On Thu, 7 Dec 2000, Alan Cox wrote:
> Andreas is looking at a slightly older kernel, and was right for that. Every
> caller to daemonize either then did the file stuff or needed to and forgot
> so I fixed daemonize
I think, there ist still a small bug.
(This time I even checked 2.4.0-test12-pre8)
In linux/arch/i386/kernel/process.c, function kernel_thread, line 453 the
flag CLONE_VM is always used.
In sched.c, function daemonize, line 1216 you call exit_mm.
Since the memory is cloned, you will take away the mem from your
user-space-application as well. So if insmod is already running at that
time, it has to segvault. If I am not wrong at this point CLONE_VM simply
has to be removed from kernel_thread. The kernel-thread will free his mem
in daemonize (calling exit_mm) and the user-space-application will free
the mem when exiting.
Bye,
-- Andreas Klein
asklein@cip.physik.uni-wuerzburg.de
root / webmaster @cip.physik.uni-wuerzburg.de
root / webmaster @www.physik.uni-wuerzburg.de
_____________________________________
| |
| Long live our gracious AMIGA! |
|___________________________________|
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
Please read the FAQ at http://www.tux.org/lkml/
This archive was generated by hypermail 2b29 : Fri Dec 15 2000 - 21:00:23 EST