| |
| |
| |
|
Page: 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51
Comments:
<0> jakub_|>> check: did you compile in support for your network card in kernel.
<1> rexp: but how is that related to not having a C preproccessor?
<2> TFKyle: look at comment #6
<3> gmem.c <--- what is that and from what package?
<4> humm
<3> that seems to be the common denominator in every error on apps that are not working
<5> Kimi_Work: dunno. google it?
<5> Kimi_Work: maybe glib?
<1> rexp: maybe theres a problem with your version of autoconf/automake then? (since it runs eautoreconf there)
<3> however, i cannot find gmem.c anywhere in sys
<4> gcc 4.1.1? isnt that ~x86..?
<5> Gidzz0r: Yeah, but it isn't package.mask'ed anymore
<4> i see..
<6> rexp: there are a series of errors in the configure logs.
<3> chaosite: that is part of the error that i get "GLib-ERROR **: gmem.c:158: failed to allocate 3214146996 bytes"
<2> TFKyle: okay, so what should I do? Recompile autoconf & automake or downgrade it?
<1> rexp: btw, it works here with gcc 4.1 (though I'm using eselect compiler here, which may work slightly different than gcc-config)
<7> The less tool no longer works as expected. When i use it on an HTML file it doesn't show me the codes. When I use it on an image it tells me dimensions. WTF?
<2> TFKyle: (last sync today 3PM)
<1> rexp: upgrade them if possible I guess
<5> Kimi_Work: thats about 3065MB
<8> dev\0 > probably not, I'll take a look at the installer again. there is no way to get the network card working with out recompiling?
<3> chaosite: this sys has had a complete overhaul (emerge -e sys and world) after changing from gcc-3.3.6 to gcc-3.4.5
<3> chaosite: exactly, no app would try to allocate that amount of mem
<5> Kimi_Work: so its obviously some error. But since you're saying that you get that on more than one lib, it might be glib that is at fault
<0> jakub_|>> hz, i didn't install gentoo with installer, so i cant say is there any other way.
<1> rexp: the path configure sees for cpp exists right?
<2> TFKyle: Before emerging X11R7 I've recompiled world with gcc-4.1.1 (emerge -e world, but I've not do emerge -e system), maybe that is reason?
<9> world is system + apps you've installed
<5> Kimi_Work: so try to recompile it, preferably with USE="debug" FEATURES="nostrip", then you can use gdb on it too
<9> so world does do system (atleast in theory :)
<1> rexp: emerge -e world should include system (unless you have a hell of a lot of orphaned packages in system)
<3> chaosite: it is possible. i did remerge glib after the first error. to see if that would help
<2> TFKyle: w8.. I'll check it in config.log
<0> jakub_|>> dont be afraid to recompile the kernel. it is easy enough%)
<10> hey boys, what are the advantages to switch to xorg 7?
<6> TFKyle: even if he has orphaned packages, world will still include system
<11> in lshal, is volume_uuid_6024_F0E8 unique for every single device? such that if i insert the device now and in a week, it will be the same, yet different from any other device si might have
<4> MoL0ToV: i wish i knew
<3> chaosite: the fact that a locate after an updatedb finds to gmem.c also has me hea scratching
<9> MoL0ToV: you can say you have xorg 7? :)
<6> MoL0ToV: LOTS of fun bug splatting
<5> Kimi_Work: well, that file is in glib source
<5> Kimi_Work: It isn't usually on your system
<10> so i stay with 6.x release :)
<6> at least based on my observations here
<5> Kimi_Work: do you know any C?
<1> MoL0ToV: well, theres evdev, and better ati opensource drivers (though 7.1 is better still for them)
<3> chaosite: ok, so a locate would not return anything
<12> hmm, maybe I should open a GLEP for a NEVERPRUNE portage variable
<9> molotov - i installed 7 just purley out of boredum.... on my system I don't notice a single difference.
<5> kojiro: what would that do?
<3> chaosite: no, unfortuantely not
<10> TFKyle, also for nvidia?
<0> MoL0ToV|>> i will have a loooot of packages instead one %)
<13> well im running rsync now.
<12> chaosite: ideally, allow you to specify certain packages that --prune would ignore, such as auto{make|conf}
<1> MoL0ToV: most people using nvidia use the closedsource drivers which aren't distributed as part of xorg so don't have much of a difference between 6.8 and 7.0
<12> chaosite: so that you could more easily prune things realistically, such as kde and gentoo-sources
<14> LinuxBlues: btw, do u know how to change the default parameters for ctorrent?
<3> chaosite: anothe oddity was that these apps were working before emerging them to get them in portage "world" list
<2> TFKyle: I've found line "checking how to run the C preprocessor... i686-pc-linux-gnu-gcc -E"
<5> kojiro: hmm. I usually use the prune feature from ecatmur's dep
<15> Mattias: read the man or ctorrent --help
<8> dev\o > thanks for the help, I'll try to reinstall again.
<1> kojiro: why would anyone want a NEVERPRUNE variable, portage doesn't prune stuff now unless you explicitly tell it to anyway
<3> chaosite: i will try the remerge first. could something have gotten flaky on boot?
<2> TFKyle: So.. it founds it?
<5> kojiro: In my exprience, its been far less harmful then portage's prune
<1> rexp: look at the line right below that
<0> jakub_|>> no problem
<12> TFKyle: yes, but if you want to remove old versions of KDE meta you have to go through the prune output very carefully
<12> TFKyle: NEVERPRUNE would make it less error-prone
<2> TFKyle: checking for egrep... grep -E
<1> kojiro: its easy to make a script to do that
<14> LinuxBlues: ye. but i want to permanently change em in some kind of config file. so i only have to do "ctorrent eg.torrent" later
<1> rexp: ok, so it doesn't find cpp, not the c preproccessor
<1> rexp: try updating autoconf/make
<16> hehe
<5> Kimi_Work: Well, if you have flaky memory or a bad processor that could certainly be it
<1> autoconf/automake rather
<3> chaosite: ughhh i just realized something. on a suggestion from a work mate trying to diagnose another problem i changed bios time to 1ry ahead. i forgot to change it back and did some merges w/the date of 2007.
<5> Kimi_Work: but that doesn't sound right if you're getting the same error everytime
<12> TFKyle: sure. It's a question of what is worth leaving for people to do on their own, and what is worth having portage do. For my part, I think --prune should automatically exclude system, and only prune world *except* for system, unless you specify system
<3> chaosite: brand new system
<15> Mattias: IIRC ~/.ctorrent sorry, I'm very busy right now....
<2> TFKyle: but some is weird ---> checking dependency style of i686-pc-linux-gnu-gcc... gcc3 when I have selected gcc-4.1.1 in gcc-config -l
<5> Kimi_Work: Sadly enough, thats no promise for working memory. =/
<7> I finally figured out that the LESSOPEN envariable was causing my problem. bug 117443
<17> purplefrog: http://bugs.gentoo.org/117443 nor, P2, All, klausman-gentoo@schwarzvogel.de->base-system@gentoo.org, RESOLVED, WONTFIX, sys-apps/less default rich processing
<5> Kimi_Work: I'm speaking from (bad) exprience
<3> chaosite: i did touch the files that i could fine that had the incorrect date. cuold that have anything to do w/it
<18> hi, there's some quick-guide to recompiling the kernel?
<1> kojiro: unless you -P world, -P system or -P the packages explicitly it won't prune system packages
<5> Kimi_Work: Um, no, don't think so
<19> I had net.eth0 during installation but now that I am using gnome it doesn't exist, any solution to this problem?
<5> Kimi_Work: Unless it works if you set the BIOS one year ahead again
<6> configure:2140: i686-pc-linux-gnu-gcc -V </dev/null >&5
<1> rexp: thats normal
<1> rexp: the dep style hasn't changed between gcc3 and gcc4
<3> chaosite: true enough. although the sys p***es all checks and had been stable form the time the company pruchased it (6 months mebbe) til today
<12> TFKyle: do you consider "emerge --prune" to be "explicitly -P world" ?
<2> TFKyle: okay, I'll go emerge -e autoconf automake
<3> chaosite: yeah i didn't think the date would have much to do with it. particularly after i corrected that on most of the files
<1> kojiro: I've never tried prune with no args before
<2> brb
<12> TFKyle: it's the same as -P world
<1> kojiro: well, just tell users not to do that then
<5> Kimi_Work: you may use some "find" hackery to fix it
<12> TFKyle: ;-)
<1> kojiro: its obvious that you should only prune packages which you want to prune
<12> TFKyle: anyway, I just mean there should be a way to say "prune everything except system"
<3> chaosite: that's what i did to touch the 2007 files :)
<5> Ah, ok =)
<5> then they're fine
<1> kojiro: and its easy enough to prune packages which are in kde-base/* and are installed
<1> kojiro: that would still be dangerous, since stuff like gtk+ isn't in system
<3> chaosite: any use in re-compiling glibc a
<12> TFKyle: and orbit
<12> TFKyle: but those can be rebuilt by portage. if you remove automake/conf you lose portage
<3> chaosite: glibc was originally built w/profile, and nls flags
<1> kojiro: I hope if this gets introduced there will be a way to override it, since pruning is often a good thing to do to kill stale packages
<5> Kimi_Work: hmm
<5> Kimi_Work: don't think so
<5> Kimi_Work: whats the full error again?
<12> TFKyle: I was suggesting an environment variable you could set in make.conf
<3> chaosite: just a sec
<1> kojiro: prune should never completely unmerge a package, just older versions of it
<12> TFKyle: it would be completely configurable
<3> chaosite: GLib-ERROR **: gmem.c:158: failed to allocate 3216943572 bytes
<5> the number changes I see.
<3> chaosite: and it's always the same gmem.c:158
<16> hey quick when I have a .diff file how can I patch this file into my kernel sources?
<16> never did this and want to give it a try
<3> chaosite: shot in the dark. but irrc the apps that are failing w/that were all running when i merged them to get them in portage world...
<1> Kimi_R: that doesn't say much, gmem.c line 158 is probably in a function that gets called alot
<3> chaosite: the allocation bytes, yeah
<1> Kimi_Work rather
<5> Kimi_Work: umm.
<5> Kimi_Work: I ***ume you restarted them by now?
<20> hostname.(none) can anyone help me with this?
<5> Kimi_Work: and remerged some of them, too?
<21> hey all... I need to run something on a cronjob to check if a proces shas died (and if it has, restart it).. I can do the check with a 'ps ax |grep -v grep |grep procname -c', but is there a better way?
<3> have tried remerging, but as for restarting. have not been able to get one to restart
<1> _AxS_: do you know the pid?
<3> TFKyle: yeah i know it's very little to go one. but that's all the error reporting that i'm getting
<16> can nobody here tell me how to patch a .diff file into the kernel sources?
<22> i am thinking about putting gentoo on my laptop (pentiumm 1.5 512 ram intel integraged graphics) and i was wondering if anyone knew how it would run?
<22> well?
<5> Kimi_Work: right
Return to
#gentoo
or
Go to some related
logs:
failed fluxbox.cat
#sdl
glbasic
#fedora
#perl
#gentoo
unsupported ALi chipset
#python
fluxbox keys tablaunch
#css
|
|