Mailing List Archive

[PATCH 0 of 4 v4] blkif.h: Document protocol and existing extensions
This patch series attempts to document the blkif PV interface and
the various extensions to it that are out in the wild.

Changes in v4:

patch 2 (blkif.h: Provide more complete documentation of the blkif interface.):
o Fix typo "discard-aligment" -> "discard-alignment"
o Fix typo "unamp" -> "unmap"
o Fix typo "formated" -> "formatted"
o Clarify the text for "params".
o Clarify the text for "sector-size".

patch 4 (blkif.h: Define and document the request number/size/segments
extension.)
o Insert space in "Red Hat".
o Clarify the text for "max-requests" in the backend section.

Changes in v3:

patch 3 (blkif.h: Document the Red Hat and Citrix blkif multi-page
ring extensions.):
o Mark the Red Hat/Amazon EC2 ring extension nodes as deprecated. While
there is nothing fundamentally wrong with the method used there, it
was not publicly documented (not even in patch form), and differs from
the "page-order" large ring XenBus node convention used for other
drivers.

patch 4 (blkif.h: Define and document the request number/size/segments
extension.)
o Fix typo "max-ring_pages" -> "max-ring-pages".

Changes in v2:

patch 2 (blkif.h: Provide more complete documentation of the blkif interface.):
o Mark backend device identification section as private to the
backend driver.
o Refer to docs/misc/vbd-interface.txt for the format of the
virtual-device front-end node.
o Correct field size for the virtual-device front-end node.

patch 3 (blkif.h: Document the Red Hat and Citrix blkif multi-page
ring extensions.):
o Correct node names for the Red Hat/Amazon multi-ring extension. The
previous patch mistakenly documented the now defunct FreeBSD extension.
o Clarify the note on multi-page ring interoperability to indicate that
identical ring parameters should be published to the XenStore for
all supported schemes.

v1 patch 4 (Deleted):
o Remove patch that added the XEN_*_MAJOR definitions.

patch 4 (blkif.h: Define and document the request number/size/segments
extension.)
o Bump __XEN_LATEST_INTERFACE_VERSION to 0x00040201 and use this version
to guard the change to BLKIF_MAX_SEGMENTS_PER_REQUEST.

--
Justin


_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xen.org
http://lists.xen.org/xen-devel
Re: [PATCH 0 of 4 v4] blkif.h: Document protocol and existing extensions [ In reply to ]
On 27/02/2012 06:27, "Justin T. Gibbs" <justing@spectralogic.com> wrote:

> This patch series attempts to document the blkif PV interface and
> the various extensions to it that are out in the wild.

The previous round of these patches was already applied. They are sitting in
the staging tree currently (http://xenbits.xen.org/staging/xen-unstable.hg).

-- Keir

> Changes in v4:
>
> patch 2 (blkif.h: Provide more complete documentation of the blkif
> interface.):
> o Fix typo "discard-aligment" -> "discard-alignment"
> o Fix typo "unamp" -> "unmap"
> o Fix typo "formated" -> "formatted"
> o Clarify the text for "params".
> o Clarify the text for "sector-size".
>
> patch 4 (blkif.h: Define and document the request number/size/segments
> extension.)
> o Insert space in "Red Hat".
> o Clarify the text for "max-requests" in the backend section.
>
> Changes in v3:
>
> patch 3 (blkif.h: Document the Red Hat and Citrix blkif multi-page
> ring extensions.):
> o Mark the Red Hat/Amazon EC2 ring extension nodes as deprecated. While
> there is nothing fundamentally wrong with the method used there, it
> was not publicly documented (not even in patch form), and differs from
> the "page-order" large ring XenBus node convention used for other
> drivers.
>
> patch 4 (blkif.h: Define and document the request number/size/segments
> extension.)
> o Fix typo "max-ring_pages" -> "max-ring-pages".
>
> Changes in v2:
>
> patch 2 (blkif.h: Provide more complete documentation of the blkif
> interface.):
> o Mark backend device identification section as private to the
> backend driver.
> o Refer to docs/misc/vbd-interface.txt for the format of the
> virtual-device front-end node.
> o Correct field size for the virtual-device front-end node.
>
> patch 3 (blkif.h: Document the Red Hat and Citrix blkif multi-page
> ring extensions.):
> o Correct node names for the Red Hat/Amazon multi-ring extension. The
> previous patch mistakenly documented the now defunct FreeBSD extension.
> o Clarify the note on multi-page ring interoperability to indicate that
> identical ring parameters should be published to the XenStore for
> all supported schemes.
>
> v1 patch 4 (Deleted):
> o Remove patch that added the XEN_*_MAJOR definitions.
>
> patch 4 (blkif.h: Define and document the request number/size/segments
> extension.)
> o Bump __XEN_LATEST_INTERFACE_VERSION to 0x00040201 and use this version
> to guard the change to BLKIF_MAX_SEGMENTS_PER_REQUEST.
>
> --
> Justin
>



_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xen.org
http://lists.xen.org/xen-devel
Re: [PATCH 0 of 4 v4] blkif.h: Document protocol and existing extensions [ In reply to ]
On Mon, 2012-02-27 at 06:50 +0000, Keir Fraser wrote:
> On 27/02/2012 06:27, "Justin T. Gibbs" <justing@spectralogic.com> wrote:
>
> > This patch series attempts to document the blkif PV interface and
> > the various extensions to it that are out in the wild.
>
> The previous round of these patches was already applied. They are sitting in
> the staging tree currently (http://xenbits.xen.org/staging/xen-unstable.hg).

Note that they are stuck there because they broke the tests. See the
"[xen-unstable bisection] complete test-amd64-i386-rhel6hvm-amd" thread.

Ian.



_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xen.org
http://lists.xen.org/xen-devel
Re: [PATCH 0 of 4 v4] blkif.h: Document protocol and existing extensions [ In reply to ]
On Feb 27, 2012, at 2:53 AM, Ian Campbell wrote:

> On Mon, 2012-02-27 at 06:50 +0000, Keir Fraser wrote:
>> On 27/02/2012 06:27, "Justin T. Gibbs" <justing@spectralogic.com> wrote:
>>
>>> This patch series attempts to document the blkif PV interface and
>>> the various extensions to it that are out in the wild.
>>
>> The previous round of these patches was already applied. They are sitting in
>> the staging tree currently (http://xenbits.xen.org/staging/xen-unstable.hg).
>
> Note that they are stuck there because they broke the tests. See the
> "[xen-unstable bisection] complete test-amd64-i386-rhel6hvm-amd" thread.
>
> Ian.

Thanks for the pointer.

It seems I've made quite a mess of it. :-(

--
Justin
_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xen.org
http://lists.xen.org/xen-devel