i40e/i40evf: Add support for padding start of frames

This patch adds padding to the start of frames to make room for headroom
for us to eventually start using build_skb.  Right now we guarantee at
least NET_SKB_PAD + NET_IP_ALIGN, however we allocate more space if more is
available.  For example on x86 the headroom should be 192 bytes.

On systems that have too large of a cache line size to support storing 1.5K
padding and shared info we default to using 3K buffers and reserve
everything that isn't used for skb_shared_info or the data buffer for
headroom.

Change-ID: I33c641c9a1ea10cf7cc484c2d20985368d2d709a
Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

6 files changed, 179 insertions, 8 deletions

diff --git a/drivers/net/ethernet/intel/i40e/i40e_main.c b/drivers/net/ethernet/intel/i40e/i40e_main.c
index 97489d69029a..b6ec9beeebff 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_main.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_main.c
@@ -3038,6 +3038,12 @@ static int i40e_configure_rx_ring(struct i40e_ring *ring)
 		return -ENOMEM;
 	}
 
+	/* configure Rx buffer alignment */
+	if (!vsi->netdev || (vsi->back->flags & I40E_FLAG_LEGACY_RX))
+		clear_ring_build_skb_enabled(ring);
+	else
+		set_ring_build_skb_enabled(ring);
+
 	/* cache tail for quicker writes, and clear the reg before use */
 	ring->tail = hw->hw_addr + I40E_QRX_TAIL(pf_q);
 	writel(0, ring->tail);
@@ -3079,7 +3085,8 @@ static int i40e_vsi_configure_rx(struct i40e_vsi *vsi)
 		vsi->max_frame = I40E_MAX_RXBUFFER;
 		vsi->rx_buf_len = I40E_RXBUFFER_2048;
 #if (PAGE_SIZE < 8192)
-	} else if (vsi->netdev->mtu <= ETH_DATA_LEN) {
+	} else if (!I40E_2K_TOO_SMALL_WITH_PADDING &&
+		   (vsi->netdev->mtu <= ETH_DATA_LEN)) {
 		vsi->max_frame = I40E_RXBUFFER_1536 - NET_IP_ALIGN;
 		vsi->rx_buf_len = I40E_RXBUFFER_1536 - NET_IP_ALIGN;
 #endif
diff --git a/drivers/net/ethernet/intel/i40e/i40e_txrx.c b/drivers/net/ethernet/intel/i40e/i40e_txrx.c
index bee16726c104..f15e1bcf3555 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_txrx.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_txrx.c
@@ -1248,6 +1248,17 @@ static inline void i40e_release_rx_desc(struct i40e_ring *rx_ring, u32 val)
 }
 
 /**
+ * i40e_rx_offset - Return expected offset into page to access data
+ * @rx_ring: Ring we are requesting offset of
+ *
+ * Returns the offset value for ring into the data buffer.
+ */
+static inline unsigned int i40e_rx_offset(struct i40e_ring *rx_ring)
+{
+	return ring_uses_build_skb(rx_ring) ? I40E_SKB_PAD : 0;
+}
+
+/**
  * i40e_alloc_mapped_page - recycle or make a new page
  * @rx_ring: ring to use
  * @bi: rx_buffer struct to modify
@@ -1291,7 +1302,7 @@ static bool i40e_alloc_mapped_page(struct i40e_ring *rx_ring,
 
 	bi->dma = dma;
 	bi->page = page;
-	bi->page_offset = 0;
+	bi->page_offset = i40e_rx_offset(rx_ring);
 
 	/* initialize pagecnt_bias to 1 representing we fully own page */
 	bi->pagecnt_bias = 1;
@@ -1696,7 +1707,7 @@ static void i40e_add_rx_frag(struct i40e_ring *rx_ring,
 #if (PAGE_SIZE < 8192)
 	unsigned int truesize = i40e_rx_pg_size(rx_ring) / 2;
 #else
-	unsigned int truesize = SKB_DATA_ALIGN(size);
+	unsigned int truesize = SKB_DATA_ALIGN(size + i40e_rx_offset(rx_ring));
 #endif
 
 	skb_add_rx_frag(skb, skb_shinfo(skb)->nr_frags, rx_buffer->page,
diff --git a/drivers/net/ethernet/intel/i40e/i40e_txrx.h b/drivers/net/ethernet/intel/i40e/i40e_txrx.h
index 2f618539a436..f5de51124cae 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_txrx.h
+++ b/drivers/net/ethernet/intel/i40e/i40e_txrx.h
@@ -135,6 +135,58 @@ enum i40e_dyn_idx_t {
 #define I40E_RX_DMA_ATTR \
 	(DMA_ATTR_SKIP_CPU_SYNC | DMA_ATTR_WEAK_ORDERING)
 
+/* Attempt to maximize the headroom available for incoming frames.  We
+ * use a 2K buffer for receives and need 1536/1534 to store the data for
+ * the frame.  This leaves us with 512 bytes of room.  From that we need
+ * to deduct the space needed for the shared info and the padding needed
+ * to IP align the frame.
+ *
+ * Note: For cache line sizes 256 or larger this value is going to end
+ *	 up negative.  In these cases we should fall back to the legacy
+ *	 receive path.
+ */
+#if (PAGE_SIZE < 8192)
+#define I40E_2K_TOO_SMALL_WITH_PADDING \
+((NET_SKB_PAD + I40E_RXBUFFER_1536) > SKB_WITH_OVERHEAD(I40E_RXBUFFER_2048))
+
+static inline int i40e_compute_pad(int rx_buf_len)
+{
+	int page_size, pad_size;
+
+	page_size = ALIGN(rx_buf_len, PAGE_SIZE / 2);
+	pad_size = SKB_WITH_OVERHEAD(page_size) - rx_buf_len;
+
+	return pad_size;
+}
+
+static inline int i40e_skb_pad(void)
+{
+	int rx_buf_len;
+
+	/* If a 2K buffer cannot handle a standard Ethernet frame then
+	 * optimize padding for a 3K buffer instead of a 1.5K buffer.
+	 *
+	 * For a 3K buffer we need to add enough padding to allow for
+	 * tailroom due to NET_IP_ALIGN possibly shifting us out of
+	 * cache-line alignment.
+	 */
+	if (I40E_2K_TOO_SMALL_WITH_PADDING)
+		rx_buf_len = I40E_RXBUFFER_3072 + SKB_DATA_ALIGN(NET_IP_ALIGN);
+	else
+		rx_buf_len = I40E_RXBUFFER_1536;
+
+	/* if needed make room for NET_IP_ALIGN */
+	rx_buf_len -= NET_IP_ALIGN;
+
+	return i40e_compute_pad(rx_buf_len);
+}
+
+#define I40E_SKB_PAD i40e_skb_pad()
+#else
+#define I40E_2K_TOO_SMALL_WITH_PADDING false
+#define I40E_SKB_PAD (NET_SKB_PAD + NET_IP_ALIGN)
+#endif
+
 /**
  * i40e_test_staterr - tests bits in Rx descriptor status and error fields
  * @rx_desc: pointer to receive descriptor (in le64 format)
@@ -341,7 +393,8 @@ struct i40e_ring {
 	u8 packet_stride;
 
 	u16 flags;
-#define I40E_TXR_FLAGS_WB_ON_ITR	BIT(0)
+#define I40E_TXR_FLAGS_WB_ON_ITR		BIT(0)
+#define I40E_RXR_FLAGS_BUILD_SKB_ENABLED	BIT(1)
 
 	/* stats structs */
 	struct i40e_queue_stats	stats;
@@ -369,6 +422,21 @@ struct i40e_ring {
 					 */
 } ____cacheline_internodealigned_in_smp;
 
+static inline bool ring_uses_build_skb(struct i40e_ring *ring)
+{
+	return !!(ring->flags & I40E_RXR_FLAGS_BUILD_SKB_ENABLED);
+}
+
+static inline void set_ring_build_skb_enabled(struct i40e_ring *ring)
+{
+	ring->flags |= I40E_RXR_FLAGS_BUILD_SKB_ENABLED;
+}
+
+static inline void clear_ring_build_skb_enabled(struct i40e_ring *ring)
+{
+	ring->flags &= ~I40E_RXR_FLAGS_BUILD_SKB_ENABLED;
+}
+
 enum i40e_latency_range {
 	I40E_LOWEST_LATENCY = 0,
 	I40E_LOW_LATENCY = 1,
diff --git a/drivers/net/ethernet/intel/i40evf/i40e_txrx.c b/drivers/net/ethernet/intel/i40evf/i40e_txrx.c
index 6b60c19a794b..a71f81a9291f 100644
--- a/drivers/net/ethernet/intel/i40evf/i40e_txrx.c
+++ b/drivers/net/ethernet/intel/i40evf/i40e_txrx.c
@@ -619,6 +619,17 @@ static inline void i40e_release_rx_desc(struct i40e_ring *rx_ring, u32 val)
 }
 
 /**
+ * i40e_rx_offset - Return expected offset into page to access data
+ * @rx_ring: Ring we are requesting offset of
+ *
+ * Returns the offset value for ring into the data buffer.
+ */
+static inline unsigned int i40e_rx_offset(struct i40e_ring *rx_ring)
+{
+	return ring_uses_build_skb(rx_ring) ? I40E_SKB_PAD : 0;
+}
+
+/**
  * i40e_alloc_mapped_page - recycle or make a new page
  * @rx_ring: ring to use
  * @bi: rx_buffer struct to modify
@@ -662,7 +673,7 @@ static bool i40e_alloc_mapped_page(struct i40e_ring *rx_ring,
 
 	bi->dma = dma;
 	bi->page = page;
-	bi->page_offset = 0;
+	bi->page_offset = i40e_rx_offset(rx_ring);
 
 	/* initialize pagecnt_bias to 1 representing we fully own page */
 	bi->pagecnt_bias = 1;
@@ -1057,7 +1068,7 @@ static void i40e_add_rx_frag(struct i40e_ring *rx_ring,
 #if (PAGE_SIZE < 8192)
 	unsigned int truesize = i40e_rx_pg_size(rx_ring) / 2;
 #else
-	unsigned int truesize = SKB_DATA_ALIGN(size);
+	unsigned int truesize = SKB_DATA_ALIGN(size + i40e_rx_offset(rx_ring));
 #endif
 
 	skb_add_rx_frag(skb, skb_shinfo(skb)->nr_frags, rx_buffer->page,
diff --git a/drivers/net/ethernet/intel/i40evf/i40e_txrx.h b/drivers/net/ethernet/intel/i40evf/i40e_txrx.h
index dc82f65267ec..901282c87cf6 100644
--- a/drivers/net/ethernet/intel/i40evf/i40e_txrx.h
+++ b/drivers/net/ethernet/intel/i40evf/i40e_txrx.h
@@ -122,6 +122,58 @@ enum i40e_dyn_idx_t {
 #define I40E_RX_DMA_ATTR \
 	(DMA_ATTR_SKIP_CPU_SYNC | DMA_ATTR_WEAK_ORDERING)
 
+/* Attempt to maximize the headroom available for incoming frames.  We
+ * use a 2K buffer for receives and need 1536/1534 to store the data for
+ * the frame.  This leaves us with 512 bytes of room.  From that we need
+ * to deduct the space needed for the shared info and the padding needed
+ * to IP align the frame.
+ *
+ * Note: For cache line sizes 256 or larger this value is going to end
+ *	 up negative.  In these cases we should fall back to the legacy
+ *	 receive path.
+ */
+#if (PAGE_SIZE < 8192)
+#define I40E_2K_TOO_SMALL_WITH_PADDING \
+((NET_SKB_PAD + I40E_RXBUFFER_1536) > SKB_WITH_OVERHEAD(I40E_RXBUFFER_2048))
+
+static inline int i40e_compute_pad(int rx_buf_len)
+{
+	int page_size, pad_size;
+
+	page_size = ALIGN(rx_buf_len, PAGE_SIZE / 2);
+	pad_size = SKB_WITH_OVERHEAD(page_size) - rx_buf_len;
+
+	return pad_size;
+}
+
+static inline int i40e_skb_pad(void)
+{
+	int rx_buf_len;
+
+	/* If a 2K buffer cannot handle a standard Ethernet frame then
+	 * optimize padding for a 3K buffer instead of a 1.5K buffer.
+	 *
+	 * For a 3K buffer we need to add enough padding to allow for
+	 * tailroom due to NET_IP_ALIGN possibly shifting us out of
+	 * cache-line alignment.
+	 */
+	if (I40E_2K_TOO_SMALL_WITH_PADDING)
+		rx_buf_len = I40E_RXBUFFER_3072 + SKB_DATA_ALIGN(NET_IP_ALIGN);
+	else
+		rx_buf_len = I40E_RXBUFFER_1536;
+
+	/* if needed make room for NET_IP_ALIGN */
+	rx_buf_len -= NET_IP_ALIGN;
+
+	return i40e_compute_pad(rx_buf_len);
+}
+
+#define I40E_SKB_PAD i40e_skb_pad()
+#else
+#define I40E_2K_TOO_SMALL_WITH_PADDING false
+#define I40E_SKB_PAD (NET_SKB_PAD + NET_IP_ALIGN)
+#endif
+
 /**
  * i40e_test_staterr - tests bits in Rx descriptor status and error fields
  * @rx_desc: pointer to receive descriptor (in le64 format)
@@ -328,7 +380,8 @@ struct i40e_ring {
 	u8 packet_stride;
 
 	u16 flags;
-#define I40E_TXR_FLAGS_WB_ON_ITR	BIT(0)
+#define I40E_TXR_FLAGS_WB_ON_ITR		BIT(0)
+#define I40E_RXR_FLAGS_BUILD_SKB_ENABLED	BIT(1)
 
 	/* stats structs */
 	struct i40e_queue_stats	stats;
@@ -356,6 +409,21 @@ struct i40e_ring {
 					 */
 } ____cacheline_internodealigned_in_smp;
 
+static inline bool ring_uses_build_skb(struct i40e_ring *ring)
+{
+	return !!(ring->flags & I40E_RXR_FLAGS_BUILD_SKB_ENABLED);
+}
+
+static inline void set_ring_build_skb_enabled(struct i40e_ring *ring)
+{
+	ring->flags |= I40E_RXR_FLAGS_BUILD_SKB_ENABLED;
+}
+
+static inline void clear_ring_build_skb_enabled(struct i40e_ring *ring)
+{
+	ring->flags &= ~I40E_RXR_FLAGS_BUILD_SKB_ENABLED;
+}
+
 enum i40e_latency_range {
 	I40E_LOWEST_LATENCY = 0,
 	I40E_LOW_LATENCY = 1,
diff --git a/drivers/net/ethernet/intel/i40evf/i40evf_main.c b/drivers/net/ethernet/intel/i40evf/i40evf_main.c
index 7d00abae6104..12a930e879af 100644
--- a/drivers/net/ethernet/intel/i40evf/i40evf_main.c
+++ b/drivers/net/ethernet/intel/i40evf/i40evf_main.c
@@ -704,7 +704,8 @@ static void i40evf_configure_rx(struct i40evf_adapter *adapter)
 		 * standard Ethernet mtu.  On x86 this gives us enough room
 		 * for shared info and 192 bytes of padding.
 		 */
-		if (netdev->mtu <= ETH_DATA_LEN)
+		if (!I40E_2K_TOO_SMALL_WITH_PADDING &&
+		    (netdev->mtu <= ETH_DATA_LEN))
 			rx_buf_len = I40E_RXBUFFER_1536 - NET_IP_ALIGN;
 	}
 #endif
@@ -712,6 +713,11 @@ static void i40evf_configure_rx(struct i40evf_adapter *adapter)
 	for (i = 0; i < adapter->num_active_queues; i++) {
 		adapter->rx_rings[i].tail = hw->hw_addr + I40E_QRX_TAIL1(i);
 		adapter->rx_rings[i].rx_buf_len = rx_buf_len;
+
+		if (adapter->flags & I40EVF_FLAG_LEGACY_RX)
+			clear_ring_build_skb_enabled(&adapter->rx_rings[i]);
+		else
+			set_ring_build_skb_enabled(&adapter->rx_rings[i]);
 	}
 }