Re: [Embedded-pv-devel] [PATCH v15] This is the ABI for the two halves of a para-virtualized sound driver to communicate with each to other.

[Oleksandr Andrushchenko <andr2000@xxxxxxxxx>]

On 02/07/2017 09:16 PM, Konrad Rzeszutek Wilk wrote:
> On Mon, Jan 30, 2017 at 08:49:50AM +0200, Oleksandr Andrushchenko wrote:
> > On 01/27/2017 08:57 PM, Konrad Rzeszutek Wilk wrote:
> > > On Fri, Jan 27, 2017 at 08:38:29PM +0200, Oleksandr Andrushchenko wrote:
> > > > On 01/27/2017 08:13 PM, Konrad Rzeszutek Wilk wrote:
> > > > > .snip..
> > > > > > I am looking at this from FAE's or integrator's POV, when one would need
> > > > > FAE?
> > > > Field Applications Engineer
> > > > > > to touch different parts of the system. "/0/0/0" makes me feel
> > > > > > sad just because either you have to keep all those numbers in mind (like you
> > > > > > do)
> > > > > > or have documentation available (and know where it is, e.g. sources
> > > > > > of Xen or kernel).
> > > > > > I have a strong feeling that if you can change configuration without
> > > > > > knowing what index stands for it is always better and fail-safer then
> > > > > > just having numbers...
> > > > > Not sure I follow that.
> > > > >
> > > > > How would you change configuration without knowing the index?
> > > > >
> > > > > ..snip..
> > > > if one looks at
> > > > .../pcm-dev-0/stream-1/...
> > > > most probably he/she will understand this w/o any documentation,
> > > > because it is human readable
> > > >
> > > > if one looks at
> > > > .../0/1/...
> > > > well, I believe you can almost do nothing w/o looking at the documentation
> > > I can see the beaty of it.
> > >
> > > I can also see the beaty of the old implied mechanism
> > > from a maintaince perspective.
> > >
> > > I am maintainer so I am leaning towards the second one
> > > as having less "special" cases.
> > >
> > > Sorry, I feel like I am mutilating your baby with this
> > > old boring view of "maintaince" and "conform to the old
> > > standard" :-(
> > np, so I will use ".../0/0/0/..." pattern here and in
> > displif as well. I will update "Addressing" section
> > so it is documented
> OK, so I will ignore v16 as of right now it has the old format.
> Please at your convience post v16.1 (or v17) with this change.
v17
> Thanks!
Thank you

_______________________________________________
Embedded-pv-devel mailing list
Embedded-pv-devel@xxxxxxxxxxxxxxxxxxxx
https://lists.xenproject.org/cgi-bin/mailman/listinfo/embedded-pv-devel