February 2011 - kernelnewbies@kernelnewbies.org

About EXPORT_SYMBOL
by kashish bhatia Feb. 10, 2011

Feb. 10, 2011

Hello All, I want to export a function named ext2_free_data() which is present in inode.c . This function is statically defined in this file. I omit the "static" keyword and exported it using EXPORT_SYMBOL(ext2_free_data). ext2_free_data() function declaration is not present in any header file , so I included its declaration in my own header file which I kept in fs/ext2 directory. Now when I insert my module using insmod command ,it shows a error : Unknown symbol in module. Kindly help me . -- Regards, Kashish

4 5

Any known memory leak issues in handling SIGEV threads in Linux?
by Sowmya Sridharan Feb. 10, 2011

Feb. 10, 2011

Hi All, In a stable system with 2.6.14 kernel, which continuously sends/receive packets, I noticed an increase in slab size of around 1.5GB in 30 hours. I see a continuous increase in number of task_struct and size-16384(thread info) object. I use an application that uses SIGEV timer thread with time-out of 20 millisecs. Each time, upon expiry of SIGEV timer(20 millisecs), a thread start routine will run "as if" it were a new thread, to process the packets in the system. Whenever the application is killed/restarted, all the task_struct objects are getting released. 1. Is there any known issue in 2.6.14 kernel for handling SIGEV timer thread ? 2.Will there be any issue in 2.6.14 kernel, in clean up of task_struct for application that creates/deletes threads at a rate of 20 millisec? Thread creation/deletion are handled properly in the application. Still I am unable to figure out reason for memory leak. Appreciate your inputs. Thanks, Sowmya =====-----=====-----===== Notice: The information contained in this e-mail message and/or attachments to it may contain confidential or privileged information. If you are not the intended recipient, any dissemination, use, review, distribution, printing or copying of the information contained in this e-mail message and/or attachments to it are strictly prohibited. If you have received this communication in error, please notify us by reply e-mail or telephone and immediately and permanently delete the message and any attachments. Thank you

2 1

Fwd: Permanent Delivery Failure
by mohit verma Feb. 10, 2011

Feb. 10, 2011

---------- Forwarded message ---------- From: MDaemon at fibcom.com <MDaemon(a)fibcom.com> Date: Thu, Feb 10, 2011 at 11:31 AM Subject: Permanent Delivery Failure To: mohit89mlnc(a)gmail.com The attached message had PERMANENT fatal delivery errors! After one or more unsuccessful delivery attempts the attached message has been removed from the mail queue on this server. The number and frequency of delivery attempts are determined by local configuration parameters. YOUR MESSAGE WAS NOT DELIVERED TO ANY OF IT'S RECIPIENTS! --- Session Transcript --- Thu 2011-02-10 11:31:36: Parsing message <xxxxxxxxxxxxxxxxxx\pd50001467261.msg> Thu 2011-02-10 11:31:36: * From: mohit89mlnc(a)gmail.com Thu 2011-02-10 11:31:36: * To: amit(a)fibcom.com Thu 2011-02-10 11:31:36: * Subject: module prob Thu 2011-02-10 11:31:36: * Message-ID: <AANLkTi= TTnOfZubXZARJByf8n82r4pXp1shg1j-JsfZR(a)mail.gmail.com> Thu 2011-02-10 11:31:36: * Route slip host: 202.71.129.104 Thu 2011-02-10 11:31:36: * Route slip port: 25 Thu 2011-02-10 11:31:36: Attempting SMTP connection to [202.71.129.104] Thu 2011-02-10 11:31:36: Attempting to send message to smart host Thu 2011-02-10 11:31:36: Attempting SMTP connection to [202.71.129.104:25] Thu 2011-02-10 11:31:36: Waiting for socket connection... Thu 2011-02-10 11:31:36: * Connection established (192.5.1.2:2559 -> 202.71.129.104:25) Thu 2011-02-10 11:31:36: Waiting for protocol to start... Thu 2011-02-10 11:31:36: <-- 220 smtp ready smtp.bizmail.net4india.com Thu 2011-02-10 11:31:36: --> EHLO fibcom.com Thu 2011-02-10 11:31:36: <-- 250-smtp.bizmail.net4india.com Hello fibcom.com [203.101.81.201] Thu 2011-02-10 11:31:36: <-- 250-SIZE 57671680 Thu 2011-02-10 11:31:36: <-- 250-PIPELINING Thu 2011-02-10 11:31:36: <-- 250 HELP Thu 2011-02-10 11:31:36: --> MAIL From:<mohit89mlnc(a)gmail.com> SIZE=15160 Thu 2011-02-10 11:31:36: <-- 250 OK Thu 2011-02-10 11:31:36: --> RCPT To:<amit(a)fibcom.com> Thu 2011-02-10 11:31:36: <-- 550 relay not permitted --- End Transcript --- -------------------------------------- LEGAL DISCLAIMER -------------------------------------- The information contained in this message may be privileged and confidential and protected from disclosure. If the reader of this message is not the intended recipient, or an employee or agent responsible for delivering this message to the intended recipient, you are hereby notified that any reproduction, dissemination or distribution of this communication is strictly prohibited. If you have received this communication in error, please notify us immediately by replying to the message and deleting it from your computer. This mail has been checked for viruses. However recipients should undertake their own virus check. We will not be liable for any losses Thank you. Fibcom India Limited ---------- Forwarded message ---------- From: mohit verma <mohit89mlnc(a)gmail.com> To: kernelnewbies(a)kernelnewbies.org, kernelnewbies < kernelnewbies(a)nl.linux.org> Date: Thu, 10 Feb 2011 11:31:32 +0530 Subject: module prob hi all, i have written a tiny module . when i install this one i check in dmesg and /proc/kallsyms. these files shows that the module is currently running . but when i go for "ohm_inode_permission" kernel security function in /proc/kallsyms it does not show that security operation but that must be registered along with the registration of module. i 've attached the source code along with........ i think the problem is while registering the security function ohm_inode_permission (IMHO). but.................. can someone tell me where am i wrong? -- ........................ *MOHIT VERMA* _______________________________________________ Kernelnewbies mailing list Kernelnewbies(a)kernelnewbies.org http://lists.kernelnewbies.org/mailman/listinfo/kernelnewbies -- ........................ *MOHIT VERMA*

2 1

Ad-hoc arm board
by Andrea Gasparini Feb. 10, 2011

Feb. 10, 2011

Hi, Working on a custom arm based board, I'm currently juggling with mach-types and machine IDs for our ARM boards, but perhaps it came the time to set them in the right way... And moreover, do anyone has a pointer on documentation/articles/somewhat on how to write a good board-*.c and how (and perhaps when) to handle platform devices instead of a classical device driver? I'm currently going by trying, and perhaps it's not the better method to achieve my purposes. thanks. Regards. -- -gaspa- ----------------------------------------------- -------- https://launchpad.net/~gaspa --------- ----- HomePage: http://gaspa.yattaweb.it ------

3 3

module prob
by mohit verma Feb. 10, 2011

Feb. 10, 2011

hi all, i have written a tiny module . when i install this one i check in dmesg and /proc/kallsyms. these files shows that the module is currently running . but when i go for "ohm_inode_permission" kernel security function in /proc/kallsyms it does not show that security operation but that must be registered along with the registration of module. i 've attached the source code along with........ i think the problem is while registering the security function ohm_inode_permission (IMHO). but.................. can someone tell me where am i wrong? -- ........................ *MOHIT VERMA*

1 0

Linux kernel network stack
by Sri Ram Vemulpali Feb. 10, 2011

Feb. 10, 2011

Hi all, I am looking to start Linux kernel network internals. I know a good book "understanding Linux Network Internals". I need something very quick to start with and remain it as reference in by library. I had gone through some online resource, which gave me very good start to work on. Now I want to delve little deeper in to kernel code on network internals side. Especially I want to learn step by step data path in kernel to user space, and how various protocols and wireless stack glued together. Can anyone please point me to moderate explanation book or online resource. Thanks in advance -- Regards, Sri.

2 1

fsck for NFS
by ridgetwo＠earthlink.net Feb. 9, 2011

Feb. 9, 2011

Hello, Is is possible to run fsck against NFS (in the unmounted state)? Thanks, A.Krishnan

3 2

Getting struct page for userspace pointer
by Nikolaus Rath Feb. 9, 2011

Feb. 9, 2011

Hello, Is there a way to get the corresponding struct page* for an userspace pointer *if it exists*? I am basically looking for a "passive" version of get_user_pages() that does not lock the page and does not put the page into memory if it's swapped out. Thanks, -Nikolaus -- »Time flies like an arrow, fruit flies like a Banana.« PGP fingerprint: 5B93 61F8 4EA2 E279 ABF6 02CF A9AD B7F8 AE4E 425C

1 0

Fwd: storing the logs
by mohit verma Feb. 9, 2011

Feb. 9, 2011

thanks greg,, a silly question........please have a look....... i create a socket to communicate to the user space app to listen to and responsible for loggin ... via my_sock=netlink_creat_sock() i create my own logging message format to send it via netlink interface. then how can i let my module know which process to bind it to? i mean to bind my socket to any user space application i need the pid of that one ,so at the very first time how can i initialize the binding process with user space application? i hope.. i am clear to u. please let me know if not....i wll try again. i 'd really appreciate any help.

3 3

Doubt regarding Minor Numbers and alloc_chrdev_region
by Sankar P Feb. 8, 2011

Feb. 8, 2011

Hi, I am going through LDD3 (scull0 section) and have a doubt regarding minor numbers. I have the following code-snippet: ============================================ #define FIRST_MINOR_NUMBER 0 #define COUNT_OF_CONTIGOUS_DEVICES 1 /* Device that we will use */ dev_t helloworld_device; /* Initialization Function for the kernel module */ static int helloworld_driver_init(void) { printk(KERN_INFO "Hello World"); /* Allocate a dynamic device number for your driver. If you want to hardcode a fixed major number for your * driver, you should use different APIs 'register_chrdev_region', 'MKDEV'. But the following is better. */ if (!alloc_chrdev_region(&helloworld_device, FIRST_MINOR_NUMBER, COUNT_OF_CONTIGOUS_DEVICES, "HelloWorldDriver")) { printk(KERN_INFO "Device Number successfully allocated."); /* If you do a cat /proc/devices you should be able to find our Driver registered. */ } else printk(KERN_ALERT "HelloWorldDriver could not get a Device number"); return 0; } ====================================== Now if I compile and insert the .ko file for this module, the driver gets registered with a major number of 250 (seen from /proc/devices). However, I don't understand the significance of the FIRST_MINOR_NUMBER and the COUNT_OF_CONTIGOUS_DEVICES that we pass to the alloc_chrdev_region function. If I try to create a file using, 'mknod /dev/scull2 c 250 7' , the character device file gets created with a minor number of 7. I have set COUNT_OF_CONTIGOUS_DEVICES to 1. However, I am able to create multiple files /dev/scull[1-10] using this driver and mknod. So, what is the significance of these two variables (first minor number and count of devices) that we pass to the alloc_chrdev_region api, if they can be over-ridden on mknod time ? Or have I misunderstood something ? Any help will be appreciated. Thanks. -- Sankar P http://psankar.blogspot.com

3 4