On Thu, Dec 10, 2015 at 02:29:18PM +0200, Haggai Eran wrote:
On 07/12/2015 22:43, Dennis Dalessandro wrote:
struct rvt_dev_info {
+ /*
+ * Prior to calling for registration the driver will be responsible for
+ * allocating space for this structure. The driver will also need to
+ * allocate space for any private device or per port data structures.
+ * Alternatively rvt could do this allocation and the registration API
+ * would then change to accept an "extra" piece to allocate.
I don't think you need rvt to allocate the private data, but even if you
decide to do that, there's no need for a comment that describes all the
alternative designs here.
Agree.
This is another of those comments which was meant to provide an early look
at the design (and alternatives) when I first posted to GitHub. The
underlying issue here is who allocates the rvt_dev_info.
The approach I went with was having the driver do the allocation. This also
means there is no need for any "private" data in the rvt_dev_info structure.
If drivers need some private struct they can just create a data structure
and embed both rvt_dev_info and that private data. See struct qib_ibdev.
This comment needs updated/removed.
+ *
+ * The driver will also be
+ * responsible for filling in certain members of dparms.props
+ */
--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html