[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [Qemu-devel] [PATCH 2/3] barriers: block-raw-posix barrier support
From: |
Christoph Hellwig |
Subject: |
Re: [Qemu-devel] [PATCH 2/3] barriers: block-raw-posix barrier support |
Date: |
Tue, 5 May 2009 15:29:44 +0200 |
User-agent: |
Mutt/1.3.28i |
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.
[Qemu-devel] [PATCH 2/3] barriers: block-raw-posix barrier support, Christoph Hellwig, 2009/05/05
[Qemu-devel] [PATCH 3/3] barriers: virtio, Christoph Hellwig, 2009/05/05
Re: [Qemu-devel] [PATCH 0/3] write barrier support, Avi Kivity, 2009/05/05