Thread (1 message) 1 message, 1 author, 2016-05-24

RE: [RFC] rdma/uverbs: Sketch for an ioctl framework

From: Liran Liss <hidden>
Date: 2016-05-24 16:02:04

From: linux-rdma-owner-u79uwXL29TY76Z2rM5mHXA@public.gmane.org [mailto:linux-rdma-
owner-u79uwXL29TY76Z2rM5mHXA@public.gmane.org] On Behalf Of Hefty, Sean
...
The following is a sketch, or outline, for an ioctl framework.
The purpose of the patch is to drive discussion and feedback on various design
decisions rather than low-level comments on the code, so that the changes can
be incorporated earlier in the development cycle.  The code is entirely untested
and some functionality is not coded.
This approach focuses on facilities for arbitrary device-specific objects (which could basically fit GPUs, NVMe, whatever...), while our primary goals are to replace write() with ioctl() and strive for a common *rdma* object model. 

In contrast, the patchset [1], focuses on the task at hand. The emphasis is on a common framework for parsing and parameter validation. It follows the agreed-upon concepts such as a single ioctl request code for common objects, natural mapping of the existing object mode (and hence easier backporting), typed attributes for extending the objects to accommodate multiple vendor features, and a unique ioctl request code for vendor-specific commands.
--Liran

[1] http://permalink.gmane.org/gmane.linux.drivers.rdma/36984
 
--
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
Keyboard shortcuts
hback out one level
jnext message in thread
kprevious message in thread
ldrill in
Escclose help / fold thread tree
?toggle this help