Re: stochastic fair queueing in the elevator [Re: [BENCHMARK] 2.4.20-ck3 / aa / rmap with contest]

[Andrea Arcangeli]

On Mon, Feb 10, 2003 at 02:47:19AM -0200, Rik van Riel wrote:
> On Sun, 9 Feb 2003, Andrew Morton wrote:
> > David Lang <[e-mail]> wrote:
> > >
> > > note that issuing a fsync should change all pending writes to 'syncronous'
> > > as should writes to any partition mounted with the sync option, or writes
> > > to a directory with the S flag set.
> >
> > We know, at I/O submission time, whether a write is to be waited upon.
> > That's in writeback_control.sync_mode.
> 
> An fsync might change already submitted asynchronous writes
> into synchronous writes, but this is not something I'm going
> to lose sleep over. ;)

correct.

> 
> > That, combined with an assumption that "all reads are synchronous" would
> > allow the outgoing BIOs to be appropriately tagged.
> >
> > It's still approximate.
> 
> Sounds like a good enough approximation to me.

yes, the approximation should be more than enough IMHO

Andrea