On Sat, Oct 27, 2001 at 02:27:03PM +0530, C. Mohanasundaram wrote:
> Hi all,
>
> i have written a kernel module while modifies the sys_call_table and inserts
> new syscalls to bring some performance counters to the userland.
my first question has to be why : perfctr has supported this ably for a long time,
and msr.c can also do this.
Or is this not x86 ?
> the process information is stored in a link list of struct task_struct.
> If i can copy the contents of the link list to the user memory
> using the macro for_each_task(p) defined in linux/sched.h
>
> for_each_task (p) {
> /*do copy and all that stuff*/
> }
don't do this - export exactly what you need, maybe as an array.
And a syscall is prolly a bad idea anyway.
regards
john
--
"If the software that a company produces isn't reliable, adding a bunch of
'Mother, may I' rules to the language and the code won't fix it."
- Pete Becker
-
Kernelnewbies: Help each other learn about the Linux kernel.
Archive: http://mail.nl.linux.org/kernelnewbies/
IRC Channel: irc.openprojects.net / #kernelnewbies
Web Page: http://www.kernelnewbies.org/
