Re: [PATCH 5/6] libxl: introduce libxl__alloc_vdev

[Ian Jackson <Ian.Jackson@eu.citrix.com>]

Stefano Stabellini writes ("Re: [Xen-devel] [PATCH 5/6] libxl: introduce libxl__alloc_vdev"):
> On Fri, 30 Mar 2012, Ian Jackson wrote:
> > Since we need to be able to allocate some vbds dynamically, the right
> > approach is to decree that some portion of the dom0 vbd space is
> > reserved for static allocations by the administrator, and that the
> > remainder is for dynamic allocations by software which picks a free
> > vbd.  Naturally the static space should come first.
> 
> When you hotplug a new disk on your system, for example a new USB disk
> to your native Linux box, usually Linux chooses the device name for
> you. I don't see why this should be any different.

It is different because Xen vbds do in practice appear in dom0 with a
stable name.  So this is something that people have reasonably come to
rely on.

> The admin is going to call instead:
>   xl block-attach 0
> and then checkout dmesg.

This is hardly automatable.  Doing the same thing with udev rules is
quite hard work.

> > I don't think that is true.  On each individual guest platform, the
> > relationship between vbd numbers (the actual interface between
> > frontend and backend), and whatever that guest has for device names,
> > needs to be statically defined.
> 
> I disagree: when we introduced docs/txt/misc/vbd-interface.txt we never
> specified what names the guest kernel is allowed to choose for a
> particular vbd encoding. See the following quote:
> 
> "The Xen interface does not specify what name a device should have in the
> guest (nor what major/minor device number it should have in the guest,
> if the guest has such a concept)."

This refers to the promises made by each side of the Xen VBD interface
to the corresponding other side.  The host's environment is not
allowed to assume things about the guest's device naming conventions.

However, that does not mean that the guest should not document its
naming conventions.  Perhaps this needs to be clarified in the
document.

> What I find confusing is that before you say that "the relationship
> between vbd numbers and whatever that guest has for device names needs
> to be statically defined", but then you say that "the device name in the
> guest is not written to xenstore anywhere.  It is private to the guest."

It is private to the guest and the guest administrator and the guest
documentation.  The guest is not required to notify the host of what
name it has chosen.  (And if it does choose a name that name might be
some crazy thing; imagine if the guest is MVS/370 or something.)

> Maybe we can slightly improve the situation moving libxl__alloc_vdev to
> an OS specific file so that we can have a Linux and a BSD implementation.

We only need a specific implementation of the final mapping from vbd
number to guest-specific device name.

Ian.