Re: [Xen-devel] [patch 0/6] xenblk: Add O_DIRECT and O_SYNC support.

[Keir Fraser <keir.fraser@xxxxxxxxxxxxx>]

On 4/11/08 08:02, "Joe Jin" <joe.jin@xxxxxxxxxx> wrote:

>> How is this better than waiting in the domU (blkfront) domain to collect the
>> required asynchronous responses? Scattering changes into Linux filesystems
>> is a dead end unless you can get the changes accepted upstream via lkml.
>> 
> 
> Do you meant implement this feature just in domU (blkfront)?

Yes. If I understand what is being done.

> To direct-io answer is yes for request have marked if write request with
> O_DIRECT flag, but to O_SYNC, filesystem implentation it at filesystem,
> layer, I think do not touch filesystem codes to implement O_SYNC support
> is impossible.

O_DIRECT has no meaning really across the xenblk interface. The intention is
that blkback talks at a raw block device below the buffer cache anyway.

For O_SYNC it can be implemented by pushing out all I/Os into the ring
buffer and then wait for all required responses. Why do you think that
doesn't work? Indeed it should work already if such logic exists in the
generic Linux block layer, since blkfront notifies that layer as responses
come in from blkback.

> And, I think the changs related filesystem maybe hard to accepted by upsteam
> for this is a special issue just related xenblk.

But xenblk (blkfront at least) is upstream now.

 -- Keir



_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxxxxxxxx
http://lists.xensource.com/xen-devel