all working quite nice however
22:43.38flyback<PROTECTED>
22:54.17flybackI bet uclibc would be nice for that 486-100 I have with 8 meg of ram
22:54.18flybackhmm

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