From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Date: Fri, 18 Mar 2022 17:06:25 +0100 From: Halil Pasic Subject: Re: [virtio-comment] Re: [PATCH v3 4/4] Add CCW configuration field "indirect_num" Message-ID: <20220318170625.0d2be174.pasic@linux.ibm.com> In-Reply-To: <11686863.Z9n2BMzBuM@silver> References: <4735344.EBYxvr1mta@silver> <3040881.qp741xram0@silver> <87cziksm5h.fsf@redhat.com> <11686863.Z9n2BMzBuM@silver> MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: quoted-printable To: Christian Schoenebeck Cc: Cornelia Huck , virtio-comment@lists.oasis-open.org, Stefan Hajnoczi , Greg Kurz , Dominique Martinet , Halil Pasic List-ID: On Fri, 18 Mar 2022 12:02:31 +0100 Christian Schoenebeck wrote: > On Donnerstag, 17. M=C3=A4rz 2022 15:12:42 CET Cornelia Huck wrote: > > On Wed, Mar 16 2022, Christian Schoenebeck wro= te: =20 > > > This new CCW configuration field allows to negotiate a more fine > > > graded maximum lenght of indirect descriptor chains. > > >=20 > > > Bump CCW virtio revision to 3 and make the existence of this new > > > field "indirect_num" dependant on revision 3. > > >=20 > > > Fixes: https://github.com/oasis-tcs/virtio-spec/issues/122 > > > Signed-off-by: Christian Schoenebeck > > > --- > > >=20 > > > content.tex | 14 +++++++++++++- > > > 1 file changed, 13 insertions(+), 1 deletion(-) =20 > >=20 > > (...) > > =20 > > > @@ -2581,12 +2583,17 @@ \subsubsection{Configuring a > > > Virtqueue}\label{sec:Virtio Transport Options / Vir>=20 > > > struct vq_config_block { > > > =20 > > > be16 index; > > > be16 max_num; > > >=20 > > > + be32 indirect_num; /* since virtio-ccw rev. 3 */ =20 > >=20 > > Maybe make it max_indirect_num (to mirror max_num?) =20 >=20 > Right, that name appears indeed more appropriate here. >=20 > > > }; > > > \end{lstlisting} > > > =20 > > > The requested number of buffers for queue \field{index} is returned = in > > > \field{max_num}. > > >=20 > > > +Since revision 3, \field{indirect_num} exists, which is supposed to > > > reflect the +Queue Indirect Size (i.e. the maximum length of indirect > > > descriptor tables) +supported by device for this queue. =20 > >=20 > > It still depends on the feature flag, though. > >=20 > > Maybe > >=20 > > "If revision 2 or lower is set, struct vq_config_block extends up to > > \field{max_num}. > >=20 > > If revision 3 or higher is set, struct vq_config_block extends to > > \field{max_indirect_num}. If VIRTIO_RING_F_INDIRECT_SIZE is negotiated, > > \field{indirect_max_num} contains the maximum Queue Indirect Size > > (i.e. the maximum length of indirect descriptor tables) supported by th= e > > device for this queue; otherwise, its contents are unpredictable." =20 >=20 > Maybe rather something like this instead: >=20 > "Actual size of struct vq_config_block depends on the virtio-ccw revision= . >=20 > If revision 2 or lower is set, struct vq_config_block extends up to > *including* \field{max_num}. >=20 > If revision 3 or higher is set, struct vq_config_block extends up to > *including* \field{max_indirect_num}. If VIRTIO_RING_F_INDIRECT_SIZE is= =20 > negotiated, \field{indirect_max_num} contains the maximum Queue Indirect = Size > (i.e. the maximum length of indirect descriptor tables) supported by the > device for this queue; otherwise, its contents are *undefined*." > =20 I agree that the "including" is important, but I'm not sure about the "its contents are undefined". I don't really understand why should we use a plural here. What speaks against specifying that in SHOULD be stored as 0 by the device, and MUST be ignored by the driver? Currently we say that \field{max_indirect_num} exists like a be32 field even if VIRTIO_RING_F_INDIRECT_SIZE is not negotiated. Which kind of implies that at least type invariants should hold. Of course, there is none here (i.e. every bits value is also a be32 value), but for something like an enum interesting corner cases can pop up. [..] Regards, Halil