Re: [PATCH v3] virtio-blk: Add support for "Force Unit Access" writes

Stefan Hajnoczi <[email protected]> Thu, 13 Nov 2025 10:41:27 -0500

Newsgroups	dev.linux.lists.virtio-comment
Message-ID	<20251113154127.GC189474@fedora>

On Wed, Nov 12, 2025 at 05:30:40AM +0000, Alberto Faria wrote:
> On Thu, Sep 18, 2025 at 2:38 PM Stefan Hajnoczi <[email protected]> wrote:
> > On Wed, Sep 17, 2025 at 08:12:49AM +0100, Alberto Faria wrote:
> > > @@ -1000,14 +1030,20 @@ \subsection{Device Operation}\label{sec:Device Types / Block Device / Device Ope
> > >    \field{writeback} field in configuration space was 0 \textbf{all the time between
> > >    the submission of the write and its completion};
> > >
> > > -\item\label{item:flush3} a VIRTIO_BLK_T_FLUSH request is sent \textbf{after the write is
> > > +\item\label{item:flush3} the VIRTIO_BLK_F_REQ_FLAGS_FUA feature was negotiated
> > > +  and the \field{fua} bit in \field{flags} was set in the write request
> > > +  (regardless of whether the VIRTIO_BLK_F_FLUSH or VIRTIO_BLK_F_CONFIG_WCE
> > > +  features were negotiated, and regardless of the current cache mode as
> > > +  expressed by the value of the \field{writeback} field in configuration space).
> >
> > Missing "and the write request is completed"?
> 
> This should be implied by the paragraph above the itemize.

You are right, write completion is already mentioned:

  A write becomes stable once it is completed and one or more of the
  following conditions is true:
  - ...

Stefan

signature.asc (application/pgp-signature, 488 B)