On Thu, Jun 20, 2002 at 09:15:41AM -0400, Aidan Cully wrote:
> On Thu, Jun 20, 2002 at 03:01:46PM +0200, Wojciech Puchar wrote:
> > > Maybe an I/O scheduler makes sense as a solution to this problem?  Or
> > > would that be Too Much Effort?  If so, another solution might be to
> > > restrict the percentage of available buffers a given process can use
> > > for read/write...  Keep it small enough, and you wouldn't need to wait
> > > seconds for dd if=/dev/zero to flush all of its buffers.
> > 
> > seconds sound cool. but does it mean per-process queues???
> 
> I don't think so.  Write ordering should be preserved, but if a process
> tries to schedule a write when no buffers are available to it (but
> would be available to other processes) then it would block.  A process
> would need keep track of how many buffers it's using for I/O so that
> it would know to block when it's using too many.  A buffer would need to
> know which process it's coming from, so that it could lower the process
> buffer count on free(), and wake the process up if necessary.

This may be one of the mechanism used by solaris.

--
Manuel Bouyer, LIP6, Universite Paris VI.           Manuel.Bouyer@lip6.fr
--