| |
| |
| |
|
Page: 1 2
Comments:
<0> hi, could anyone tell me what this process is? [migration/1/1] I see several of them with different #/#'s
<1> bare_foot: Could?Perhaps;Will?Probably Not.
<2> I compiled the 2.6.15.4 kenel today. the experience was great. I can't wait till there's a new kernel version :) when is a new version expected
<2> ?
<0> did I miss the magic word? :)
<0> please
<0> im not asking what exactly is running that process, more like that what process does.
<3> heyo
<3> is there _some_ way to enforce that multiple threads that wait on a mutex/signal are awakened in the same order as they put themselves on wait?
<3> if possible, which is portable?
<4> bare_foot: is that smp kernel?
<0> yes
<4> bare_foot: i might be wrong but search "migration" in http://www.samspublishing.com/articles/article.asp?p=101760&seqNum=5
<0> hmm, so if there is alot of activity, maybe a bad cpu would cause alot of migration processes?
<5> you guys know when you cat /proc/acpi/battery/BAT1/state, and it pulls information like remaining capacity, is there a way to 'just access' the remaining capacity?
<6> In 2.6.13.2, I get this error "<1>Unable to handle kernel NULL pointer dereference at virtual address 00000110" with in the call trace: wake_bit_function and do_generic_mapping_read. What would the source of the problem be?
<6> If I try to sync(1), I get a further: "kernel BUG at lib/radix-tree.c:592!"...
<6> This is with an ext3 file system.
<1> baaren't you lucky
<6> I never had any problem with ext3 (much less than reiserfs).
<1> hehe same
<6> Ok, I'll reboot -n, and try to make a new file system... (it's not the first time and this one is fsck'ed to death).
<1> restore frombackup :)
<1> you sure itsfile system n not hardware?
<1> in the past I've had disks cause ext2 issues for which no amount of fscks or mkfs's would fix them
<6> smartctl doesn't report any problem. It's a RAID-1.
<1> hmm ok, maybe it's time you went raid5 :)
<1> lets you plug in countless usb2 n 1 n letsyou use it like pnp drives
<1> be nice too play with raid on it, especially if like meall your hardware is non usb2.0compliant
<7> hello, i want to delieve the error and data from driver layer to applicaion layer in socket programming,how?
<1> mm
<8> Hello, can linux modules define new system calls in 2.6?
<9> Jack-Bauer, TheBonsai_ gb morning
<7> hello is there anyone familiar with network driver?
<1> gday _RADIOhead
<10> Is the kernel patch 2.6.14 cumulative - i.e. the same as all patches 2.6.13.x together?
<9> gb, dude
<10> was that "yes" or "no"?
<11> i have a question concerning the socket state flag in the socket structure ...
<11> Is it possible to disconnect / release a socket in state SS_NOTCONNECTED ?
<11> when i use the shutdown method in "proto_ops" i get a NOCON (107) error code
<1> mpg: a patch to kernel.x is an upgrade from kernel x-1 too kernel.x
<1> you work it out for yourself from there
<10> gb, uhm, I knew that. The question was different
<1> uhuh
<10> and the answer seems to be: yes, 2.6.x+1 is cumulative - 2.6.x.a + 2.6.x.b + ...
<1> mpg: whatever floats your boat
<10> one more question : if I set up 2 IP adresses for eth0 - one using C-type network adresses and other one using B-type network adresses - do I need to do something special with samba to see machines on both networks?
<1> mpg: yes
<1> Please refer samba configuration questions too #samba
<10> gb, you are right. #samba seems to be dead right now, but I'll try later. thanks.
<1> mpg: of course, alternatively you might want too consider swig or webmin
<1> I personally wouldn't, even when It was my job, but I've heard they're both good alternatives too writing a configuration file from scratch.
<10> I do have a working configuration file. I just wanted to extend the functionality of my machine - make it talk to another network - by "ip address add x.y.z.q/24 dev eth0". The TCP/IP level works well. But samba seems to have a problem. From the comments in smb.conf and in man page it should work. But it does not.
<10> I wonder if the problem can be that samba is using obsoloete aproach - similar to ifconfig being replaced by ip (from iproute2)
<12> iproute2 - that? cool :)
<13> Hi, using alsamixer or kde kmix the master volume doesn't work instead the headphones do. In all groups they told me it has to do with the driver. All I did is activate the driver CONFIG_SND_AC97_CODEC=y and CONFIG_SND_AC97_BUS=y. Any idea?
<14> Hi @ all. I have a little trouble, with programming a kernel block driver. The driver is for mmc/sd cards on mips hardware (adm5120). The mmc card is connected with gpio pins.
<14> My problem is, only if i add a printk("X") statement in the function, that do the gpio stuff (get/set the serial bits) the driver works
<14> without the printf the sampled bit's are not correct, What is the problem here ? Have someone a idea ?
<14> if i change the printf with a simple delay, it doesn't work
<14> Kernel is 2.4.32
<14> is noone here, or as nobody an idea ?
<12> you can?t use printf - it?s a c-library function.....you may only use printk for debugging
<12> am I right? :)
<14> sorry, is printk !
<14> yes, aou are right !
<14> you
<12> so...what your question is?
<14> if i add a printk statement in my io funtion for set outport bit and sample inport bits the driver works
<14> if i remove the printk statement, the driver work not. They samples the inport bits not right
<14> but why ?
<12> what dmesg says?
<12> the kernel module isn?t loade?
<12> or it?s crashe when your io function called?
<14> no, the kernel mod is loaded, since i see, the work of the driver (printk), dmesg does not have to do anything thereby
<14> no, the module don't crash
<14> it is running, but in a false way. I read directly registers in the cpu (gpio registers) to get the state of the gpio pin
<14> and this sampling is not correct if i don't use the printk statement inside
<14> it is possible, that the sampling is breaked by a interupt ?
<14> do printk disable the interrupts ?
<12> may be..
<14> where can i find good info about driver development for kernel 2.4 ?
<14> any idea ?
<12> http://www.oreilly.com/catalog/linuxdrive2/
<14> tanks
<14> thanks
<12> np )
<15> hi all
<15> ci sono italiani?
<16> hello
<16> i want to permit the bootsplash to be load at the same time of the kernel
<16> do you know where i need to add the corresponding line ?
<17> So what is the fundamental difference between tty and ttyS?
<18> ttyS is serial iirc
<19> if the fstab location determined inside the kernel or outside?
<19> ...
<20> outside
<20> cblack: i think it's hardcoded in the /sbin/init executable
<18> ok.
<20> well, depending on what /sbin/init you're using
<18> yeah
<19> thx
<19> I'll look into that
<20> /sbin/init can be replaced by an sh script or anything you want
<18> Overall Grade .1% A
<18> - Category: Test & Quiz .5% B+
<18> - Category: Homework .0% A+
<18> - Completed Work Avg. .1% A
<18> :-)
<19> well I don't know what else it does... I just need a way to boot from two different fstab files without the overhead of having different partitions for them
<18> cblack: umm
<20> hmm
<19> myea I'm outta ideas
<20> cblack: i'd probably leave fstab empty then, and add the real mounting to some script in /etc/init.d
<20> to avoid wirting your own /sbin/init
<18> cblack: yeah, what kind of differences are in the fstabs?
<19> just harddrives, one boots from sdb, one from hdb (weird setup, I know)
<18> two different partitions, I'd say... thats the least overhead
<18> one boot partition
<18> and supply the root as sdb or hdb at boot time
<19> sometimes the system gets loaded through vmware, it's setup to autodetect all the hardware right, depending on if it raelly boots or not, but this part is tricky
<18> then on those disks you have the /etc/fstab
<19> yeah, that'd work
<19> I'll give it a shot, thx
<18> that was simple...
<19> I'm not that good at scripting but I'll see what I can do
<19> at least I have an idea what way I'm heading in now
<20> the root partition doesn't need to be in /etc/fstab
<19> eh, I define it in grub too, but removing that doesn't change anything as far as I can tell anyway
<19> does it just boot without the root line in fstab?
<20> root fs is mounted before /etc/fstab is even accessible to the kernel
<19> isn't it mounted ro first? (not sure)
<20> it's either hardcoded in the kernel or taken from the "root" parameter
<19> must be hardcoded, I tried commenting out the root parameter and it boots fine
<21> Hi
<22> i have a surprise for you guys.
<22> Linus Torvalds built the linux kernel off of AmigaOS.
<12> explain plz)
<23> that's lies, linus didnt build the kernel off of amigaOS
<23> he roundhouse kicked a computer and it appeard
<23> then he ate 10 newly born babies
<20> AmigaOS is opensource?
<21> Hi, my labtop freez on ACPI: PM-Timer IO when booting, no kernel panic
<24> hi
<24> i want to know if is true that linux use the algoritm of Ostrich for deadlocks ?
<24> Ostrich's Algoritm
<24> mmm?
Return to
#kernel
or
Go to some related
logs:
#debian
gentoo disable pcspkr
#math
gnomebaker power calibration failed
#physics
print_devices(): didn't get a reply from NetworkManager.
#web
#ubuntu
#perl
dante block tcp/accept
|
|