X-Git-Url: http://git.droids-corp.org/?a=blobdiff_plain;f=drivers%2Fnet%2Fmlx5%2Fmlx5_rxtx_vec_sse.h;h=7fee4355cfe7fb2d23a030a145cb5912aa71c419;hb=6e695b0cda4728ec37bd0aa2310d4b967f4ca86d;hp=d4df9816aa96dd3ffd7534fccb248f47075a442b;hpb=54c2d46b160f8ad0bff0977812bf871ca5dd8241;p=dpdk.git

diff --git a/drivers/net/mlx5/mlx5_rxtx_vec_sse.h b/drivers/net/mlx5/mlx5_rxtx_vec_sse.h
index d4df9816aa..7fee4355cf 100644
--- a/drivers/net/mlx5/mlx5_rxtx_vec_sse.h
+++ b/drivers/net/mlx5/mlx5_rxtx_vec_sse.h
@@ -197,8 +197,8 @@ rxq_cq_decompress_v(struct mlx5_rxq_data *rxq, volatile struct mlx5_cqe *cq,
 				const __m128i flow_mark_adj =
 					_mm_set_epi32(-1, -1, -1, -1);
 				const __m128i flow_mark_shuf =
-					_mm_set_epi8(-1,  1,  0,  4,
-						     -1,  9,  8, 12,
+					_mm_set_epi8(-1,  9,  8, 12,
+						     -1,  1,  0,  4,
 						     -1, -1, -1, -1,
 						     -1, -1, -1, -1);
 				const __m128i ft_mask =
@@ -565,7 +565,7 @@ rxq_cq_process_v(struct mlx5_rxq_data *rxq, volatile struct mlx5_cqe *cq,
 	const __m128i flow_mark_adj = _mm_set_epi32(rxq->mark * (-1), 0, 0, 0);
 	/*
 	 * A. load first Qword (8bytes) in one loop.
-	 * B. copy 4 mbuf pointers from elts ring to returing pkts.
+	 * B. copy 4 mbuf pointers from elts ring to returning pkts.
 	 * C. load remained CQE data and extract necessary fields.
 	 *    Final 16bytes cqes[] extracted from original 64bytes CQE has the
 	 *    following structure:
@@ -768,15 +768,26 @@ rxq_cq_process_v(struct mlx5_rxq_data *rxq, volatile struct mlx5_cqe *cq,
 		if (rxq->dynf_meta) {
 			/* This code is subject for futher optimization. */
 			int32_t offs = rxq->flow_meta_offset;
+			uint32_t mask = rxq->flow_meta_port_mask;
+			uint32_t shift =
+				__builtin_popcount(rxq->flow_meta_port_mask);
 
 			*RTE_MBUF_DYNFIELD(pkts[pos], offs, uint32_t *) =
-				cq[pos].flow_table_metadata;
+				(rte_be_to_cpu_32
+				(cq[pos].flow_table_metadata) >> shift) &
+				mask;
 			*RTE_MBUF_DYNFIELD(pkts[pos + 1], offs, uint32_t *) =
-				cq[pos + p1].flow_table_metadata;
+				(rte_be_to_cpu_32
+				(cq[pos + p1].flow_table_metadata) >> shift) &
+				mask;
 			*RTE_MBUF_DYNFIELD(pkts[pos + 2], offs, uint32_t *) =
-				cq[pos + p2].flow_table_metadata;
+				(rte_be_to_cpu_32
+				(cq[pos + p2].flow_table_metadata) >> shift) &
+				mask;
 			*RTE_MBUF_DYNFIELD(pkts[pos + 3], offs, uint32_t *) =
-				cq[pos + p3].flow_table_metadata;
+				(rte_be_to_cpu_32
+				(cq[pos + p3].flow_table_metadata) >> shift) &
+				mask;
 			if (*RTE_MBUF_DYNFIELD(pkts[pos], offs, uint32_t *))
 				pkts[pos]->ol_flags |= rxq->flow_meta_mask;
 			if (*RTE_MBUF_DYNFIELD(pkts[pos + 1], offs, uint32_t *))