Re: [PATCH] barrier patch set
From: Chris Mason
Date: Tue Mar 30 2004 - 17:17:50 EST
On Tue, 2004-03-30 at 16:50, Stephen C. Tweedie wrote:
> On Tue, 2004-03-30 at 20:19, Chris Mason wrote:
> > I think we're mixing a few concepts together. submit_bh(WRITE_BARRIER,
> > bh) gives us an ordered write in whatever form the lower layers can
> > provide. It also ensures that if you happen to call wait_on_buffer()
> > for the barrier buffer, the wait won't return until the data is on
> > media.
> Right, but that's just how it works right now --- one doesn't _have_ to
> imply the other. You could easily imagine an implementation that
> implements barriers and flushing separately, and which does not do
> automatic flushing on completion of WRITE_BARRIER IOs. SCSI with
> writeback caching enabled might be one example of that. NBD/DRBD would
> be another likely candidate --- if you've got network latencies in the
> way, then a flushing sync may be far more expensive than a barrier
Yes, that's true, although the barriers don't really imply a flush, it
just implies that if you do use wait_on_* for flushing, it will report
> Unfortunately, a lot of the cases we care about really have to do the
> barrier via flushing, so the benefit of keeping them separate is
> limited. For LVM/raid0, for example, we've got no way of preserving
> ordering between IOs on different drives, so a flush is necessary there
> unless we start journaling the low-level IOs to preserve order.
> Yep. It scares me to think what performance characteristics we'll start
> seeing once that gets used everywhere it's needed, though. If every raw
> or O_DIRECT write needs a flush after it, databases are going to become
> very sensitive to flush performance. I guess disabling the flushing and
> using disks which tell the truth about data hitting the platter is the
> sane answer there.
Most database benchmarks are done on scsi, and the blkdev_flush should
be a noop there. For IDE based database and mail server benchmarks, the
results won't be pretty.
The reiserfs fsync code tries hard to only flush once, so if a commit is
done then blkdev_flush isn't called. We might have to do a few other
tricks to queue up multiple synchronous ios and only flush once.
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/