Re: [Qemu-devel] [PATCH 2/3] barriers: block-raw-posix barrier support

[Christoph Hellwig]

On Tue, May 05, 2009 at 01:33:11PM +0100, Jamie Lokier wrote:
> You don't need two fdatasyncs if the barrier request is just a
> barrier, no data write, used only to flush previously written data by
> a guest's fsync/fdatasync implementation.

Yeah.  I'll put that optimization in after some testing.

> This is the best argument yet for having distinct "barrier" and "sync"
> operations.  "Barrier" is for ordering I/O, such as journalling
> filesystems.

Doesn't really help as long as we're using the normal Posix filesystem
APIs on the host.  The only way to guarantee ordering of multiple
*write* systen calls is to call f(data)sync between them.