diff mbox

[next,S23,01/13] i40e: Fix Rx hash reported to the stack by our driver

Message ID 1449705033-4968-2-git-send-email-joshua.a.hay@intel.com
State Accepted
Delegated to: Jeff Kirsher
Headers show

Commit Message

Joshua Hay Dec. 9, 2015, 11:50 p.m. UTC 
From: Anjali Singhai Jain <anjali.singhai@intel.com>

If the driver calls skb_set_hash even with a zero hash, that
indicates to the stack that the hash calculation is offloaded
in hardware. So the Stack doesn't do a SW hash which is required
for load balancing if the user decides to turn of rx-hashing
on our device.

This patch fixes the path so that we do not call skb_set_hash
if the feature is disabled.

Signed-off-by: Anjali Singhai Jain <anjali.singhai@intel.com>
Change-ID: Ic4debfa4ff91b5a72e447348a75768ed7a2d3e1b
---
Testing Hints:Test interrupt overload x710 Apart from the
interrupt overload, the user sees a problem where the stack
is not calculating the hash when rx-hasing is turned off which
is resulting in tpacket_v3 not balancing the load across
multiple threads.

 drivers/net/ethernet/intel/i40e/i40e_txrx.c   | 54 ++++++++++++++-------------
 drivers/net/ethernet/intel/i40evf/i40e_txrx.c | 54 ++++++++++++++-------------
 2 files changed, 58 insertions(+), 50 deletions(-)

Comments

Bowers, AndrewX Dec. 16, 2015, 6:31 p.m. UTC | #1 
> -----Original Message-----
> From: Intel-wired-lan [mailto:intel-wired-lan-bounces@lists.osuosl.org] On
> Behalf Of Joshua Hay
> Sent: Wednesday, December 09, 2015 3:50 PM
> To: intel-wired-lan@lists.osuosl.org
> Subject: [Intel-wired-lan] [next PATCH S23 01/13] i40e: Fix Rx hash reported
> to the stack by our driver
> 
> From: Anjali Singhai Jain <anjali.singhai@intel.com>
> 
> If the driver calls skb_set_hash even with a zero hash, that indicates to the
> stack that the hash calculation is offloaded in hardware. So the Stack doesn't
> do a SW hash which is required for load balancing if the user decides to turn
> of rx-hashing on our device.
> 
> This patch fixes the path so that we do not call skb_set_hash if the feature is
> disabled.
> 
> Signed-off-by: Anjali Singhai Jain <anjali.singhai@intel.com>
> Change-ID: Ic4debfa4ff91b5a72e447348a75768ed7a2d3e1b
> ---
> Testing Hints:Test interrupt overload x710 Apart from the interrupt overload,
> the user sees a problem where the stack is not calculating the hash when rx-
> hasing is turned off which is resulting in tpacket_v3 not balancing the load
> across multiple threads.
> 
>  drivers/net/ethernet/intel/i40e/i40e_txrx.c   | 54 ++++++++++++++----------
> ---
>  drivers/net/ethernet/intel/i40evf/i40e_txrx.c | 54 ++++++++++++++---------
> ----
>  2 files changed, 58 insertions(+), 50 deletions(-)

Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Patch code changes correctly applied, load properly balances across queues with hashing enabled and disabled
diff mbox

Patch

diff --git a/drivers/net/ethernet/intel/i40e/i40e_txrx.c b/drivers/net/ethernet/intel/i40e/i40e_txrx.c
index e2ab7a6..28d2943 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_txrx.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_txrx.c
@@ -1422,31 +1422,12 @@  checksum_fail:
 }
 
 /**
- * i40e_rx_hash - returns the hash value from the Rx descriptor
- * @ring: descriptor ring
- * @rx_desc: specific descriptor
- **/
-static inline u32 i40e_rx_hash(struct i40e_ring *ring,
-			       union i40e_rx_desc *rx_desc)
-{
-	const __le64 rss_mask =
-		cpu_to_le64((u64)I40E_RX_DESC_FLTSTAT_RSS_HASH <<
-			    I40E_RX_DESC_STATUS_FLTSTAT_SHIFT);
-
-	if ((ring->netdev->features & NETIF_F_RXHASH) &&
-	    (rx_desc->wb.qword1.status_error_len & rss_mask) == rss_mask)
-		return le32_to_cpu(rx_desc->wb.qword0.hi_dword.rss);
-	else
-		return 0;
-}
-
-/**
- * i40e_ptype_to_hash - get a hash type
+ * i40e_ptype_to_htype - get a hash type
  * @ptype: the ptype value from the descriptor
  *
  * Returns a hash type to be used by skb_set_hash
  **/
-static inline enum pkt_hash_types i40e_ptype_to_hash(u8 ptype)
+static inline enum pkt_hash_types i40e_ptype_to_htype(u8 ptype)
 {
 	struct i40e_rx_ptype_decoded decoded = decode_rx_desc_ptype(ptype);
 
@@ -1464,6 +1445,30 @@  static inline enum pkt_hash_types i40e_ptype_to_hash(u8 ptype)
 }
 
 /**
+ * i40e_rx_hash - set the hash value in the skb
+ * @ring: descriptor ring
+ * @rx_desc: specific descriptor
+ **/
+static inline void i40e_rx_hash(struct i40e_ring *ring,
+				union i40e_rx_desc *rx_desc,
+				struct sk_buff *skb,
+				u8 rx_ptype)
+{
+	u32 hash;
+	const __le64 rss_mask  =
+		cpu_to_le64((u64)I40E_RX_DESC_FLTSTAT_RSS_HASH <<
+			    I40E_RX_DESC_STATUS_FLTSTAT_SHIFT);
+
+	if (ring->netdev->features & NETIF_F_RXHASH)
+		return;
+
+	if ((rx_desc->wb.qword1.status_error_len & rss_mask) == rss_mask) {
+		hash = le32_to_cpu(rx_desc->wb.qword0.hi_dword.rss);
+		skb_set_hash(skb, hash, i40e_ptype_to_htype(rx_ptype));
+	}
+}
+
+/**
  * i40e_clean_rx_irq_ps - Reclaim resources after receive; packet split
  * @rx_ring:  rx ring to clean
  * @budget:   how many cleans we're allowed
@@ -1612,8 +1617,8 @@  static int i40e_clean_rx_irq_ps(struct i40e_ring *rx_ring, int budget)
 			continue;
 		}
 
-		skb_set_hash(skb, i40e_rx_hash(rx_ring, rx_desc),
-			     i40e_ptype_to_hash(rx_ptype));
+		i40e_rx_hash(rx_ring, rx_desc, skb, rx_ptype);
+
 		if (unlikely(rx_status & I40E_RXD_QW1_STATUS_TSYNVALID_MASK)) {
 			i40e_ptp_rx_hwtstamp(vsi->back, skb, (rx_status &
 					   I40E_RXD_QW1_STATUS_TSYNINDX_MASK) >>
@@ -1741,8 +1746,7 @@  static int i40e_clean_rx_irq_1buf(struct i40e_ring *rx_ring, int budget)
 			continue;
 		}
 
-		skb_set_hash(skb, i40e_rx_hash(rx_ring, rx_desc),
-			     i40e_ptype_to_hash(rx_ptype));
+		i40e_rx_hash(rx_ring, rx_desc, skb, rx_ptype);
 		if (unlikely(rx_status & I40E_RXD_QW1_STATUS_TSYNVALID_MASK)) {
 			i40e_ptp_rx_hwtstamp(vsi->back, skb, (rx_status &
 					   I40E_RXD_QW1_STATUS_TSYNINDX_MASK) >>
diff --git a/drivers/net/ethernet/intel/i40evf/i40e_txrx.c b/drivers/net/ethernet/intel/i40evf/i40e_txrx.c
index 4ca4065..7a00657 100644
--- a/drivers/net/ethernet/intel/i40evf/i40e_txrx.c
+++ b/drivers/net/ethernet/intel/i40evf/i40e_txrx.c
@@ -886,31 +886,12 @@  checksum_fail:
 }
 
 /**
- * i40e_rx_hash - returns the hash value from the Rx descriptor
- * @ring: descriptor ring
- * @rx_desc: specific descriptor
- **/
-static inline u32 i40e_rx_hash(struct i40e_ring *ring,
-			       union i40e_rx_desc *rx_desc)
-{
-	const __le64 rss_mask =
-		cpu_to_le64((u64)I40E_RX_DESC_FLTSTAT_RSS_HASH <<
-			    I40E_RX_DESC_STATUS_FLTSTAT_SHIFT);
-
-	if ((ring->netdev->features & NETIF_F_RXHASH) &&
-	    (rx_desc->wb.qword1.status_error_len & rss_mask) == rss_mask)
-		return le32_to_cpu(rx_desc->wb.qword0.hi_dword.rss);
-	else
-		return 0;
-}
-
-/**
- * i40e_ptype_to_hash - get a hash type
+ * i40e_ptype_to_htype - get a hash type
  * @ptype: the ptype value from the descriptor
  *
  * Returns a hash type to be used by skb_set_hash
  **/
-static inline enum pkt_hash_types i40e_ptype_to_hash(u8 ptype)
+static inline enum pkt_hash_types i40e_ptype_to_htype(u8 ptype)
 {
 	struct i40e_rx_ptype_decoded decoded = decode_rx_desc_ptype(ptype);
 
@@ -928,6 +909,30 @@  static inline enum pkt_hash_types i40e_ptype_to_hash(u8 ptype)
 }
 
 /**
+ * i40e_rx_hash - set the hash value in the skb
+ * @ring: descriptor ring
+ * @rx_desc: specific descriptor
+ **/
+static inline void i40e_rx_hash(struct i40e_ring *ring,
+				union i40e_rx_desc *rx_desc,
+				struct sk_buff *skb,
+				u8 rx_ptype)
+{
+	u32 hash;
+	const __le64 rss_mask  =
+		cpu_to_le64((u64)I40E_RX_DESC_FLTSTAT_RSS_HASH <<
+			    I40E_RX_DESC_STATUS_FLTSTAT_SHIFT);
+
+	if (ring->netdev->features & NETIF_F_RXHASH)
+		return;
+
+	if ((rx_desc->wb.qword1.status_error_len & rss_mask) == rss_mask) {
+		hash = le32_to_cpu(rx_desc->wb.qword0.hi_dword.rss);
+		skb_set_hash(skb, hash, i40e_ptype_to_htype(rx_ptype));
+	}
+}
+
+/**
  * i40e_clean_rx_irq_ps - Reclaim resources after receive; packet split
  * @rx_ring:  rx ring to clean
  * @budget:   how many cleans we're allowed
@@ -1068,8 +1073,8 @@  static int i40e_clean_rx_irq_ps(struct i40e_ring *rx_ring, int budget)
 			continue;
 		}
 
-		skb_set_hash(skb, i40e_rx_hash(rx_ring, rx_desc),
-			     i40e_ptype_to_hash(rx_ptype));
+		i40e_rx_hash(rx_ring, rx_desc, skb, rx_ptype);
+
 		/* probably a little skewed due to removing CRC */
 		total_rx_bytes += skb->len;
 		total_rx_packets++;
@@ -1185,8 +1190,7 @@  static int i40e_clean_rx_irq_1buf(struct i40e_ring *rx_ring, int budget)
 			continue;
 		}
 
-		skb_set_hash(skb, i40e_rx_hash(rx_ring, rx_desc),
-			     i40e_ptype_to_hash(rx_ptype));
+		i40e_rx_hash(rx_ring, rx_desc, skb, rx_ptype);
 		/* probably a little skewed due to removing CRC */
 		total_rx_bytes += skb->len;
 		total_rx_packets++;