hello,
i wrote a driver for a pci board an created a file in sysfs,
which i use for communication from userspace. so far
the path to the file was hardcoded in my userspace
client but now i want it to be independent of whatever
system and pci slot this card is installed on. it seems
like no matter
You could read "OReilly.Understanding.the.Linux.Kernel.3rd.Edition.Nov.2005".
The name says everything =D.
A direct link:
http://eduunix.ccut.edu.cn/index/html/linux/OReilly.Understanding.the.Linux.Kernel.3rd.Edition.Nov.2005.HAPPY.NEW.YEAR/
On Dec 7, 2007 5:14 AM, Amogh Hooshdar <[EMAIL PROTECTED
1)Can read the appendix of the book Understanding Linux Kernel 3rd Ed as
chap said erlier. However if you just want to print a message while booting,
just follow your intuition in the / directory.You just need to add a line
with your text somewhere ;-).
2)Can start here , though you will need
On Fri, 7 Dec 2007 17:23:17 +0100
[CC'ing Rik]
Thomas Petazzoni <[EMAIL PROTECTED]> wrote:
[...]
Okay, is something wrong somewhere?
Why am i getting mails out of order?
I got Thomas's reply before the parent post, see mail headers -
From: Thomas Petazzoni <[EMAIL PROTECTED]>
To: kernelnewbies@
I went through this and it started me in the right direction:
http://www.freesoftwaremagazine.com/articles/drivers_linux
On Dec 7, 2007 1:14 AM, Amogh Hooshdar <[EMAIL PROTECTED]> wrote:
> Hi,
>
> I have knowledge of assembly language. I am good at C programming. I
> have learnt to write "hello w
Hi Thomas,
> As far as I've understood the SLUB development, I'd say the SLUB
> intends to be a drop-in replacement for SLAB, usable in all situations,
> with performances at least equal to the ones of SLAB, but with a better
> scalability.
Yes, I intended it in this way, too.
But I read some d
Le Fri, 7 Dec 2007 19:58:26 +0530,
"Adil Mujeeb" <[EMAIL PROTECTED]> a écrit :
> void fastcall __lock_page(struct page *page)
>
> {
>
> DEFINE_WAIT_BIT(wait, &page->flags, PG_locked);
>
> __wait_on_bit_lock(page_waitqueue(page), &wait, sync_page,
> TASK_UNINTERRUPTIBLE);
>
> }
Hi All,
In Oreilly book "Understanding Linux kernel", following text written under
section "15.1.3.1. Finding a page":
"The lock_page( ) function, in turn, blocks the current process if the page
is already locked. To that end, it invokes the _ _wait_on_bit_lock( )
function on the PG_locked bi
Hi
> This function usually sleeps a couple of times, so it may change cpu.
> If get_cycles is called the first time (function start) on one cpu and the
> second one on another one, will this difference of get_cycle values be
> meaningful?
IMHO, they are getting the same "jiffies" variable (or TSC
Le Thu, 6 Dec 2007 11:54:47 +0100,
Andrea Gasparini <[EMAIL PROTECTED]> a écrit :
> there's someone with experience on Slub/slab performance?
> I would understand when to use the first, when the latter...
>
> I found that on numa architecture is surely bettere SLUB.[1]
> But this is a rare case,
Hallo,
thanks for the answer. yes you're right, it is just because i called
enable_irq() function more than one time without any disable_irq() function.
it works now. btw, i am using 2.4 kernel on PPC target.
regards,
On Dec 7, 2007 2:17 PM, Thomas Petazzoni <[EMAIL PROTECTED]> wrote:
> Le Fri
Hi
I am a newbie too and I faced this problem a few months back when I
was trying to understand the process of booting. I got two articles
from an IRC channel which helped me a lot.
1. http://en.skelix.org/skelixos/tutorial01.php
2. http://susam.in/articles/boot-sector-code.php
Hope these articl
Le Fri, 7 Dec 2007 12:00:35 +0100,
"Leo Hendrawan" <[EMAIL PROTECTED]> a écrit :
> i am working on a project with linux on embedded platform (MPC5200
> processor).
> i have this problem that the kernel says "enable_irq(some_number)
> unbalaced". what does it mean?
> i tried to google but haven't f
Le Fri, 7 Dec 2007 13:29:06 +0330,
"Iman Darabi" <[EMAIL PROTECTED]> a écrit :
> i'm trying to find some tiny modules
> searched personal pages , blogs and ... , but can't find useful
> samples . do you have any link , sample , even your hobbies (or what
> ever you think which is good for newbie)
On Fri, Dec 07, 2007 at 04:20:45PM +0530, balzar tondo wrote:
> i am not sure whether this is the correct forum to ask. i am a newbie to
> this forum and please pardon me if I violate any rule. i want to write a
> small code in Linux that will boot and print something. this is something
> very basi
Le Fri, 7 Dec 2007 14:04:16 +0530,
"Rajat Jain" <[EMAIL PROTECTED]> a écrit :
> You need request_mem_region().
I'm not sure it's enough in Gavin's case.
From what I understand, request_mem_region() simply allows to reserve a
memory region, to say « I'm using it », so that nobody else will use it
Le Thu, 6 Dec 2007 16:18:24 +0530,
"Sandeep K Sinha" <[EMAIL PROTECTED]> a écrit :
> Am I logically wrong somewhere ??
Yes, you should not open files from the kernel:
http://kernelnewbies.org/FAQ/WhyWritingFilesFromKernelIsBad
Thomas
--
Thomas Petazzoni, [EMAIL PROTECTED], http://thomas.enix.o
Le Thu, 6 Dec 2007 22:47:35 +0900,
"Sumudu Nishantha" <[EMAIL PROTECTED]> a écrit :
> My guess is a hardware problem.
> can it be a problem of the driver or anything to do with
> the kernel??
> There is no register to reset dma.(no idea whether it is
> necessary).
It can be a driver problem, but
Le Fri, 07 Dec 2007 10:28:53 +0100,
Peter Poulsen <[EMAIL PROTECTED]> a écrit :
> I have noticed __user in some code, when a pointer points to
> userspace code. What does that do exaclty?
AFAIK, they are used by sparse[1] to verify that the kernel code doesn't
access directly to pointers pointing
Hi all,
i am working on a project with linux on embedded platform (MPC5200
processor).
i have this problem that the kernel says "enable_irq(some_number)
unbalaced". what does it mean?
i tried to google but haven't found any clear hint.
thanks for any answer.
regards,
i am not sure whether this is the correct forum to ask. i am a newbie to
this forum and please pardon me if I violate any rule. i want to write a
small code in Linux that will boot and print something. this is something
very basic but I have no clue how this can be done. i searched the web and
got
hi
i'm trying to find some tiny modules
searched personal pages , blogs and ... , but can't find useful samples .
do you have any link , sample , even your hobbies (or what ever you think
which is good for newbie) to share with me ?
thx for your attention .
--
Nothing but perfection
http://ima
On Fri, 7 Dec 2007, Peter Poulsen wrote:
> I have noticed __user in some code, when a pointer points to
> userspace code. What does that do exaclty?
it's meant to denote that the pointer is a pointer to *userspace*, not
kernel space. it doesn't actually get involved in the build process,
it's fo
I have noticed __user in some code, when a pointer points to userspace
code. What does that do exaclty?
/peter
--
To unsubscribe from this list: send an email with
"unsubscribe kernelnewbies" to [EMAIL PROTECTED]
Please read the FAQ at http://kernelnewbies.org/FAQ
Oops... I had missed the setting of fs to the oldfs...
Thanks.
On Dec 7, 2007 1:29 PM, Vivek Kutal <[EMAIL PROTECTED]> wrote:
> Sandeep K Sinha wrote:
> > It returns with ENOENT after using set_fs.
> >
> >
> same code works for me.
> take a look at this http://www.linuxjournal.com/article/8110
> t
Hi,
I have knowledge of assembly language. I am good at C programming. I
have learnt to write "hello world" type of kernel modules using
printk() statements. I know nothing else about kernel programming. I
want to become a kernel hacker. I am a newbie at the moment. To begin
with, I have 3 goals.
Hi,
> Anyone got any ideas? I've found some functions, ioremap, kmap and
> mem_request_region. Not really sure what I ought to be doing though.
> Any thoughts greatly appreciated.
You need request_mem_region().
Thanks,
Rajat
--
To unsubscribe from this list: send an email with
"unsubscribe
Sandeep K Sinha wrote:
It returns with ENOENT after using set_fs.
same code works for me.
take a look at this http://www.linuxjournal.com/article/8110
there is a example given at the end.
HTH
--
Vivek Kutal
--
To unsubscribe from this list: send an email with
"unsubscribe kernelnewbies
28 matches
Mail list logo