On Tue, Oct 31, 2017 at 04:02:07PM +0200, Yuval Shaia wrote:
> > + if (is_srq) {
> > + if (dev->dsr->caps.max_srq == 0) {
>
> Let us be consist, so if in function pvrdma_init_device the assumption is
> that max_srq can have negative value then also check it here. Or on the
> other hands if not (and i assume it is not) then modify the check on
> pvrdma_init_device.
>
> Also, suggesting:
> if (is_srq && !dev->dsr->caps.max_srq)
> Or (if you do not accept my comment):
> if (is_srq && dev->dsr->caps.max_srq == 0)
>
Sure. I'll take your comment and fix the check in
pvrdma_register_device as well.
> > + if (init_attr->srq_type != IB_SRQT_BASIC)
> > + return ERR_PTR(-EINVAL);
> > +
> > + if (init_attr->attr.max_wr > dev->dsr->caps.max_srq_wr ||
> > + init_attr->attr.max_sge > dev->dsr->caps.max_srq_sge)
>
> Again - let us be consist, if we got warning in pvrdma_create_qp we should
> get one here as well.
>
Added a warning here, as well as in the srq_type check above.
> > + if (!(pd->uobject && udata)) {
> > + /* No support for kernel clients. */
> > + ret = -EOPNOTSUPP;
> > + goto err_srq;
> > + }
>
> Why this check cannot be performed before the above initialization??
> Also, suggesting a debug message here.
>
Moved check to beginning, and added a warning here.
Thanks for the review!
Bryan
--
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
