[PATCH -next 1/6] bnx2: Check BNX2_FLAG_USING_MSIX flag when setting up MSIX.

[PATCH -next 1/6] bnx2: Check BNX2_FLAG_USING_MSIX flag when setting up MSIX.

[Michael Chan]

Checking the flag is more correct than checking bp->irq_nvecs. By
accident it is not a problem because we always have more than 1
vectors when using MSIX mode.

Signed-off-by: Michael Chan <mchan@broadcom.com>
Signed-off-by: Benjamin Li <benli@broadcom.com>
---
 drivers/net/bnx2.c |    2 +-
 1 files changed, 1 insertions(+), 1 deletions(-)

diff --git a/drivers/net/bnx2.c b/drivers/net/bnx2.c
index c7f5515..48bc578 100644
--- a/drivers/net/bnx2.c
+++ b/drivers/net/bnx2.c
@@ -4940,7 +4940,7 @@ bnx2_init_chip(struct bnx2 *bp)
 		      BNX2_HC_CONFIG_COLLECT_STATS;
 	}
 
-	if (bp->irq_nvecs > 1) {
+	if (bp->flags & BNX2_FLAG_USING_MSIX) {
 		REG_WR(bp, BNX2_HC_MSIX_BIT_VECTOR,
 		       BNX2_HC_MSIX_BIT_VECTOR_VAL);
 
-- 
1.6.4.GIT

[PATCH -next 2/6] bnx2: Need to call cnic_setup_cnic_irq_info() after MTU change.

[Michael Chan]

New status blocks are allocated during MTU change so we need to
update this information for the cnic driver.

Signed-off-by: Michael Chan <mchan@broadcom.com>
Signed-off-by: Benjamin Li <benli@broadcom.com>
---
 drivers/net/bnx2.c |    7 +++++++
 1 files changed, 7 insertions(+), 0 deletions(-)

diff --git a/drivers/net/bnx2.c b/drivers/net/bnx2.c
index 48bc578..063f283 100644
--- a/drivers/net/bnx2.c
+++ b/drivers/net/bnx2.c
@@ -7141,6 +7141,13 @@ bnx2_change_ring_size(struct bnx2 *bp, u32 rx, u32 tx)
 			dev_close(bp->dev);
 			return rc;
 		}
+#ifdef BCM_CNIC
+		mutex_lock(&bp->cnic_lock);
+		/* Let cnic know about the new status block. */
+		if (bp->cnic_eth_dev.drv_state & CNIC_DRV_STATE_REGD)
+			bnx2_setup_cnic_irq_info(bp);
+		mutex_unlock(&bp->cnic_lock);
+#endif
 		bnx2_netif_start(bp);
 	}
 	return 0;
-- 
1.6.4.GIT

[PATCH -next 3/6] bnx2: Adjust flow control water marks.

[Michael Chan]

The current water marks are too high and can cause unnecessary flow
control frames.

Signed-off-by: Michael Chan <mchan@broadcom.com>
Signed-off-by: Benjamin Li <benli@broadcom.com>
---
 drivers/net/bnx2.c |    2 +-
 drivers/net/bnx2.h |    2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/net/bnx2.c b/drivers/net/bnx2.c
index 063f283..54d43d8 100644
--- a/drivers/net/bnx2.c
+++ b/drivers/net/bnx2.c
@@ -1277,7 +1277,7 @@ bnx2_init_rx_context(struct bnx2 *bp, u32 cid)
 		if (lo_water >= bp->rx_ring_size)
 			lo_water = 0;
 
-		hi_water = bp->rx_ring_size / 4;
+		hi_water = min_t(int, bp->rx_ring_size / 4, lo_water + 16);
 
 		if (hi_water <= lo_water)
 			lo_water = 0;
diff --git a/drivers/net/bnx2.h b/drivers/net/bnx2.h
index b860fbb..cd4b0e4 100644
--- a/drivers/net/bnx2.h
+++ b/drivers/net/bnx2.h
@@ -349,7 +349,7 @@ struct l2_fhdr {
 #define BNX2_L2CTX_BD_PRE_READ				0x00000000
 #define BNX2_L2CTX_CTX_SIZE				0x00000000
 #define BNX2_L2CTX_CTX_TYPE				0x00000000
-#define BNX2_L2CTX_LO_WATER_MARK_DEFAULT		 32
+#define BNX2_L2CTX_LO_WATER_MARK_DEFAULT		 4
 #define BNX2_L2CTX_LO_WATER_MARK_SCALE			 4
 #define BNX2_L2CTX_LO_WATER_MARK_DIS			 0
 #define BNX2_L2CTX_HI_WATER_MARK_SHIFT			 4
-- 
1.6.4.GIT

[PATCH -next 4/6] bnx2: Allow user-specified multiple advertisement speed values.

[Michael Chan]

Remove unnecessary code that works around older versions of ethtool
that can pass down invalid advertisement speed values.  This old
code prevents the user from specifying multiple advertisement values.
The new code uses simple masking to mask out invalid advertisment bits.

Reported-by: Vlad Yasevich <vladislav.yasevich@hp.com>
Signed-off-by: Michael Chan <mchan@broadcom.com>
Signed-off-by: Benjamin Li <benli@broadcom.com>
---
 drivers/net/bnx2.c |   33 ++++++++-------------------------
 1 files changed, 8 insertions(+), 25 deletions(-)

diff --git a/drivers/net/bnx2.c b/drivers/net/bnx2.c
index 54d43d8..2c75fd7 100644
--- a/drivers/net/bnx2.c
+++ b/drivers/net/bnx2.c
@@ -6751,32 +6751,15 @@ bnx2_set_settings(struct net_device *dev, struct ethtool_cmd *cmd)
 	if (cmd->autoneg == AUTONEG_ENABLE) {
 		autoneg |= AUTONEG_SPEED;
 
-		cmd->advertising &= ETHTOOL_ALL_COPPER_SPEED;
-
-		/* allow advertising 1 speed */
-		if ((cmd->advertising == ADVERTISED_10baseT_Half) ||
-			(cmd->advertising == ADVERTISED_10baseT_Full) ||
-			(cmd->advertising == ADVERTISED_100baseT_Half) ||
-			(cmd->advertising == ADVERTISED_100baseT_Full)) {
-
-			if (cmd->port == PORT_FIBRE)
-				goto err_out_unlock;
-
-			advertising = cmd->advertising;
-
-		} else if (cmd->advertising == ADVERTISED_2500baseX_Full) {
-			if (!(bp->phy_flags & BNX2_PHY_FLAG_2_5G_CAPABLE) ||
-			    (cmd->port == PORT_TP))
-				goto err_out_unlock;
-		} else if (cmd->advertising == ADVERTISED_1000baseT_Full)
-			advertising = cmd->advertising;
-		else if (cmd->advertising == ADVERTISED_1000baseT_Half)
-			goto err_out_unlock;
-		else {
-			if (cmd->port == PORT_FIBRE)
-				advertising = ETHTOOL_ALL_FIBRE_SPEED;
-			else
+		advertising = cmd->advertising;
+		if (cmd->port == PORT_TP) {
+			advertising &= ETHTOOL_ALL_COPPER_SPEED;
+			if (!advertising)
 				advertising = ETHTOOL_ALL_COPPER_SPEED;
+		} else {
+			advertising &= ETHTOOL_ALL_FIBRE_SPEED;
+			if (!advertising)
+				advertising = ETHTOOL_ALL_FIBRE_SPEED;
 		}
 		advertising |= ADVERTISED_Autoneg;
 	}
-- 
1.6.4.GIT

[PATCH -next 5/6] bnx2: Fix bug when saving statistics.

[Michael Chan]

From: Patrick Rabau <pr2345@gmail.com>

This fixes the problem of dropping the carry when adding 2 32-bit values.
Switch to use array indexing for better readability.

Reported by and fix provided by Patrick Rabau.

Signed-off-by: Michael Chan <mchan@broadcom.com>
Signed-off-by: Benjamin Li <benli@broadcom.com>
---
 drivers/net/bnx2.c |   10 +++++-----
 1 files changed, 5 insertions(+), 5 deletions(-)

diff --git a/drivers/net/bnx2.c b/drivers/net/bnx2.c
index 2c75fd7..dda815a 100644
--- a/drivers/net/bnx2.c
+++ b/drivers/net/bnx2.c
@@ -6555,16 +6555,16 @@ bnx2_save_stats(struct bnx2 *bp)
 		u32 hi;
 		u64 lo;
 
-		hi = *(temp_stats + i) + *(hw_stats + i);
-		lo = *(temp_stats + i + 1) + *(hw_stats + i + 1);
+		hi = temp_stats[i] + hw_stats[i];
+		lo = (u64) temp_stats[i + 1] + (u64) hw_stats[i + 1];
 		if (lo > 0xffffffff)
 			hi++;
-		*(temp_stats + i) = hi;
-		*(temp_stats + i + 1) = lo & 0xffffffff;
+		temp_stats[i] = hi;
+		temp_stats[i + 1] = lo & 0xffffffff;
 	}
 
 	for ( ; i < sizeof(struct statistics_block) / 4; i++)
-		*(temp_stats + i) = *(temp_stats + i) + *(hw_stats + i);
+		temp_stats[i] += hw_stats[i];
 }
 
 #define GET_64BIT_NET_STATS64(ctr)				\
-- 
1.6.4.GIT

Re: [PATCH -next 5/6] bnx2: Fix bug when saving statistics.

[David Miller]

From: "Michael Chan" <mchan@broadcom.com>
Date: Mon, 15 Feb 2010 21:42:11 -0800

> From: Patrick Rabau <pr2345@gmail.com>
> 
> This fixes the problem of dropping the carry when adding 2 32-bit values.
> Switch to use array indexing for better readability.
> 
> Reported by and fix provided by Patrick Rabau.
> 
> Signed-off-by: Michael Chan <mchan@broadcom.com>
> Signed-off-by: Benjamin Li <benli@broadcom.com>

Applied.

Re: [PATCH -next 6/6] bnx2: Update firmwares and update version to 2.0.8.

[David Miller]

From: "Michael Chan" <mchan@broadcom.com>
Date: Mon, 15 Feb 2010 21:42:12 -0800

> - Increase FTQ depth to 256 to ehnabce performance.
> - Fix RV2P context corruption on 5709 when flow control is enabled.
> 
> Signed-off-by: Michael Chan <mchan@broadcom.com>

Also applied, thanks!

Re: [PATCH -next 4/6] bnx2: Allow user-specified multiple advertisement speed values.

[David Miller]

From: "Michael Chan" <mchan@broadcom.com>
Date: Mon, 15 Feb 2010 21:42:10 -0800

> Remove unnecessary code that works around older versions of ethtool
> that can pass down invalid advertisement speed values.  This old
> code prevents the user from specifying multiple advertisement values.
> The new code uses simple masking to mask out invalid advertisment bits.
> 
> Reported-by: Vlad Yasevich <vladislav.yasevich@hp.com>
> Signed-off-by: Michael Chan <mchan@broadcom.com>
> Signed-off-by: Benjamin Li <benli@broadcom.com>

Applied.

Re: [PATCH -next 3/6] bnx2: Adjust flow control water marks.

[David Miller]

From: "Michael Chan" <mchan@broadcom.com>
Date: Mon, 15 Feb 2010 21:42:09 -0800

> The current water marks are too high and can cause unnecessary flow
> control frames.
> 
> Signed-off-by: Michael Chan <mchan@broadcom.com>
> Signed-off-by: Benjamin Li <benli@broadcom.com>

Applied.

Re: [PATCH -next 2/6] bnx2: Need to call cnic_setup_cnic_irq_info() after MTU change.

[David Miller]

From: "Michael Chan" <mchan@broadcom.com>
Date: Mon, 15 Feb 2010 21:42:08 -0800

> New status blocks are allocated during MTU change so we need to
> update this information for the cnic driver.
> 
> Signed-off-by: Michael Chan <mchan@broadcom.com>
> Signed-off-by: Benjamin Li <benli@broadcom.com>

Applied.

Re: [PATCH -next 1/6] bnx2: Check BNX2_FLAG_USING_MSIX flag when setting up MSIX.

[David Miller]

From: "Michael Chan" <mchan@broadcom.com>
Date: Mon, 15 Feb 2010 21:42:07 -0800

> Checking the flag is more correct than checking bp->irq_nvecs. By
> accident it is not a problem because we always have more than 1
> vectors when using MSIX mode.
> 
> Signed-off-by: Michael Chan <mchan@broadcom.com>
> Signed-off-by: Benjamin Li <benli@broadcom.com>

Applied.