Netdev List
 help / color / mirror / Atom feed
* Re: [net-next-2.6 PATCH 20/20] ixgbe: rewrite ethtool test to use standard config functions
From: David Miller @ 2010-08-19 23:47 UTC (permalink / raw)
  To: jeffrey.t.kirsher; +Cc: netdev, gospo, bphilips, alexander.h.duyck
In-Reply-To: <20100819234052.10335.57519.stgit@localhost.localdomain>

From: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Date: Thu, 19 Aug 2010 16:40:54 -0700

> From: Alexander Duyck <alexander.h.duyck@intel.com>
> 
> This change makes it so that the ethtool loopback test uses the standard
> ring configuration and allocation functions.  As a result the loopback test
> will be much more effective at testing core driver functionality.
> 
> Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com>
> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

Applied.

^ permalink raw reply

* Re: [net-next-2.6 PATCH 19/20] ixgbe: update all DESC_ADV macros to accept a ring pointer
From: David Miller @ 2010-08-19 23:47 UTC (permalink / raw)
  To: jeffrey.t.kirsher; +Cc: netdev, gospo, bphilips, alexander.h.duyck
In-Reply-To: <20100819234028.10335.95335.stgit@localhost.localdomain>

From: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Date: Thu, 19 Aug 2010 16:40:31 -0700

> From: Alexander Duyck <alexander.h.duyck@intel.com>
> 
> All of the DESC_ADV macros are currently needing the pointers to be
> de-referenced before accessing the ring.  Instead of having to add all of
> the asterisks it is easier to just update the macro to expect a pointer to
> the ring.
> 
> Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com>
> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

Applied.

^ permalink raw reply

* Re: [net-next-2.6 PATCH 18/20] ixgbe: combine Rx into into ixgbe_configure_rx
From: David Miller @ 2010-08-19 23:47 UTC (permalink / raw)
  To: jeffrey.t.kirsher; +Cc: netdev, gospo, bphilips, alexander.h.duyck
In-Reply-To: <20100819234004.10335.44104.stgit@localhost.localdomain>

From: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Date: Thu, 19 Aug 2010 16:40:06 -0700

> From: Alexander Duyck <alexander.h.duyck@intel.com>
> 
> The Rx init is currently split over ixgbe_configure, ixgbe_configure_rx,
> and ixgbe_up_complete.  Instead of leaving it split over 3 function it is
> easier to consolidate them all into ixgbe_configure_rx.
> 
> Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com>
> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

Applied.

^ permalink raw reply

* Re: [net-next-2.6 PATCH 17/20] ixgbe: pull all Tx init into ixgbe_configure_tx
From: David Miller @ 2010-08-19 23:47 UTC (permalink / raw)
  To: jeffrey.t.kirsher; +Cc: netdev, gospo, bphilips, alexander.h.duyck
In-Reply-To: <20100819233940.10335.28076.stgit@localhost.localdomain>

From: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Date: Thu, 19 Aug 2010 16:39:43 -0700

> From: Alexander Duyck <alexander.h.duyck@intel.com>
> 
> The Tx init was spread out over ixgbe_configure, ixgbe_configure_tx, and
> ixgbe_up_complete.  This change combines all of that into the
> ixgbe_configure_tx function in order to simplify the Tx init path.
> 
> Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com>
> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

Applied.

^ permalink raw reply

* Re: [net-next-2.6 PATCH 16/20] ixgbe: move all GPIE register config into a single function
From: David Miller @ 2010-08-19 23:47 UTC (permalink / raw)
  To: jeffrey.t.kirsher; +Cc: netdev, gospo, bphilips, alexander.h.duyck
In-Reply-To: <20100819233917.10335.91882.stgit@localhost.localdomain>

From: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Date: Thu, 19 Aug 2010 16:39:20 -0700

> From: Alexander Duyck <alexander.h.duyck@intel.com>
> 
> This change moves all GPIE register configuration into a single function.
> The advantage of this is that we can avoid a number of unnecessary
> read/modify/write cycles on the register.
> 
> Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com>
> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

Applied.

^ permalink raw reply

* Re: [net-next-2.6 PATCH 15/20] ixgbe: Move virtualization config into a separate function
From: David Miller @ 2010-08-19 23:47 UTC (permalink / raw)
  To: jeffrey.t.kirsher; +Cc: netdev, gospo, bphilips, alexander.h.duyck
In-Reply-To: <20100819233854.10335.15538.stgit@localhost.localdomain>

From: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Date: Thu, 19 Aug 2010 16:38:57 -0700

> From: Alexander Duyck <alexander.h.duyck@intel.com>
> 
> This change moves the configuration that was done in configure_rx into a
> separate virtualization configuration function.
> 
> Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com>
> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

Applied.

^ permalink raw reply

* Re: [net-next-2.6 PATCH 14/20] ixgbe: move all Rx DMA control register writes to one central location
From: David Miller @ 2010-08-19 23:47 UTC (permalink / raw)
  To: jeffrey.t.kirsher; +Cc: netdev, gospo, bphilips, alexander.h.duyck
In-Reply-To: <20100819233831.10335.23092.stgit@localhost.localdomain>

From: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Date: Thu, 19 Aug 2010 16:38:34 -0700

> From: Alexander Duyck <alexander.h.duyck@intel.com>
> 
> This change moves all of the Rx DMA control register writes to one central
> location.  This should help to avoid accidentally overwriting existing
> settings.
> 
> Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com>
> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

Applied.

^ permalink raw reply

* Re: [net-next-2.6 PATCH 13/20] ixgbe: Move max frame size and Rx buffer length configuration into a function
From: David Miller @ 2010-08-19 23:46 UTC (permalink / raw)
  To: jeffrey.t.kirsher; +Cc: netdev, gospo, bphilips, alexander.h.duyck
In-Reply-To: <20100819233808.10335.36102.stgit@localhost.localdomain>

From: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Date: Thu, 19 Aug 2010 16:38:11 -0700

> From: Alexander Duyck <alexander.h.duyck@intel.com>
> 
> This change consolidates all of the Rx max frame size and Rx buffer length
> configuration into a single function.
> 
> Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com>
> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

Applied.

^ permalink raw reply

* Re: [net-next-2.6 PATCH 12/20] ixgbe: remove redundant configuration of vmolr, rename generic variable
From: David Miller @ 2010-08-19 23:46 UTC (permalink / raw)
  To: jeffrey.t.kirsher; +Cc: netdev, gospo, bphilips, alexander.h.duyck
In-Reply-To: <20100819233745.10335.58155.stgit@localhost.localdomain>

From: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Date: Thu, 19 Aug 2010 16:37:48 -0700

> From: Alexander Duyck <alexander.h.duyck@intel.com>
> 
> The vmolr is configured already in ixgbe_set_rx_mode for the PF so there is
> no need to set it again in ixgbe_configure_rx.
> 
> Instead of using the variable name reg, it is easier to just rename it to
> gcr_ext to reflect the register contents that the variable holds.
> 
> Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com>
> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

Applied.

^ permalink raw reply

* Re: [net-next-2.6 PATCH 11/20] ixgbe: bump PS header size to 512 bytes
From: David Miller @ 2010-08-19 23:46 UTC (permalink / raw)
  To: jeffrey.t.kirsher; +Cc: netdev, gospo, bphilips, alexander.h.duyck
In-Reply-To: <20100819233719.10335.25526.stgit@localhost.localdomain>

From: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Date: Thu, 19 Aug 2010 16:37:21 -0700

> From: Alexander Duyck <alexander.h.duyck@intel.com>
> 
> Bump the header size for packet split to 512 bytes since this makes the
> best use of the 1k buffer that is allocated for any skb 512 bytes or
> smaller.
> 
> Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com>
> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

Applied.

^ permalink raw reply

* Re: [net-next-2.6 PATCH 10/20] ixgbe: combine accesses to FCTRL register into ixgbe_set_rx_mode
From: David Miller @ 2010-08-19 23:46 UTC (permalink / raw)
  To: jeffrey.t.kirsher; +Cc: netdev, gospo, bphilips, alexander.h.duyck
In-Reply-To: <20100819233647.10335.20152.stgit@localhost.localdomain>

From: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Date: Thu, 19 Aug 2010 16:36:49 -0700

> From: Alexander Duyck <alexander.h.duyck@intel.com>
> 
> We are accessing the FCTRL register in multiple spots in the init path and
> we can simplify things by combining the configuration all into
> ixgbe_set_rx_mode.
> 
> Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com>
> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

Applied.

^ permalink raw reply

* Re: [net-next-2.6 PATCH 09/20] ixgbe: pull PSRTYPE configuration into a separate function
From: David Miller @ 2010-08-19 23:46 UTC (permalink / raw)
  To: jeffrey.t.kirsher; +Cc: netdev, gospo, bphilips, alexander.h.duyck
In-Reply-To: <20100819233624.10335.67259.stgit@localhost.localdomain>

From: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Date: Thu, 19 Aug 2010 16:36:27 -0700

> From: Alexander Duyck <alexander.h.duyck@intel.com>
> 
> The configuration of PSRTYPE was being done conditionally on if packet
> split is enabled or not.  It can be configured always since it will not
> have any effect when packet split is not enabled.
> 
> Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com>
> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

Applied.

^ permalink raw reply

* Re: [net-next-2.6 PATCH 08/20] ixgbe: pull ring configuration into it's own function
From: David Miller @ 2010-08-19 23:46 UTC (permalink / raw)
  To: jeffrey.t.kirsher; +Cc: netdev, gospo, bphilips, alexander.h.duyck
In-Reply-To: <20100819233602.10335.80879.stgit@localhost.localdomain>

From: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Date: Thu, 19 Aug 2010 16:36:05 -0700

> From: Alexander Duyck <alexander.h.duyck@intel.com>
> 
> These changes add ixgbe_configure_rx_ring which is used to setup the base
> function pointers for the ring.
> 
> Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com>
> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

Applied.

^ permalink raw reply

* Re: [net-next-2.6 PATCH 07/20] ixgbe: consolidate all setting of MRQC into one function
From: David Miller @ 2010-08-19 23:46 UTC (permalink / raw)
  To: jeffrey.t.kirsher; +Cc: netdev, gospo, bphilips, alexander.h.duyck
In-Reply-To: <20100819233539.10335.93283.stgit@localhost.localdomain>

From: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Date: Thu, 19 Aug 2010 16:35:41 -0700

> From: Alexander Duyck <alexander.h.duyck@intel.com>
> 
> This change simplifies the configuration of MRQC by consolidating the
> setting of it into one function. As such the register is no longer set in
> multiple places which should make any future changes easier to work with.
> In addition we can combine RSS related register writes into the call since
> enabling all of those bits without enabling RSS itself in MRQC should have
> no effect.
> 
> Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com>
> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

Applied.

^ permalink raw reply

* Re: [net-next-2.6 PATCH 06/20] ixgbe: move Tx ring configuration into a separate function
From: David Miller @ 2010-08-19 23:46 UTC (permalink / raw)
  To: jeffrey.t.kirsher; +Cc: netdev, gospo, bphilips, alexander.h.duyck
In-Reply-To: <20100819233509.10335.23922.stgit@localhost.localdomain>

From: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Date: Thu, 19 Aug 2010 16:35:12 -0700

> From: Alexander Duyck <alexander.h.duyck@intel.com>
> 
> This patch moves the Tx ring configuration into a separate function.  In
> addition the function drops the setting of the head writeback RO bit since
> head writeback is no longer used within ixgbe.
> 
> Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com>
> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

Applied.

^ permalink raw reply

* Re: [net-next-2.6 PATCH 05/20] ixgbe: move configuration of the MTQC register into it's own function
From: David Miller @ 2010-08-19 23:46 UTC (permalink / raw)
  To: jeffrey.t.kirsher; +Cc: netdev, gospo, bphilips, alexander.h.duyck
In-Reply-To: <20100819233447.10335.56588.stgit@localhost.localdomain>

From: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Date: Thu, 19 Aug 2010 16:34:50 -0700

> From: Alexander Duyck <alexander.h.duyck@intel.com>
> 
> This patch moves the configuration of the MTQC register into it's own
> function call similar to ixgbe_setup_mrqc.
> 
> Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com>
> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

Applied.

^ permalink raw reply

* Re: [net-next-2.6 PATCH 04/20] ixgbe: combine two modifications of TXDCTL into one
From: David Miller @ 2010-08-19 23:46 UTC (permalink / raw)
  To: jeffrey.t.kirsher; +Cc: netdev, gospo, bphilips, alexander.h.duyck
In-Reply-To: <20100819233425.10335.67211.stgit@localhost.localdomain>

From: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Date: Thu, 19 Aug 2010 16:34:27 -0700

> From: Alexander Duyck <alexander.h.duyck@intel.com>
> 
> In ixgbe_up_complete we were doing a read-modify-write of TXDCTL followed
> by another one just a few lines further down.  Instead of performing two
> separate read-modify-writes it would make more sense to combine the two
> into one.
> 
> Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com>
> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

Applied.

^ permalink raw reply

* Re: [net-next-2.6 PATCH 03/20] ixgbe: move setting of GSO size for 82598 into ixgbe_configure_dcb
From: David Miller @ 2010-08-19 23:46 UTC (permalink / raw)
  To: jeffrey.t.kirsher; +Cc: netdev, gospo, bphilips, alexander.h.duyck
In-Reply-To: <20100819233402.10335.66985.stgit@localhost.localdomain>

From: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Date: Thu, 19 Aug 2010 16:34:04 -0700

> From: Alexander Duyck <alexander.h.duyck@intel.com>
> 
> We are unnecessarily modifying the GSO size for all HW when we don't need
> to.  The code can be simplified by moving the check for DCB and the
> adjustment of the GSO size for 82598 into ixgbe_configure_dcb.
> 
> Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com>
> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

Applied.

^ permalink raw reply

* Re: [net-next-2.6 PATCH 02/20] ixgbe: remove redundant DMA alignment code
From: David Miller @ 2010-08-19 23:45 UTC (permalink / raw)
  To: jeffrey.t.kirsher; +Cc: netdev, gospo, bphilips, alexander.h.duyck
In-Reply-To: <20100819233339.10335.17698.stgit@localhost.localdomain>

From: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Date: Thu, 19 Aug 2010 16:33:41 -0700

> From: Alexander Duyck <alexander.h.duyck@intel.com>
> 
> This patch removes the redundant DMA alignment code from the Rx buffer
> allocation path.  This code is no longer necessary since all x86 buffers
> are now DMA aligned due to recent changes to NET_IP_ALIGN and NET_SKB_PAD.
> 
> It also moves the setting of the Rx queue value into the allocation path
> since it is more likely that the queue mapping will still be in the cache
> at the time of allocation.
> Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com>
> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

Applied.

^ permalink raw reply

* Re: [net-next-2.6 PATCH 01/20] ixgbe: cleanup ixgbe_get_drvinfo to be extra careful with buffer boundaries
From: David Miller @ 2010-08-19 23:45 UTC (permalink / raw)
  To: jeffrey.t.kirsher; +Cc: netdev, gospo, bphilips, donald.c.skidmore
In-Reply-To: <20100819233052.10335.13176.stgit@localhost.localdomain>

From: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Date: Thu, 19 Aug 2010 16:33:16 -0700

> From: Don Skidmore <donald.c.skidmore@intel.com>
> 
> Make the code in ixgbe_get_drvinfo more protected against buffer overflow
> boundaries by using snprintf and sizeof with snprintf.
> 
> Signed-off-by: Don Skidmore <donald.c.skidmore@intel.com>
> Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>

Applied.

^ permalink raw reply

* [net-next-2.6 PATCH 20/20] ixgbe: rewrite ethtool test to use standard config functions
From: Jeff Kirsher @ 2010-08-19 23:40 UTC (permalink / raw)
  To: davem; +Cc: netdev, gospo, bphilips, Alexander Duyck, Jeff Kirsher
In-Reply-To: <20100819233052.10335.13176.stgit@localhost.localdomain>

From: Alexander Duyck <alexander.h.duyck@intel.com>

This change makes it so that the ethtool loopback test uses the standard
ring configuration and allocation functions.  As a result the loopback test
will be much more effective at testing core driver functionality.

Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---

 drivers/net/ixgbe/ixgbe.h         |   11 +
 drivers/net/ixgbe/ixgbe_ethtool.c |  362 +++++++++++++------------------------
 drivers/net/ixgbe/ixgbe_main.c    |   49 +++--
 3 files changed, 169 insertions(+), 253 deletions(-)

diff --git a/drivers/net/ixgbe/ixgbe.h b/drivers/net/ixgbe/ixgbe.h
index ac4b90e..5cebc37 100644
--- a/drivers/net/ixgbe/ixgbe.h
+++ b/drivers/net/ixgbe/ixgbe.h
@@ -453,9 +453,20 @@ extern int ixgbe_setup_rx_resources(struct ixgbe_adapter *, struct ixgbe_ring *)
 extern int ixgbe_setup_tx_resources(struct ixgbe_adapter *, struct ixgbe_ring *);
 extern void ixgbe_free_rx_resources(struct ixgbe_adapter *, struct ixgbe_ring *);
 extern void ixgbe_free_tx_resources(struct ixgbe_adapter *, struct ixgbe_ring *);
+extern void ixgbe_configure_rx_ring(struct ixgbe_adapter *,struct ixgbe_ring *);
+extern void ixgbe_configure_tx_ring(struct ixgbe_adapter *,struct ixgbe_ring *);
 extern void ixgbe_update_stats(struct ixgbe_adapter *adapter);
 extern int ixgbe_init_interrupt_scheme(struct ixgbe_adapter *adapter);
 extern void ixgbe_clear_interrupt_scheme(struct ixgbe_adapter *adapter);
+extern netdev_tx_t ixgbe_xmit_frame_ring(struct sk_buff *,
+					 struct net_device *,
+					 struct ixgbe_adapter *,
+					 struct ixgbe_ring *);
+extern void ixgbe_unmap_and_free_tx_resource(struct ixgbe_adapter *,
+                                             struct ixgbe_tx_buffer *);
+extern void ixgbe_alloc_rx_buffers(struct ixgbe_adapter *adapter,
+                                   struct ixgbe_ring *rx_ring,
+                                   int cleaned_count);
 extern void ixgbe_write_eitr(struct ixgbe_q_vector *);
 extern int ethtool_ioctl(struct ifreq *ifr);
 extern s32 ixgbe_reinit_fdir_tables_82599(struct ixgbe_hw *hw);
diff --git a/drivers/net/ixgbe/ixgbe_ethtool.c b/drivers/net/ixgbe/ixgbe_ethtool.c
index 4d74f4b..25ef8b1 100644
--- a/drivers/net/ixgbe/ixgbe_ethtool.c
+++ b/drivers/net/ixgbe/ixgbe_ethtool.c
@@ -1438,9 +1438,7 @@ static void ixgbe_free_desc_rings(struct ixgbe_adapter *adapter)
 	struct ixgbe_ring *tx_ring = &adapter->test_tx_ring;
 	struct ixgbe_ring *rx_ring = &adapter->test_rx_ring;
 	struct ixgbe_hw *hw = &adapter->hw;
-	struct pci_dev *pdev = adapter->pdev;
 	u32 reg_ctl;
-	int i;
 
 	/* shut down the DMA engines now so they can be reinitialized later */
 
@@ -1448,14 +1446,15 @@ static void ixgbe_free_desc_rings(struct ixgbe_adapter *adapter)
 	reg_ctl = IXGBE_READ_REG(hw, IXGBE_RXCTRL);
 	reg_ctl &= ~IXGBE_RXCTRL_RXEN;
 	IXGBE_WRITE_REG(hw, IXGBE_RXCTRL, reg_ctl);
-	reg_ctl = IXGBE_READ_REG(hw, IXGBE_RXDCTL(0));
+	reg_ctl = IXGBE_READ_REG(hw, IXGBE_RXDCTL(rx_ring->reg_idx));
 	reg_ctl &= ~IXGBE_RXDCTL_ENABLE;
-	IXGBE_WRITE_REG(hw, IXGBE_RXDCTL(0), reg_ctl);
+	IXGBE_WRITE_REG(hw, IXGBE_RXDCTL(rx_ring->reg_idx), reg_ctl);
 
 	/* now Tx */
-	reg_ctl = IXGBE_READ_REG(hw, IXGBE_TXDCTL(0));
+	reg_ctl = IXGBE_READ_REG(hw, IXGBE_TXDCTL(tx_ring->reg_idx));
 	reg_ctl &= ~IXGBE_TXDCTL_ENABLE;
-	IXGBE_WRITE_REG(hw, IXGBE_TXDCTL(0), reg_ctl);
+	IXGBE_WRITE_REG(hw, IXGBE_TXDCTL(tx_ring->reg_idx), reg_ctl);
+
 	if (hw->mac.type == ixgbe_mac_82599EB) {
 		reg_ctl = IXGBE_READ_REG(hw, IXGBE_DMATXCTL);
 		reg_ctl &= ~IXGBE_DMATXCTL_TE;
@@ -1464,221 +1463,57 @@ static void ixgbe_free_desc_rings(struct ixgbe_adapter *adapter)
 
 	ixgbe_reset(adapter);
 
-	if (tx_ring->desc && tx_ring->tx_buffer_info) {
-		for (i = 0; i < tx_ring->count; i++) {
-			struct ixgbe_tx_buffer *buf =
-					&(tx_ring->tx_buffer_info[i]);
-			if (buf->dma)
-				dma_unmap_single(&pdev->dev, buf->dma,
-						 buf->length, DMA_TO_DEVICE);
-			if (buf->skb)
-				dev_kfree_skb(buf->skb);
-		}
-	}
-
-	if (rx_ring->desc && rx_ring->rx_buffer_info) {
-		for (i = 0; i < rx_ring->count; i++) {
-			struct ixgbe_rx_buffer *buf =
-					&(rx_ring->rx_buffer_info[i]);
-			if (buf->dma)
-				dma_unmap_single(&pdev->dev, buf->dma,
-						 IXGBE_RXBUFFER_2048,
-						 DMA_FROM_DEVICE);
-			if (buf->skb)
-				dev_kfree_skb(buf->skb);
-		}
-	}
-
-	if (tx_ring->desc) {
-		dma_free_coherent(&pdev->dev, tx_ring->size, tx_ring->desc,
-				  tx_ring->dma);
-		tx_ring->desc = NULL;
-	}
-	if (rx_ring->desc) {
-		dma_free_coherent(&pdev->dev, rx_ring->size, rx_ring->desc,
-				  rx_ring->dma);
-		rx_ring->desc = NULL;
-	}
-
-	kfree(tx_ring->tx_buffer_info);
-	tx_ring->tx_buffer_info = NULL;
-	kfree(rx_ring->rx_buffer_info);
-	rx_ring->rx_buffer_info = NULL;
+	ixgbe_free_tx_resources(adapter, &adapter->test_tx_ring);
+	ixgbe_free_rx_resources(adapter, &adapter->test_rx_ring);
 }
 
 static int ixgbe_setup_desc_rings(struct ixgbe_adapter *adapter)
 {
 	struct ixgbe_ring *tx_ring = &adapter->test_tx_ring;
 	struct ixgbe_ring *rx_ring = &adapter->test_rx_ring;
-	struct pci_dev *pdev = adapter->pdev;
 	u32 rctl, reg_data;
-	int i, ret_val;
+	int ret_val;
+	int err;
 
 	/* Setup Tx descriptor ring and Tx buffers */
+	tx_ring->count = IXGBE_DEFAULT_TXD;
+	tx_ring->queue_index = 0;
+	tx_ring->reg_idx = adapter->tx_ring[0]->reg_idx;
+	tx_ring->numa_node = adapter->node;
 
-	if (!tx_ring->count)
-		tx_ring->count = IXGBE_DEFAULT_TXD;
-
-	tx_ring->tx_buffer_info = kcalloc(tx_ring->count,
-	                                  sizeof(struct ixgbe_tx_buffer),
-	                                  GFP_KERNEL);
-	if (!(tx_ring->tx_buffer_info)) {
-		ret_val = 1;
-		goto err_nomem;
-	}
-
-	tx_ring->size = tx_ring->count * sizeof(union ixgbe_adv_tx_desc);
-	tx_ring->size = ALIGN(tx_ring->size, 4096);
-	tx_ring->desc = dma_alloc_coherent(&pdev->dev, tx_ring->size,
-					   &tx_ring->dma, GFP_KERNEL);
-	if (!(tx_ring->desc)) {
-		ret_val = 2;
-		goto err_nomem;
-	}
-	tx_ring->next_to_use = tx_ring->next_to_clean = 0;
-
-	IXGBE_WRITE_REG(&adapter->hw, IXGBE_TDBAL(0),
-			((u64) tx_ring->dma & 0x00000000FFFFFFFF));
-	IXGBE_WRITE_REG(&adapter->hw, IXGBE_TDBAH(0),
-			((u64) tx_ring->dma >> 32));
-	IXGBE_WRITE_REG(&adapter->hw, IXGBE_TDLEN(0),
-			tx_ring->count * sizeof(union ixgbe_adv_tx_desc));
-	IXGBE_WRITE_REG(&adapter->hw, IXGBE_TDH(0), 0);
-	IXGBE_WRITE_REG(&adapter->hw, IXGBE_TDT(0), 0);
-
-	reg_data = IXGBE_READ_REG(&adapter->hw, IXGBE_HLREG0);
-	reg_data |= IXGBE_HLREG0_TXPADEN;
-	IXGBE_WRITE_REG(&adapter->hw, IXGBE_HLREG0, reg_data);
+	err = ixgbe_setup_tx_resources(adapter, tx_ring);
+	if (err)
+		return 1;
 
 	if (adapter->hw.mac.type == ixgbe_mac_82599EB) {
 		reg_data = IXGBE_READ_REG(&adapter->hw, IXGBE_DMATXCTL);
 		reg_data |= IXGBE_DMATXCTL_TE;
 		IXGBE_WRITE_REG(&adapter->hw, IXGBE_DMATXCTL, reg_data);
 	}
-	reg_data = IXGBE_READ_REG(&adapter->hw, IXGBE_TXDCTL(0));
-	reg_data |= IXGBE_TXDCTL_ENABLE;
-	IXGBE_WRITE_REG(&adapter->hw, IXGBE_TXDCTL(0), reg_data);
-
-	for (i = 0; i < tx_ring->count; i++) {
-		union ixgbe_adv_tx_desc *desc = IXGBE_TX_DESC_ADV(tx_ring, i);
-		struct sk_buff *skb;
-		unsigned int size = 1024;
-
-		skb = alloc_skb(size, GFP_KERNEL);
-		if (!skb) {
-			ret_val = 3;
-			goto err_nomem;
-		}
-		skb_put(skb, size);
-		tx_ring->tx_buffer_info[i].skb = skb;
-		tx_ring->tx_buffer_info[i].length = skb->len;
-		tx_ring->tx_buffer_info[i].dma =
-			dma_map_single(&pdev->dev, skb->data, skb->len,
-				       DMA_TO_DEVICE);
-		desc->read.buffer_addr =
-		                    cpu_to_le64(tx_ring->tx_buffer_info[i].dma);
-		desc->read.cmd_type_len = cpu_to_le32(skb->len);
-		desc->read.cmd_type_len |= cpu_to_le32(IXGBE_TXD_CMD_EOP |
-		                                       IXGBE_TXD_CMD_IFCS |
-		                                       IXGBE_TXD_CMD_RS);
-		desc->read.olinfo_status = 0;
-		if (adapter->hw.mac.type == ixgbe_mac_82599EB)
-			desc->read.olinfo_status |=
-			                (skb->len << IXGBE_ADVTXD_PAYLEN_SHIFT);
 
-	}
+	ixgbe_configure_tx_ring(adapter, tx_ring);
 
 	/* Setup Rx Descriptor ring and Rx buffers */
-
-	if (!rx_ring->count)
-		rx_ring->count = IXGBE_DEFAULT_RXD;
-
-	rx_ring->rx_buffer_info = kcalloc(rx_ring->count,
-	                                  sizeof(struct ixgbe_rx_buffer),
-	                                  GFP_KERNEL);
-	if (!(rx_ring->rx_buffer_info)) {
+	rx_ring->count = IXGBE_DEFAULT_RXD;
+	rx_ring->queue_index = 0;
+	rx_ring->reg_idx = adapter->rx_ring[0]->reg_idx;
+	rx_ring->rx_buf_len = IXGBE_RXBUFFER_2048;
+	rx_ring->numa_node = adapter->node;
+
+	err = ixgbe_setup_rx_resources(adapter, rx_ring);
+	if (err) {
 		ret_val = 4;
 		goto err_nomem;
 	}
 
-	rx_ring->size = rx_ring->count * sizeof(union ixgbe_adv_rx_desc);
-	rx_ring->size = ALIGN(rx_ring->size, 4096);
-	rx_ring->desc = dma_alloc_coherent(&pdev->dev, rx_ring->size,
-					   &rx_ring->dma, GFP_KERNEL);
-	if (!(rx_ring->desc)) {
-		ret_val = 5;
-		goto err_nomem;
-	}
-	rx_ring->next_to_use = rx_ring->next_to_clean = 0;
-
 	rctl = IXGBE_READ_REG(&adapter->hw, IXGBE_RXCTRL);
 	IXGBE_WRITE_REG(&adapter->hw, IXGBE_RXCTRL, rctl & ~IXGBE_RXCTRL_RXEN);
-	IXGBE_WRITE_REG(&adapter->hw, IXGBE_RDBAL(0),
-			((u64)rx_ring->dma & 0xFFFFFFFF));
-	IXGBE_WRITE_REG(&adapter->hw, IXGBE_RDBAH(0),
-			((u64) rx_ring->dma >> 32));
-	IXGBE_WRITE_REG(&adapter->hw, IXGBE_RDLEN(0), rx_ring->size);
-	IXGBE_WRITE_REG(&adapter->hw, IXGBE_RDH(0), 0);
-	IXGBE_WRITE_REG(&adapter->hw, IXGBE_RDT(0), 0);
 
-	reg_data = IXGBE_READ_REG(&adapter->hw, IXGBE_FCTRL);
-	reg_data |= IXGBE_FCTRL_BAM | IXGBE_FCTRL_SBP | IXGBE_FCTRL_MPE;
-	IXGBE_WRITE_REG(&adapter->hw, IXGBE_FCTRL, reg_data);
-
-	reg_data = IXGBE_READ_REG(&adapter->hw, IXGBE_HLREG0);
-	reg_data &= ~IXGBE_HLREG0_LPBK;
-	IXGBE_WRITE_REG(&adapter->hw, IXGBE_HLREG0, reg_data);
-
-	reg_data = IXGBE_READ_REG(&adapter->hw, IXGBE_RDRXCTL);
-#define IXGBE_RDRXCTL_RDMTS_MASK    0x00000003 /* Receive Descriptor Minimum
-                                                  Threshold Size mask */
-	reg_data &= ~IXGBE_RDRXCTL_RDMTS_MASK;
-	IXGBE_WRITE_REG(&adapter->hw, IXGBE_RDRXCTL, reg_data);
-
-	reg_data = IXGBE_READ_REG(&adapter->hw, IXGBE_MCSTCTRL);
-#define IXGBE_MCSTCTRL_MO_MASK      0x00000003 /* Multicast Offset mask */
-	reg_data &= ~IXGBE_MCSTCTRL_MO_MASK;
-	reg_data |= adapter->hw.mac.mc_filter_type;
-	IXGBE_WRITE_REG(&adapter->hw, IXGBE_MCSTCTRL, reg_data);
-
-	reg_data = IXGBE_READ_REG(&adapter->hw, IXGBE_RXDCTL(0));
-	reg_data |= IXGBE_RXDCTL_ENABLE;
-	IXGBE_WRITE_REG(&adapter->hw, IXGBE_RXDCTL(0), reg_data);
-	if (adapter->hw.mac.type == ixgbe_mac_82599EB) {
-		int j = adapter->rx_ring[0]->reg_idx;
-		u32 k;
-		for (k = 0; k < 10; k++) {
-			if (IXGBE_READ_REG(&adapter->hw,
-			                   IXGBE_RXDCTL(j)) & IXGBE_RXDCTL_ENABLE)
-				break;
-			else
-				msleep(1);
-		}
-	}
+	ixgbe_configure_rx_ring(adapter, rx_ring);
 
 	rctl |= IXGBE_RXCTRL_RXEN | IXGBE_RXCTRL_DMBYPS;
 	IXGBE_WRITE_REG(&adapter->hw, IXGBE_RXCTRL, rctl);
 
-	for (i = 0; i < rx_ring->count; i++) {
-		union ixgbe_adv_rx_desc *rx_desc =
-		                                 IXGBE_RX_DESC_ADV(rx_ring, i);
-		struct sk_buff *skb;
-
-		skb = alloc_skb(IXGBE_RXBUFFER_2048 + NET_IP_ALIGN, GFP_KERNEL);
-		if (!skb) {
-			ret_val = 6;
-			goto err_nomem;
-		}
-		skb_reserve(skb, NET_IP_ALIGN);
-		rx_ring->rx_buffer_info[i].skb = skb;
-		rx_ring->rx_buffer_info[i].dma =
-			dma_map_single(&pdev->dev, skb->data,
-				       IXGBE_RXBUFFER_2048, DMA_FROM_DEVICE);
-		rx_desc->read.pkt_addr =
-				cpu_to_le64(rx_ring->rx_buffer_info[i].dma);
-		memset(skb->data, 0x00, skb->len);
-	}
-
 	return 0;
 
 err_nomem:
@@ -1692,16 +1527,21 @@ static int ixgbe_setup_loopback_test(struct ixgbe_adapter *adapter)
 	u32 reg_data;
 
 	/* right now we only support MAC loopback in the driver */
-
-	/* Setup MAC loopback */
 	reg_data = IXGBE_READ_REG(&adapter->hw, IXGBE_HLREG0);
+	/* Setup MAC loopback */
 	reg_data |= IXGBE_HLREG0_LPBK;
 	IXGBE_WRITE_REG(&adapter->hw, IXGBE_HLREG0, reg_data);
 
+	reg_data = IXGBE_READ_REG(&adapter->hw, IXGBE_FCTRL);
+	reg_data |= IXGBE_FCTRL_BAM | IXGBE_FCTRL_SBP | IXGBE_FCTRL_MPE;
+	IXGBE_WRITE_REG(&adapter->hw, IXGBE_FCTRL, reg_data);
+
 	reg_data = IXGBE_READ_REG(&adapter->hw, IXGBE_AUTOC);
 	reg_data &= ~IXGBE_AUTOC_LMS_MASK;
 	reg_data |= IXGBE_AUTOC_LMS_10G_LINK_NO_AN | IXGBE_AUTOC_FLU;
 	IXGBE_WRITE_REG(&adapter->hw, IXGBE_AUTOC, reg_data);
+	IXGBE_WRITE_FLUSH(&adapter->hw);
+	msleep(10);
 
 	/* Disable Atlas Tx lanes; re-enabled in reset path */
 	if (hw->mac.type == ixgbe_mac_82598EB) {
@@ -1759,15 +1599,81 @@ static int ixgbe_check_lbtest_frame(struct sk_buff *skb,
 	return 13;
 }
 
+static u16 ixgbe_clean_test_rings(struct ixgbe_adapter *adapter,
+                                  struct ixgbe_ring *rx_ring,
+                                  struct ixgbe_ring *tx_ring,
+                                  unsigned int size)
+{
+	union ixgbe_adv_rx_desc *rx_desc;
+	struct ixgbe_rx_buffer *rx_buffer_info;
+	struct ixgbe_tx_buffer *tx_buffer_info;
+	const int bufsz = rx_ring->rx_buf_len;
+	u32 staterr;
+	u16 rx_ntc, tx_ntc, count = 0;
+
+	/* initialize next to clean and descriptor values */
+	rx_ntc = rx_ring->next_to_clean;
+	tx_ntc = tx_ring->next_to_clean;
+	rx_desc = IXGBE_RX_DESC_ADV(rx_ring, rx_ntc);
+	staterr = le32_to_cpu(rx_desc->wb.upper.status_error);
+
+	while (staterr & IXGBE_RXD_STAT_DD) {
+		/* check Rx buffer */
+		rx_buffer_info = &rx_ring->rx_buffer_info[rx_ntc];
+
+		/* unmap Rx buffer, will be remapped by alloc_rx_buffers */
+		dma_unmap_single(&adapter->pdev->dev,
+		                 rx_buffer_info->dma,
+				 bufsz,
+				 DMA_FROM_DEVICE);
+		rx_buffer_info->dma = 0;
+
+		/* verify contents of skb */
+		if (!ixgbe_check_lbtest_frame(rx_buffer_info->skb, size))
+			count++;
+
+		/* unmap buffer on Tx side */
+		tx_buffer_info = &tx_ring->tx_buffer_info[tx_ntc];
+		ixgbe_unmap_and_free_tx_resource(adapter, tx_buffer_info);
+
+		/* increment Rx/Tx next to clean counters */
+		rx_ntc++;
+		if (rx_ntc == rx_ring->count)
+			rx_ntc = 0;
+		tx_ntc++;
+		if (tx_ntc == tx_ring->count)
+			tx_ntc = 0;
+
+		/* fetch next descriptor */
+		rx_desc = IXGBE_RX_DESC_ADV(rx_ring, rx_ntc);
+		staterr = le32_to_cpu(rx_desc->wb.upper.status_error);
+	}
+
+	/* re-map buffers to ring, store next to clean values */
+	ixgbe_alloc_rx_buffers(adapter, rx_ring, count);
+	rx_ring->next_to_clean = rx_ntc;
+	tx_ring->next_to_clean = tx_ntc;
+
+	return count;
+}
+
 static int ixgbe_run_loopback_test(struct ixgbe_adapter *adapter)
 {
 	struct ixgbe_ring *tx_ring = &adapter->test_tx_ring;
 	struct ixgbe_ring *rx_ring = &adapter->test_rx_ring;
-	struct pci_dev *pdev = adapter->pdev;
-	int i, j, k, l, lc, good_cnt, ret_val = 0;
-	unsigned long time;
+	int i, j, lc, good_cnt, ret_val = 0;
+	unsigned int size = 1024;
+	netdev_tx_t tx_ret_val;
+	struct sk_buff *skb;
+
+	/* allocate test skb */
+	skb = alloc_skb(size, GFP_KERNEL);
+	if (!skb)
+		return 11;
 
-	IXGBE_WRITE_REG(&adapter->hw, IXGBE_RDT(0), rx_ring->count - 1);
+	/* place data into test skb */
+	ixgbe_create_lbtest_frame(skb, size);
+	skb_put(skb, size);
 
 	/*
 	 * Calculate the loop count based on the largest descriptor ring
@@ -1780,54 +1686,40 @@ static int ixgbe_run_loopback_test(struct ixgbe_adapter *adapter)
 	else
 		lc = ((rx_ring->count / 64) * 2) + 1;
 
-	k = l = 0;
 	for (j = 0; j <= lc; j++) {
-		for (i = 0; i < 64; i++) {
-			ixgbe_create_lbtest_frame(
-					tx_ring->tx_buffer_info[k].skb,
-					1024);
-			dma_sync_single_for_device(&pdev->dev,
-				tx_ring->tx_buffer_info[k].dma,
-				tx_ring->tx_buffer_info[k].length,
-				DMA_TO_DEVICE);
-			if (unlikely(++k == tx_ring->count))
-				k = 0;
-		}
-		IXGBE_WRITE_REG(&adapter->hw, IXGBE_TDT(0), k);
-		msleep(200);
-		/* set the start time for the receive */
-		time = jiffies;
+		/* reset count of good packets */
 		good_cnt = 0;
-		do {
-			/* receive the sent packets */
-			dma_sync_single_for_cpu(&pdev->dev,
-					rx_ring->rx_buffer_info[l].dma,
-					IXGBE_RXBUFFER_2048,
-					DMA_FROM_DEVICE);
-			ret_val = ixgbe_check_lbtest_frame(
-					rx_ring->rx_buffer_info[l].skb, 1024);
-			if (!ret_val)
+
+		/* place 64 packets on the transmit queue*/
+		for (i = 0; i < 64; i++) {
+			skb_get(skb);
+			tx_ret_val = ixgbe_xmit_frame_ring(skb,
+							   adapter->netdev,
+							   adapter,
+							   tx_ring);
+			if (tx_ret_val == NETDEV_TX_OK)
 				good_cnt++;
-			if (++l == rx_ring->count)
-				l = 0;
-			/*
-			 * time + 20 msecs (200 msecs on 2.4) is more than
-			 * enough time to complete the receives, if it's
-			 * exceeded, break and error off
-			 */
-		} while (good_cnt < 64 && jiffies < (time + 20));
+		}
+
 		if (good_cnt != 64) {
-			/* ret_val is the same as mis-compare */
-			ret_val = 13;
+			ret_val = 12;
 			break;
 		}
-		if (jiffies >= (time + 20)) {
-			/* Error code for time out error */
-			ret_val = 14;
+
+		/* allow 200 milliseconds for packets to go from Tx to Rx */
+		msleep(200);
+
+		good_cnt = ixgbe_clean_test_rings(adapter, rx_ring,
+						  tx_ring, size);
+		if (good_cnt != 64) {
+			ret_val = 13;
 			break;
 		}
 	}
 
+	/* free the original skb */
+	kfree_skb(skb);
+
 	return ret_val;
 }
 
diff --git a/drivers/net/ixgbe/ixgbe_main.c b/drivers/net/ixgbe/ixgbe_main.c
index 85ecf0e..18163e3 100644
--- a/drivers/net/ixgbe/ixgbe_main.c
+++ b/drivers/net/ixgbe/ixgbe_main.c
@@ -601,9 +601,9 @@ static inline void ixgbe_irq_rearm_queues(struct ixgbe_adapter *adapter,
 	}
 }
 
-static void ixgbe_unmap_and_free_tx_resource(struct ixgbe_adapter *adapter,
-                                             struct ixgbe_tx_buffer
-                                             *tx_buffer_info)
+void ixgbe_unmap_and_free_tx_resource(struct ixgbe_adapter *adapter,
+                                      struct ixgbe_tx_buffer
+                                      *tx_buffer_info)
 {
 	if (tx_buffer_info->dma) {
 		if (tx_buffer_info->mapped_as_page)
@@ -1032,9 +1032,9 @@ static inline void ixgbe_release_rx_desc(struct ixgbe_hw *hw,
  * ixgbe_alloc_rx_buffers - Replace used receive buffers; packet split
  * @adapter: address of board private structure
  **/
-static void ixgbe_alloc_rx_buffers(struct ixgbe_adapter *adapter,
-                                   struct ixgbe_ring *rx_ring,
-                                   int cleaned_count)
+void ixgbe_alloc_rx_buffers(struct ixgbe_adapter *adapter,
+                            struct ixgbe_ring *rx_ring,
+                            int cleaned_count)
 {
 	struct net_device *netdev = adapter->netdev;
 	struct pci_dev *pdev = adapter->pdev;
@@ -1095,6 +1095,7 @@ static void ixgbe_alloc_rx_buffers(struct ixgbe_adapter *adapter,
 			rx_desc->read.hdr_addr = cpu_to_le64(bi->dma);
 		} else {
 			rx_desc->read.pkt_addr = cpu_to_le64(bi->dma);
+			rx_desc->read.hdr_addr = 0;
 		}
 
 		i++;
@@ -2431,8 +2432,8 @@ static void ixgbe_configure_msi_and_legacy(struct ixgbe_adapter *adapter)
  *
  * Configure the Tx descriptor ring after a reset.
  **/
- static void ixgbe_configure_tx_ring(struct ixgbe_adapter *adapter,
-				     struct ixgbe_ring *ring)
+void ixgbe_configure_tx_ring(struct ixgbe_adapter *adapter,
+			     struct ixgbe_ring *ring)
 {
 	struct ixgbe_hw *hw = &adapter->hw;
 	u64 tdba = ring->dma;
@@ -2759,8 +2760,8 @@ static void ixgbe_rx_desc_queue_enable(struct ixgbe_adapter *adapter,
 	}
 }
 
-static void ixgbe_configure_rx_ring(struct ixgbe_adapter *adapter,
-				    struct ixgbe_ring *ring)
+void ixgbe_configure_rx_ring(struct ixgbe_adapter *adapter,
+			     struct ixgbe_ring *ring)
 {
 	struct ixgbe_hw *hw = &adapter->hw;
 	u64 rdba = ring->dma;
@@ -3671,8 +3672,11 @@ static void ixgbe_clean_rx_ring(struct ixgbe_adapter *adapter,
 	unsigned long size;
 	unsigned int i;
 
-	/* Free all the Rx ring sk_buffs */
+	/* ring already cleared, nothing to do */
+	if (!rx_ring->rx_buffer_info)
+		return;
 
+	/* Free all the Rx ring sk_buffs */
 	for (i = 0; i < rx_ring->count; i++) {
 		struct ixgbe_rx_buffer *rx_buffer_info;
 
@@ -3739,8 +3743,11 @@ static void ixgbe_clean_tx_ring(struct ixgbe_adapter *adapter,
 	unsigned long size;
 	unsigned int i;
 
-	/* Free all the Tx ring sk_buffs */
+	/* ring already cleared, nothing to do */
+	if (!tx_ring->tx_buffer_info)
+		return;
 
+	/* Free all the Tx ring sk_buffs */
 	for (i = 0; i < tx_ring->count; i++) {
 		tx_buffer_info = &tx_ring->tx_buffer_info[i];
 		ixgbe_unmap_and_free_tx_resource(adapter, tx_buffer_info);
@@ -6239,11 +6246,10 @@ static u16 ixgbe_select_queue(struct net_device *dev, struct sk_buff *skb)
 	return skb_tx_hash(dev, skb);
 }
 
-static netdev_tx_t ixgbe_xmit_frame(struct sk_buff *skb,
-				    struct net_device *netdev)
+netdev_tx_t ixgbe_xmit_frame_ring(struct sk_buff *skb, struct net_device *netdev,
+			  struct ixgbe_adapter *adapter,
+			  struct ixgbe_ring *tx_ring)
 {
-	struct ixgbe_adapter *adapter = netdev_priv(netdev);
-	struct ixgbe_ring *tx_ring;
 	struct netdev_queue *txq;
 	unsigned int first;
 	unsigned int tx_flags = 0;
@@ -6267,8 +6273,6 @@ static netdev_tx_t ixgbe_xmit_frame(struct sk_buff *skb,
 		tx_flags |= IXGBE_TX_FLAGS_VLAN;
 	}
 
-	tx_ring = adapter->tx_ring[skb->queue_mapping];
-
 #ifdef IXGBE_FCOE
 	/* for FCoE with DCB, we force the priority to what
 	 * was specified by the switch */
@@ -6362,6 +6366,15 @@ static netdev_tx_t ixgbe_xmit_frame(struct sk_buff *skb,
 	return NETDEV_TX_OK;
 }
 
+static netdev_tx_t ixgbe_xmit_frame(struct sk_buff *skb, struct net_device *netdev)
+{
+	struct ixgbe_adapter *adapter = netdev_priv(netdev);
+	struct ixgbe_ring *tx_ring;
+
+	tx_ring = adapter->tx_ring[skb->queue_mapping];
+	return ixgbe_xmit_frame_ring(skb, netdev, adapter, tx_ring);
+}
+
 /**
  * ixgbe_set_mac - Change the Ethernet Address of the NIC
  * @netdev: network interface device structure


^ permalink raw reply related

* [net-next-2.6 PATCH 19/20] ixgbe: update all DESC_ADV macros to accept a ring pointer
From: Jeff Kirsher @ 2010-08-19 23:40 UTC (permalink / raw)
  To: davem; +Cc: netdev, gospo, bphilips, Alexander Duyck, Jeff Kirsher
In-Reply-To: <20100819233052.10335.13176.stgit@localhost.localdomain>

From: Alexander Duyck <alexander.h.duyck@intel.com>

All of the DESC_ADV macros are currently needing the pointers to be
de-referenced before accessing the ring.  Instead of having to add all of
the asterisks it is easier to just update the macro to expect a pointer to
the ring.

Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---

 drivers/net/ixgbe/ixgbe.h         |    6 +++---
 drivers/net/ixgbe/ixgbe_ethtool.c |    4 ++--
 drivers/net/ixgbe/ixgbe_fcoe.c    |    2 +-
 drivers/net/ixgbe/ixgbe_main.c    |   24 ++++++++++++------------
 4 files changed, 18 insertions(+), 18 deletions(-)

diff --git a/drivers/net/ixgbe/ixgbe.h b/drivers/net/ixgbe/ixgbe.h
index 7233fef..ac4b90e 100644
--- a/drivers/net/ixgbe/ixgbe.h
+++ b/drivers/net/ixgbe/ixgbe.h
@@ -256,11 +256,11 @@ struct ixgbe_q_vector {
 	(R)->next_to_clean - (R)->next_to_use - 1)
 
 #define IXGBE_RX_DESC_ADV(R, i)	    \
-	(&(((union ixgbe_adv_rx_desc *)((R).desc))[i]))
+	(&(((union ixgbe_adv_rx_desc *)((R)->desc))[i]))
 #define IXGBE_TX_DESC_ADV(R, i)	    \
-	(&(((union ixgbe_adv_tx_desc *)((R).desc))[i]))
+	(&(((union ixgbe_adv_tx_desc *)((R)->desc))[i]))
 #define IXGBE_TX_CTXTDESC_ADV(R, i)	    \
-	(&(((struct ixgbe_adv_tx_context_desc *)((R).desc))[i]))
+	(&(((struct ixgbe_adv_tx_context_desc *)((R)->desc))[i]))
 
 #define IXGBE_MAX_JUMBO_FRAME_SIZE        16128
 #ifdef IXGBE_FCOE
diff --git a/drivers/net/ixgbe/ixgbe_ethtool.c b/drivers/net/ixgbe/ixgbe_ethtool.c
index fb6d8e8..4d74f4b 100644
--- a/drivers/net/ixgbe/ixgbe_ethtool.c
+++ b/drivers/net/ixgbe/ixgbe_ethtool.c
@@ -1560,7 +1560,7 @@ static int ixgbe_setup_desc_rings(struct ixgbe_adapter *adapter)
 	IXGBE_WRITE_REG(&adapter->hw, IXGBE_TXDCTL(0), reg_data);
 
 	for (i = 0; i < tx_ring->count; i++) {
-		union ixgbe_adv_tx_desc *desc = IXGBE_TX_DESC_ADV(*tx_ring, i);
+		union ixgbe_adv_tx_desc *desc = IXGBE_TX_DESC_ADV(tx_ring, i);
 		struct sk_buff *skb;
 		unsigned int size = 1024;
 
@@ -1661,7 +1661,7 @@ static int ixgbe_setup_desc_rings(struct ixgbe_adapter *adapter)
 
 	for (i = 0; i < rx_ring->count; i++) {
 		union ixgbe_adv_rx_desc *rx_desc =
-		                                 IXGBE_RX_DESC_ADV(*rx_ring, i);
+		                                 IXGBE_RX_DESC_ADV(rx_ring, i);
 		struct sk_buff *skb;
 
 		skb = alloc_skb(IXGBE_RXBUFFER_2048 + NET_IP_ALIGN, GFP_KERNEL);
diff --git a/drivers/net/ixgbe/ixgbe_fcoe.c b/drivers/net/ixgbe/ixgbe_fcoe.c
index 072327c..86fa07c 100644
--- a/drivers/net/ixgbe/ixgbe_fcoe.c
+++ b/drivers/net/ixgbe/ixgbe_fcoe.c
@@ -471,7 +471,7 @@ int ixgbe_fso(struct ixgbe_adapter *adapter,
 
 	/* write context desc */
 	i = tx_ring->next_to_use;
-	context_desc = IXGBE_TX_CTXTDESC_ADV(*tx_ring, i);
+	context_desc = IXGBE_TX_CTXTDESC_ADV(tx_ring, i);
 	context_desc->vlan_macip_lens	= cpu_to_le32(vlan_macip_lens);
 	context_desc->seqnum_seed	= cpu_to_le32(fcoe_sof_eof);
 	context_desc->type_tucmd_mlhl	= cpu_to_le32(type_tucmd);
diff --git a/drivers/net/ixgbe/ixgbe_main.c b/drivers/net/ixgbe/ixgbe_main.c
index c88ba13..85ecf0e 100644
--- a/drivers/net/ixgbe/ixgbe_main.c
+++ b/drivers/net/ixgbe/ixgbe_main.c
@@ -385,7 +385,7 @@ static void ixgbe_dump(struct ixgbe_adapter *adapter)
 			"leng  ntw timestamp        bi->skb\n");
 
 		for (i = 0; tx_ring->desc && (i < tx_ring->count); i++) {
-			tx_desc = IXGBE_TX_DESC_ADV(*tx_ring, i);
+			tx_desc = IXGBE_TX_DESC_ADV(tx_ring, i);
 			tx_buffer_info = &tx_ring->tx_buffer_info[i];
 			u0 = (struct my_u0 *)tx_desc;
 			printk(KERN_INFO "T [0x%03X]    %016llX %016llX %016llX"
@@ -466,7 +466,7 @@ rx_ring_summary:
 
 		for (i = 0; i < rx_ring->count; i++) {
 			rx_buffer_info = &rx_ring->rx_buffer_info[i];
-			rx_desc = IXGBE_RX_DESC_ADV(*rx_ring, i);
+			rx_desc = IXGBE_RX_DESC_ADV(rx_ring, i);
 			u0 = (struct my_u0 *)rx_desc;
 			staterr = le32_to_cpu(rx_desc->wb.upper.status_error);
 			if (staterr & IXGBE_RXD_STAT_DD) {
@@ -695,7 +695,7 @@ static inline bool ixgbe_check_tx_hang(struct ixgbe_adapter *adapter,
 	    ixgbe_tx_xon_state(adapter, tx_ring)) {
 		/* detected Tx unit hang */
 		union ixgbe_adv_tx_desc *tx_desc;
-		tx_desc = IXGBE_TX_DESC_ADV(*tx_ring, eop);
+		tx_desc = IXGBE_TX_DESC_ADV(tx_ring, eop);
 		e_err(drv, "Detected Tx Unit Hang\n"
 		      "  Tx Queue             <%d>\n"
 		      "  TDH, TDT             <%x>, <%x>\n"
@@ -743,7 +743,7 @@ static bool ixgbe_clean_tx_irq(struct ixgbe_q_vector *q_vector,
 
 	i = tx_ring->next_to_clean;
 	eop = tx_ring->tx_buffer_info[i].next_to_watch;
-	eop_desc = IXGBE_TX_DESC_ADV(*tx_ring, eop);
+	eop_desc = IXGBE_TX_DESC_ADV(tx_ring, eop);
 
 	while ((eop_desc->wb.status & cpu_to_le32(IXGBE_TXD_STAT_DD)) &&
 	       (count < tx_ring->work_limit)) {
@@ -751,7 +751,7 @@ static bool ixgbe_clean_tx_irq(struct ixgbe_q_vector *q_vector,
 		rmb(); /* read buffer_info after eop_desc */
 		for ( ; !cleaned; count++) {
 			struct sk_buff *skb;
-			tx_desc = IXGBE_TX_DESC_ADV(*tx_ring, i);
+			tx_desc = IXGBE_TX_DESC_ADV(tx_ring, i);
 			tx_buffer_info = &tx_ring->tx_buffer_info[i];
 			cleaned = (i == eop);
 			skb = tx_buffer_info->skb;
@@ -791,7 +791,7 @@ static bool ixgbe_clean_tx_irq(struct ixgbe_q_vector *q_vector,
 		}
 
 		eop = tx_ring->tx_buffer_info[i].next_to_watch;
-		eop_desc = IXGBE_TX_DESC_ADV(*tx_ring, eop);
+		eop_desc = IXGBE_TX_DESC_ADV(tx_ring, eop);
 	}
 
 	tx_ring->next_to_clean = i;
@@ -1047,7 +1047,7 @@ static void ixgbe_alloc_rx_buffers(struct ixgbe_adapter *adapter,
 	bi = &rx_ring->rx_buffer_info[i];
 
 	while (cleaned_count--) {
-		rx_desc = IXGBE_RX_DESC_ADV(*rx_ring, i);
+		rx_desc = IXGBE_RX_DESC_ADV(rx_ring, i);
 
 		if (!bi->page_dma &&
 		    (rx_ring->flags & IXGBE_RING_RX_PS_ENABLED)) {
@@ -1188,7 +1188,7 @@ static bool ixgbe_clean_rx_irq(struct ixgbe_q_vector *q_vector,
 #endif /* IXGBE_FCOE */
 
 	i = rx_ring->next_to_clean;
-	rx_desc = IXGBE_RX_DESC_ADV(*rx_ring, i);
+	rx_desc = IXGBE_RX_DESC_ADV(rx_ring, i);
 	staterr = le32_to_cpu(rx_desc->wb.upper.status_error);
 	rx_buffer_info = &rx_ring->rx_buffer_info[i];
 
@@ -1263,7 +1263,7 @@ static bool ixgbe_clean_rx_irq(struct ixgbe_q_vector *q_vector,
 		if (i == rx_ring->count)
 			i = 0;
 
-		next_rxd = IXGBE_RX_DESC_ADV(*rx_ring, i);
+		next_rxd = IXGBE_RX_DESC_ADV(rx_ring, i);
 		prefetch(next_rxd);
 		cleaned_count++;
 
@@ -5813,7 +5813,7 @@ static int ixgbe_tso(struct ixgbe_adapter *adapter,
 		i = tx_ring->next_to_use;
 
 		tx_buffer_info = &tx_ring->tx_buffer_info[i];
-		context_desc = IXGBE_TX_CTXTDESC_ADV(*tx_ring, i);
+		context_desc = IXGBE_TX_CTXTDESC_ADV(tx_ring, i);
 
 		/* VLAN MACLEN IPLEN */
 		if (tx_flags & IXGBE_TX_FLAGS_VLAN)
@@ -5872,7 +5872,7 @@ static bool ixgbe_tx_csum(struct ixgbe_adapter *adapter,
 	    (tx_flags & IXGBE_TX_FLAGS_VLAN)) {
 		i = tx_ring->next_to_use;
 		tx_buffer_info = &tx_ring->tx_buffer_info[i];
-		context_desc = IXGBE_TX_CTXTDESC_ADV(*tx_ring, i);
+		context_desc = IXGBE_TX_CTXTDESC_ADV(tx_ring, i);
 
 		if (tx_flags & IXGBE_TX_FLAGS_VLAN)
 			vlan_macip_lens |=
@@ -6101,7 +6101,7 @@ static void ixgbe_tx_queue(struct ixgbe_adapter *adapter,
 	i = tx_ring->next_to_use;
 	while (count--) {
 		tx_buffer_info = &tx_ring->tx_buffer_info[i];
-		tx_desc = IXGBE_TX_DESC_ADV(*tx_ring, i);
+		tx_desc = IXGBE_TX_DESC_ADV(tx_ring, i);
 		tx_desc->read.buffer_addr = cpu_to_le64(tx_buffer_info->dma);
 		tx_desc->read.cmd_type_len =
 		        cpu_to_le32(cmd_type_len | tx_buffer_info->length);


^ permalink raw reply related

* [net-next-2.6 PATCH 18/20] ixgbe: combine Rx into into ixgbe_configure_rx
From: Jeff Kirsher @ 2010-08-19 23:40 UTC (permalink / raw)
  To: davem; +Cc: netdev, gospo, bphilips, Alexander Duyck, Jeff Kirsher
In-Reply-To: <20100819233052.10335.13176.stgit@localhost.localdomain>

From: Alexander Duyck <alexander.h.duyck@intel.com>

The Rx init is currently split over ixgbe_configure, ixgbe_configure_rx,
and ixgbe_up_complete.  Instead of leaving it split over 3 function it is
easier to consolidate them all into ixgbe_configure_rx.

Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---

 drivers/net/ixgbe/ixgbe_main.c |  152 +++++++++++++++++++++++++---------------
 1 files changed, 95 insertions(+), 57 deletions(-)

diff --git a/drivers/net/ixgbe/ixgbe_main.c b/drivers/net/ixgbe/ixgbe_main.c
index fd2026e..c88ba13 100644
--- a/drivers/net/ixgbe/ixgbe_main.c
+++ b/drivers/net/ixgbe/ixgbe_main.c
@@ -2574,6 +2574,8 @@ static void ixgbe_configure_srrctl(struct ixgbe_adapter *adapter,
 
 	srrctl &= ~IXGBE_SRRCTL_BSIZEHDR_MASK;
 	srrctl &= ~IXGBE_SRRCTL_BSIZEPKT_MASK;
+	if (adapter->num_vfs)
+		srrctl |= IXGBE_SRRCTL_DROP_EN;
 
 	srrctl |= (IXGBE_RX_HDR_SIZE << IXGBE_SRRCTL_BSIZEHDRSIZE_SHIFT) &
 		  IXGBE_SRRCTL_BSIZEHDR_MASK;
@@ -2705,13 +2707,72 @@ static void ixgbe_configure_rscctl(struct ixgbe_adapter *adapter,
 	IXGBE_WRITE_REG(hw, IXGBE_RSCCTL(reg_idx), rscctrl);
 }
 
+/**
+ *  ixgbe_set_uta - Set unicast filter table address
+ *  @adapter: board private structure
+ *
+ *  The unicast table address is a register array of 32-bit registers.
+ *  The table is meant to be used in a way similar to how the MTA is used
+ *  however due to certain limitations in the hardware it is necessary to
+ *  set all the hash bits to 1 and use the VMOLR ROPE bit as a promiscuous
+ *  enable bit to allow vlan tag stripping when promiscuous mode is enabled
+ **/
+static void ixgbe_set_uta(struct ixgbe_adapter *adapter)
+{
+	struct ixgbe_hw *hw = &adapter->hw;
+	int i;
+
+	/* The UTA table only exists on 82599 hardware and newer */
+	if (hw->mac.type < ixgbe_mac_82599EB)
+		return;
+
+	/* we only need to do this if VMDq is enabled */
+	if (!(adapter->flags & IXGBE_FLAG_SRIOV_ENABLED))
+		return;
+
+	for (i = 0; i < 128; i++)
+		IXGBE_WRITE_REG(hw, IXGBE_UTA(i), ~0);
+}
+
+#define IXGBE_MAX_RX_DESC_POLL 10
+static void ixgbe_rx_desc_queue_enable(struct ixgbe_adapter *adapter,
+				       struct ixgbe_ring *ring)
+{
+	struct ixgbe_hw *hw = &adapter->hw;
+	int reg_idx = ring->reg_idx;
+	int wait_loop = IXGBE_MAX_RX_DESC_POLL;
+	u32 rxdctl;
+
+	/* RXDCTL.EN will return 0 on 82598 if link is down, so skip it */
+	if (hw->mac.type == ixgbe_mac_82598EB &&
+	    !(IXGBE_READ_REG(hw, IXGBE_LINKS) & IXGBE_LINKS_UP))
+		return;
+
+	do {
+		msleep(1);
+		rxdctl = IXGBE_READ_REG(hw, IXGBE_RXDCTL(reg_idx));
+	} while (--wait_loop && !(rxdctl & IXGBE_RXDCTL_ENABLE));
+
+	if (!wait_loop) {
+		e_err(drv, "RXDCTL.ENABLE on Rx queue %d not set within "
+		      "the polling period\n", reg_idx);
+	}
+}
+
 static void ixgbe_configure_rx_ring(struct ixgbe_adapter *adapter,
 				    struct ixgbe_ring *ring)
 {
 	struct ixgbe_hw *hw = &adapter->hw;
 	u64 rdba = ring->dma;
+	u32 rxdctl;
 	u16 reg_idx = ring->reg_idx;
 
+	/* disable queue to avoid issues while updating state */
+	rxdctl = IXGBE_READ_REG(hw, IXGBE_RXDCTL(reg_idx));
+	IXGBE_WRITE_REG(hw, IXGBE_RXDCTL(reg_idx),
+			rxdctl & ~IXGBE_RXDCTL_ENABLE);
+	IXGBE_WRITE_FLUSH(hw);
+
 	IXGBE_WRITE_REG(hw, IXGBE_RDBAL(reg_idx), (rdba & DMA_BIT_MASK(32)));
 	IXGBE_WRITE_REG(hw, IXGBE_RDBAH(reg_idx), (rdba >> 32));
 	IXGBE_WRITE_REG(hw, IXGBE_RDLEN(reg_idx),
@@ -2720,6 +2781,28 @@ static void ixgbe_configure_rx_ring(struct ixgbe_adapter *adapter,
 	IXGBE_WRITE_REG(hw, IXGBE_RDT(reg_idx), 0);
 	ring->head = IXGBE_RDH(reg_idx);
 	ring->tail = IXGBE_RDT(reg_idx);
+
+	ixgbe_configure_srrctl(adapter, ring);
+	ixgbe_configure_rscctl(adapter, ring);
+
+	if (hw->mac.type == ixgbe_mac_82598EB) {
+		/*
+		 * enable cache line friendly hardware writes:
+		 * PTHRESH=32 descriptors (half the internal cache),
+		 * this also removes ugly rx_no_buffer_count increment
+		 * HTHRESH=4 descriptors (to minimize latency on fetch)
+		 * WTHRESH=8 burst writeback up to two cache lines
+		 */
+		rxdctl &= ~0x3FFFFF;
+		rxdctl |=  0x080420;
+	}
+
+	/* enable receive descriptor ring */
+	rxdctl |= IXGBE_RXDCTL_ENABLE;
+	IXGBE_WRITE_REG(hw, IXGBE_RXDCTL(reg_idx), rxdctl);
+
+	ixgbe_rx_desc_queue_enable(adapter, ring);
+	ixgbe_alloc_rx_buffers(adapter, ring, IXGBE_DESC_UNUSED(ring));
 }
 
 static void ixgbe_setup_psrtype(struct ixgbe_adapter *adapter)
@@ -2908,7 +2991,6 @@ static void ixgbe_setup_rdrxctl(struct ixgbe_adapter *adapter)
 static void ixgbe_configure_rx(struct ixgbe_adapter *adapter)
 {
 	struct ixgbe_hw *hw = &adapter->hw;
-	struct ixgbe_ring *rx_ring;
 	int i;
 	u32 rxctrl;
 
@@ -2919,10 +3001,12 @@ static void ixgbe_configure_rx(struct ixgbe_adapter *adapter)
 	ixgbe_setup_psrtype(adapter);
 	ixgbe_setup_rdrxctl(adapter);
 
-	/* Program MRQC for the distribution of queues */
+	/* Program registers for the distribution of queues */
 	ixgbe_setup_mrqc(adapter);
 	ixgbe_configure_virtualization(adapter);
 
+	ixgbe_set_uta(adapter);
+
 	/* set_rx_buffer_len must be called before ring initialization */
 	ixgbe_set_rx_buffer_len(adapter);
 
@@ -2930,13 +3014,16 @@ static void ixgbe_configure_rx(struct ixgbe_adapter *adapter)
 	 * Setup the HW Rx Head and Tail Descriptor Pointers and
 	 * the Base and Length of the Rx Descriptor Ring
 	 */
-	for (i = 0; i < adapter->num_rx_queues; i++) {
-		rx_ring = adapter->rx_ring[i];
-		ixgbe_configure_rx_ring(adapter, rx_ring);
-		ixgbe_configure_srrctl(adapter, rx_ring);
-		ixgbe_configure_rscctl(adapter, rx_ring);
-	}
+	for (i = 0; i < adapter->num_rx_queues; i++)
+		ixgbe_configure_rx_ring(adapter, adapter->rx_ring[i]);
 
+	/* disable drop enable for 82598 parts */
+	if (hw->mac.type == ixgbe_mac_82598EB)
+		rxctrl |= IXGBE_RXCTRL_DMBYPS;
+
+	/* enable all receives */
+	rxctrl |= IXGBE_RXCTRL_RXEN;
+	hw->mac.ops.enable_rx_dma(hw, rxctrl);
 }
 
 static void ixgbe_vlan_rx_add_vid(struct net_device *netdev, u16 vid)
@@ -3306,9 +3393,6 @@ static void ixgbe_configure(struct ixgbe_adapter *adapter)
 
 	ixgbe_configure_tx(adapter);
 	ixgbe_configure_rx(adapter);
-	for (i = 0; i < adapter->num_rx_queues; i++)
-		ixgbe_alloc_rx_buffers(adapter, adapter->rx_ring[i],
-		                       (adapter->rx_ring[i]->count - 1));
 }
 
 static inline bool ixgbe_is_sfp(struct ixgbe_hw *hw)
@@ -3389,28 +3473,6 @@ link_cfg_out:
 	return ret;
 }
 
-#define IXGBE_MAX_RX_DESC_POLL 10
-static inline void ixgbe_rx_desc_queue_enable(struct ixgbe_adapter *adapter,
-	                                      int rxr)
-{
-	int j = adapter->rx_ring[rxr]->reg_idx;
-	int k;
-
-	for (k = 0; k < IXGBE_MAX_RX_DESC_POLL; k++) {
-		if (IXGBE_READ_REG(&adapter->hw,
-		                   IXGBE_RXDCTL(j)) & IXGBE_RXDCTL_ENABLE)
-			break;
-		else
-			msleep(1);
-	}
-	if (k >= IXGBE_MAX_RX_DESC_POLL) {
-		e_err(drv, "RXDCTL.ENABLE on Rx queue %d not set within "
-		      "the polling period\n", rxr);
-	}
-	ixgbe_release_rx_desc(&adapter->hw, adapter->rx_ring[rxr],
-	                      (adapter->rx_ring[rxr]->count - 1));
-}
-
 static void ixgbe_setup_gpie(struct ixgbe_adapter *adapter)
 {
 	struct ixgbe_hw *hw = &adapter->hw;
@@ -3462,35 +3524,12 @@ static void ixgbe_setup_gpie(struct ixgbe_adapter *adapter)
 static int ixgbe_up_complete(struct ixgbe_adapter *adapter)
 {
 	struct ixgbe_hw *hw = &adapter->hw;
-	int i, j = 0;
-	int num_rx_rings = adapter->num_rx_queues;
 	int err;
-	u32 rxdctl;
 	u32 ctrl_ext;
 
 	ixgbe_get_hw_control(adapter);
 	ixgbe_setup_gpie(adapter);
 
-	for (i = 0; i < num_rx_rings; i++) {
-		j = adapter->rx_ring[i]->reg_idx;
-		rxdctl = IXGBE_READ_REG(hw, IXGBE_RXDCTL(j));
-		/* enable PTHRESH=32 descriptors (half the internal cache)
-		 * and HTHRESH=0 descriptors (to minimize latency on fetch),
-		 * this also removes a pesky rx_no_buffer_count increment */
-		rxdctl |= 0x0020;
-		rxdctl |= IXGBE_RXDCTL_ENABLE;
-		IXGBE_WRITE_REG(hw, IXGBE_RXDCTL(j), rxdctl);
-		if (hw->mac.type == ixgbe_mac_82599EB)
-			ixgbe_rx_desc_queue_enable(adapter, i);
-	}
-	/* enable all receives */
-	rxdctl = IXGBE_READ_REG(hw, IXGBE_RXCTRL);
-	if (hw->mac.type == ixgbe_mac_82598EB)
-		rxdctl |= (IXGBE_RXCTRL_DMBYPS | IXGBE_RXCTRL_RXEN);
-	else
-		rxdctl |= IXGBE_RXCTRL_RXEN;
-	hw->mac.ops.enable_rx_dma(hw, rxdctl);
-
 	if (adapter->flags & IXGBE_FLAG_MSIX_ENABLED)
 		ixgbe_configure_msix(adapter);
 	else
@@ -3505,7 +3544,6 @@ static int ixgbe_up_complete(struct ixgbe_adapter *adapter)
 
 	/* clear any pending interrupts, may auto mask */
 	IXGBE_READ_REG(hw, IXGBE_EICR);
-
 	ixgbe_irq_enable(adapter);
 
 	/*


^ permalink raw reply related

* [net-next-2.6 PATCH 17/20] ixgbe: pull all Tx init into ixgbe_configure_tx
From: Jeff Kirsher @ 2010-08-19 23:39 UTC (permalink / raw)
  To: davem; +Cc: netdev, gospo, bphilips, Alexander Duyck, Jeff Kirsher
In-Reply-To: <20100819233052.10335.13176.stgit@localhost.localdomain>

From: Alexander Duyck <alexander.h.duyck@intel.com>

The Tx init was spread out over ixgbe_configure, ixgbe_configure_tx, and
ixgbe_up_complete.  This change combines all of that into the
ixgbe_configure_tx function in order to simplify the Tx init path.

Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---

 drivers/net/ixgbe/ixgbe_main.c |   91 +++++++++++++++++++++++-----------------
 1 files changed, 52 insertions(+), 39 deletions(-)

diff --git a/drivers/net/ixgbe/ixgbe_main.c b/drivers/net/ixgbe/ixgbe_main.c
index 0b23522..fd2026e 100644
--- a/drivers/net/ixgbe/ixgbe_main.c
+++ b/drivers/net/ixgbe/ixgbe_main.c
@@ -2436,8 +2436,16 @@ static void ixgbe_configure_msi_and_legacy(struct ixgbe_adapter *adapter)
 {
 	struct ixgbe_hw *hw = &adapter->hw;
 	u64 tdba = ring->dma;
+	int wait_loop = 10;
+	u32 txdctl;
 	u16 reg_idx = ring->reg_idx;
 
+	/* disable queue to avoid issues while updating state */
+	txdctl = IXGBE_READ_REG(hw, IXGBE_TXDCTL(reg_idx));
+	IXGBE_WRITE_REG(hw, IXGBE_TXDCTL(reg_idx),
+			txdctl & ~IXGBE_TXDCTL_ENABLE);
+	IXGBE_WRITE_FLUSH(hw);
+
 	IXGBE_WRITE_REG(hw, IXGBE_TDBAL(reg_idx),
 	                (tdba & DMA_BIT_MASK(32)));
 	IXGBE_WRITE_REG(hw, IXGBE_TDBAH(reg_idx), (tdba >> 32));
@@ -2448,6 +2456,38 @@ static void ixgbe_configure_msi_and_legacy(struct ixgbe_adapter *adapter)
 	ring->head = IXGBE_TDH(reg_idx);
 	ring->tail = IXGBE_TDT(reg_idx);
 
+	/* configure fetching thresholds */
+	if (adapter->rx_itr_setting == 0) {
+		/* cannot set wthresh when itr==0 */
+		txdctl &= ~0x007F0000;
+	} else {
+		/* enable WTHRESH=8 descriptors, to encourage burst writeback */
+		txdctl |= (8 << 16);
+	}
+	if (adapter->flags & IXGBE_FLAG_DCB_ENABLED) {
+		/* PThresh workaround for Tx hang with DFP enabled. */
+		txdctl |= 32;
+	}
+
+	/* reinitialize flowdirector state */
+	set_bit(__IXGBE_FDIR_INIT_DONE, &ring->reinit_state);
+
+	/* enable queue */
+	txdctl |= IXGBE_TXDCTL_ENABLE;
+	IXGBE_WRITE_REG(hw, IXGBE_TXDCTL(reg_idx), txdctl);
+
+	/* TXDCTL.EN will return 0 on 82598 if link is down, so skip it */
+	if (hw->mac.type == ixgbe_mac_82598EB &&
+	    !(IXGBE_READ_REG(hw, IXGBE_LINKS) & IXGBE_LINKS_UP))
+		return;
+
+	/* poll to verify queue is enabled */
+	do {
+		msleep(1);
+		txdctl = IXGBE_READ_REG(hw, IXGBE_TXDCTL(reg_idx));
+	} while (--wait_loop && !(txdctl & IXGBE_TXDCTL_ENABLE));
+	if (!wait_loop)
+		e_err(drv, "Could not enable Tx Queue %d\n", reg_idx);
 }
 
 static void ixgbe_setup_mtqc(struct ixgbe_adapter *adapter)
@@ -2497,13 +2537,22 @@ static void ixgbe_setup_mtqc(struct ixgbe_adapter *adapter)
  **/
 static void ixgbe_configure_tx(struct ixgbe_adapter *adapter)
 {
+	struct ixgbe_hw *hw = &adapter->hw;
+	u32 dmatxctl;
 	u32 i;
 
+	ixgbe_setup_mtqc(adapter);
+
+	if (hw->mac.type != ixgbe_mac_82598EB) {
+		/* DMATXCTL.EN must be before Tx queues are enabled */
+		dmatxctl = IXGBE_READ_REG(hw, IXGBE_DMATXCTL);
+		dmatxctl |= IXGBE_DMATXCTL_TE;
+		IXGBE_WRITE_REG(hw, IXGBE_DMATXCTL, dmatxctl);
+	}
+
 	/* Setup the HW Tx Head and Tail descriptor pointers */
 	for (i = 0; i < adapter->num_tx_queues; i++)
 		ixgbe_configure_tx_ring(adapter, adapter->tx_ring[i]);
-
-	ixgbe_setup_mtqc(adapter);
 }
 
 #define IXGBE_SRRCTL_BSIZEHDRSIZE_SHIFT 2
@@ -3416,44 +3465,12 @@ static int ixgbe_up_complete(struct ixgbe_adapter *adapter)
 	int i, j = 0;
 	int num_rx_rings = adapter->num_rx_queues;
 	int err;
-	u32 txdctl, rxdctl;
-	u32 dmatxctl;
+	u32 rxdctl;
 	u32 ctrl_ext;
 
 	ixgbe_get_hw_control(adapter);
 	ixgbe_setup_gpie(adapter);
 
-	if (hw->mac.type == ixgbe_mac_82599EB) {
-		/* DMATXCTL.EN must be set after all Tx queue config is done */
-		dmatxctl = IXGBE_READ_REG(hw, IXGBE_DMATXCTL);
-		dmatxctl |= IXGBE_DMATXCTL_TE;
-		IXGBE_WRITE_REG(hw, IXGBE_DMATXCTL, dmatxctl);
-	}
-	for (i = 0; i < adapter->num_tx_queues; i++) {
-		j = adapter->tx_ring[i]->reg_idx;
-		txdctl = IXGBE_READ_REG(hw, IXGBE_TXDCTL(j));
-		if (adapter->rx_itr_setting == 0) {
-			/* cannot set wthresh when itr==0 */
-			txdctl &= ~0x007F0000;
-		} else {
-			/* enable WTHRESH=8 descriptors, to encourage burst writeback */
-			txdctl |= (8 << 16);
-		}
-		txdctl |= IXGBE_TXDCTL_ENABLE;
-		IXGBE_WRITE_REG(hw, IXGBE_TXDCTL(j), txdctl);
-		if (hw->mac.type == ixgbe_mac_82599EB) {
-			int wait_loop = 10;
-			/* poll for Tx Enable ready */
-			do {
-				msleep(1);
-				txdctl = IXGBE_READ_REG(hw, IXGBE_TXDCTL(j));
-			} while (--wait_loop &&
-			         !(txdctl & IXGBE_TXDCTL_ENABLE));
-			if (!wait_loop)
-				e_err(drv, "Could not enable Tx Queue %d\n", j);
-		}
-	}
-
 	for (i = 0; i < num_rx_rings; i++) {
 		j = adapter->rx_ring[i]->reg_idx;
 		rxdctl = IXGBE_READ_REG(hw, IXGBE_RXDCTL(j));
@@ -3530,10 +3547,6 @@ static int ixgbe_up_complete(struct ixgbe_adapter *adapter)
 			e_err(probe, "link_config FAILED %d\n", err);
 	}
 
-	for (i = 0; i < adapter->num_tx_queues; i++)
-		set_bit(__IXGBE_FDIR_INIT_DONE,
-		        &(adapter->tx_ring[i]->reinit_state));
-
 	/* enable transmits */
 	netif_tx_start_all_queues(adapter->netdev);
 


^ permalink raw reply related

* [net-next-2.6 PATCH 16/20] ixgbe: move all GPIE register config into a single function
From: Jeff Kirsher @ 2010-08-19 23:39 UTC (permalink / raw)
  To: davem; +Cc: netdev, gospo, bphilips, Alexander Duyck, Jeff Kirsher
In-Reply-To: <20100819233052.10335.13176.stgit@localhost.localdomain>

From: Alexander Duyck <alexander.h.duyck@intel.com>

This change moves all GPIE register configuration into a single function.
The advantage of this is that we can avoid a number of unnecessary
read/modify/write cycles on the register.

Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
---

 drivers/net/ixgbe/ixgbe_main.c |   72 +++++++++++++++++-----------------------
 1 files changed, 30 insertions(+), 42 deletions(-)

diff --git a/drivers/net/ixgbe/ixgbe_main.c b/drivers/net/ixgbe/ixgbe_main.c
index d358bef..0b23522 100644
--- a/drivers/net/ixgbe/ixgbe_main.c
+++ b/drivers/net/ixgbe/ixgbe_main.c
@@ -3362,38 +3362,15 @@ static inline void ixgbe_rx_desc_queue_enable(struct ixgbe_adapter *adapter,
 	                      (adapter->rx_ring[rxr]->count - 1));
 }
 
-static int ixgbe_up_complete(struct ixgbe_adapter *adapter)
+static void ixgbe_setup_gpie(struct ixgbe_adapter *adapter)
 {
 	struct ixgbe_hw *hw = &adapter->hw;
-	int i, j = 0;
-	int num_rx_rings = adapter->num_rx_queues;
-	int err;
-	u32 txdctl, rxdctl;
-	u32 dmatxctl;
-	u32 gpie;
-	u32 ctrl_ext;
-
-	ixgbe_get_hw_control(adapter);
-
-	if ((adapter->flags & IXGBE_FLAG_MSIX_ENABLED) ||
-	    (adapter->flags & IXGBE_FLAG_MSI_ENABLED)) {
-		if (adapter->flags & IXGBE_FLAG_MSIX_ENABLED) {
-			gpie = (IXGBE_GPIE_MSIX_MODE | IXGBE_GPIE_EIAME |
-			        IXGBE_GPIE_PBA_SUPPORT | IXGBE_GPIE_OCD);
-		} else {
-			/* MSI only */
-			gpie = 0;
-		}
-		if (adapter->flags & IXGBE_FLAG_SRIOV_ENABLED) {
-			gpie &= ~IXGBE_GPIE_VTMODE_MASK;
-			gpie |= IXGBE_GPIE_VTMODE_64;
-		}
-		/* XXX: to interrupt immediately for EICS writes, enable this */
-		/* gpie |= IXGBE_GPIE_EIMEN; */
-		IXGBE_WRITE_REG(hw, IXGBE_GPIE, gpie);
-	}
+	u32 gpie = 0;
 
 	if (adapter->flags & IXGBE_FLAG_MSIX_ENABLED) {
+		gpie = IXGBE_GPIE_MSIX_MODE | IXGBE_GPIE_PBA_SUPPORT |
+		       IXGBE_GPIE_OCD;
+		gpie |= IXGBE_GPIE_EIAME;
 		/*
 		 * use EIAM to auto-mask when MSI-X interrupt is asserted
 		 * this saves a register write for every interrupt
@@ -3414,26 +3391,37 @@ static int ixgbe_up_complete(struct ixgbe_adapter *adapter)
 		IXGBE_WRITE_REG(hw, IXGBE_EIAM, IXGBE_EICS_RTX_QUEUE);
 	}
 
-	/* Enable Thermal over heat sensor interrupt */
-	if (adapter->flags2 & IXGBE_FLAG2_TEMP_SENSOR_CAPABLE) {
-		gpie = IXGBE_READ_REG(hw, IXGBE_GPIE);
-		gpie |= IXGBE_SDP0_GPIEN;
-		IXGBE_WRITE_REG(hw, IXGBE_GPIE, gpie);
+	/* XXX: to interrupt immediately for EICS writes, enable this */
+	/* gpie |= IXGBE_GPIE_EIMEN; */
+
+	if (adapter->flags & IXGBE_FLAG_SRIOV_ENABLED) {
+		gpie &= ~IXGBE_GPIE_VTMODE_MASK;
+		gpie |= IXGBE_GPIE_VTMODE_64;
 	}
 
-	/* Enable fan failure interrupt if media type is copper */
-	if (adapter->flags & IXGBE_FLAG_FAN_FAIL_CAPABLE) {
-		gpie = IXGBE_READ_REG(hw, IXGBE_GPIE);
+	/* Enable fan failure interrupt */
+	if (adapter->flags & IXGBE_FLAG_FAN_FAIL_CAPABLE)
 		gpie |= IXGBE_SDP1_GPIEN;
-		IXGBE_WRITE_REG(hw, IXGBE_GPIE, gpie);
-	}
 
-	if (hw->mac.type == ixgbe_mac_82599EB) {
-		gpie = IXGBE_READ_REG(hw, IXGBE_GPIE);
+	if (hw->mac.type == ixgbe_mac_82599EB)
 		gpie |= IXGBE_SDP1_GPIEN;
 		gpie |= IXGBE_SDP2_GPIEN;
-		IXGBE_WRITE_REG(hw, IXGBE_GPIE, gpie);
-	}
+
+	IXGBE_WRITE_REG(hw, IXGBE_GPIE, gpie);
+}
+
+static int ixgbe_up_complete(struct ixgbe_adapter *adapter)
+{
+	struct ixgbe_hw *hw = &adapter->hw;
+	int i, j = 0;
+	int num_rx_rings = adapter->num_rx_queues;
+	int err;
+	u32 txdctl, rxdctl;
+	u32 dmatxctl;
+	u32 ctrl_ext;
+
+	ixgbe_get_hw_control(adapter);
+	ixgbe_setup_gpie(adapter);
 
 	if (hw->mac.type == ixgbe_mac_82599EB) {
 		/* DMATXCTL.EN must be set after all Tx queue config is done */


^ permalink raw reply related


This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox