Now that no-data barriers are supported/supportable, does it make sense to have blkfront call blk_queue_issue_flush_fn to register a flush method, which would use a no-data barrier request? Is there any reason why this shouldn''t work? We have code which instead of using a bio barrier, calls blkdev_issue_flush() in the linux guest, which returns -EOPNOTSUPP because that support has never been in blkfront. - Bruce Rogers _______________________________________________ Xen-devel mailing list Xen-devel@lists.xensource.com http://lists.xensource.com/xen-devel
It sounds reasonable to me, although I''m no Linux blkdev interface expert. -- Keir On 12/6/08 23:35, "Bruce Rogers" <BROGERS@novell.com> wrote:> Now that no-data barriers are supported/supportable, does it make sense to > have blkfront call blk_queue_issue_flush_fn to register a flush method, which > would use a no-data barrier request? > > Is there any reason why this shouldn''t work? > > We have code which instead of using a bio barrier, calls blkdev_issue_flush() > in the linux guest, which returns -EOPNOTSUPP because that support has never > been in blkfront. > > - Bruce Rogers > > > _______________________________________________ > Xen-devel mailing list > Xen-devel@lists.xensource.com > http://lists.xensource.com/xen-devel_______________________________________________ Xen-devel mailing list Xen-devel@lists.xensource.com http://lists.xensource.com/xen-devel
On Fri, Jun 13, 2008 at 08:27:37AM +0100, Keir Fraser wrote:> It sounds reasonable to me, although I''m no Linux blkdev interface expert.Does this mean that Linux can support BLKIF_OP_FLUSH_DISKCACHE ? thanks john _______________________________________________ Xen-devel mailing list Xen-devel@lists.xensource.com http://lists.xensource.com/xen-devel