From b36c598c999c628130f6743dc2362585360de65c Mon Sep 17 00:00:00 2001 From: Anirudh Venkataramanan Date: Thu, 9 Aug 2018 06:29:45 -0700 Subject: ice: Updates to Tx scheduler code 1) The maximum device nodes is a global value and shared by the whole device. Add element AQ command would fail if there is no space to add new nodes so the check for max nodes isn't required. So remove ice_sched_get_num_nodes_per_layer and ice_sched_val_max_nodes. 2) In ice_sched_add_elems, set default node's CIR/EIR bandwidth weight. 3) Fix default scheduler topology buffer size as the firmware expects a 4KB buffer at all times, and will error out if one of any other size is provided. 4) In the latest spec, max children per node per layer is replaced by max sibling group size. Now it provides the max children of the below layer node, not the current layer node. 5) Fix some newline/whitespace issues for consistency. Signed-off-by: Anirudh Venkataramanan Tested-by: Tony Brelinski Signed-off-by: Jeff Kirsher --- drivers/net/ethernet/intel/ice/ice_common.c | 7 +++++++ 1 file changed, 7 insertions(+) (limited to 'drivers/net/ethernet/intel/ice/ice_common.c') diff --git a/drivers/net/ethernet/intel/ice/ice_common.c b/drivers/net/ethernet/intel/ice/ice_common.c index 53cbfd942d03..b315655eab27 100644 --- a/drivers/net/ethernet/intel/ice/ice_common.c +++ b/drivers/net/ethernet/intel/ice/ice_common.c @@ -527,6 +527,13 @@ enum ice_status ice_init_hw(struct ice_hw *hw) if (status) goto err_unroll_sched; + /* need a valid SW entry point to build a Tx tree */ + if (!hw->sw_entry_point_layer) { + ice_debug(hw, ICE_DBG_SCHED, "invalid sw entry point\n"); + status = ICE_ERR_CFG; + goto err_unroll_sched; + } + status = ice_init_fltr_mgmt_struct(hw); if (status) goto err_unroll_sched; -- cgit v1.2.3