[dm-devel] Re: [PATCH 12/17] bottom-layer barrier support
Mikulas Patocka
mpatocka at redhat.com
Mon Apr 27 23:48:30 UTC 2009
On Tue, 28 Apr 2009, Alasdair G Kergon wrote:
> On Mon, Apr 27, 2009 at 07:41:47PM -0400, Mikulas Patocka wrote:
> > There is separate read_delay and write_delay. Flushes are governed by
> > write_delay.
>
> And the value of read_delay has no impact on the flush?
No.
> (Because of the "wait for all outstanding I/O" bit that eventually I
> want to see removed to facilitate true barrier passing?)
With dm-delay, "true barrier passing" doesn't have sense. Dm-delay is
designed to delay requests, so no one wants high performance from it :)
> Alasdair
> --
> agk at redhat.com
Mikulas
More information about the dm-devel
mailing list