similar to: [RFC] virtio_pci draft spec

Add an rpmsg appendix to the virtio spec and tag it as Appendix H. Push the SCSI host appendix to 'I' so the appendices order stays according to the virtio device ids. Patch is against: http://ozlabs.org/~rusty/virtio-spec/virtio-0.9.4.lyx Tested with LyX 2.0.0. Signed-off-by: Ohad Ben-Cohen <ohad at wizery.com> Cc: Rusty Russell <rusty at rustcorp.com.au> Cc:

Add an rpmsg appendix to the virtio spec and tag it as Appendix H. Push the SCSI host appendix to 'I' so the appendices order stays according to the virtio device ids. Patch is against: http://ozlabs.org/~rusty/virtio-spec/virtio-0.9.4.lyx Tested with LyX 2.0.0. Signed-off-by: Ohad Ben-Cohen <ohad at wizery.com> Cc: Rusty Russell <rusty at rustcorp.com.au> Cc:

virtio header resizes in the first I/O region, not in PCI configuration space as the spec v0.8.1 implied. Correct this. Signed-off-by: Michael S. Tsirkin <mst at redhat.com> --- I started looking at virtio spec with the view to updating it with new features and possibly documenting more devices. I have used latex in the past but not lyx. As a start, here's a patch to correct a nit in

virtio header resizes in the first I/O region, not in PCI configuration space as the spec v0.8.1 implied. Correct this. Signed-off-by: Michael S. Tsirkin <mst at redhat.com> --- I started looking at virtio spec with the view to updating it with new features and possibly documenting more devices. I have used latex in the past but not lyx. As a start, here's a patch to correct a nit in

"A number of virtqueues are created" seems clearer than "the number of virtqueues": it's virtqueues that are created not the number. Signed-off-by: Michael S. Tsirkin <mst at redhat.com> --- I'm not a native english speaker but the below seems correct to me. Rusty? virtio-spec.lyx | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git

"A number of virtqueues are created" seems clearer than "the number of virtqueues": it's virtqueues that are created not the number. Signed-off-by: Michael S. Tsirkin <mst at redhat.com> --- I'm not a native english speaker but the below seems correct to me. Rusty? virtio-spec.lyx | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git

control vq is only used if feature bit is negotiated - it's not enough that device supports it. Clarify this in the spec. Signed-off-by: Michael S. Tsirkin <mst at redhat.com> --- virtio-spec.lyx | 9 +++++++-- 1 file changed, 7 insertions(+), 2 deletions(-) diff --git a/virtio-spec.lyx b/virtio-spec.lyx index 921388b..3ef96a9 100644 --- a/virtio-spec.lyx +++ b/virtio-spec.lyx @@

control vq is only used if feature bit is negotiated - it's not enough that device supports it. Clarify this in the spec. Signed-off-by: Michael S. Tsirkin <mst at redhat.com> --- virtio-spec.lyx | 9 +++++++-- 1 file changed, 7 insertions(+), 2 deletions(-) diff --git a/virtio-spec.lyx b/virtio-spec.lyx index 921388b..3ef96a9 100644 --- a/virtio-spec.lyx +++ b/virtio-spec.lyx @@

The lengcy guests don't have mac programming command, we don't know when it's safe to use MAC. We can change QEMU to make MAC change effect when the last byte of MAC is written to config space. Signed-off-by: Amos Kong <akong at redhat.com> --- virtio-spec.lyx | 11 ++++++++++- 1 file changed, 10 insertions(+), 1 deletion(-) diff --git a/virtio-spec.lyx b/virtio-spec.lyx index

The lengcy guests don't have mac programming command, we don't know when it's safe to use MAC. We can change QEMU to make MAC change effect when the last byte of MAC is written to config space. Signed-off-by: Amos Kong <akong at redhat.com> --- virtio-spec.lyx | 11 ++++++++++- 1 file changed, 10 insertions(+), 1 deletion(-) diff --git a/virtio-spec.lyx b/virtio-spec.lyx index

This patch splits the device status field of virtio-net into ro and rw byte. This would simplify the implementation of both host and guest and make the layout more clean. As VIRTIO_NET_S_ANNOUNCE is a rw bit, it was moved to bit 8 (0x100). btw. looks like there's no implementation that depends on VIRTIO_NET_S_ANNOUNCE, so the move is safe. Signed-off-by: Jason Wang <jasowang at

This patch splits the device status field of virtio-net into ro and rw byte. This would simplify the implementation of both host and guest and make the layout more clean. As VIRTIO_NET_S_ANNOUNCE is a rw bit, it was moved to bit 8 (0x100). btw. looks like there's no implementation that depends on VIRTIO_NET_S_ANNOUNCE, so the move is safe. Signed-off-by: Jason Wang <jasowang at

VIRTIO_BALLOON_F_MUST_TELL_HOST cannot be used properly because it is a "negative" feature: it tells you that silent defalte is not supported. Right now, QEMU refuses migration if the target does not support all the features that were negotiated. But then: - a migration from non-MUST_TELL_HOST to MUST_TELL_HOST will succeed, which is wrong; - a migration from MUST_TELL_HOST to

VIRTIO_BALLOON_F_MUST_TELL_HOST cannot be used properly because it is a "negative" feature: it tells you that silent defalte is not supported. Right now, QEMU refuses migration if the target does not support all the features that were negotiated. But then: - a migration from non-MUST_TELL_HOST to MUST_TELL_HOST will succeed, which is wrong; - a migration from MUST_TELL_HOST to

Drivers treat MUST_TELL_HOST as optional: windows drivers do not ack it and expect this means they can tell host *after* deflate. This was not the intent but the documentation was not very clear on this point. Luckily hyprevisors did not implement this feature yet so to provide guidance for future devices make spec match drivers expectations, and clarify that this feature only has effect if

Drivers treat MUST_TELL_HOST as optional: windows drivers do not ack it and expect this means they can tell host *after* deflate. This was not the intent but the documentation was not very clear on this point. Luckily hyprevisors did not implement this feature yet so to provide guidance for future devices make spec match drivers expectations, and clarify that this feature only has effect if

Add field for guest to specify command size for virtio-blk. Signed-off-by: Michael S. Tsirkin <mst at redhat.com> --- virtio-spec.lyx | 83 +++++++++++++++++++++++++++++++++++++++++++++++++++++---- 1 file changed, 78 insertions(+), 5 deletions(-) diff --git a/virtio-spec.lyx b/virtio-spec.lyx index a8ce3f9..fea97ed 100644 --- a/virtio-spec.lyx +++ b/virtio-spec.lyx @@ -5826,6 +5826,16 @@

Add field for guest to specify command size for virtio-blk. Signed-off-by: Michael S. Tsirkin <mst at redhat.com> --- virtio-spec.lyx | 83 +++++++++++++++++++++++++++++++++++++++++++++++++++++---- 1 file changed, 78 insertions(+), 5 deletions(-) diff --git a/virtio-spec.lyx b/virtio-spec.lyx index a8ce3f9..fea97ed 100644 --- a/virtio-spec.lyx +++ b/virtio-spec.lyx @@ -5826,6 +5826,16 @@

Hello Rusty, A few trivial fixes on top of virtio-spec-0.8.9 (patch -p0): - grammatical errors - mark '-' in 'avail->flags' as non-breakable hyphen - add a footnote for console device to indicate queues 2 onwards are avl. if a feature was negotiated Amit --- virtio-spec-0.8.9.lyx 2010-10-05 17:26:11.318836266 +0530 +++ virtio-spec-0.8.9-mod.lyx 2010-10-05

Hello Rusty, A few trivial fixes on top of virtio-spec-0.8.9 (patch -p0): - grammatical errors - mark '-' in 'avail->flags' as non-breakable hyphen - add a footnote for console device to indicate queues 2 onwards are avl. if a feature was negotiated Amit --- virtio-spec-0.8.9.lyx 2010-10-05 17:26:11.318836266 +0530 +++ virtio-spec-0.8.9-mod.lyx 2010-10-05