[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
[Newsgroup Home]
|
[news.eclipse.dsdp.vpp] Re: VPP And Eclipse - minutes of the first meeting (3rd June 08)
|
The next meeting is on Monday - Please use the same contact numbers as
previously.There were many discussions at DAC about this, I think we now need to
move forward.
Lets discuss this on Monday.
Cheers
Mark.
Mark Burton<mark@xxxxxxxxxxxxx> wrote:
>
>
>
>Short Minutes:
>
>Present:
>Aaron Spear (mentor)
>Anthony Berent (arm)
>Kris Keyeyser (CoWare)
>Gary Delp (LSI)
>Laurent Gerard (ST)
>Mark Burton (GreenSocs)
>
>Our next meeting is scheduled for 23rd June at 5pm CEST, on the same
>numbers.
>
>Anthony gave us an overview of IPXACT, and told us about the ARM IPXACT
>component descriptions etc, before making use of the components inside a
>debugger setup (for instance).
>IPXATC 1.4 now has a API called the "TGI". This interface would allow an
>IPXACT editor, and other tools, to share a common IPXACT database,
>rather than having to move files between them, which would allow for a much
>neater user interaction - this would require an IPXACT server. - However
>2/ the current TGI is possible too "low level" to be helpful for
>debuggers.
>Gary suggested that SPIRIT would like feedback on the TGI. SPIRIT is in
>a release phase, where 1.4 will only become ratified once 3 independent
>tools have used it. Once this is done, work will start on the next
>revisions, and feedback on the TGI could be taken into consideration then.
>Aaron asked what the plans for future developments of the ARM IPXAT
>editor were - Anthony said nothing much more than today. Aaron suggested
>that making more of a "framework" would be good, to enable EDA companies to
>expand the capabilities - Anthony suggested that, for now, he didn't
>see many more places that there could be "hooks" provided than were
>already there. Possibly a "navigation" API could be provided. - This was left
>open.
>Laurent talked about the ST addition to the IPXACT editor. Their use
>model is more viewing than editing, and their extension is for memory maps.
>
>The conclusion from this discussion is one clear candidate that this
>group should work on - an IPXACT 'server'."Business" case:
>Clearly IPXACT servers exist buried inside EDA tools - however, for many
>use cases, it is inappropriate to be using an eda tool simply to
>provide an IPXAC server. The server itself is of little or no value add to the
>tool.
>
>The __KEY__ interoperability interface is the SPIRIT TGI interface.
>However, to achieve adoption, providing the server, and e.g. the IPXACT
>editing tool within Eclipse will greatly help.
>From the EDA companies point of view, they can choose to adopt the
>IPXACT server that is an 'exemplar' implementation, or use their own - the
>choice is theres, and probably invisible to the user. Their own solution
>may have a much richer interface than the one provided as an exemplar
>tool - which - at least initially - will be focused on supporting the
>Eclipse IPXACT tool.
>So - this is the "vision" I have.
>
>How can we make it a reality?
>
>I would encourage informal discussions over the next couple of weeks - I
>will try and get round to talking to as many of you as I can, by our
>next meeting I will try to bring some concrete proposals to the table.
>Our next meeting is scheduled for 23rd June at 5pm CEST, on the same
>numbers.
>Cheers
>
>Mark
>
>
>
>
>--
>
>
--