Re: [PATCH] ide write barrier support

[Jens Axboe <axboe@suse.de>]

On Thu, Oct 16 2003, Mudama, Eric wrote:
> 
> On Tue, Oct 14 2003, Greg Stark wrote:
> > Jens Axboe <axboe@suse.de> writes:
> > > There's also the case of files opened with O_SYNC. Would inserting a
> > > write barrier after every write to such a file destroy performance?
> > 
> > If it's mainly sequential io, then no it won't destroy performance. It
> > will be lower than without the cache flush of course.
> 
> If you flush a cache after every command in a sequential write, yes, you'll
> destroy performance.  How much you destroy it is a function of command size
> relative to track size, and the RPM of the drive.

Yes you are right, my logic was a bit backwards. But you don't have to
issue a flush after every write, thankfully. But lets still not forget
that performance means nothing if integrity isn't guarenteed.

-- 
Jens Axboe