On Wed, Apr 28, 1999 at 07:32:26PM +0100, Alan Cox wrote:
> > We are seeing some weird memory problems after upgrading to
> > 2.0.37pre11.  For example, xntpd consistently gets stuck forever in
> > disk wait, sometimes on a call to brk() and sometimes on a call to
> > mmap().  Other processes seem to be dying in mmap() as well but are
> > harder to reproduce.
> 
> Does 2.0.37pre10 show the same problem >
> 

No experience with this, but removing the semaphore protection in merge_segments
has side-effects in some other syscalls, especially sys_shm, sys_mlock,
sys_mprotect, sys_mremap, i.e.

No idea, of the actions taken in merge_segments has to be protected in these
syscalls. Or if all these syscalls should get the semaphore before calling
merge_segments or even bevor using functions like do_mmap, ...

What ever, it seems to be a rather big chance for a pre11.

Wolfgang Walter


-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/