[net-next 00/15][pull request] 40GbE Intel Wired LAN Driver Updates 2015-12-03

[net-next 00/15][pull request] 40GbE Intel Wired LAN Driver Updates 2015-12-03

[Jeff Kirsher]

This series contains updates to i40e and i40evf only.

Mitch updates the i40evf driver by increasing the maximum number of queues,
since future devices will allow for more queue pairs.  Cleans up a
duplicate printing of the driver info string done in init, since it is
already done in probe.  Cleaned up the several allocations which did
not need to be at atomic level, where GFP_KERNEL would work just fine.
Then makes i40e_sync_vsi_filters() a more mature function, make having
a common exit point so it will properly release the busy lock on the VSI
and propagate errors to the callers.  Then does some whitespace
housekeeping in i40evf.

Kiran moves and updates the detection/recovery of transmit queue hang code
to service_task from tx_timeout function.  Also fixed memory leak when
users program flow-director filter using ethtool (sideband filter
programming), the cause being the check of 'tx_buffer->skb' was preventing
'raw_buf' from being freed as part of the cleanup.

Jesse enabled the ability to turn off/on packet split using ethtool priv
flags.  Then does some housekeeping for both the i40e and i40evf drivers
which includes: remove unused/useless code, correct whitespace, remove
duplicate #include, fix incorrect comment, etc...

Neerav cleans up functions to gather Flow Control Rx XOFF stats, since
the recent change in the driver logic for checking transmit hang has been
moved, so these functions do not do anything meaningful any longer.

The following are changes since commit 6f24e5d599896b5091af72e4b3edfce6307627f7:
  sfc: use ALIGN macro for aligning frame sizes
and are available in the git repository at:
  git://git.kernel.org/pub/scm/linux/kernel/git/jkirsher/next-queue 40GbE

Catherine Sullivan (1):
  i40e/i40evf: Bump version to 1.4.7 for i40e and 1.4.3 for i40evf

Jesse Brandeburg (3):
  i40e: use priv flags to control packet split
  i40e: remove CONFIG_I40E_VXLAN
  i40e: trivial fixes

Kiran Patil (2):
  i40e: Detection and recovery of TX queue hung logic moved to
    service_task from tx_timeout
  i40e: Fix memory leaks, sideband filter programming

Mitch Williams (8):
  i40evf: increase max number of queues
  i40evf: set real num queues
  i40evf: remove duplicate string
  i40evf: don't use atomic allocation
  i40e: propagate properly
  i40evf: use correct types
  i40e: fix whitespace
  i40e/i40evf: use logical operator

Neerav Parikh (1):
  i40e: Remove separate functions gathering XOFF Rx stats

 drivers/net/ethernet/intel/Kconfig                 |  11 -
 drivers/net/ethernet/intel/i40e/i40e.h             |  15 +-
 drivers/net/ethernet/intel/i40e/i40e_ethtool.c     |  27 +++
 drivers/net/ethernet/intel/i40e/i40e_main.c        | 240 ++++++++++-----------
 drivers/net/ethernet/intel/i40e/i40e_txrx.c        |  29 ++-
 drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c |  18 +-
 drivers/net/ethernet/intel/i40evf/i40e_txrx.c      |  48 ++---
 drivers/net/ethernet/intel/i40evf/i40e_txrx.h      |  15 ++
 drivers/net/ethernet/intel/i40evf/i40evf.h         |   3 +-
 drivers/net/ethernet/intel/i40evf/i40evf_main.c    |  11 +-
 .../net/ethernet/intel/i40evf/i40evf_virtchnl.c    |  12 +-
 11 files changed, 226 insertions(+), 203 deletions(-)

-- 
2.5.0

[net-next 01/15] i40evf: increase max number of queues

[Jeff Kirsher]

From: Mitch Williams <mitch.a.williams@intel.com>

Future devices will allow for more queue pairs, so allocate a netdev
that can handle them. While we're at it, get rid of the separate
MAX_TX/MAX_RX defines. Since we always get matched queue pairs, having
these makes no sense.

Change-ID: I0e3556cd9a962506e509eb7c0afa36b329e8cb51
Signed-off-by: Mitch Williams <mitch.a.williams@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/i40evf/i40evf.h      | 3 +--
 drivers/net/ethernet/intel/i40evf/i40evf_main.c | 3 +--
 2 files changed, 2 insertions(+), 4 deletions(-)

diff --git a/drivers/net/ethernet/intel/i40evf/i40evf.h b/drivers/net/ethernet/intel/i40evf/i40evf.h
index a6318c4..be1b72b 100644
--- a/drivers/net/ethernet/intel/i40evf/i40evf.h
+++ b/drivers/net/ethernet/intel/i40evf/i40evf.h
@@ -97,8 +97,7 @@ struct i40e_vsi {
 #define I40E_TX_DESC(R, i) (&(((struct i40e_tx_desc *)((R)->desc))[i]))
 #define I40E_TX_CTXTDESC(R, i) \
 	(&(((struct i40e_tx_context_desc *)((R)->desc))[i]))
-#define MAX_RX_QUEUES 8
-#define MAX_TX_QUEUES MAX_RX_QUEUES
+#define MAX_QUEUES 16
 
 #define I40EVF_HKEY_ARRAY_SIZE ((I40E_VFQF_HKEY_MAX_INDEX + 1) * 4)
 #define I40EVF_HLUT_ARRAY_SIZE ((I40E_VFQF_HLUT_MAX_INDEX + 1) * 4)
diff --git a/drivers/net/ethernet/intel/i40evf/i40evf_main.c b/drivers/net/ethernet/intel/i40evf/i40evf_main.c
index 7ef1042..38f882c 100644
--- a/drivers/net/ethernet/intel/i40evf/i40evf_main.c
+++ b/drivers/net/ethernet/intel/i40evf/i40evf_main.c
@@ -2615,8 +2615,7 @@ static int i40evf_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 
 	pci_set_master(pdev);
 
-	netdev = alloc_etherdev_mq(sizeof(struct i40evf_adapter),
-				   MAX_TX_QUEUES);
+	netdev = alloc_etherdev_mq(sizeof(struct i40evf_adapter), MAX_QUEUES);
 	if (!netdev) {
 		err = -ENOMEM;
 		goto err_alloc_etherdev;
-- 
2.5.0

[net-next 02/15] i40evf: set real num queues

[Jeff Kirsher]

From: Mitch Williams <mitch.a.williams@intel.com>

Use the helper function to set the real number of RX queues, and also
set the real number of TX queues.

Change-ID: I67982799de3f248fb4158ccdc9b1a74385f42ddd
Signed-off-by: Mitch Williams <mitch.a.williams@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/i40evf/i40evf_main.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/intel/i40evf/i40evf_main.c b/drivers/net/ethernet/intel/i40evf/i40evf_main.c
index 38f882c..ba03605 100644
--- a/drivers/net/ethernet/intel/i40evf/i40evf_main.c
+++ b/drivers/net/ethernet/intel/i40evf/i40evf_main.c
@@ -1205,7 +1205,8 @@ static int i40evf_set_interrupt_capability(struct i40evf_adapter *adapter)
 	err = i40evf_acquire_msix_vectors(adapter, v_budget);
 
 out:
-	adapter->netdev->real_num_tx_queues = pairs;
+	netif_set_real_num_rx_queues(adapter->netdev, pairs);
+	netif_set_real_num_tx_queues(adapter->netdev, pairs);
 	return err;
 }
 
-- 
2.5.0

[net-next 03/15] i40evf: remove duplicate string

[Jeff Kirsher]

From: Mitch Williams <mitch.a.williams@intel.com>

We already print the driver info string in probe, so don't print
it again in init. No need to repeat. No need to repeat.

Change-ID: Ief597997f580a8c54d5950e3a84c29f2075be66b
Signed-off-by: Mitch Williams <mitch.a.williams@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/i40evf/i40evf_main.c | 1 -
 1 file changed, 1 deletion(-)

diff --git a/drivers/net/ethernet/intel/i40evf/i40evf_main.c b/drivers/net/ethernet/intel/i40evf/i40evf_main.c
index ba03605..82c9adf 100644
--- a/drivers/net/ethernet/intel/i40evf/i40evf_main.c
+++ b/drivers/net/ethernet/intel/i40evf/i40evf_main.c
@@ -2512,7 +2512,6 @@ static void i40evf_init_task(struct work_struct *work)
 	if (netdev->features & NETIF_F_GRO)
 		dev_info(&pdev->dev, "GRO is enabled\n");
 
-	dev_info(&pdev->dev, "%s\n", i40evf_driver_string);
 	adapter->state = __I40EVF_DOWN;
 	set_bit(__I40E_DOWN, &adapter->vsi.state);
 	i40evf_misc_irq_enable(adapter);
-- 
2.5.0

[net-next 04/15] i40e: Detection and recovery of TX queue hung logic moved to service_task from tx_timeout

[Jeff Kirsher]

From: Kiran Patil <kiran.patil@intel.com>

This patch contains following changes:
   - detection and recovery logic (issue SW interrupt) has been moved to
     service_task from timeout function.
   - added some more debug info from tx_timeout.

Logic to detect and recover TX queue hung is now two step process:
  - service_task detects TX queue hung and sets a bit(hung_detected) if
    it was not set.
  - if bit was set (means this is back-back hung condition detected),
    issue SW interrupt and clear the bit.
  - napi_poll clears the bit unconditionally since it cleans TX/RX queues.

Change-ID: Ieed03a48927c845a988b3ff375090bf37caeb903
Signed-off-by: Kiran Patil <kiran.patil@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/i40e/i40e.h        |  3 +++
 drivers/net/ethernet/intel/i40e/i40e_main.c   | 36 ++++++++++++++++++++++-----
 drivers/net/ethernet/intel/i40e/i40e_txrx.c   |  2 ++
 drivers/net/ethernet/intel/i40evf/i40e_txrx.c | 21 ++++++++++------
 drivers/net/ethernet/intel/i40evf/i40e_txrx.h | 15 +++++++++++
 5 files changed, 64 insertions(+), 13 deletions(-)

diff --git a/drivers/net/ethernet/intel/i40e/i40e.h b/drivers/net/ethernet/intel/i40e/i40e.h
index 0b9537b..dde7ae7 100644
--- a/drivers/net/ethernet/intel/i40e/i40e.h
+++ b/drivers/net/ethernet/intel/i40e/i40e.h
@@ -579,6 +579,9 @@ struct i40e_q_vector {
 
 	u8 num_ringpairs;	/* total number of ring pairs in vector */
 
+#define I40E_Q_VECTOR_HUNG_DETECT 0 /* Bit Index for hung detection logic */
+	unsigned long hung_detected; /* Set/Reset for hung_detection logic */
+
 	cpumask_t affinity_mask;
 	struct rcu_head rcu;	/* to avoid race with update stats on free */
 	char name[I40E_INT_NAME_STR_LEN];
diff --git a/drivers/net/ethernet/intel/i40e/i40e_main.c b/drivers/net/ethernet/intel/i40e/i40e_main.c
index 9e6268b..e19a579 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_main.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_main.c
@@ -4368,17 +4368,41 @@ static void i40e_detect_recover_hung_queue(int q_idx, struct i40e_vsi *vsi)
 	else
 		val = rd32(&pf->hw, I40E_PFINT_DYN_CTL0);
 
+	/* Bail out if interrupts are disabled because napi_poll
+	 * execution in-progress or will get scheduled soon.
+	 * napi_poll cleans TX and RX queues and updates 'next_to_clean'.
+	 */
+	if (!(val & I40E_PFINT_DYN_CTLN_INTENA_MASK))
+		return;
+
 	head = i40e_get_head(tx_ring);
 
 	tx_pending = i40e_get_tx_pending(tx_ring);
 
-	/* Interrupts are disabled and TX pending is non-zero,
-	 * trigger the SW interrupt (don't wait). Worst case
-	 * there will be one extra interrupt which may result
-	 * into not cleaning any queues because queues are cleaned.
+	/* HW is done executing descriptors, updated HEAD write back,
+	 * but SW hasn't processed those descriptors. If interrupt is
+	 * not generated from this point ON, it could result into
+	 * dev_watchdog detecting timeout on those netdev_queue,
+	 * hence proactively trigger SW interrupt.
 	 */
-	if (tx_pending && (!(val & I40E_PFINT_DYN_CTLN_INTENA_MASK)))
-		i40e_force_wb(vsi, tx_ring->q_vector);
+	if (tx_pending) {
+		/* NAPI Poll didn't run and clear since it was set */
+		if (test_and_clear_bit(I40E_Q_VECTOR_HUNG_DETECT,
+				       &tx_ring->q_vector->hung_detected)) {
+			netdev_info(vsi->netdev, "VSI_seid %d, Hung TX queue %d, tx_pending: %d, NTC:0x%x, HWB: 0x%x, NTU: 0x%x, TAIL: 0x%x\n",
+				    vsi->seid, q_idx, tx_pending,
+				    tx_ring->next_to_clean, head,
+				    tx_ring->next_to_use,
+				    readl(tx_ring->tail));
+			netdev_info(vsi->netdev, "VSI_seid %d, Issuing force_wb for TX queue %d, Interrupt Reg: 0x%x\n",
+				    vsi->seid, q_idx, val);
+			i40e_force_wb(vsi, tx_ring->q_vector);
+		} else {
+			/* First Chance - detected possible hung */
+			set_bit(I40E_Q_VECTOR_HUNG_DETECT,
+				&tx_ring->q_vector->hung_detected);
+		}
+	}
 }
 
 /**
diff --git a/drivers/net/ethernet/intel/i40e/i40e_txrx.c b/drivers/net/ethernet/intel/i40e/i40e_txrx.c
index 7371df7..1d7d01c 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_txrx.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_txrx.c
@@ -1891,6 +1891,8 @@ int i40e_napi_poll(struct napi_struct *napi, int budget)
 		return 0;
 	}
 
+	/* Clear hung_detected bit */
+	clear_bit(I40E_Q_VECTOR_HUNG_DETECT, &q_vector->hung_detected);
 	/* Since the actual Tx work is minimal, we can give the Tx a larger
 	 * budget and be more aggressive about cleaning up the Tx descriptors.
 	 */
diff --git a/drivers/net/ethernet/intel/i40evf/i40e_txrx.c b/drivers/net/ethernet/intel/i40evf/i40e_txrx.c
index 11561b6..06b6636 100644
--- a/drivers/net/ethernet/intel/i40evf/i40e_txrx.c
+++ b/drivers/net/ethernet/intel/i40evf/i40e_txrx.c
@@ -127,17 +127,24 @@ void i40evf_free_tx_resources(struct i40e_ring *tx_ring)
 }
 
 /**
- * i40e_get_head - Retrieve head from head writeback
- * @tx_ring:  tx ring to fetch head of
+ * i40evf_get_tx_pending - how many Tx descriptors not processed
+ * @tx_ring: the ring of descriptors
  *
- * Returns value of Tx ring head based on value stored
- * in head write-back location
+ * Since there is no access to the ring head register
+ * in XL710, we need to use our local copies
  **/
-static inline u32 i40e_get_head(struct i40e_ring *tx_ring)
+u32 i40evf_get_tx_pending(struct i40e_ring *ring)
 {
-	void *head = (struct i40e_tx_desc *)tx_ring->desc + tx_ring->count;
+	u32 head, tail;
 
-	return le32_to_cpu(*(volatile __le32 *)head);
+	head = i40e_get_head(ring);
+	tail = readl(ring->tail);
+
+	if (head != tail)
+		return (head < tail) ?
+			tail - head : (tail + ring->count - head);
+
+	return 0;
 }
 
 #define WB_STRIDE 0x3
diff --git a/drivers/net/ethernet/intel/i40evf/i40e_txrx.h b/drivers/net/ethernet/intel/i40evf/i40e_txrx.h
index 929ddd9..e29bb3e 100644
--- a/drivers/net/ethernet/intel/i40evf/i40e_txrx.h
+++ b/drivers/net/ethernet/intel/i40evf/i40e_txrx.h
@@ -324,4 +324,19 @@ int i40evf_setup_rx_descriptors(struct i40e_ring *rx_ring);
 void i40evf_free_tx_resources(struct i40e_ring *tx_ring);
 void i40evf_free_rx_resources(struct i40e_ring *rx_ring);
 int i40evf_napi_poll(struct napi_struct *napi, int budget);
+u32 i40evf_get_tx_pending(struct i40e_ring *ring);
+
+/**
+ * i40e_get_head - Retrieve head from head writeback
+ * @tx_ring: Tx ring to fetch head of
+ *
+ * Returns value of Tx ring head based on value stored
+ * in head write-back location
+ **/
+static inline u32 i40e_get_head(struct i40e_ring *tx_ring)
+{
+	void *head = (struct i40e_tx_desc *)tx_ring->desc + tx_ring->count;
+
+	return le32_to_cpu(*(volatile __le32 *)head);
+}
 #endif /* _I40E_TXRX_H_ */
-- 
2.5.0

[net-next 05/15] i40e: Fix memory leaks, sideband filter programming

[Jeff Kirsher]

From: Kiran Patil <kiran.patil@intel.com>

This patch fixes the memory leak which would be seen otherwise when user
programs flow-director filter using ethtool (sideband filter programming).

When ethtool is used to program flow directory filter, 'raw_buf' gets
allocated and it is supposed to be freed as part of queue cleanup. But
check of 'tx_buffer->skb' was preventing it from being freed.

Change-ID: Ief4f0a1a32a653180498bf6e987c1b4342ab8923
Signed-off-by: Kiran Patil <kiran.patil@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/i40e/i40e_txrx.c   | 19 ++++++++++++++-----
 drivers/net/ethernet/intel/i40evf/i40e_txrx.c | 10 +++++-----
 2 files changed, 19 insertions(+), 10 deletions(-)

diff --git a/drivers/net/ethernet/intel/i40e/i40e_txrx.c b/drivers/net/ethernet/intel/i40e/i40e_txrx.c
index 1d7d01c..18a493e 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_txrx.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_txrx.c
@@ -235,6 +235,9 @@ static int i40e_add_del_fdir_udpv4(struct i40e_vsi *vsi,
 				 "Filter deleted for PCTYPE %d loc = %d\n",
 				 fd_data->pctype, fd_data->fd_id);
 	}
+	if (err)
+		kfree(raw_packet);
+
 	return err ? -EOPNOTSUPP : 0;
 }
 
@@ -312,6 +315,9 @@ static int i40e_add_del_fdir_tcpv4(struct i40e_vsi *vsi,
 				 fd_data->pctype, fd_data->fd_id);
 	}
 
+	if (err)
+		kfree(raw_packet);
+
 	return err ? -EOPNOTSUPP : 0;
 }
 
@@ -387,6 +393,9 @@ static int i40e_add_del_fdir_ipv4(struct i40e_vsi *vsi,
 		}
 	}
 
+	if (err)
+		kfree(raw_packet);
+
 	return err ? -EOPNOTSUPP : 0;
 }
 
@@ -526,11 +535,7 @@ static void i40e_unmap_and_free_tx_resource(struct i40e_ring *ring,
 					    struct i40e_tx_buffer *tx_buffer)
 {
 	if (tx_buffer->skb) {
-		if (tx_buffer->tx_flags & I40E_TX_FLAGS_FD_SB)
-			kfree(tx_buffer->raw_buf);
-		else
-			dev_kfree_skb_any(tx_buffer->skb);
-
+		dev_kfree_skb_any(tx_buffer->skb);
 		if (dma_unmap_len(tx_buffer, len))
 			dma_unmap_single(ring->dev,
 					 dma_unmap_addr(tx_buffer, dma),
@@ -542,6 +547,10 @@ static void i40e_unmap_and_free_tx_resource(struct i40e_ring *ring,
 			       dma_unmap_len(tx_buffer, len),
 			       DMA_TO_DEVICE);
 	}
+
+	if (tx_buffer->tx_flags & I40E_TX_FLAGS_FD_SB)
+		kfree(tx_buffer->raw_buf);
+
 	tx_buffer->next_to_watch = NULL;
 	tx_buffer->skb = NULL;
 	dma_unmap_len_set(tx_buffer, len, 0);
diff --git a/drivers/net/ethernet/intel/i40evf/i40e_txrx.c b/drivers/net/ethernet/intel/i40evf/i40e_txrx.c
index 06b6636..4f1ac86 100644
--- a/drivers/net/ethernet/intel/i40evf/i40e_txrx.c
+++ b/drivers/net/ethernet/intel/i40evf/i40e_txrx.c
@@ -51,11 +51,7 @@ static void i40e_unmap_and_free_tx_resource(struct i40e_ring *ring,
 					    struct i40e_tx_buffer *tx_buffer)
 {
 	if (tx_buffer->skb) {
-		if (tx_buffer->tx_flags & I40E_TX_FLAGS_FD_SB)
-			kfree(tx_buffer->raw_buf);
-		else
-			dev_kfree_skb_any(tx_buffer->skb);
-
+		dev_kfree_skb_any(tx_buffer->skb);
 		if (dma_unmap_len(tx_buffer, len))
 			dma_unmap_single(ring->dev,
 					 dma_unmap_addr(tx_buffer, dma),
@@ -67,6 +63,10 @@ static void i40e_unmap_and_free_tx_resource(struct i40e_ring *ring,
 			       dma_unmap_len(tx_buffer, len),
 			       DMA_TO_DEVICE);
 	}
+
+	if (tx_buffer->tx_flags & I40E_TX_FLAGS_FD_SB)
+		kfree(tx_buffer->raw_buf);
+
 	tx_buffer->next_to_watch = NULL;
 	tx_buffer->skb = NULL;
 	dma_unmap_len_set(tx_buffer, len, 0);
-- 
2.5.0

[net-next 06/15] i40evf: don't use atomic allocation

[Jeff Kirsher]

From: Mitch Williams <mitch.a.williams@intel.com>

These allocations don't need to be at atomic level. GFP_KERNEL is fine
and they'll reduce stress on the allocator when the system is starved
for memory.

Change-ID: I3561d0399a681de0ad25291b6c848b224c1fde12
Signed-off-by: Mitch Williams <mitch.a.williams@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/i40evf/i40evf_virtchnl.c | 12 ++++++------
 1 file changed, 6 insertions(+), 6 deletions(-)

diff --git a/drivers/net/ethernet/intel/i40evf/i40evf_virtchnl.c b/drivers/net/ethernet/intel/i40evf/i40evf_virtchnl.c
index 9b55576..3c9c008 100644
--- a/drivers/net/ethernet/intel/i40evf/i40evf_virtchnl.c
+++ b/drivers/net/ethernet/intel/i40evf/i40evf_virtchnl.c
@@ -242,7 +242,7 @@ void i40evf_configure_queues(struct i40evf_adapter *adapter)
 	adapter->current_op = I40E_VIRTCHNL_OP_CONFIG_VSI_QUEUES;
 	len = sizeof(struct i40e_virtchnl_vsi_queue_config_info) +
 		       (sizeof(struct i40e_virtchnl_queue_pair_info) * pairs);
-	vqci = kzalloc(len, GFP_ATOMIC);
+	vqci = kzalloc(len, GFP_KERNEL);
 	if (!vqci)
 		return;
 
@@ -353,7 +353,7 @@ void i40evf_map_queues(struct i40evf_adapter *adapter)
 	len = sizeof(struct i40e_virtchnl_irq_map_info) +
 	      (adapter->num_msix_vectors *
 		sizeof(struct i40e_virtchnl_vector_map));
-	vimi = kzalloc(len, GFP_ATOMIC);
+	vimi = kzalloc(len, GFP_KERNEL);
 	if (!vimi)
 		return;
 
@@ -421,7 +421,7 @@ void i40evf_add_ether_addrs(struct i40evf_adapter *adapter)
 		more = true;
 	}
 
-	veal = kzalloc(len, GFP_ATOMIC);
+	veal = kzalloc(len, GFP_KERNEL);
 	if (!veal)
 		return;
 
@@ -483,7 +483,7 @@ void i40evf_del_ether_addrs(struct i40evf_adapter *adapter)
 		      (count * sizeof(struct i40e_virtchnl_ether_addr));
 		more = true;
 	}
-	veal = kzalloc(len, GFP_ATOMIC);
+	veal = kzalloc(len, GFP_KERNEL);
 	if (!veal)
 		return;
 
@@ -547,7 +547,7 @@ void i40evf_add_vlans(struct i40evf_adapter *adapter)
 		      (count * sizeof(u16));
 		more = true;
 	}
-	vvfl = kzalloc(len, GFP_ATOMIC);
+	vvfl = kzalloc(len, GFP_KERNEL);
 	if (!vvfl)
 		return;
 
@@ -609,7 +609,7 @@ void i40evf_del_vlans(struct i40evf_adapter *adapter)
 		      (count * sizeof(u16));
 		more = true;
 	}
-	vvfl = kzalloc(len, GFP_ATOMIC);
+	vvfl = kzalloc(len, GFP_KERNEL);
 	if (!vvfl)
 		return;
 
-- 
2.5.0

[net-next 07/15] i40e: propagate properly

[Jeff Kirsher]

From: Mitch Williams <mitch.a.williams@intel.com>

i40e_sync_vsi_filters() is the surly teenager of this driver. It says
it's going to report errors, but it doesn't actually do that most of the
time. And when it does, it leaves a mess.

Change this function to have a common exit point so it will properly
release the busy lock on the VSI. Propagate errors to the callers.
Finally, adjust a few callers to check for and deal with errors from
this function.

Change-ID: Ic6af4956491e72402ebb3c538a3c31a0ad7f8667
Signed-off-by: Mitch Williams <mitch.a.williams@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/i40e/i40e_main.c        | 113 +++++++++++++--------
 drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c |  14 +--
 2 files changed, 76 insertions(+), 51 deletions(-)

diff --git a/drivers/net/ethernet/intel/i40e/i40e_main.c b/drivers/net/ethernet/intel/i40e/i40e_main.c
index e19a579..c5a24fe 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_main.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_main.c
@@ -1553,11 +1553,8 @@ static int i40e_set_mac(struct net_device *netdev, void *p)
 	}
 
 	ether_addr_copy(netdev->dev_addr, addr->sa_data);
-	/* schedule our worker thread which will take care of
-	 * applying the new filter changes
-	 */
-	i40e_service_event_schedule(vsi->back);
-	return 0;
+
+	return i40e_sync_vsi_filters(vsi);
 }
 
 /**
@@ -1872,8 +1869,9 @@ int i40e_sync_vsi_filters(struct i40e_vsi *vsi)
 	bool add_happened = false;
 	int filter_list_len = 0;
 	u32 changed_flags = 0;
+	i40e_status aq_ret = 0;
 	bool err_cond = false;
-	i40e_status ret = 0;
+	int retval = 0;
 	struct i40e_pf *pf;
 	int num_add = 0;
 	int num_del = 0;
@@ -1936,8 +1934,11 @@ int i40e_sync_vsi_filters(struct i40e_vsi *vsi)
 		}
 		spin_unlock_bh(&vsi->mac_filter_list_lock);
 
-		if (err_cond)
+		if (err_cond) {
 			i40e_cleanup_add_list(&tmp_add_list);
+			retval = -ENOMEM;
+			goto out;
+		}
 	}
 
 	/* Now process 'del_list' outside the lock */
@@ -1955,7 +1956,8 @@ int i40e_sync_vsi_filters(struct i40e_vsi *vsi)
 			i40e_undo_del_filter_entries(vsi, &tmp_del_list);
 			i40e_undo_add_filter_entries(vsi);
 			spin_unlock_bh(&vsi->mac_filter_list_lock);
-			return -ENOMEM;
+			retval = -ENOMEM;
+			goto out;
 		}
 
 		list_for_each_entry_safe(f, ftmp, &tmp_del_list, list) {
@@ -1973,18 +1975,22 @@ int i40e_sync_vsi_filters(struct i40e_vsi *vsi)
 
 			/* flush a full buffer */
 			if (num_del == filter_list_len) {
-				ret = i40e_aq_remove_macvlan(&pf->hw,
-						  vsi->seid, del_list, num_del,
-						  NULL);
+				aq_ret = i40e_aq_remove_macvlan(&pf->hw,
+								vsi->seid,
+								del_list,
+								num_del,
+								NULL);
 				aq_err = pf->hw.aq.asq_last_status;
 				num_del = 0;
 				memset(del_list, 0, sizeof(*del_list));
 
-				if (ret && aq_err != I40E_AQ_RC_ENOENT)
+				if (aq_ret && aq_err != I40E_AQ_RC_ENOENT) {
+					retval = -EIO;
 					dev_err(&pf->pdev->dev,
 						"ignoring delete macvlan error, err %s, aq_err %s while flushing a full buffer\n",
-						i40e_stat_str(&pf->hw, ret),
+						i40e_stat_str(&pf->hw, aq_ret),
 						i40e_aq_str(&pf->hw, aq_err));
+				}
 			}
 			/* Release memory for MAC filter entries which were
 			 * synced up with HW.
@@ -1994,15 +2000,16 @@ int i40e_sync_vsi_filters(struct i40e_vsi *vsi)
 		}
 
 		if (num_del) {
-			ret = i40e_aq_remove_macvlan(&pf->hw, vsi->seid,
-						     del_list, num_del, NULL);
+			aq_ret = i40e_aq_remove_macvlan(&pf->hw, vsi->seid,
+							del_list, num_del,
+							NULL);
 			aq_err = pf->hw.aq.asq_last_status;
 			num_del = 0;
 
-			if (ret && aq_err != I40E_AQ_RC_ENOENT)
+			if (aq_ret && aq_err != I40E_AQ_RC_ENOENT)
 				dev_info(&pf->pdev->dev,
 					 "ignoring delete macvlan error, err %s aq_err %s\n",
-					 i40e_stat_str(&pf->hw, ret),
+					 i40e_stat_str(&pf->hw, aq_ret),
 					 i40e_aq_str(&pf->hw, aq_err));
 		}
 
@@ -2026,7 +2033,8 @@ int i40e_sync_vsi_filters(struct i40e_vsi *vsi)
 			spin_lock_bh(&vsi->mac_filter_list_lock);
 			i40e_undo_add_filter_entries(vsi);
 			spin_unlock_bh(&vsi->mac_filter_list_lock);
-			return -ENOMEM;
+			retval = -ENOMEM;
+			goto out;
 		}
 
 		list_for_each_entry_safe(f, ftmp, &tmp_add_list, list) {
@@ -2047,13 +2055,13 @@ int i40e_sync_vsi_filters(struct i40e_vsi *vsi)
 
 			/* flush a full buffer */
 			if (num_add == filter_list_len) {
-				ret = i40e_aq_add_macvlan(&pf->hw, vsi->seid,
-							  add_list, num_add,
-							  NULL);
+				aq_ret = i40e_aq_add_macvlan(&pf->hw, vsi->seid,
+							     add_list, num_add,
+							     NULL);
 				aq_err = pf->hw.aq.asq_last_status;
 				num_add = 0;
 
-				if (ret)
+				if (aq_ret)
 					break;
 				memset(add_list, 0, sizeof(*add_list));
 			}
@@ -2065,18 +2073,19 @@ int i40e_sync_vsi_filters(struct i40e_vsi *vsi)
 		}
 
 		if (num_add) {
-			ret = i40e_aq_add_macvlan(&pf->hw, vsi->seid,
-						  add_list, num_add, NULL);
+			aq_ret = i40e_aq_add_macvlan(&pf->hw, vsi->seid,
+						     add_list, num_add, NULL);
 			aq_err = pf->hw.aq.asq_last_status;
 			num_add = 0;
 		}
 		kfree(add_list);
 		add_list = NULL;
 
-		if (add_happened && ret && aq_err != I40E_AQ_RC_EINVAL) {
+		if (add_happened && aq_ret && aq_err != I40E_AQ_RC_EINVAL) {
+			retval = i40e_aq_rc_to_posix(aq_ret, aq_err);
 			dev_info(&pf->pdev->dev,
 				 "add filter failed, err %s aq_err %s\n",
-				 i40e_stat_str(&pf->hw, ret),
+				 i40e_stat_str(&pf->hw, aq_ret),
 				 i40e_aq_str(&pf->hw, aq_err));
 			if ((pf->hw.aq.asq_last_status == I40E_AQ_RC_ENOSPC) &&
 			    !test_bit(__I40E_FILTER_OVERFLOW_PROMISC,
@@ -2094,16 +2103,19 @@ int i40e_sync_vsi_filters(struct i40e_vsi *vsi)
 		bool cur_multipromisc;
 
 		cur_multipromisc = !!(vsi->current_netdev_flags & IFF_ALLMULTI);
-		ret = i40e_aq_set_vsi_multicast_promiscuous(&vsi->back->hw,
-							    vsi->seid,
-							    cur_multipromisc,
-							    NULL);
-		if (ret)
+		aq_ret = i40e_aq_set_vsi_multicast_promiscuous(&vsi->back->hw,
+							       vsi->seid,
+							       cur_multipromisc,
+							       NULL);
+		if (aq_ret) {
+			retval = i40e_aq_rc_to_posix(aq_ret,
+						     pf->hw.aq.asq_last_status);
 			dev_info(&pf->pdev->dev,
 				 "set multi promisc failed, err %s aq_err %s\n",
-				 i40e_stat_str(&pf->hw, ret),
+				 i40e_stat_str(&pf->hw, aq_ret),
 				 i40e_aq_str(&pf->hw,
 					     pf->hw.aq.asq_last_status));
+		}
 	}
 	if ((changed_flags & IFF_PROMISC) || promisc_forced_on) {
 		bool cur_promisc;
@@ -2122,36 +2134,47 @@ int i40e_sync_vsi_filters(struct i40e_vsi *vsi)
 				set_bit(__I40E_PF_RESET_REQUESTED, &pf->state);
 			}
 		} else {
-			ret = i40e_aq_set_vsi_unicast_promiscuous(
+			aq_ret = i40e_aq_set_vsi_unicast_promiscuous(
 							  &vsi->back->hw,
 							  vsi->seid,
 							  cur_promisc, NULL);
-			if (ret)
+			if (aq_ret) {
+				retval =
+				i40e_aq_rc_to_posix(aq_ret,
+						    pf->hw.aq.asq_last_status);
 				dev_info(&pf->pdev->dev,
 					 "set unicast promisc failed, err %d, aq_err %d\n",
-					 ret, pf->hw.aq.asq_last_status);
-			ret = i40e_aq_set_vsi_multicast_promiscuous(
+					 aq_ret, pf->hw.aq.asq_last_status);
+			}
+			aq_ret = i40e_aq_set_vsi_multicast_promiscuous(
 							  &vsi->back->hw,
 							  vsi->seid,
 							  cur_promisc, NULL);
-			if (ret)
+			if (aq_ret) {
+				retval =
+				i40e_aq_rc_to_posix(aq_ret,
+						    pf->hw.aq.asq_last_status);
 				dev_info(&pf->pdev->dev,
 					 "set multicast promisc failed, err %d, aq_err %d\n",
-					 ret, pf->hw.aq.asq_last_status);
+					 aq_ret, pf->hw.aq.asq_last_status);
+			}
 		}
-		ret = i40e_aq_set_vsi_broadcast(&vsi->back->hw,
-						vsi->seid,
-						cur_promisc, NULL);
-		if (ret)
+		aq_ret = i40e_aq_set_vsi_broadcast(&vsi->back->hw,
+						   vsi->seid,
+						   cur_promisc, NULL);
+		if (aq_ret) {
+			retval = i40e_aq_rc_to_posix(aq_ret,
+						     pf->hw.aq.asq_last_status);
 			dev_info(&pf->pdev->dev,
 				 "set brdcast promisc failed, err %s, aq_err %s\n",
-				 i40e_stat_str(&pf->hw, ret),
+				 i40e_stat_str(&pf->hw, aq_ret),
 				 i40e_aq_str(&pf->hw,
 					     pf->hw.aq.asq_last_status));
+		}
 	}
-
+out:
 	clear_bit(__I40E_CONFIG_BUSY, &vsi->state);
-	return 0;
+	return retval;
 }
 
 /**
diff --git a/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c b/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c
index 819803c8..30a1d30 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c
@@ -1633,9 +1633,10 @@ static int i40e_vc_add_mac_addr_msg(struct i40e_vf *vf, u8 *msg, u16 msglen)
 	spin_unlock_bh(&vsi->mac_filter_list_lock);
 
 	/* program the updated filter list */
-	if (i40e_sync_vsi_filters(vsi))
-		dev_err(&pf->pdev->dev, "Unable to program VF %d MAC filters\n",
-			vf->vf_id);
+	ret = i40e_sync_vsi_filters(vsi);
+	if (ret)
+		dev_err(&pf->pdev->dev, "Unable to program VF %d MAC filters, error %d\n",
+			vf->vf_id, ret);
 
 error_param:
 	/* send the response to the VF */
@@ -1687,9 +1688,10 @@ static int i40e_vc_del_mac_addr_msg(struct i40e_vf *vf, u8 *msg, u16 msglen)
 	spin_unlock_bh(&vsi->mac_filter_list_lock);
 
 	/* program the updated filter list */
-	if (i40e_sync_vsi_filters(vsi))
-		dev_err(&pf->pdev->dev, "Unable to program VF %d MAC filters\n",
-			vf->vf_id);
+	ret = i40e_sync_vsi_filters(vsi);
+	if (ret)
+		dev_err(&pf->pdev->dev, "Unable to program VF %d MAC filters, error %d\n",
+			vf->vf_id, ret);
 
 error_param:
 	/* send the response to the VF */
-- 
2.5.0

[net-next 08/15] i40evf: use correct types

[Jeff Kirsher]

From: Mitch Williams <mitch.a.williams@intel.com>

Don't use uint32_t type the kernel. Use u32 instead. No functional
change.

Change-ID: I77bbf3b6464edaef747c7104b43534032a4dba63
Signed-off-by: Mitch Williams <mitch.a.williams@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/i40evf/i40evf_main.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/intel/i40evf/i40evf_main.c b/drivers/net/ethernet/intel/i40evf/i40evf_main.c
index 82c9adf..2577757 100644
--- a/drivers/net/ethernet/intel/i40evf/i40evf_main.c
+++ b/drivers/net/ethernet/intel/i40evf/i40evf_main.c
@@ -259,7 +259,7 @@ static void i40evf_fire_sw_int(struct i40evf_adapter *adapter, u32 mask)
 {
 	struct i40e_hw *hw = &adapter->hw;
 	int i;
-	uint32_t dyn_ctl;
+	u32 dyn_ctl;
 
 	if (mask & 1) {
 		dyn_ctl = rd32(hw, I40E_VFINT_DYN_CTL01);
-- 
2.5.0

[net-next 09/15] i40e: use priv flags to control packet split

[Jeff Kirsher]

From: Jesse Brandeburg <jesse.brandeburg@intel.com>

Ethtool priv flags implementation to enable or disable packet split, which
is a hardware feature that inspects headers and will put headers in a
separate DMA buffer from the payload data.  The driver was automatically
choosing to enable packet split in some cases and this gives the user the
ability to turn it off/on explicitly.

to query state:
ethtool --show-priv-flags ethx

to enable:
ethtool --set-priv-flags ethx packet-split on
to disable:
ethtool --set-priv-flags ethx packet-split off

Why would anyone want this?
	Because some environments benefit from header/data split in the receive
	buffer, and the driver defaults to one or the other depending on
	environment/kernel parameters.

Why didn't you implement a generic ethtool control for this feature?
	Because Intel hardware is the only hardware that supports header/data
	split.

Change-ID: I803121e1eecc9ccb2884031fd85dd1110b3af66d
Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/i40e/i40e.h         |  1 +
 drivers/net/ethernet/intel/i40e/i40e_ethtool.c | 27 ++++++++++++++++++++++++++
 2 files changed, 28 insertions(+)

diff --git a/drivers/net/ethernet/intel/i40e/i40e.h b/drivers/net/ethernet/intel/i40e/i40e.h
index dde7ae7..4ca85d5 100644
--- a/drivers/net/ethernet/intel/i40e/i40e.h
+++ b/drivers/net/ethernet/intel/i40e/i40e.h
@@ -104,6 +104,7 @@
 #define I40E_PRIV_FLAGS_LINKPOLL_FLAG	BIT(1)
 #define I40E_PRIV_FLAGS_FD_ATR		BIT(2)
 #define I40E_PRIV_FLAGS_VEB_STATS	BIT(3)
+#define I40E_PRIV_FLAGS_PS		BIT(4)
 
 #define I40E_NVM_VERSION_LO_SHIFT  0
 #define I40E_NVM_VERSION_LO_MASK   (0xff << I40E_NVM_VERSION_LO_SHIFT)
diff --git a/drivers/net/ethernet/intel/i40e/i40e_ethtool.c b/drivers/net/ethernet/intel/i40e/i40e_ethtool.c
index b52c509..29d5833 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_ethtool.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_ethtool.c
@@ -231,6 +231,7 @@ static const char i40e_priv_flags_strings[][ETH_GSTRING_LEN] = {
 	"LinkPolling",
 	"flow-director-atr",
 	"veb-stats",
+	"packet-split",
 };
 
 #define I40E_PRIV_FLAGS_STR_LEN ARRAY_SIZE(i40e_priv_flags_strings)
@@ -2709,6 +2710,8 @@ static u32 i40e_get_priv_flags(struct net_device *dev)
 		I40E_PRIV_FLAGS_FD_ATR : 0;
 	ret_flags |= pf->flags & I40E_FLAG_VEB_STATS_ENABLED ?
 		I40E_PRIV_FLAGS_VEB_STATS : 0;
+	ret_flags |= pf->flags & I40E_FLAG_RX_PS_ENABLED ?
+		I40E_PRIV_FLAGS_PS : 0;
 
 	return ret_flags;
 }
@@ -2723,6 +2726,26 @@ static int i40e_set_priv_flags(struct net_device *dev, u32 flags)
 	struct i40e_netdev_priv *np = netdev_priv(dev);
 	struct i40e_vsi *vsi = np->vsi;
 	struct i40e_pf *pf = vsi->back;
+	bool reset_required = false;
+
+	/* NOTE: MFP is not settable */
+
+	/* allow the user to control the method of receive
+	 * buffer DMA, whether the packet is split at header
+	 * boundaries into two separate buffers.  In some cases
+	 * one routine or the other will perform better.
+	 */
+	if ((flags & I40E_PRIV_FLAGS_PS) &&
+	    !(pf->flags & I40E_FLAG_RX_PS_ENABLED)) {
+		pf->flags |= I40E_FLAG_RX_PS_ENABLED;
+		pf->flags &= ~I40E_FLAG_RX_1BUF_ENABLED;
+		reset_required = true;
+	} else if (!(flags & I40E_PRIV_FLAGS_PS) &&
+		   (pf->flags & I40E_FLAG_RX_PS_ENABLED)) {
+		pf->flags &= ~I40E_FLAG_RX_PS_ENABLED;
+		pf->flags |= I40E_FLAG_RX_1BUF_ENABLED;
+		reset_required = true;
+	}
 
 	if (flags & I40E_PRIV_FLAGS_LINKPOLL_FLAG)
 		pf->flags |= I40E_FLAG_LINK_POLLING_ENABLED;
@@ -2745,6 +2768,10 @@ static int i40e_set_priv_flags(struct net_device *dev, u32 flags)
 	else
 		pf->flags &= ~I40E_FLAG_VEB_STATS_ENABLED;
 
+	/* if needed, issue reset to cause things to take effect */
+	if (reset_required)
+		i40e_do_reset(pf, BIT(__I40E_PF_RESET_REQUESTED));
+
 	return 0;
 }
 
-- 
2.5.0

[net-next 10/15] i40e: remove CONFIG_I40E_VXLAN

[Jeff Kirsher]

From: Jesse Brandeburg <jesse.brandeburg@intel.com>

Instead of having our own custom symbol, we can just rely
on whether or not the kernel has the feature enabled.

In this case use IS_ENABLED(CONFIG_VXLAN) in order to handle
built-in or module in the current BKM way.

Change-ID: I5890fbb518ff8ed6bb07c3362fb0a8a829f9b241
Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/Kconfig          | 11 -----------
 drivers/net/ethernet/intel/i40e/i40e.h      |  4 +---
 drivers/net/ethernet/intel/i40e/i40e_main.c | 12 ++++++------
 3 files changed, 7 insertions(+), 20 deletions(-)

diff --git a/drivers/net/ethernet/intel/Kconfig b/drivers/net/ethernet/intel/Kconfig
index 4163b16..061e4e0 100644
--- a/drivers/net/ethernet/intel/Kconfig
+++ b/drivers/net/ethernet/intel/Kconfig
@@ -269,17 +269,6 @@ config I40E
 	  To compile this driver as a module, choose M here. The module
 	  will be called i40e.
 
-config I40E_VXLAN
-	bool "Virtual eXtensible Local Area Network Support"
-	default n
-	depends on I40E && VXLAN && !(I40E=y && VXLAN=m)
-	---help---
-	  This allows one to create VXLAN virtual interfaces that provide
-	  Layer 2 Networks over Layer 3 Networks. VXLAN is often used
-	  to tunnel virtual network infrastructure in virtualized environments.
-	  Say Y here if you want to use Virtual eXtensible Local Area Network
-	  (VXLAN) in the driver.
-
 config I40E_DCB
 	bool "Data Center Bridging (DCB) Support"
 	default n
diff --git a/drivers/net/ethernet/intel/i40e/i40e.h b/drivers/net/ethernet/intel/i40e/i40e.h
index 4ca85d5..8ed759e 100644
--- a/drivers/net/ethernet/intel/i40e/i40e.h
+++ b/drivers/net/ethernet/intel/i40e/i40e.h
@@ -282,7 +282,7 @@ struct i40e_pf {
 	u32 fd_atr_cnt;
 	u32 fd_tcp_rule;
 
-#ifdef CONFIG_I40E_VXLAN
+#if IS_ENABLED(CONFIG_VXLAN)
 	__be16  vxlan_ports[I40E_MAX_PF_UDP_OFFLOAD_PORTS];
 	u16 pending_vxlan_bitmap;
 
@@ -323,9 +323,7 @@ struct i40e_pf {
 #define I40E_FLAG_FD_ATR_ENABLED		BIT_ULL(22)
 #define I40E_FLAG_PTP				BIT_ULL(25)
 #define I40E_FLAG_MFP_ENABLED			BIT_ULL(26)
-#ifdef CONFIG_I40E_VXLAN
 #define I40E_FLAG_VXLAN_FILTER_SYNC		BIT_ULL(27)
-#endif
 #define I40E_FLAG_PORT_ID_VALID			BIT_ULL(28)
 #define I40E_FLAG_DCB_CAPABLE			BIT_ULL(29)
 #define I40E_FLAG_RSS_AQ_CAPABLE		BIT_ULL(31)
diff --git a/drivers/net/ethernet/intel/i40e/i40e_main.c b/drivers/net/ethernet/intel/i40e/i40e_main.c
index c5a24fe..b731a82 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_main.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_main.c
@@ -27,7 +27,7 @@
 /* Local includes */
 #include "i40e.h"
 #include "i40e_diag.h"
-#ifdef CONFIG_I40E_VXLAN
+#if IS_ENABLED(CONFIG_VXLAN)
 #include <net/vxlan.h>
 #endif
 
@@ -5357,7 +5357,7 @@ int i40e_open(struct net_device *netdev)
 						       TCP_FLAG_CWR) >> 16);
 	wr32(&pf->hw, I40E_GLLAN_TSOMSK_L, be32_to_cpu(TCP_FLAG_CWR) >> 16);
 
-#ifdef CONFIG_I40E_VXLAN
+#if IS_ENABLED(CONFIG_VXLAN)
 	vxlan_get_rx_port(netdev);
 #endif
 
@@ -7053,7 +7053,7 @@ static void i40e_handle_mdd_event(struct i40e_pf *pf)
 	i40e_flush(hw);
 }
 
-#ifdef CONFIG_I40E_VXLAN
+#if IS_ENABLED(CONFIG_VXLAN)
 /**
  * i40e_sync_vxlan_filters_subtask - Sync the VSI filter list with HW
  * @pf: board private structure
@@ -7120,7 +7120,7 @@ static void i40e_service_task(struct work_struct *work)
 	i40e_watchdog_subtask(pf);
 	i40e_fdir_reinit_subtask(pf);
 	i40e_sync_filters_subtask(pf);
-#ifdef CONFIG_I40E_VXLAN
+#if IS_ENABLED(CONFIG_VXLAN)
 	i40e_sync_vxlan_filters_subtask(pf);
 #endif
 	i40e_clean_adminq_subtask(pf);
@@ -8496,7 +8496,7 @@ static int i40e_set_features(struct net_device *netdev,
 	return 0;
 }
 
-#ifdef CONFIG_I40E_VXLAN
+#if IS_ENABLED(CONFIG_VXLAN)
 /**
  * i40e_get_vxlan_port_idx - Lookup a possibly offloaded for Rx UDP port
  * @pf: board private structure
@@ -8816,7 +8816,7 @@ static const struct net_device_ops i40e_netdev_ops = {
 	.ndo_get_vf_config	= i40e_ndo_get_vf_config,
 	.ndo_set_vf_link_state	= i40e_ndo_set_vf_link_state,
 	.ndo_set_vf_spoofchk	= i40e_ndo_set_vf_spoofchk,
-#ifdef CONFIG_I40E_VXLAN
+#if IS_ENABLED(CONFIG_VXLAN)
 	.ndo_add_vxlan_port	= i40e_add_vxlan_port,
 	.ndo_del_vxlan_port	= i40e_del_vxlan_port,
 #endif
-- 
2.5.0

[net-next 11/15] i40e: Remove separate functions gathering XOFF Rx stats

[Jeff Kirsher]

From: Neerav Parikh <neerav.parikh@intel.com>

The separate functions to gather Flow control Rx XOFF stats was to
determine if the Tx for a queue was paused due to Link Flow Control(LFC)
or Priority Flow Control(PFC).

But, with recent change in the i40e driver the logic for checking th Tx
hang has been removed and these functions don't do anything meaningful.
Hence, there is no need to keep these separate functions to gather Rx
XOFF stats for LFC or PFC.

This patch removes these functions and moves the stat collection for
XOFF Rx to the i40e_update_pf_stats() that collects all the PF stats.

Change-ID: Iec1452dac3a6766f0d968e754cb407530d7c60cd
Signed-off-by: Neerav Parikh <neerav.parikh@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/i40e/i40e_main.c | 77 +++--------------------------
 1 file changed, 7 insertions(+), 70 deletions(-)

diff --git a/drivers/net/ethernet/intel/i40e/i40e_main.c b/drivers/net/ethernet/intel/i40e/i40e_main.c
index b731a82..1c5c0fd 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_main.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_main.c
@@ -792,75 +792,6 @@ static void i40e_update_fcoe_stats(struct i40e_vsi *vsi)
 
 #endif
 /**
- * i40e_update_link_xoff_rx - Update XOFF received in link flow control mode
- * @pf: the corresponding PF
- *
- * Update the Rx XOFF counter (PAUSE frames) in link flow control mode
- **/
-static void i40e_update_link_xoff_rx(struct i40e_pf *pf)
-{
-	struct i40e_hw_port_stats *osd = &pf->stats_offsets;
-	struct i40e_hw_port_stats *nsd = &pf->stats;
-	struct i40e_hw *hw = &pf->hw;
-	u64 xoff = 0;
-
-	if ((hw->fc.current_mode != I40E_FC_FULL) &&
-	    (hw->fc.current_mode != I40E_FC_RX_PAUSE))
-		return;
-
-	xoff = nsd->link_xoff_rx;
-	i40e_stat_update32(hw, I40E_GLPRT_LXOFFRXC(hw->port),
-			   pf->stat_offsets_loaded,
-			   &osd->link_xoff_rx, &nsd->link_xoff_rx);
-
-	/* No new LFC xoff rx */
-	if (!(nsd->link_xoff_rx - xoff))
-		return;
-
-}
-
-/**
- * i40e_update_prio_xoff_rx - Update XOFF received in PFC mode
- * @pf: the corresponding PF
- *
- * Update the Rx XOFF counter (PAUSE frames) in PFC mode
- **/
-static void i40e_update_prio_xoff_rx(struct i40e_pf *pf)
-{
-	struct i40e_hw_port_stats *osd = &pf->stats_offsets;
-	struct i40e_hw_port_stats *nsd = &pf->stats;
-	bool xoff[I40E_MAX_TRAFFIC_CLASS] = {false};
-	struct i40e_dcbx_config *dcb_cfg;
-	struct i40e_hw *hw = &pf->hw;
-	u16 i;
-	u8 tc;
-
-	dcb_cfg = &hw->local_dcbx_config;
-
-	/* Collect Link XOFF stats when PFC is disabled */
-	if (!dcb_cfg->pfc.pfcenable) {
-		i40e_update_link_xoff_rx(pf);
-		return;
-	}
-
-	for (i = 0; i < I40E_MAX_USER_PRIORITY; i++) {
-		u64 prio_xoff = nsd->priority_xoff_rx[i];
-
-		i40e_stat_update32(hw, I40E_GLPRT_PXOFFRXC(hw->port, i),
-				   pf->stat_offsets_loaded,
-				   &osd->priority_xoff_rx[i],
-				   &nsd->priority_xoff_rx[i]);
-
-		/* No new PFC xoff rx */
-		if (!(nsd->priority_xoff_rx[i] - prio_xoff))
-			continue;
-		/* Get the TC for given priority */
-		tc = dcb_cfg->etscfg.prioritytable[i];
-		xoff[tc] = true;
-	}
-}
-
-/**
  * i40e_update_vsi_stats - Update the vsi statistics counters.
  * @vsi: the VSI to be updated
  *
@@ -1054,12 +985,18 @@ static void i40e_update_pf_stats(struct i40e_pf *pf)
 	i40e_stat_update32(hw, I40E_GLPRT_LXONTXC(hw->port),
 			   pf->stat_offsets_loaded,
 			   &osd->link_xon_tx, &nsd->link_xon_tx);
-	i40e_update_prio_xoff_rx(pf);  /* handles I40E_GLPRT_LXOFFRXC */
+	i40e_stat_update32(hw, I40E_GLPRT_LXOFFRXC(hw->port),
+			   pf->stat_offsets_loaded,
+			   &osd->link_xoff_rx, &nsd->link_xoff_rx);
 	i40e_stat_update32(hw, I40E_GLPRT_LXOFFTXC(hw->port),
 			   pf->stat_offsets_loaded,
 			   &osd->link_xoff_tx, &nsd->link_xoff_tx);
 
 	for (i = 0; i < 8; i++) {
+		i40e_stat_update32(hw, I40E_GLPRT_PXOFFRXC(hw->port, i),
+				   pf->stat_offsets_loaded,
+				   &osd->priority_xoff_rx[i],
+				   &nsd->priority_xoff_rx[i]);
 		i40e_stat_update32(hw, I40E_GLPRT_PXONRXC(hw->port, i),
 				   pf->stat_offsets_loaded,
 				   &osd->priority_xon_rx[i],
-- 
2.5.0

[net-next 12/15] i40e: fix whitespace

[Jeff Kirsher]

From: Mitch Williams <mitch.a.williams@intel.com>

Operators should have spaces around them.

Change-ID: I64735e9aa8618b9a5059a87ace1c999d6d3bfcfb
Signed-off-by: Mitch Williams <mitch.a.williams@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c b/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c
index 30a1d30..b3bd81c 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_virtchnl_pf.c
@@ -290,8 +290,8 @@ static void i40e_config_irq_link_list(struct i40e_vf *vf, u16 vsi_id,
 	next_q = find_first_bit(&linklistmap,
 				(I40E_MAX_VSI_QP *
 				 I40E_VIRTCHNL_SUPPORTED_QTYPES));
-	vsi_queue_id = next_q/I40E_VIRTCHNL_SUPPORTED_QTYPES;
-	qtype = next_q%I40E_VIRTCHNL_SUPPORTED_QTYPES;
+	vsi_queue_id = next_q / I40E_VIRTCHNL_SUPPORTED_QTYPES;
+	qtype = next_q % I40E_VIRTCHNL_SUPPORTED_QTYPES;
 	pf_queue_id = i40e_vc_get_pf_queue_id(vf, vsi_id, vsi_queue_id);
 	reg = ((qtype << I40E_VPINT_LNKLSTN_FIRSTQ_TYPE_SHIFT) | pf_queue_id);
 
-- 
2.5.0

[net-next 13/15] i40e/i40evf: use logical operator

[Jeff Kirsher]

From: Mitch Williams <mitch.a.williams@intel.com>

We shouldn't be using a bitwise operator here; it's not a bitwise
operation. Use a logical operator instead. Why doesn't c have a
logical-or-and-assign operator?

Change-ID: Id84f3ca884910bed7073c84b1e16a102e958d0de
Signed-off-by: Mitch Williams <mitch.a.williams@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/i40e/i40e_txrx.c   | 2 +-
 drivers/net/ethernet/intel/i40evf/i40e_txrx.c | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/net/ethernet/intel/i40e/i40e_txrx.c b/drivers/net/ethernet/intel/i40e/i40e_txrx.c
index 18a493e..be129d3 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_txrx.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_txrx.c
@@ -1907,7 +1907,7 @@ int i40e_napi_poll(struct napi_struct *napi, int budget)
 	 */
 	i40e_for_each_ring(ring, q_vector->tx) {
 		clean_complete &= i40e_clean_tx_irq(ring, vsi->work_limit);
-		arm_wb |= ring->arm_wb;
+		arm_wb = arm_wb || ring->arm_wb;
 		ring->arm_wb = false;
 	}
 
diff --git a/drivers/net/ethernet/intel/i40evf/i40e_txrx.c b/drivers/net/ethernet/intel/i40evf/i40e_txrx.c
index 4f1ac86..b53333a 100644
--- a/drivers/net/ethernet/intel/i40evf/i40e_txrx.c
+++ b/drivers/net/ethernet/intel/i40evf/i40e_txrx.c
@@ -1331,7 +1331,7 @@ int i40evf_napi_poll(struct napi_struct *napi, int budget)
 	 */
 	i40e_for_each_ring(ring, q_vector->tx) {
 		clean_complete &= i40e_clean_tx_irq(ring, vsi->work_limit);
-		arm_wb |= ring->arm_wb;
+		arm_wb = arm_wb || ring->arm_wb;
 		ring->arm_wb = false;
 	}
 
-- 
2.5.0

[net-next 14/15] i40e/i40evf: Bump version to 1.4.7 for i40e and 1.4.3 for i40evf

[Jeff Kirsher]

From: Catherine Sullivan <catherine.sullivan@intel.com>

Bump.

Change-ID: Id8c83c64c973349a722bab40d285ad8ded8c28f7
Signed-off-by: Catherine Sullivan <catherine.sullivan@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/i40e/i40e_main.c     | 2 +-
 drivers/net/ethernet/intel/i40evf/i40evf_main.c | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/net/ethernet/intel/i40e/i40e_main.c b/drivers/net/ethernet/intel/i40e/i40e_main.c
index 1c5c0fd..b447af6 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_main.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_main.c
@@ -39,7 +39,7 @@ static const char i40e_driver_string[] =
 
 #define DRV_VERSION_MAJOR 1
 #define DRV_VERSION_MINOR 4
-#define DRV_VERSION_BUILD 4
+#define DRV_VERSION_BUILD 7
 #define DRV_VERSION __stringify(DRV_VERSION_MAJOR) "." \
 	     __stringify(DRV_VERSION_MINOR) "." \
 	     __stringify(DRV_VERSION_BUILD)    DRV_KERN
diff --git a/drivers/net/ethernet/intel/i40evf/i40evf_main.c b/drivers/net/ethernet/intel/i40evf/i40evf_main.c
index 2577757..b4c632f 100644
--- a/drivers/net/ethernet/intel/i40evf/i40evf_main.c
+++ b/drivers/net/ethernet/intel/i40evf/i40evf_main.c
@@ -34,7 +34,7 @@ char i40evf_driver_name[] = "i40evf";
 static const char i40evf_driver_string[] =
 	"Intel(R) XL710/X710 Virtual Function Network Driver";
 
-#define DRV_VERSION "1.4.1"
+#define DRV_VERSION "1.4.3"
 const char i40evf_driver_version[] = DRV_VERSION;
 static const char i40evf_copyright[] =
 	"Copyright (c) 2013 - 2015 Intel Corporation.";
-- 
2.5.0

[net-next 15/15] i40e: trivial fixes

[Jeff Kirsher]

From: Jesse Brandeburg <jesse.brandeburg@intel.com>

1) remove duplicate include of tcp.h
2) put an ampersand at the end of a line instead of the beginning
3) remove a useless dev_info
4) match declaration of function to the implementation
5) repair incorrect comment
6) correct whitespace
7) remove unused define

Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---
 drivers/net/ethernet/intel/i40e/i40e.h        |  7 +++----
 drivers/net/ethernet/intel/i40e/i40e_txrx.c   |  6 +-----
 drivers/net/ethernet/intel/i40evf/i40e_txrx.c | 15 ++++-----------
 3 files changed, 8 insertions(+), 20 deletions(-)

diff --git a/drivers/net/ethernet/intel/i40e/i40e.h b/drivers/net/ethernet/intel/i40e/i40e.h
index 8ed759e..23b4580 100644
--- a/drivers/net/ethernet/intel/i40e/i40e.h
+++ b/drivers/net/ethernet/intel/i40e/i40e.h
@@ -42,7 +42,6 @@
 #include <linux/string.h>
 #include <linux/in.h>
 #include <linux/ip.h>
-#include <linux/tcp.h>
 #include <linux/sctp.h>
 #include <linux/pkt_sched.h>
 #include <linux/ipv6.h>
@@ -608,8 +607,8 @@ static inline char *i40e_nvm_version_str(struct i40e_hw *hw)
 
 	full_ver = hw->nvm.oem_ver;
 	ver = (u8)(full_ver >> I40E_OEM_VER_SHIFT);
-	build = (u16)((full_ver >> I40E_OEM_VER_BUILD_SHIFT)
-		 & I40E_OEM_VER_BUILD_MASK);
+	build = (u16)((full_ver >> I40E_OEM_VER_BUILD_SHIFT) &
+		 I40E_OEM_VER_BUILD_MASK);
 	patch = (u8)(full_ver & I40E_OEM_VER_PATCH_MASK);
 
 	snprintf(buf, sizeof(buf),
@@ -717,7 +716,7 @@ struct i40e_veb *i40e_veb_setup(struct i40e_pf *pf, u16 flags, u16 uplink_seid,
 void i40e_veb_release(struct i40e_veb *veb);
 
 int i40e_veb_config_tc(struct i40e_veb *veb, u8 enabled_tc);
-i40e_status i40e_vsi_add_pvid(struct i40e_vsi *vsi, u16 vid);
+int i40e_vsi_add_pvid(struct i40e_vsi *vsi, u16 vid);
 void i40e_vsi_remove_pvid(struct i40e_vsi *vsi);
 void i40e_vsi_reset_stats(struct i40e_vsi *vsi);
 void i40e_pf_reset_stats(struct i40e_pf *pf);
diff --git a/drivers/net/ethernet/intel/i40e/i40e_txrx.c b/drivers/net/ethernet/intel/i40e/i40e_txrx.c
index be129d3..b0ae3e6 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_txrx.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_txrx.c
@@ -328,7 +328,7 @@ static int i40e_add_del_fdir_tcpv4(struct i40e_vsi *vsi,
  * @fd_data: the flow director data required for the FDir descriptor
  * @add: true adds a filter, false removes it
  *
- * Always returns -EOPNOTSUPP
+ * Returns 0 if the filters were successfully added or removed
  **/
 static int i40e_add_del_fdir_sctpv4(struct i40e_vsi *vsi,
 				    struct i40e_fdir_filter *fd_data,
@@ -515,9 +515,6 @@ static void i40e_fd_handle_status(struct i40e_ring *rx_ring,
 				pf->auto_disable_flags |=
 							I40E_FLAG_FD_SB_ENABLED;
 			}
-		} else {
-			dev_info(&pdev->dev,
-				"FD filter programming failed due to incorrect filter parameters\n");
 		}
 	} else if (error == BIT(I40E_RX_PROG_STATUS_DESC_NO_FD_ENTRY_SHIFT)) {
 		if (I40E_DEBUG_FD & pf->hw.debug_mask)
@@ -1872,7 +1869,6 @@ enable_int:
 		q_vector->itr_countdown--;
 	else
 		q_vector->itr_countdown = ITR_COUNTDOWN_START;
-
 }
 
 /**
diff --git a/drivers/net/ethernet/intel/i40evf/i40e_txrx.c b/drivers/net/ethernet/intel/i40evf/i40e_txrx.c
index b53333a..4ca4065 100644
--- a/drivers/net/ethernet/intel/i40evf/i40e_txrx.c
+++ b/drivers/net/ethernet/intel/i40evf/i40e_txrx.c
@@ -411,7 +411,7 @@ static bool i40e_set_new_dynamic_itr(struct i40e_ring_container *rc)
 	return false;
 }
 
-/*
+/**
  * i40evf_setup_tx_descriptors - Allocate the Tx descriptors
  * @tx_ring: the tx ring to set up
  *
@@ -1259,10 +1259,12 @@ static inline void i40e_update_enable_itr(struct i40e_vsi *vsi,
 		rx = i40e_set_new_dynamic_itr(&q_vector->rx);
 		rxval = i40e_buildreg_itr(I40E_RX_ITR, q_vector->rx.itr);
 	}
+
 	if (ITR_IS_DYNAMIC(vsi->tx_itr_setting)) {
 		tx = i40e_set_new_dynamic_itr(&q_vector->tx);
 		txval = i40e_buildreg_itr(I40E_TX_ITR, q_vector->tx.itr);
 	}
+
 	if (rx || tx) {
 		/* get the higher of the two ITR adjustments and
 		 * use the same value for both ITR registers
@@ -1298,7 +1300,6 @@ enable_int:
 		q_vector->itr_countdown--;
 	else
 		q_vector->itr_countdown = ITR_COUNTDOWN_START;
-
 }
 
 /**
@@ -1552,7 +1553,6 @@ static void i40e_tx_enable_csum(struct sk_buff *skb, u32 *tx_flags,
 			*tx_flags |= I40E_TX_FLAGS_IPV6;
 		}
 
-
 		if ((tx_ring->flags & I40E_TXR_FLAGS_OUTER_UDP_CSUM) &&
 		    (l4_tunnel == I40E_TXD_CTX_UDP_TUNNELING)        &&
 		    (*cd_tunneling & I40E_TXD_CTX_QW0_EXT_IP_MASK)) {
@@ -1651,7 +1651,7 @@ static void i40e_create_tx_ctx(struct i40e_ring *tx_ring,
 	context_desc->type_cmd_tso_mss = cpu_to_le64(cd_type_cmd_tso_mss);
 }
 
- /**
+/**
  * i40e_chk_linearize - Check if there are more than 8 fragments per packet
  * @skb:      send buffer
  * @tx_flags: collected send information
@@ -1847,7 +1847,6 @@ static inline void i40evf_tx_map(struct i40e_ring *tx_ring, struct sk_buff *skb,
 		tx_bi = &tx_ring->tx_bi[i];
 	}
 
-#define WB_STRIDE 0x3
 	/* set next_to_watch value indicating a packet is present */
 	first->next_to_watch = tx_desc;
 
@@ -1874,12 +1873,6 @@ static inline void i40evf_tx_map(struct i40e_ring *tx_ring, struct sk_buff *skb,
 	 *			update tail and set RS bit on every packet.
 	 *	if xmit_more is false and last_xmit_more was true
 	 *		update tail and set RS bit.
-	 * else (kernel < 3.18)
-	 *	if every packet spanned less than 4 desc
-	 *		then set RS bit on 4th packet and update tail
-	 *		on every packet
-	 *	else
-	 *		set RS bit on EOP for every packet and update tail
 	 *
 	 * Optimization: wmb to be issued only in case of tail update.
 	 * Also optimize the Descriptor WB path for RS bit with the same
-- 
2.5.0

Re: [net-next 00/15][pull request] 40GbE Intel Wired LAN Driver Updates 2015-12-03

[David Miller]

From: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Date: Thu,  3 Dec 2015 05:25:45 -0800

> This series contains updates to i40e and i40evf only.

Pulled, thanks Jeff.

Re: [net-next 09/15] i40e: use priv flags to control packet split

[Or Gerlitz]

On Thu, Dec 3, 2015 at 3:25 PM, Jeff Kirsher
<jeffrey.t.kirsher@intel.com> wrote:
> From: Jesse Brandeburg <jesse.brandeburg@intel.com>
>
> Ethtool priv flags implementation to enable or disable packet split, which
> is a hardware feature that inspects headers and will put headers in a
> separate DMA buffer from the payload data.  The driver was automatically
> choosing to enable packet split in some cases and this gives the user the
> ability to turn it off/on explicitly.
>
> to query state:
> ethtool --show-priv-flags ethx
>
> to enable:
> ethtool --set-priv-flags ethx packet-split on
> to disable:
> ethtool --set-priv-flags ethx packet-split off
>
> Why would anyone want this?
>         Because some environments benefit from header/data split in the receive
>         buffer, and the driver defaults to one or the other depending on
>         environment/kernel parameters.
>
> Why didn't you implement a generic ethtool control for this feature?
>         Because Intel hardware is the only hardware that supports header/data
>         split.


Hi Jesse,

HDS is a generic feature,and even if you think only Intel hardware is
the only hardware that supports that, you have multiple drivers..
right? why not go the right way of adding new netdev feature and
enable/disable it through the conventional ethtool feature on/off
directive?

Or.

Re: [net-next 09/15] i40e: use priv flags to control packet split

[Jesse Brandeburg]

On Sat, Dec 5, 2015 at 1:25 PM, Or Gerlitz <gerlitz.or@gmail.com> wrote:
> HDS is a generic feature,and even if you think only Intel hardware is
> the only hardware that supports that, you have multiple drivers..
> right? why not go the right way of adding new netdev feature and
> enable/disable it through the conventional ethtool feature on/off
> directive?

In this case, the return on the investment in the new feature is too
low, IMO. If your hardware also supports this feature we can discuss
adding this via ethtool, and I'd be glad to review any patches, and
implement for Intel devices if necessary.

Re: [net-next 09/15] i40e: use priv flags to control packet split

[Or Gerlitz]

On Wed, Dec 9, 2015 at 1:24 AM, Jesse Brandeburg
<jesse.brandeburg@gmail.com> wrote:
> On Sat, Dec 5, 2015 at 1:25 PM, Or Gerlitz <gerlitz.or@gmail.com> wrote:
>> HDS is a generic feature,and even if you think only Intel hardware is
>> the only hardware that supports that, you have multiple drivers..
>> right? why not go the right way of adding new netdev feature and
>> enable/disable it through the conventional ethtool feature on/off
>> directive?
>
> In this case, the return on the investment in the new feature is too
> low, IMO. If your hardware also supports this feature we can discuss
> adding this via ethtool, and I'd be glad to review any patches, and
> implement for Intel devices if necessary.

Wasn't sure to follow on the ROI comment, but anyway, we can renew
this discussion if/when Mellanox and/or more HWs to open that. BTW do
you iSCSI/NFS or alike benchmark of (say) 4KB payload to demo the gain
of the patch? where does your HW does the HDSplit in case of these or
other IP based storage protocols?

Or.

Or.