Keith M Wesolowski wrote:
> On Tue, Dec 18, 2007 at 11:13:25PM -0500, Richard Lowe wrote:
> 
>> I'm sure the point's being raised to you in private will be raised to
>> the ARC and the C-Team (out of the fear of precise duplication), and
>> be argued on their technical merit.
> 
> Duplication is certainly bad.  I was thinking the team should really
> go engage with the other driver's Project Team first, but I couldn't
> find them in the project directory[0].  Did the Device Drivers Group
> create this Project and we just failed to set it up?
> 
> [0] http://www.opensolaris.org/os/projects/
> 

The problem is that 'drivers from third parties' is a very informal, very 
touch-and-go, million-times-harder-to-predict process than 'drivers from Sun'
(or apparently 'drivers from the OSS community').  Many companies are just not 
very forthcoming about their progress, and even when they are, there are legal 
forests that apparently simply cannot be bypassed or traversed quickly.

that doesn't make it right, and it would also be awesome if "two drivers for 
the 
same device" was an easier thing to handle in Solaris, but the fact is, we've 
been burned in the past by "let's just do this driver now, and then later 
replace it with a better one"...so there's a strong desire to do the complete 
driver first.  The stumbling block is always upgrade, because you need to 
translate from the old physical paths to the new physical paths, and need to 
dig 
around in all the places on the disk that have the old paths stored, and sadly, 
that's even beyond some of the Solaris places (rumor has it, for instance, that 
physical paths are stored inside some third-party disk management or 
third-party 
database configurations).

...which still doesn't make it right.  I'm just trying to provide some 
alternative perspective, and I don't know any details of the LSI situation 
directly, so I don't know what's dominating here.

It's probably best to try to work this internally to Sun, however, since I'm 
assuming, and I'll bet correctly, that the details of the relationship are not 
things that LSI/Sun want to be public.

David, I'm thrilled that your work is done, and I wish there were an easier 
answer that we could all support without reservation.  Maybe if we can make the 
argument that "on that halcyon day, when LSI provides the complete solution, 
upgrade will magically work correctly without extra effort", it would help, but 
I'm afraid that's technically far-fetched.
_______________________________________________
driver-discuss mailing list
[email protected]
http://mail.opensolaris.org/mailman/listinfo/driver-discuss

Reply via email to