June 2012 - kernelnewbies@kernelnewbies.org - Kernelnewbies Mailing Lists

Journey of a network packet through linux 2.6 network stack
by AFT June 19, 2012

June 19, 2012

Hi, A document with this heading can be found for 2.4 network stack. But i have not found any document regarding 2.6 network stack. When i was searching netfilter mailing list archive, i got the following post : http://lists.netfilter.org/pipermail/netfilter/2007-august/069583.html in this post author mentioned about a document named after this. but does not gave any link to that document. so i'm wondering if any similar document can be found? cheers. ------ --aft

2 1

no-op error in dmesg
by Pritam Bankar June 19, 2012

June 19, 2012

Hi, I have one structure called foo. Address of object of foo(say objfoo) is passed to ioctl call status = ioctl(fd, arg, &objfoo); But during manipulation I am not getting ioctl error 5 (input/output error). I am more confused about error in dmesg which is *foo: no-op * What does no-op indicate? If it is no-operation, how can we relate this to foo structure? Thanks, Pritam

2 1

Continuous Interrupt Problem
by siddharth saxena June 16, 2012

June 16, 2012

Hi all I need help with an issue. I have written a touch driver for a device and used the flag IRQF_TRIGGER_LOW to request irq. Now, when I boot the device, the touch interrupts are coming already without touching the screen. Continuous interrupts are occurring without any touch. -- Regards Siddharth Saxena

6 13

is a task_struct vfork_done completion special only for kthreads?
by Robert P. J. Day June 16, 2012

June 16, 2012

(what is probably the first in a plethora of silly questions as i dig into the ugly details of various kernel features. you've been warned.) clawing my way through the kernel code for kthreads and i notice the following. in kernel/kthread.c, we have: === struct kthread { int should_stop; void *data; struct completion exited; }; #define to_kthread(tsk) \ container_of((tsk)->vfork_done, struct kthread, exited) === in other words, the kthread to associated "task_struct" mapping is based on the fact that this member field in task_struct: struct completion *vfork_done; /* for vfork() */ will point at the "exited" completion object in the corresponding "struct kthread". is the above construct only true for kthreads, in the sense of, unless you've specifically created a kthread, the vfork_done pointer in task_struct is to a simple completion. but if the task is a kthread, the vfork_done pointer is *still* to a completion, but it's a completion that's encapsulated in a larger structure that's meant to be manipulated by kthread routines, obviously. as a test, i did this: === $ grep -r "container_of.*vfork_done" * kernel/kthread.c: container_of((tsk)->vfork_done, struct kthread, exited) $ === so it appears that that vfork_done field is only ever "container_of"ed with respect to kthreads and nowhere else. thoughts? rday -- ======================================================================== Robert P. J. Day Ottawa, Ontario, CANADA http://crashcourse.ca Twitter: http://twitter.com/rpjday LinkedIn: http://ca.linkedin.com/in/rpjday ========================================================================

2 1

initrd insmod: error inserting '/lib/foo.ko' : -1 operation not permitted
by leo liang June 14, 2012

June 14, 2012

Hi, I am trying to add one driver to my kernel. I have the kernel source and use make all build all the modules and bzImage. I already have initrd.img in my machine. I extract it to a tmp folder. Then I copy the driver module into the lib folder. I edited the init file to insert the module when machine boot. After that, I recreated the initrd.img by cpio and gzip. When machine boot, I got "insmod: error inserting '/lib/foo.ko' : -1 operation not permitted" error. Did I missed any step? What caused the error? Thanks.

2 2

modules in Initramfs
by beyond.hack June 14, 2012

June 14, 2012

hello experts,, I am on Fedora 16 and My current kernel is as per the uname -r output is 3.1.0-7.fc16.i686.PAE I am compiling a kernel 3.4.1 and after the compilation the bzimage and initramfs are successfully generated.. then doing a #make install is giving error like this:: sh-4.2# make install sh /usr/local/src/linux-3.4.1/arch/x86/boot/install.sh 3.4.1 arch/x86/boot/bzImage \ System.map "/boot" WARNING: Couldn't open directory /lib/modules/3.4.1: No such file or directory FATAL: Could not open /lib/modules/3.4.1/modules.dep.temp for writing: No such file or directory find: `/lib/modules/3.4.1/': No such file or directory F: No modules.builtin.bin and modules.builtin found! E: FATAL: Could not load /lib/modules/3.4.1/modules.dep: No such file or directory E: FATAL: Could not load /lib/modules/3.4.1/modules.dep: No such file or directory E: FATAL: Could not load /lib/modules/3.4.1/modules.dep: No such file or directory E: FATAL: Could not load /lib/modules/3.4.1/modules.dep: No such file or directory E: FATAL: Could not load /lib/modules/3.4.1/modules.dep: No such file or directory E: FATAL: Could not load /lib/modules/3.4.1/modules.dep: No such file or directory E: FATAL: Could not load /lib/modules/3.4.1/modules.dep: No such file or directory E: FATAL: Could not load /lib/modules/3.4.1/modules.dep: No such file or directory E: FATAL: Could not load /lib/modules/3.4.1/modules.dep: No such file or directory E: FATAL: Could not load /lib/modules/3.4.1/modules.dep: No such file or directory E: FATAL: Could not load /lib/modules/3.4.1/modules.dep: No such file or directory E: FATAL: Could not load /lib/modules/3.4.1/modules.dep: No such file or directory E: FATAL: Could not load /lib/modules/3.4.1/modules.dep: No such file or directory E: FATAL: Could not load /lib/modules/3.4.1/modules.dep: No such file or directory E: FATAL: Could not load /lib/modules/3.4.1/modules.dep: No such file or directory E: FATAL: Could not load /lib/modules/3.4.1/modules.dep: No such file or directory E: FATAL: Could not load /lib/modules/3.4.1/modules.dep: No such file or directory E: FATAL: Could not load /lib/modules/3.4.1/modules.dep: No such file or directory E: FATAL: Could not load /lib/modules/3.4.1/modules.dep: No such file or directory E: FATAL: Could not load /lib/modules/3.4.1/modules.dep: No such file or directory E: FATAL: Could not load /lib/modules/3.4.1/modules.dep: No such file or directory E: FATAL: Could not load /lib/modules/3.4.1/modules.dep: No such file or directory find: `/lib/modules/3.4.1/': No such file or directory find: `/lib/modules/3.4.1/': No such file or directory find: `/lib/modules/3.4.1/': No such file or directory E: FATAL: Could not load /lib/modules/3.4.1/modules.dep: No such file or directory E: FATAL: Could not load /lib/modules/3.4.1/modules.dep: No such file or directory sh-4.2# 2.Do all the modules needed by the kernel(specified at the configure time) are built into the initramfs compressed image?? thanks in advance...plz. bear with me if this is a real simple thing...but I am just starting out.... Regards

3 3

is memcpy a system call?
by J.Hwan Kim June 12, 2012

June 12, 2012

Hi, everyone Is memcpy a system call? Thanks in advnace, Best Regards, J.Hwna Kim

6 6

confusion about reading time from cmos
by 王哲 June 12, 2012

June 12, 2012

Hello everyone: env: Ubuntu 11.04 I write a program reading time from cmos,and the program is as followed: #include <stdio.h> #include <sys/io.h> unsigned char cmos_read(unsigned char addr) { unsigned char tmp = 0; iopl(3); outb(addr,0x70); tmp = inb(0x71); iopl(0); return tmp; } int main(void) { printf("seconds = %x\n",cmos_read(0)); printf("minute = %x\n",cmos_read(2)); printf("hour = %x\n",cmos_read(4)); system("date"); return 0; } and run the program and the result is as followed: wanny@wanny-C-Notebook-XXXX:~$ sudo ./a.out seconds = 9 minute = 42 hour = 7 2012年 06月 12日星期二 15:42:09 CST question: why the second is right,minute is right , but the hour is different? Thanks in advance!!

2 2

Regarding ext4 allocation and deallocation of blocks
by Akshay Nehe June 11, 2012

June 11, 2012

Hi, I've been trying to understand ext4 block allocation and deallocation but I've not been able to figure it out completely and hopefully someone else knows better so please help me on this. I'm using kernel 2.6.35.6 and I have the problem as we try to modify file on ext4 file system based partition, system deallocates all previous blocks and allocate new blocks with modified data, so with this behaviour system also changes inode number of file. As I see it in the code, the ext4_free_blocks function is used to deallocate the blocks. I have read and understood this behaviour, now I want to protect some blocks from deallocation i.e. not allowing blocks say 8705 from deallocation, Is it possible? If yes what is the correct way to do so. I have written a module to find out block numbers used by the file before modification( lets take a.txt with block number as 8705), now if I modify file surely new block will be allocated(say 9705) I want to protect 8705 from deallocation, how to do it, if possible? I was trying to put code in ext4_free_block function to avoid deallocation of block 8705 but when I insert module, I am not able to reach code in ext4_free_blocks. I am observing following Behaviour, Initially file(a.txt) with block number 8705 then I inserted my module so came to know block number. Then I open file for modification so after modification new block was 9217. But I was not able to protect 8705 block, as kernel should deallocate it using ext4_free_blocks function, but this was not the case. I again open file for modification( file with 9217 block) and new block is 8708 but this time kernel used ext4_free_blocks to deallocate 9217 block. What is the reason behind this behaviour? Please help me in this issue, if anyone familiar with such behaviour. -- Regards, Akshay Nehe.

2 1

Delete IPv6 address
by Rohan Doshi June 11, 2012

June 11, 2012

Hi, I am trying to modify host ID generation process for IPv6 addresses. Currently it generates host ID using ipv6_generate_eui64() in addrconf.c and start DAD for newly generated address. If DAD fails, it calls addrconf_dad_failure() and disables IPv6 on the interface using idev->cnf.disable_ipv6=1. I have used different algorithm for generating the host ID. What I want to do now is that if DAD fails, it should remove previously generated IPv6 address (host ID) and start over. Any idea how to do that? -- Rohan D. Doshi

1 0