Re: memory mapped store entries

On Wed, Aug 26, 1998 at 07:11:21PM +0200, Stephen R. van den Berg wrote:

> The ideal situation would be if we could tell the kernel never to
> flush the dirty pages automatically, but force a flush using
> msync(MS_ASYNC) once every night at say 5AM.

Since dirty pages probably won't be flushed by the kernel until they
are paged out of explicitly msync'd - it should be more often, much
more often.

We don't have to sync the whole file either, we can get away with
sync parts of it, and roll a window over the swap so that we sync it
bit by bit.

-cw
Received on Tue Jul 29 2003 - 13:15:53 MDT