ice: split ice_ring onto Tx/Rx separate structs

While it was convenient to have a generic ring structure that served
both Tx and Rx sides, next commits are going to introduce several
Tx-specific fields, so in order to avoid hurting the Rx side, let's
pull out the Tx ring onto new ice_tx_ring and ice_rx_ring structs.

Rx ring could be handled by the old ice_ring which would reduce the code
churn within this patch, but this would make things asymmetric.

Make the union out of the ring container within ice_q_vector so that it
is possible to iterate over newly introduced ice_tx_ring.

Remove the @size as it's only accessed from control path and it can be
calculated pretty easily.

Change definitions of ice_update_ring_stats and
ice_fetch_u64_stats_per_ring so that they are ring agnostic and can be
used for both Rx and Tx rings.

Sizes of Rx and Tx ring structs are 256 and 192 bytes, respectively. In
Rx ring xdp_rxq_info occupies its own cacheline, so it's the major
difference now.

Signed-off-by: Maciej Fijalkowski <maciej.fijalkowski@intel.com>
Tested-by: Gurucharan G <gurucharanx.g@intel.com>
Signed-off-by: Tony Nguyen <anthony.l.nguyen@intel.com>

1 files changed, 69 insertions, 40 deletions

diff --git a/drivers/net/ethernet/intel/ice/ice_txrx.h b/drivers/net/ethernet/intel/ice/ice_txrx.h
index bf40608f9bac..dcbdfe649a09 100644
--- a/drivers/net/ethernet/intel/ice/ice_txrx.h
+++ b/drivers/net/ethernet/intel/ice/ice_txrx.h
@@ -154,7 +154,7 @@ struct ice_tx_buf {
 
 struct ice_tx_offload_params {
 	u64 cd_qw1;
-	struct ice_ring *tx_ring;
+	struct ice_tx_ring *tx_ring;
 	u32 td_cmd;
 	u32 td_offset;
 	u32 td_l2tag1;
@@ -251,9 +251,9 @@ enum ice_dynamic_itr {
 #define ICE_TX_LEGACY	1
 
 /* descriptor ring, associated with a VSI */
-struct ice_ring {
+struct ice_rx_ring {
 	/* CL1 - 1st cacheline starts here */
-	struct ice_ring *next;		/* pointer to next ring in q_vector */
+	struct ice_rx_ring *next;	/* pointer to next ring in q_vector */
 	void *desc;			/* Descriptor ring memory */
 	struct device *dev;		/* Used for DMA mapping */
 	struct net_device *netdev;	/* netdev ring maps to */
@@ -261,13 +261,13 @@ struct ice_ring {
 	struct ice_q_vector *q_vector;	/* Backreference to associated vector */
 	u8 __iomem *tail;
 	union {
-		struct ice_tx_buf *tx_buf;
 		struct ice_rx_buf *rx_buf;
 		struct xdp_buff **xdp_buf;
 	};
 	/* CL2 - 2nd cacheline starts here */
+	struct xdp_rxq_info xdp_rxq;
+	/* CL3 - 3rd cacheline starts here */
 	u16 q_index;			/* Queue number of ring */
-	u16 q_handle;			/* Queue handle per TC */
 
 	u16 count;			/* Number of descriptors */
 	u16 reg_idx;			/* HW register index of the ring */
@@ -276,56 +276,79 @@ struct ice_ring {
 	u16 next_to_use;
 	u16 next_to_clean;
 	u16 next_to_alloc;
+	u16 rx_offset;
+	u16 rx_buf_len;
 
 	/* stats structs */
+	struct ice_rxq_stats rx_stats;
 	struct ice_q_stats	stats;
 	struct u64_stats_sync syncp;
-	union {
-		struct ice_txq_stats tx_stats;
-		struct ice_rxq_stats rx_stats;
-	};
 
 	struct rcu_head rcu;		/* to avoid race on free */
-	DECLARE_BITMAP(xps_state, ICE_TX_NBITS);	/* XPS Config State */
+	/* CL4 - 3rd cacheline starts here */
 	struct bpf_prog *xdp_prog;
 	struct xsk_buff_pool *xsk_pool;
-	u16 rx_offset;
-	/* CL3 - 3rd cacheline starts here */
-	struct xdp_rxq_info xdp_rxq;
 	struct sk_buff *skb;
-	/* CLX - the below items are only accessed infrequently and should be
-	 * in their own cache line if possible
-	 */
-#define ICE_TX_FLAGS_RING_XDP		BIT(0)
+	dma_addr_t dma;			/* physical address of ring */
 #define ICE_RX_FLAGS_RING_BUILD_SKB	BIT(1)
+	u64 cached_phctime;
+	u8 dcb_tc;			/* Traffic class of ring */
+	u8 ptp_rx;
 	u8 flags;
+} ____cacheline_internodealigned_in_smp;
+
+struct ice_tx_ring {
+	/* CL1 - 1st cacheline starts here */
+	struct ice_tx_ring *next;	/* pointer to next ring in q_vector */
+	void *desc;			/* Descriptor ring memory */
+	struct device *dev;		/* Used for DMA mapping */
+	u8 __iomem *tail;
+	struct ice_tx_buf *tx_buf;
+	struct ice_q_vector *q_vector;	/* Backreference to associated vector */
+	struct net_device *netdev;	/* netdev ring maps to */
+	struct ice_vsi *vsi;		/* Backreference to associated VSI */
+	/* CL2 - 2nd cacheline starts here */
 	dma_addr_t dma;			/* physical address of ring */
-	unsigned int size;		/* length of descriptor ring in bytes */
+	u16 next_to_use;
+	u16 next_to_clean;
+	u16 count;			/* Number of descriptors */
+	u16 q_index;			/* Queue number of ring */
+	struct xsk_buff_pool *xsk_pool;
+
+	/* stats structs */
+	struct ice_q_stats	stats;
+	struct u64_stats_sync syncp;
+	struct ice_txq_stats tx_stats;
+
+	/* CL3 - 3rd cacheline starts here */
+	struct rcu_head rcu;		/* to avoid race on free */
+	DECLARE_BITMAP(xps_state, ICE_TX_NBITS);	/* XPS Config State */
+	struct ice_ptp_tx *tx_tstamps;
 	u32 txq_teid;			/* Added Tx queue TEID */
-	u16 rx_buf_len;
+	u16 q_handle;			/* Queue handle per TC */
+	u16 reg_idx;			/* HW register index of the ring */
+#define ICE_TX_FLAGS_RING_XDP		BIT(0)
+	u8 flags;
 	u8 dcb_tc;			/* Traffic class of ring */
-	struct ice_ptp_tx *tx_tstamps;
-	u64 cached_phctime;
-	u8 ptp_rx:1;
-	u8 ptp_tx:1;
+	u8 ptp_tx;
 } ____cacheline_internodealigned_in_smp;
 
-static inline bool ice_ring_uses_build_skb(struct ice_ring *ring)
+static inline bool ice_ring_uses_build_skb(struct ice_rx_ring *ring)
 {
 	return !!(ring->flags & ICE_RX_FLAGS_RING_BUILD_SKB);
 }
 
-static inline void ice_set_ring_build_skb_ena(struct ice_ring *ring)
+static inline void ice_set_ring_build_skb_ena(struct ice_rx_ring *ring)
 {
 	ring->flags |= ICE_RX_FLAGS_RING_BUILD_SKB;
 }
 
-static inline void ice_clear_ring_build_skb_ena(struct ice_ring *ring)
+static inline void ice_clear_ring_build_skb_ena(struct ice_rx_ring *ring)
 {
 	ring->flags &= ~ICE_RX_FLAGS_RING_BUILD_SKB;
 }
 
-static inline bool ice_ring_is_xdp(struct ice_ring *ring)
+static inline bool ice_ring_is_xdp(struct ice_tx_ring *ring)
 {
 	return !!(ring->flags & ICE_TX_FLAGS_RING_XDP);
 }
@@ -337,7 +360,10 @@ enum ice_container_type {
 
 struct ice_ring_container {
 	/* head of linked-list of rings */
-	struct ice_ring *ring;
+	union {
+		struct ice_rx_ring *rx_ring;
+		struct ice_tx_ring *tx_ring;
+	};
 	struct dim dim;		/* data for net_dim algorithm */
 	u16 itr_idx;		/* index in the interrupt vector */
 	/* this matches the maximum number of ITR bits, but in usec
@@ -358,10 +384,13 @@ struct ice_coalesce_stored {
 };
 
 /* iterator for handling rings in ring container */
-#define ice_for_each_ring(pos, head) \
-	for (pos = (head).ring; pos; pos = pos->next)
+#define ice_for_each_rx_ring(pos, head) \
+	for (pos = (head).rx_ring; pos; pos = pos->next)
+
+#define ice_for_each_tx_ring(pos, head) \
+	for (pos = (head).tx_ring; pos; pos = pos->next)
 
-static inline unsigned int ice_rx_pg_order(struct ice_ring *ring)
+static inline unsigned int ice_rx_pg_order(struct ice_rx_ring *ring)
 {
 #if (PAGE_SIZE < 8192)
 	if (ring->rx_buf_len > (PAGE_SIZE / 2))
@@ -374,21 +403,21 @@ static inline unsigned int ice_rx_pg_order(struct ice_ring *ring)
 
 union ice_32b_rx_flex_desc;
 
-bool ice_alloc_rx_bufs(struct ice_ring *rxr, u16 cleaned_count);
+bool ice_alloc_rx_bufs(struct ice_rx_ring *rxr, u16 cleaned_count);
 netdev_tx_t ice_start_xmit(struct sk_buff *skb, struct net_device *netdev);
 u16
 ice_select_queue(struct net_device *dev, struct sk_buff *skb,
 		 struct net_device *sb_dev);
-void ice_clean_tx_ring(struct ice_ring *tx_ring);
-void ice_clean_rx_ring(struct ice_ring *rx_ring);
-int ice_setup_tx_ring(struct ice_ring *tx_ring);
-int ice_setup_rx_ring(struct ice_ring *rx_ring);
-void ice_free_tx_ring(struct ice_ring *tx_ring);
-void ice_free_rx_ring(struct ice_ring *rx_ring);
+void ice_clean_tx_ring(struct ice_tx_ring *tx_ring);
+void ice_clean_rx_ring(struct ice_rx_ring *rx_ring);
+int ice_setup_tx_ring(struct ice_tx_ring *tx_ring);
+int ice_setup_rx_ring(struct ice_rx_ring *rx_ring);
+void ice_free_tx_ring(struct ice_tx_ring *tx_ring);
+void ice_free_rx_ring(struct ice_rx_ring *rx_ring);
 int ice_napi_poll(struct napi_struct *napi, int budget);
 int
 ice_prgm_fdir_fltr(struct ice_vsi *vsi, struct ice_fltr_desc *fdir_desc,
 		   u8 *raw_packet);
-int ice_clean_rx_irq(struct ice_ring *rx_ring, int budget);
-void ice_clean_ctrl_tx_irq(struct ice_ring *tx_ring);
+int ice_clean_rx_irq(struct ice_rx_ring *rx_ring, int budget);
+void ice_clean_ctrl_tx_irq(struct ice_tx_ring *tx_ring);
 #endif /* _ICE_TXRX_H_ */