Alan DuBoff wrote:
> On Tue, 25 Dec 2007, David Gwynne wrote:
>
>> Yes, Sun is very active, which is great, but not relevant here.
>
> Actually it is...read on...
>
>> Sun and OpenSolaris are supposed to be separate and I'm trying to deal
>> with the OpenSolaris community. There is no project in OpenSolaris
>> relating to LSI or the megasas driver. If you think there should be then
>> submit a proposal like I did for the integration of the mfi driver.
>
> They are seperate, but the community is fortunate that Sun had been and is
> still working on it. Existing ARC cases can prevent you from getting a
> similar driver putback, and Sun does have ARC cases for the work that has
> been going on with the LSI driver, AFAIK. So, while it is seperate,
> history can't be changed and there was work going on and much of that work
> is still going on. Unfortunately this is not an either/or situation.
Which particular ARC cases are you referring to here? The ones for mpt
and its follow-ons such as the ones I've been involved with? Or some
other cases/fasttracks which haven't actually been logged with PSARC?
I certainly could not find any in the archives which appear to relate
to LSI1078, let alone the actual MegaRAID SAS chip which mfi drives.
>> Whether mfi is integrated is up to the OpenSolaris community. From what
>> I can tell they are rather positive about mfi but the apparent confusion
>> about Sun's role in OpenSolaris, and this ongoing negotiation they're
>> doing with LSI seems to keep anything from happening when it really
>> shouldn't matter.
>
> I would say that some people are positive.
Some people are positive, some people are negative, and some
people (not you, Alan) don't seem to understand the concept of
this whole "Open" Solaris and how it's different to how Sun
and LSI/{third party vendor du jour} have operated in the past.
> The ARC is still the deciding factor of allowing something from even being
> able to go in.
No, it's the CTeam which is shaping up to be the gating
entity in this case. The relevant ARC decides on the
architectural merits, not the "will we allow this into
the consolidation" questions.
> BTW, does your driver support sparc as well?
Neither David nor I have tested it on sparc, since we don't
currently have the hardware to do that with. Having gone over
the code quite thoroughly I don't see any glaring issues which
would prevent mfi working on sparc too.
James C. McPherson
--
Senior Kernel Software Engineer, Solaris
Sun Microsystems
http://blogs.sun.com/jmcp http://www.jmcp.homeunix.com/blog
_______________________________________________
driver-discuss mailing list
driver-discuss@opensolaris.org
http://mail.opensolaris.org/mailman/listinfo/driver-discuss
