@@ -619,14 +619,10 @@ struct mlx5e_wqe_frag_info {
 	bool last_in_page;
 };
 
-struct mlx5e_umr_dma_info {
-	struct mlx5e_dma_info  dma_info[MLX5_MPWRQ_PAGES_PER_WQE];
-};
-
 struct mlx5e_mpw_info {
-	struct mlx5e_umr_dma_info umr;
 	u16 consumed_strides;
 	DECLARE_BITMAP(xdp_xmit_bitmap, MLX5_MPWRQ_PAGES_PER_WQE);
+	struct mlx5e_dma_info dma_info[];
 };
 
 #define MLX5E_MAX_RX_FRAGS 4

@@ -30,7 +30,7 @@ struct sk_buff *mlx5e_xsk_skb_from_cqe_mpwrq_linear(struct mlx5e_rq *rq,
 						    u32 head_offset,
 						    u32 page_idx)
 {
-	struct xdp_buff *xdp = wi->umr.dma_info[page_idx].xsk;
+	struct xdp_buff *xdp = wi->dma_info[page_idx].xsk;
 	struct bpf_prog *prog;
 
 	/* Check packet size. Note LRO doesn't use linear SKB */

@@ -260,10 +260,12 @@ static void mlx5e_rq_shampo_hd_info_free(struct mlx5e_rq *rq)
 static int mlx5e_rq_alloc_mpwqe_info(struct mlx5e_rq *rq, int node)
 {
 	int wq_sz = mlx5_wq_ll_get_size(&rq->mpwqe.wq);
+	size_t alloc_size;
 
-	rq->mpwqe.info = kvzalloc_node(array_size(wq_sz,
-						  sizeof(*rq->mpwqe.info)),
-				       GFP_KERNEL, node);
+	alloc_size = array_size(wq_sz, struct_size(rq->mpwqe.info, dma_info,
+						   MLX5_MPWRQ_PAGES_PER_WQE));
+
+	rq->mpwqe.info = kvzalloc_node(alloc_size, GFP_KERNEL, node);
 	if (!rq->mpwqe.info)
 		return -ENOMEM;
 

@@ -75,6 +75,13 @@ const struct mlx5e_rx_handlers mlx5e_rx_handlers_nic = {
 	.handle_rx_cqe_mpwqe_shampo = mlx5e_handle_rx_cqe_mpwrq_shampo,
 };
 
+static struct mlx5e_mpw_info *mlx5e_get_mpw_info(struct mlx5e_rq *rq, int i)
+{
+	size_t isz = struct_size(rq->mpwqe.info, dma_info, MLX5_MPWRQ_PAGES_PER_WQE);
+
+	return (struct mlx5e_mpw_info *)((char *)rq->mpwqe.info + array_size(i, isz));
+}
+
 static inline bool mlx5e_rx_hw_stamp(struct hwtstamp_config *config)
 {
 	return config->rx_filter == HWTSTAMP_FILTER_ALL;

@@ -478,7 +485,7 @@ static void
 mlx5e_free_rx_mpwqe(struct mlx5e_rq *rq, struct mlx5e_mpw_info *wi, bool recycle)
 {
 	bool no_xdp_xmit;
-	struct mlx5e_dma_info *dma_info = wi->umr.dma_info;
+	struct mlx5e_dma_info *dma_info = wi->dma_info;
 	int i;
 
 	/* A common case for AF_XDP. */

@@ -660,8 +667,8 @@ static int mlx5e_alloc_rx_hd_mpwqe(struct mlx5e_rq *rq)
 
 static int mlx5e_alloc_rx_mpwqe(struct mlx5e_rq *rq, u16 ix)
 {
-	struct mlx5e_mpw_info *wi = &rq->mpwqe.info[ix];
-	struct mlx5e_dma_info *dma_info = &wi->umr.dma_info[0];
+	struct mlx5e_mpw_info *wi = mlx5e_get_mpw_info(rq, ix);
+	struct mlx5e_dma_info *dma_info = &wi->dma_info[0];
 	struct mlx5e_icosq *sq = rq->icosq;
 	struct mlx5_wq_cyc *wq = &sq->wq;
 	struct mlx5e_umr_wqe *umr_wqe;

@@ -768,7 +775,7 @@ void mlx5e_shampo_dealloc_hd(struct mlx5e_rq *rq, u16 len, u16 start, bool close
 
 static void mlx5e_dealloc_rx_mpwqe(struct mlx5e_rq *rq, u16 ix)
 {
-	struct mlx5e_mpw_info *wi = &rq->mpwqe.info[ix];
+	struct mlx5e_mpw_info *wi = mlx5e_get_mpw_info(rq, ix);
 	/* Don't recycle, this function is called on rq/netdev close */
 	mlx5e_free_rx_mpwqe(rq, wi, false);
 }

@@ -1795,7 +1802,7 @@ static void mlx5e_handle_rx_cqe_mpwrq_rep(struct mlx5e_rq *rq, struct mlx5_cqe64
 {
 	u16 cstrides       = mpwrq_get_cqe_consumed_strides(cqe);
 	u16 wqe_id         = be16_to_cpu(cqe->wqe_id);
-	struct mlx5e_mpw_info *wi = &rq->mpwqe.info[wqe_id];
+	struct mlx5e_mpw_info *wi = mlx5e_get_mpw_info(rq, wqe_id);
 	u16 stride_ix      = mpwrq_get_cqe_stride_index(cqe);
 	u32 wqe_offset     = stride_ix << rq->mpwqe.log_stride_sz;
 	u32 head_offset    = wqe_offset & (PAGE_SIZE - 1);

@@ -1878,7 +1885,7 @@ mlx5e_skb_from_cqe_mpwrq_nonlinear(struct mlx5e_rq *rq, struct mlx5e_mpw_info *w
 				   u16 cqe_bcnt, u32 head_offset, u32 page_idx)
 {
 	u16 headlen = min_t(u16, MLX5E_RX_MAX_HEAD, cqe_bcnt);
-	struct mlx5e_dma_info *di = &wi->umr.dma_info[page_idx];
+	struct mlx5e_dma_info *di = &wi->dma_info[page_idx];
 	u32 frag_offset    = head_offset + headlen;
 	u32 byte_cnt       = cqe_bcnt - headlen;
 	struct mlx5e_dma_info *head_di = di;

@@ -1912,7 +1919,7 @@ static struct sk_buff *
 mlx5e_skb_from_cqe_mpwrq_linear(struct mlx5e_rq *rq, struct mlx5e_mpw_info *wi,
 				u16 cqe_bcnt, u32 head_offset, u32 page_idx)
 {
-	struct mlx5e_dma_info *di = &wi->umr.dma_info[page_idx];
+	struct mlx5e_dma_info *di = &wi->dma_info[page_idx];
 	u16 rx_headroom = rq->buff.headroom;
 	struct bpf_prog *prog;
 	struct sk_buff *skb;

@@ -2078,7 +2085,7 @@ static void mlx5e_handle_rx_cqe_mpwrq_shampo(struct mlx5e_rq *rq, struct mlx5_cq
 	struct mlx5e_mpw_info *wi;
 	struct mlx5_wq_ll *wq;
 
-	wi = &rq->mpwqe.info[wqe_id];
+	wi = mlx5e_get_mpw_info(rq, wqe_id);
 	wi->consumed_strides += cstrides;
 
 	if (unlikely(MLX5E_RX_ERR_CQE(cqe))) {

@@ -2124,7 +2131,7 @@ static void mlx5e_handle_rx_cqe_mpwrq_shampo(struct mlx5e_rq *rq, struct mlx5_cq
 	}
 
 	if (likely(head_size)) {
-		di = &wi->umr.dma_info[page_idx];
+		di = &wi->dma_info[page_idx];
 		mlx5e_fill_skb_data(*skb, rq, di, data_bcnt, data_offset);
 	}
 

@@ -2147,7 +2154,7 @@ static void mlx5e_handle_rx_cqe_mpwrq(struct mlx5e_rq *rq, struct mlx5_cqe64 *cq
 {
 	u16 cstrides       = mpwrq_get_cqe_consumed_strides(cqe);
 	u16 wqe_id         = be16_to_cpu(cqe->wqe_id);
-	struct mlx5e_mpw_info *wi = &rq->mpwqe.info[wqe_id];
+	struct mlx5e_mpw_info *wi = mlx5e_get_mpw_info(rq, wqe_id);
 	u16 stride_ix      = mpwrq_get_cqe_stride_index(cqe);
 	u32 wqe_offset     = stride_ix << rq->mpwqe.log_stride_sz;
 	u32 head_offset    = wqe_offset & (PAGE_SIZE - 1);