Re: kernel memory registration (was: RDMA/core: Transport-independent access flags)

['Christoph Hellwig' <hch-wEGCiKHe2LqWVfeAwA7xHQ@public.gmane.org>]

On Fri, Jul 10, 2015 at 11:55:29AM +0300, Sagi Grimberg wrote:
> If there is one thing worse than a complicated API, it is a restrictive
> one. I'd much rather ULPs just having a simple API for registering
> memory.

Quite to the contrary.  The complex API almost asks for weird abuses
and twists.  The too restrictive one means people that want to extend
it need to start a discussion on how we extent the API, which improves
the chances to come up with something sensible dramatically.

> >I expect all these calls would be function pointers, and each driver
> >would provide a function pointer that is optimal for it's use. Eg mlx4
> >would provide a pointer that used the S/G list, then falls back to
> >FRMR if the S/G list is exhausted. The core code would provide a
> >toolbox of common functions the drivers can use here.
> 
> Maybe it's just me, but I can't help but wander if this is facilitating
> an atmosphere where drivers will keep finding new ways to abuse even
> the most simple operations.

I'm not too excited about moving the code in the drivers.  The RDMA
subsystem actually has a lot more hardware drivers than ULDs, so moving
logic into them seems like a major step backwards.  From my journeys
into the drivers it rather seems like they are doing too much work
already.

Having a few schemes availabe in the core code that the driver can chose
from seems like a much more sensible option.
--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html