- kernelnewbies@kernelnewbies.org - Kernelnewbies Mailing Lists

books and pointers...
by Srinidhi Rao Aug. 20, 2011

Aug. 20, 2011

Hello Kernel Hackers, As you can notice I am a kernel newbee!, I believe this is the platform to ask any thing about the Linux kernel. Well, I am extremely interested to know more about kernel programming especially in the networks and the PCI subsystems, I reall want to know where and how to start, I have some basic knowledge of IPC and I want to build on this... Kindly guide me with our valuable pointers to start with. -- Thanks and Regards, Srinidhi. ---- All our knowledge has its origins in our perception. -- Leonardo da Vinci. ---- Tell me and I forget. Teach me and I remember. Involve me and I learn. -- Benjamin Franklin.

2 1

Port Linux multiprocessor support
by Giorgis Georgakoudis Aug. 20, 2011

Aug. 20, 2011

Hello all, I would like to ask for some pointers regarding porting Linux kernel multiprocessor support in new platforms. More specifically, what are the HW requirements that a platform must have to enable Linux multiprocessing and some info on kernel internals with respect to the booting process, scheduling/migration etc. I've been looking around for relevant documenta- tion but found little to none. Best regards, Giorgis

2 3

How to measure the performance hit of changes in kernel code
by john k Aug. 20, 2011

Aug. 20, 2011

What is the optimal way to measure the performance hit a change in the kernel brings to the system under normal conditions? Are there tools for this job or a standard procedure kernel developers follow or everyone comes up with his own metrics? I know it sounds vague and maybe the measurement must be related with where the change is on the kernel code. Say for example that I implemented some extra security checks in the copy_from_user function, or some other critical code. Whats the best way to measure the performance hit it causes? I have used measurements of the time a full kernel compile takes, but I dont think this is the best metric. Also, is there a test suite that somehow emulates actions of an average user or another for an average web server? So that you can take average measurements etc (In trying to establish an overall performance hit, so that the metric may be unrelated to the part of the kernel code that is affected)

2 1

multiple dependent out-of-tree modules
by Jorgyano Bruno Aug. 20, 2011

Aug. 20, 2011

Hello guys! I'm working with three out-of-tree kernel modules, one of them export symbols used by the others. At compile time, MODPOST warns about undefined symbols, which are defined on the first module and used in the second and third modules. When I try to load the modules* "insmod" *returns the error "Unknown Symbol in module" for those symbols warned by MODPOST. How can I to compile several out-of-tree modules which depends on one another and make to the dependant modules look for the needed symbols on the directories of the modules which have the symbols? I have found a similar question here http://www.spinics.net/lists/newbies/msg22926.html, but unanswered. ------------------------- Jorgyano

3 3

Re: ohci_hcd leak ed
by DiP Aug. 19, 2011

Aug. 19, 2011

I found where it is. It seems to appear on my desktop box when I connect cams to the same bus. Two connectors next to each other. But on the notebook it does matter. Any bus, any port. Problem still alive with a kernel 3.0.2-pf-di #1 SMP PREEMPT Fri Aug 19 00:21:22 EEST 2011 x86_64 Intel(R) Core(TM)2 Duo CPU E8200 @ 2.66GHz GenuineIntel GNU/Linux //D

2 4

Is it possible to relocate a kernel page in physical memory?
by Arvid Brodin Aug. 19, 2011

Aug. 19, 2011

I want to do memory checks (à la memtest86) on a running system. I know using ECC memory would be a better approach, but unfortunately that is not an option in this case. Also, this seems like an excellent opportunity to learn more about the memory system in Linux. This is on a AVR32 embedded system, BTW. One approach is to use the AVR32's built-in SRAM pages to store and run kernel code that periodically takes a spinlock and checks a small part of the physical memory, before writing back its original value. This has numerous problems, among others that the SRAM pages are also used by the power management code on these processors. And of course, it's not exactly portable... Another approach is to run the test as a part of the normal kernel. For memory allocated by user processes, this should be pretty straight-forward (I think): just copy the page data to another physical page and modify the page tables (right?). When in comes to the kernel, I might be able to skip testing of the (read-only) text segment by doing md5summing over it instead (would this have any drawbacks)? But how would one test writable pages allocated by the kernel? Is there a way to "relocate" a kernel page in physical memory? -- Arvid Brodin Enea Services Stockholm AB

2 3

Is it needed to reinstall custom kernel every time, after rebuilding?
by Amirali Shambayati Aug. 19, 2011

Aug. 19, 2011

Hi all, Last time, I emailed to linux-kernel and kernenewbies mailing-lists about how to speed up kernel compilation. I received very useful peace of advice about using "ccache". Now I want to modify and compile a unique custom kernel several times. I used https://help.ubuntu.com/community/Kernel/Compile to compile custom kernel. This tutorial uses "kpkg" command to build kernel and it produces two .deb files: "linux-image" and "linux-header" as result. So for the first time I followed this tutorial. For next times, according to to experts' advice, I use "make all" in custom kernel root directory to rebuild kernel. I am not sure that it would be enough to use changed kernel after rebooting. As I asked from some experts, It is also required to install kernel again, but they were not familiar with "kpkg" and "dbkg",... commands for kernel installation. Should I use "dbkg" command after every time I rebuild kernel, for re-installation? If anybody is familiar with these method of kernel installation, please help me. It is very urgent, because it is related to my B.S. thesis :P Regards, -- Amirali Shambayati Bachelor Student Computer Engineering Department Sharif University of Technology Tehran, Iran

3 4

Re: Is it needed to reinstall custom kernel every time, after rebuilding?
by Mulyadi Santosa Aug. 19, 2011

Aug. 19, 2011

On Fri, Aug 19, 2011 at 01:02, Amirali Shambayati <amirali.shambayati(a)gmail.com> wrote: > Would you say, I got what you mean right or not? Because I have very > restricted time to present my BS. thesis:-s I am not really sure I really catch what you mean by "reinstall" :) IMO reinstall is only a proper term when related to package management. So, I think if you have to use the term "reinstall", perhaps you confuse it with package management ? Sorry, maybe you're under pressure but I can't help much with that. -- regards, Mulyadi Santosa Freelance Linux trainer and consultant blog: the-hydra.blogspot.com training: mulyaditraining.blogspot.com

1 0

ohci_hcd leak ed
by DiP Aug. 18, 2011

Aug. 18, 2011

Hi all When I take my two sonix-based webcams to my box, and started vlc on both I got <code> Aug 19 02:14:45 mixer kernel: [ 558.384176] ohci_hcd 0000:00:0b.0: leak ed f53dd580 (#81) state 2 [repeated many times...] Aug 19 02:14:45 mixer kernel: [ 558.406651] gspca: no transfer endpoint found </code> That with standard archlinux kernel 3.0 Linux mixer 3.0-ARCH #1 SMP PREEMPT Sat Aug 6 16:49:00 CEST 2011 i686 Mobile AMD Sempron(tm) Processor 3800+ AuthenticAMD GNU/Linux Also I got almost the same with my other box (kernel is hand-compiled) Linux di 3.0.1-pf-di #10 SMP PREEMPT Wed Aug 17 19:23:37 EEST 2011 x86_64 Intel (R) Core(TM)2 Duo CPU E8200 @ 2.66GHz GenuineIntel GNU/Linux Aug 18 23:36:37 di kernel: gspca: no transfer endpoint found Aug 18 23:36:38 di kernel: gspca: no transfer endpoint found Aug 18 23:37:22 di kernel: gspca: no transfer endpoint found Aug 18 23:37:23 di kernel: gspca: no transfer endpoint found With that (hand-compiled kernel) Linux belu4ka 2.6.38-pf7-nusq #1 SMP PREEMPT Thu May 5 11:27:48 EEST 2011 i686 Mobile AMD Sempron(tm) Processor 3800+ AuthenticAMD GNU/Linux I got Aug 19 00:30:13 belu4ka kernel: [ 353.140330] gspca: bandwidth not wide enough - trying again Aug 19 00:30:13 belu4ka kernel: [ 353.409184] ohci_hcd 0000:00:0b.0: leak ed f560c400 (#81) state 2 modules: gspca_sonixb 10863 0 gspca_main 19539 1 gspca_sonixb webcams: idVendor=0c45, idProduct=60b0 idVendor=0c45, idProduct=6029 The problem persists only when I make vlc show me both cams in the same time. Video from the first camera started still displayed, but the second cam is switched off. I wonder if somebody tell me if I should bugreport that and where to do that? I didn`t found anything similar to my problem in linux-usb, linux-video and linux-kernel lists? //D

1 0

Proc fs Guide
by Ezequiel García Aug. 18, 2011

Aug. 18, 2011

Hi, Pardon for the newbie question, but since this is newbie forum anyway, here it goes Is this guide http://buffer.antifork.org/linux/procfs-guide.pdf (2001) updated (or outdated)? Is proc suffering any movement? I heard the prefered way to comunicate is sysfs, but I am working on a dated embedded environment and sysfs is not present. Btw, I am into that because I need to know the overall thread count. Greetings!

1 0