On Saturday 04 October 2008 22:58:33 Jeff Epler wrote:
> On Sat, Oct 04, 2008 at 08:36:56PM +0200, Michael Buesch wrote:
> > Isn't it possible to implement a check and exit with an error message?
> > Where is it segfaulting? Does the kernel segfault us if we try to memlock
> > a range bigger than the allowed one? Or does it crash due to some NULL
> > pointer deref later?
> 
> I'm not 100% sure what goes on.  When I deliberately configure the
> locked memory amount to a low value using umask, my *dapper*, 32-bit
> machine gives a much better message, something like
>     RTAPI: ERROR: failed to map shmem
>     HAL: ERROR: could not open shared memory
>     halcmd: hal_init() failed: -9
>     NOTE: 'rtapi' kernel module must be loaded

Oh, this is interesting, indeed.

> .. however the result is different either on post-dapper systems or
> 64-bit systems.  I haven't tried to debug this lately, but my
> recollection is that the calls into the realtime layer appear to
> succeed, giving a segfault when the memory region is accessed.  However,
> this is months ago so I'm fuzzy on the details.  Maybe I'll find time
> this weekend to look into it again.

Thanks a lot. :)
(I'm using a vanilla kernel.org kernel with latest rtai, btw, if that
matters. So it's _not_ something special with the lenny kernel.)

-- 
Greetings Michael.

-------------------------------------------------------------------------
This SF.Net email is sponsored by the Moblin Your Move Developer's challenge
Build the coolest Linux based applications with Moblin SDK & win great prizes
Grand prize is a trip for two to an Open Source event anywhere in the world
http://moblin-contest.org/redirect.php?banner_id=100&url=/
_______________________________________________
Emc-developers mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/emc-developers

Reply via email to