[openib-general] [PATCH v4 01/13] Linux RDMA Core Changes

[Steve Wise]

On Thu, 2007-01-04 at 13:34 -0800, Roland Dreier wrote:
> OK, I'm back from vacation today.
> 
> Anyway I don't have a definitive statement on this right now.  I guess
> I agree that I don't like having an extra parameter to a function that
> should be pretty fast (although req notify isn't quite as hot as
> something like posting a send request or polling a cq), given that it
> adds measurable overhead.  (And I am surprised that the overhead is
> measurable, since 3 arguments still fit in registers, but OK).
> 
> I also agree that adding an extra entry point just to pass in the user
> data is ugly, and also racy.
> 
> Giving the kernel driver a pointer it can read seems OK I guess,
> although it's a little ugly to have a backdoor channel like that.
> 
> I'm somewhat surprised the driver has to go into the kernel to rearm a
> CQ -- what makes the operation need kernel privileges?  (Sorry for not
> reading the code)
> -

Rearming the CQ requires reading and writing to a global adapter
register that is shared and thus needs to be protected.  They didn't
architect the rearm to be a direct user operation.

Steve.