On 05/27/2016 10:44 AM, Leon Romanovsky wrote:

On Fri, May 27, 2016 at 09:32:33AM -0400, Doug Ledford wrote:

quoted

On 5/27/2016 9:13 AM, Leon Romanovsky wrote:

quoted

On Fri, May 27, 2016 at 07:44:15AM -0400, Dennis Dalessandro wrote:

quoted

quoted

Separate char device? IOCTLs per-device vs. global IOCTLs per subsystem?
Role of the IB CORE code in the driver management?

Really Leon?  The qib driver has the *exact* same issue, and it sits out
of staging.  If moving this driver out of staging somehow stops us from
making the new ABI while the qib driver not being in staging doesn't
prevent it, then we are a bunch of idiots.  This would appear to me to
be a very disingenuous complaint on your part.

Doug,
Did you read my question?

I feel that I need to repeat my main point again: "It is OK to move hfi1 driver
out-of-staging"

Good, I'm glad that's settled.

and repeat the question too: "Will the hfi1 interface
decisions limit our ABI work?".

And I answered this already: If it does, we are idiots.  If you would
prefer the answer a different way.  Of course not.  Why would you think
it would?

There is no politics here, just my PERSONAL opinion and believe that ABI work can be
done in months time frame. And again, "It is OK to move hfi1 driver out-of-staging".

If you really didn't object to the driver coming out of staging, then
maybe asking these questions in response to my pull request, thereby
throwing this conversation in Linus' lap, was not the most appropriate
thing to do.

-- 
Doug Ledford [off-list ref]
              GPG KeyID: 0E572FDD