The detection of an uneven number of queues on the given memory windows
was not correct. The mw_num is zero based and the mod should be
division to spread them evenly over the mw's.
Signed-off-by: Jon Mason <jon.mason@intel.com>
WARN_ON(nt->mw[mw_num].virt_addr == NULL);
- if (nt->max_qps % mw_max && mw_num < nt->max_qps % mw_max)
+ if (nt->max_qps % mw_max && mw_num + 1 < nt->max_qps / mw_max)
num_qps_mw = nt->max_qps / mw_max + 1;
else
num_qps_mw = nt->max_qps / mw_max;
qp->client_ready = NTB_LINK_DOWN;
qp->event_handler = NULL;
- if (nt->max_qps % mw_max && mw_num < nt->max_qps % mw_max)
+ if (nt->max_qps % mw_max && mw_num + 1 < nt->max_qps / mw_max)
num_qps_mw = nt->max_qps / mw_max + 1;
else
num_qps_mw = nt->max_qps / mw_max;