| |
| |
| |
|
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 52 53 54 55 56 57 58 59 60 61
Comments:
<0> Okay, while editing conky, I put a few extra scripts into it, and I put them in ~/bin. Whenever I try to run conky, they don't work; it says permission denied on all of the files. what did I do wrong?
<1> the rating AMD uses for it's athlon CPUs suggests that you are just 20% faster with a dual core than with a single core (with equal caches, clocks, ...) is this correct in reality?
<2> mbishop: there's "distfiles"
<2> expose: depends
<3> helo, is there possible in ssh save session? Like I connect to my linux box throught ssh and put there to do some job, disconnect from it and later connect back to see what is done or to set do somthing next? (for example downloading some stuff from ftp or something else). I know it would be possible to send process to background, but how could I see its output later?
<4> Blyat, chmod +x ?
<2> expose: some apps will get up to +80 or 90%, some straight 0.
<0> what's the x option do?
<5> eXecute
<1> DexterF: compiling should be faster than 20%, shouldnt it?
<4> spuhpointer, screen/nohup
<1> DexterF: get me two examples, one for faster one for equal
<2> compiling is quite different. with gcc you can expect perf gains, yes. possibly more than 20%
<2> depends on the code
<1> DexterF: depends upon wether you actually run with more than one process or not...
<2> depends on whether code parts depend on other code parts a lot
<6> andrew______: ah, sorted it out yet?
<7> johntramp: what version of gcc is that?
<1> DexterF: which would prevent to use more than one gcc instance...thats just what i meant
<2> you can have four cores and a 1000 threads, no go if all threads wait for a fat wad of code to compile they depend on
<0> okay, I used chmod +x on all of the files and it still says permission denied
<1> it's the make -jX option
<8> kerframil nope
<8> just going 2 try it once more
<2> why would one wanna use more than one gcc instance at all?
<2> *at once
<9> marienz, 4.0.3 it is a different distro but gentooers know a lot more about compilation problems :)
<4> Blyat, pastebin ls -lh in the ~/bin directory
<10> DexterF: er right sorry, here http://mirror.espri.arizona.edu/gentoo/rsync/
<1> DexterF: parallel compilation of two independant parts of the source?
<7> johntramp: I was about to say "use the ~arch version, it has a patch for probably exactly that issue"
<11> anyone here know how I can config syslog-ng to log /dev/tty8?
<7> johntramp: lemme dig the patch out of sources.g.o
<12> * Failed Patch: baselayout-1.11.14-depscan.patch !
<12> anyone with this issue?
<2> expose: man gcc: -j
<7> Bejgli: emerge sync unless you did so less than about an hour ago
<9> marienz, is it a gcc bug?
<2> you dont call multi gccs but threads
<1> DexterF: i know of the -j, mentioned it myself a few lines up in just this discussion
<12> marienz: did sync 5 mins ago
<7> johntramp: http://sources.gentoo.org/viewcvs.py/gentoo-x86/dev-python/ctypes/files/ctypes-gcc4?rev=1.1&view=markup
<4> Bejgli, yes, same problem here - 2005.1 > 2006.0 ?
<7> johntramp: fixed in the latest version of ctypes upstream too iirc.
<12> scott___: no, only did an emerge sync, and this happens
<7> johntramp: no, ctypes bug, gcc 4 is just a bit pickier than previous versions.
<5> Could I safely go higher than -j3?
<12> when i give an emerge -uva baselayout
<2> mbishop: ok, then what I'm lookign for is not in the there. bummer.
<2> thanks anyway
<9> marienz, ok so I need to get a newer release of ctypes?
<7> Bejgli: that is quite odd. Searched bugzie? there was a broken patch for baselayout fixed pretty recently.
<13> when I do modprobe snd_hda_intel occurs this error:
<13> FATAL: Error inserting snd_hda_intel (/lib/modules/2.6.13/kernel/sound/pci/hda/snd-hda-intel.ko): Unknown symbol in module, or unknown parameter (see dmesg)
<8> kerframil
<10> DexterF: what are you looking for anyway?
<13> anyone can help me?
<7> johntramp: either that or apply the patch I just linked you to.
<8> kerframil, can i set the rootp*** the same as my user?
<1> DexterF: you know wether the athlon x2 supports "binding", so, that one process/thread is forced to stay on core number 0/1
<2> mbishop: patches for ati's fglrx driver, v8.22.5
<4> uein, see dmesg
<10> DexterF: most patches aren't gentoo specific, even if the gentoo project makes them, they are still sent upstream
<10> DexterF: to do what?
<12> marienz: i did sync only 5 mins ago on 2 of my servers and they synced from different mirrors and the same happens on both
<8> kerframil, would that work? untill i log onto root and change the p***
<6> andrew______: if you've got a way to do it you can set it any which way you want
<7> uein: do as it says: see dmesg. Usually this is caused by a mismatch between the kernel you're running and the module you're trying to load though.
<14> s0ulslack: you could get it from viewcvs, http://www.gentoo.org/cgi-bin/viewcvs.cgi/sys-apps/portage/Attic/portage-2.1_pre4.ebuild?rev=1.4&content-type=text/vnd.viewcvs-markup
<13> scott___: dmesg_:
<13> snd_timer: Unknown symbol snd_kmalloc_strdup
<13> snd_pcm: Unknown symbol snd_timer_notify
<13> snd_pcm: Unknown symbol snd_timer_interrupt
<13> snd_pcm: Unknown symbol snd_verbose_printk
<9> ok, so I patch the ctypes source with that? the last time i patched anything was a long time ago
<8> kerframil would that work tho?
<7> johntramp: yes.
<2> mbishop: to prevent sys freeze on logout to kdm/gdm. and yes, I googled it already. long story. hard to find.
<6> andrew______: what is is that you're proposing exactly? did you try the earlier "p***wd -d" suggestion from someone else? or are you still trying to put in a hash?
<4> uein, pastebin
<14> s0ulslack: er, http://www.gentoo.org/cgi-bin/viewcvs.cgi/*checkout*/sys-apps/portage/Attic/portage-2.1_pre4.ebuild?rev=1.3 rather :)
<7> johntramp: this is not the right place to ask for further instructions though :)
<9> marienz, ok i'll see what I can do. thanks for the help
<9> yea
<1> ciao
<2> expose: no clue about binding. I'd guess so, but I don't know
<13> sorry
<8> im still trying editing /etc/shadow
<7> mzbot: huh?
<15> TFKyle: w00t, you rock :-) thanks man
<8> hrm p***wd -d andrew
<8> what wud that do?
<12> marienz: so any idea?
<7> Bejgli: gimme a sec (box just crashed, I'm still restarting stuff)
<16> andrew______: man p***wd and find out
<12> marienz: ok, don't have to hurry
<8> i aint connected to t he box
<17> hi johnm .. are you about?
<8> :D
<8> at ****ing last
<7> andrew______: language
<8> sorry marienz
<8> now to start ftp
<8> :$
<6> andrew______: don't leave it blank though
<8> ive changed it
<17> andrew______: that's a terrible nick!
<8> i no
<8> my real nickname is taken
<7> Bejgli: there's a "Fixed patch - again!" ChangeLog entry about half an hour old. Recommend you try to sync again in a short while.
<17> couldn't you remove some of the _s?
<12> marienz: ok, thank you! will do tomorrow again
<8> :D
<8> you guys, thank you very much
<18> is upgrading to gcc-3.4.5 going to be a problem?
<12> no
<19> wepy: not if you do it properly
<18> properly = emerge -uD world?
<19> wepy: not really
<19> wepy: http://www.gentoo.org/doc/en/gcc-upgrading.xml
<20> is there a way to use dd to output to not only 1 location, but perhaps, 2?
<19> sleek: in what way?
<19> sleek: do you mean like piping it to two places?
<20> Kyuu: for example dd if=somefile of=file1,file2
<21> sleek: not sure about dd, but you can use tee for that.. somecommand | tee /some/file > /some/other/file
<20> ah, cool
<22> how does one print the lines in a file to a term? what command?
<23> pontusen: 'cat [filename]'
<18> Kyuu: i see.. apparently libtool is important
<8> hrm is there a bin in gentoo
<8> ?
<8> cuz i think i deleted sumit i needed
<19> sleek: named piping will probably do what you want
<24> does anyone know when kde 3.5 will be marked stable ?
<20> Kyuu: how does that work?
<24> (on amd64)
<8> .... is there a bin on gentoo where all the files get deleted to
<14> andrew______: not if you use rm, though gnome or kde they typically have a trash
<19> sleek: mkfifo *fifoname* p
<8> what about ssh lol
<8> where would the bin be?
<19> sleek: then have dd output to the filename you made.. and then cat filename > file1 & ; cat filename > file2 &
<25> That's odd...
<19> sleek: i think that will work
<24> andrew______, if you deleted a file with "rm" via ssh
<24> it's lost.
<24> stop.
<14> andrew______: you can't delete files using ssh itself :P
<8> or is there a restore? the file i lost
<26> andrew______: File recovery.
Return to
#gentoo
or
Go to some related
logs:
hash hack
#fedora
postfix catchall-address and virtual alias
#qemu
delete files from dvd
VLAN utilities not installed
scp largefiles
grub ubuntu autoboot
#osdev
k3b-mad source
|
|