smsc95xx: Fix MAC address programming and some minor issues

PATCH Superseded Wolfgang Grandegger 5 participants 17 messages Nov 11, 2011 – Nov 15, 2011 USB

Share & tools

Wolfgang Grandegger

author

2011-11-11 10:59 UTC #0 lore raw

cover smsc95xx: Fix MAC address programming and some minor issues

Wolfgang Grandegger (3):
smsc95xx: Fix MAC address programming
smsc95xx: in smsc95xx_set_multicast write to reg
smsc95xx: remove an unecessary debug messages

 drivers/usb/eth/smsc95xx.c |   16 +++++++++-------
 1 files changed, 9 insertions(+), 7 deletions(-)

Signature · 1 line

1.7.4.1

Wolfgang Grandegger

author

2011-11-11 10:59 UTC #1 lore raw

PATCH 1/3 smsc95xx: Fix MAC address programming

Commit 79ad54400932d6484178a372fb3b659e3437473b broke the MAC address programming. Fix this by using the method from Linux' smsc95xx_set_mac_address().

Signed-off-by Wolfgang Grandegger <[email protected]>

Cc Marek Vasut <[email protected]>

Cc Simon Glass <[email protected]>

---
 drivers/usb/eth/smsc95xx.c |    7 ++++---
 1 files changed, 4 insertions(+), 3 deletions(-)

drivers/usb/eth/smsc95xx.c +4 -3

index 7ee4f87..eb529f1 100644
@@ -372,13 +372,14 @@ static int smsc95xx_init_mac_address(struct eth_device *eth,
 static int smsc95xx_write_hwaddr(struct eth_device *eth)
 {
 	struct ueth_data *dev = (struct ueth_data *)eth->priv;
-	u32 addr_lo, addr_hi;
+	u32 addr_lo = eth->enetaddr[0] | eth->enetaddr[1] << 8 |
+		eth->enetaddr[2] << 16 | eth->enetaddr[3] << 24;
+	u32 addr_hi = eth->enetaddr[4] | eth->enetaddr[5] << 8;
 	int ret;
 
 	/* set hardware address */
 	debug("** %s()\n", __func__);
-	addr_lo = cpu_to_le32(*eth->enetaddr);
-	addr_hi = cpu_to_le16(*((u16 *)(eth->enetaddr + 4)));
+
 	ret = smsc95xx_write_reg(dev, ADDRL, addr_lo);
 	if (ret < 0) {
 		debug("Failed to write ADDRL: %d\n", ret);

Signature · 1 line

1.7.4.1

Marek Vasut

2011-11-11 11:04 UTC #2 lore raw

PATCH 1/3 smsc95xx: Fix MAC address programming

Quoted 23 lines

Commit 79ad54400932d6484178a372fb3b659e3437473b broke the MAC address programming. Fix this by using the method from Linux' smsc95xx_set_mac_address().
Signed-off-by Wolfgang Grandegger <[email protected]>
Cc Marek Vasut <[email protected]>
Cc Simon Glass <[email protected]>
---
 drivers/usb/eth/smsc95xx.c |    7 ++++---
 1 files changed, 4 insertions(+), 3 deletions(-)
drivers/usb/eth/smsc95xx.c +3 -1
index 7ee4f87..eb529f1 100644
@@ -372,13 +372,14 @@ static int smsc95xx_init_mac_address(struct
 eth_device *eth, static int smsc95xx_write_hwaddr(struct eth_device *eth)
 {
 	struct ueth_data *dev = (struct ueth_data *)eth->priv;
-	u32 addr_lo, addr_hi;
+	u32 addr_lo = eth->enetaddr[0] | eth->enetaddr[1] << 8 |
+		eth->enetaddr[2] << 16 | eth->enetaddr[3] << 24;
+	u32 addr_hi = eth->enetaddr[4] | eth->enetaddr[5] << 8;

drivers/usb/eth/smsc95xx.c static int smsc95xx_init_mac_address(struct · line 385

Quoted 10 lines

drivers/usb/eth/smsc95xx.c +1 -2

 	int ret;
 
 	/* set hardware address */
 	debug("** %s()\n", __func__);
-	addr_lo = cpu_to_le32(*eth->enetaddr);
-	addr_hi = cpu_to_le16(*((u16 *)(eth->enetaddr + 4)));
+
 	ret = smsc95xx_write_reg(dev, ADDRL, addr_lo);
 	if (ret < 0) {
 		debug("Failed to write ADDRL: %d\n", ret);

Hey,

didn't Mike Frysinger send similar patch yesterday? Also, do you have the hardware? If so, that's good, we know you can provide tested fix.

Mike, are you good with using this fix instead or can you two negotiate?

Wolfgang Grandegger

2011-11-11 12:33 UTC #3 lore raw

PATCH 1/3 smsc95xx: Fix MAC address programming

Quoted 36 lines · Marek Vasut

Quoted 23 lines
Commit 79ad54400932d6484178a372fb3b659e3437473b broke the MAC address programming. Fix this by using the method from Linux' smsc95xx_set_mac_address().
Signed-off-by Wolfgang Grandegger <[email protected]>
Cc Marek Vasut <[email protected]>
Cc Simon Glass <[email protected]>
---
 drivers/usb/eth/smsc95xx.c |    7 ++++---
 1 files changed, 4 insertions(+), 3 deletions(-)
drivers/usb/eth/smsc95xx.c +3 -1
index 7ee4f87..eb529f1 100644
@@ -372,13 +372,14 @@ static int smsc95xx_init_mac_address(struct
 eth_device *eth, static int smsc95xx_write_hwaddr(struct eth_device *eth)
 {
 	struct ueth_data *dev = (struct ueth_data *)eth->priv;
-	u32 addr_lo, addr_hi;
+	u32 addr_lo = eth->enetaddr[0] | eth->enetaddr[1] << 8 |
+		eth->enetaddr[2] << 16 | eth->enetaddr[3] << 24;
+	u32 addr_hi = eth->enetaddr[4] | eth->enetaddr[5] << 8;
drivers/usb/eth/smsc95xx.c static int smsc95xx_init_mac_address(struct · line 385
Quoted 10 lines
drivers/usb/eth/smsc95xx.c +1 -2
 	int ret;
 
 	/* set hardware address */
 	debug("** %s()\n", __func__);
-	addr_lo = cpu_to_le32(*eth->enetaddr);
-	addr_hi = cpu_to_le16(*((u16 *)(eth->enetaddr + 4)));
+
 	ret = smsc95xx_write_reg(dev, ADDRL, addr_lo);
 	if (ret < 0) {
 		debug("Failed to write ADDRL: %d\n", ret);
Hey,

Quoted 1 line

didn't Mike Frysinger send similar patch yesterday? Also, do you have the

Ah, I obviously missed that.

Quoted 1 line

hardware? If so, that's good, we know you can provide tested fix.

Well, yes, I know somebody who has smsc95xx hardware ;-).

Quoted 1 line

Mike, are you good with using this fix instead or can you two negotiate?

Let's await Mike's answer (now on CC as well). I need to re-send the series anyway due to a stupid typo.

Wolfgang

Mike Frysinger

2011-11-11 15:18 UTC #4 lore raw

PATCH 1/3 smsc95xx: Fix MAC address programming

Quoted 27 lines · On Friday 11 November 2011 07:33:00 Wolfgang Grandegger wrote:

Quoted 17 lines · Marek Vasut
Quoted 17 lines
Commit 79ad54400932d6484178a372fb3b659e3437473b broke the MAC address programming. Fix this by using the method from Linux' smsc95xx_set_mac_address().
drivers/usb/eth/smsc95xx.c +3 -1
 
-	u32 addr_lo, addr_hi;
+	u32 addr_lo = eth->enetaddr[0] | eth->enetaddr[1] << 8 |
+		eth->enetaddr[2] << 16 | eth->enetaddr[3] << 24;
+	u32 addr_hi = eth->enetaddr[4] | eth->enetaddr[5] << 8;
 
 	int ret;
 
 	/* set hardware address */
 	debug("** %s()\n", __func__);
Quoted 6 lines
drivers/usb/eth/smsc95xx.c static int smsc95xx_init_mac_address(struct eth_device *eth, · line 382
Quoted 4 lines
drivers/usb/eth/smsc95xx.c +1 -2
-	addr_lo = cpu_to_le32(*eth->enetaddr);
-	addr_hi = cpu_to_le16(*((u16 *)(eth->enetaddr + 4)));
+
didn't Mike Frysinger send similar patch yesterday? Also, do you have the
Ah, I obviously missed that.

Quoted 8 lines

Quoted 1 line
hardware? If so, that's good, we know you can provide tested fix.
Well, yes, I know somebody who has smsc95xx hardware ;-).
Quoted 1 line
Mike, are you good with using this fix instead or can you two negotiate?
Let's await Mike's answer (now on CC as well). I need to re-send the series anyway due to a stupid typo.

mine might take a little longer due to wrangling with wolfgang. i'd suggest we go with your patch (although i have feedback on it).
-mike
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 836 bytes
Desc: This is a digitally signed message part.
Url : http://lists.denx.de/pipermail/u-boot/attachments/20111111/8f9764e6/attachment.pgp

Wolfgang Grandegger

2011-11-14 08:25 UTC #5 lore raw

PATCH 1/3 smsc95xx: Fix MAC address programming

Quoted 32 lines · Mike Frysinger

Quoted 31 lines · On Friday 11 November 2011 07:33:00 Wolfgang Grandegger wrote:
Quoted 17 lines · Marek Vasut
Quoted 17 lines
Commit 79ad54400932d6484178a372fb3b659e3437473b broke the MAC address programming. Fix this by using the method from Linux' smsc95xx_set_mac_address().
drivers/usb/eth/smsc95xx.c +3 -1
 
-	u32 addr_lo, addr_hi;
+	u32 addr_lo = eth->enetaddr[0] | eth->enetaddr[1] << 8 |
+		eth->enetaddr[2] << 16 | eth->enetaddr[3] << 24;
+	u32 addr_hi = eth->enetaddr[4] | eth->enetaddr[5] << 8;
 
 	int ret;
 
 	/* set hardware address */
 	debug("** %s()\n", __func__);
Quoted 6 lines
drivers/usb/eth/smsc95xx.c static int smsc95xx_init_mac_address(struct eth_device *eth, · line 382
Quoted 4 lines
drivers/usb/eth/smsc95xx.c +1 -2
-	addr_lo = cpu_to_le32(*eth->enetaddr);
-	addr_hi = cpu_to_le16(*((u16 *)(eth->enetaddr + 4)));
+
didn't Mike Frysinger send similar patch yesterday? Also, do you have the
Ah, I obviously missed that.
Quoted 1 line
hardware? If so, that's good, we know you can provide tested fix.
Well, yes, I know somebody who has smsc95xx hardware ;-).

Quoted 8 lines

Quoted 4 lines
Quoted 1 line
Mike, are you good with using this fix instead or can you two negotiate?
Let's await Mike's answer (now on CC as well). I need to re-send the series anyway due to a stupid typo.
mine might take a little longer due to wrangling with wolfgang. i'd suggest we go with your patch (although i have feedback on it).
-mike

Well, I have no problem with your patch if it fixes the same issue correctly... and preparing and testing a new patch series takes time. I will simply add my "Tested-by" when I have feedback and we then can drop my other patches.

Wolfgang.

Mike Frysinger

2011-11-14 16:50 UTC #6 lore raw

PATCH 1/3 smsc95xx: Fix MAC address programming

Quoted 8 lines · On Monday 14 November 2011 03:25:37 Wolfgang Grandegger wrote:

Quoted 2 lines · Mike Frysinger
mine might take a little longer due to wrangling with wolfgang. i'd suggest we go with your patch (although i have feedback on it).
Well, I have no problem with your patch if it fixes the same issue correctly... and preparing and testing a new patch series takes time. I will simply add my "Tested-by" when I have feedback and we then can drop my other patches.

let's not go with the one i submitted earlier. tweak your patches with the feedback posted and we'll go with those.
-mike
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 836 bytes
Desc: This is a digitally signed message part.
Url : http://lists.denx.de/pipermail/u-boot/attachments/20111114/d32c6d25/attachment.pgp

Wolfgang Grandegger

2011-11-15 09:25 UTC #7 lore raw

PATCH 1/3 smsc95xx: Fix MAC address programming

Quoted 5 lines · Mike Frysinger

Quoted 4 lines · On Monday 14 November 2011 03:25:37 Wolfgang Grandegger wrote:
Quoted 2 lines · Mike Frysinger
mine might take a little longer due to wrangling with wolfgang. i'd suggest we go with your patch (although i have feedback on it).

Quoted 7 lines

Quoted 4 lines
Well, I have no problem with your patch if it fixes the same issue correctly... and preparing and testing a new patch series takes time. I will simply add my "Tested-by" when I have feedback and we then can drop my other patches.
let's not go with the one i submitted earlier. tweak your patches with the feedback posted and we'll go with those.

OK, just sent v2.

Wolfgang.

Mike Frysinger

2011-11-11 15:20 UTC #8 lore raw

PATCH 1/3 smsc95xx: Fix MAC address programming

In reply to Wolfgang Grandegger

drivers/usb/eth/smsc95xx.c static int smsc95xx_init_mac_address(struct eth_device *eth, · line 377

Quoted 11 lines · On Friday 11 November 2011 05:59:56 Wolfgang Grandegger wrote:

Commit 79ad54400932d6484178a372fb3b659e3437473b broke the MAC address programming. Fix this by using the method from Linux' smsc95xx_set_mac_address().
drivers/usb/eth/smsc95xx.c +3 -1
 
-	u32 addr_lo, addr_hi;
+	u32 addr_lo = eth->enetaddr[0] | eth->enetaddr[1] << 8 |
+		eth->enetaddr[2] << 16 | eth->enetaddr[3] << 24;
+	u32 addr_hi = eth->enetaddr[4] | eth->enetaddr[5] << 8;

please use:

	#include <asm/unaligned.h>
	u32 addr_lo = __get_unaligned_le32(&eth->enetaddr[0]);
	u32 addr_hi = __get_unaligned_le16(&eth->enetaddr[4]);

-mike
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 836 bytes
Desc: This is a digitally signed message part.
Url : http://lists.denx.de/pipermail/u-boot/attachments/20111111/1228de54/attachment.pgp

Mike Frysinger

2011-11-11 15:22 UTC #9 lore raw

PATCH 1/3 smsc95xx: Fix MAC address programming

In reply to Wolfgang Grandegger

Quoted 3 lines · On Friday 11 November 2011 05:59:56 Wolfgang Grandegger wrote:

Commit 79ad54400932d6484178a372fb3b659e3437473b broke the MAC address programming. Fix this by using the method from Linux' smsc95xx_set_mac_address().

oh, and while you're here, test out this fix:
- debug("MAC %02x:%02x:%02x:%02x:%02x:%02x\n",

-       eth->enetaddr[0], eth->enetaddr[1],
-       eth->enetaddr[2], eth->enetaddr[3],
-       eth->enetaddr[4], eth->enetaddr[5]);

+ debug("MAC %pM\n", eth->enetaddr);
-mike
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 836 bytes
Desc: This is a digitally signed message part.
Url : http://lists.denx.de/pipermail/u-boot/attachments/20111111/c34a3d49/attachment.pgp

Wolfgang Grandegger

author

2011-11-11 10:59 UTC #10 lore raw

PATCH 2/3 smsc95xx: in smsc95xx_set_multicast write to reg

In reply to Wolfgang Grandegger

The write to the mac_cr register was missing. This usually not cause an issue before, since the next function writing the register's shadow copy into the register would do it as a side effect.

Signed-off-by Wolfgang Grandegger <[email protected]>

Cc Simon Glass <[email protected]>

---
 drivers/usb/eth/smsc95xx.c |    2 ++
 1 files changed, 2 insertions(+), 0 deletions(-)

drivers/usb/eth/smsc95xx.c +2 -0

index eb529f1..16e24bd 100644
@@ -428,6 +428,8 @@ static void smsc95xx_set_multicast(struct ueth_data *dev)
 {
 	/* No multicast in u-boot */
 	dev->mac_cr &= ~(MAC_CR_PRMS_ | MAC_CR_MCPAS_ | MAC_CR_HPFILT_);
+
+	smsc95xx_write_reg(dev, MAC_CR, dev->mac_cr);
 }
 
 /* starts the TX path */

Signature · 1 line

1.7.4.1

Simon Glass

2011-11-11 15:35 UTC #11 lore raw

PATCH 2/3 smsc95xx: in smsc95xx_set_multicast write to reg

Hi Wolfgang,

Quoted 17 lines · Wolfgang Grandegger

The write to the mac_cr register was missing. This usually not cause an issue before, since the next function writing the register's shadow copy into the register would do it as a side effect.
Signed-off-by Wolfgang Grandegger <[email protected]>
Cc Simon Glass <[email protected]>
---
?drivers/usb/eth/smsc95xx.c | ? ?2 ++
?1 files changed, 2 insertions(+), 0 deletions(-)
drivers/usb/eth/smsc95xx.c +0 -0
index eb529f1..16e24bd 100644
@@ -428,6 +428,8 @@ static void smsc95xx_set_multicast(struct ueth_data *dev)
 ?{

Quoted 4 lines

? ? ? ?/* No multicast in u-boot */
? ? ? ?dev->mac_cr &= ~(MAC_CR_PRMS_ | MAC_CR_MCPAS_ | MAC_CR_HPFILT_); +
+ ? ? ? smsc95xx_write_reg(dev, MAC_CR, dev->mac_cr);

It seems a bit odd - what problem does your addition actually fix? At present both smsc95xx_start_tx_path() and smsc95xx_start_rx_path() write to this register, so it will be written three times in quick succession. If there are no other callers to smsc95xx_set_multicast() outside init, then perhaps we should just write it once in init, after calling the three functions?

Regards,
Simon

Quoted 7 lines

?}
?/* starts the TX path */
--
1.7.4.1

Wolfgang Grandegger

2011-11-14 12:18 UTC #12 lore raw

PATCH 2/3 smsc95xx: in smsc95xx_set_multicast write to reg

Hi Simon,

Quoted 23 lines · Simon Glass

Hi Wolfgang,
Quoted 20 lines · Wolfgang Grandegger
The write to the mac_cr register was missing. This usually not cause an issue before, since the next function writing the register's shadow copy into the register would do it as a side effect.
Signed-off-by Wolfgang Grandegger <[email protected]>
Cc Simon Glass <[email protected]>
---
 drivers/usb/eth/smsc95xx.c |    2 ++
 1 files changed, 2 insertions(+), 0 deletions(-)
drivers/usb/eth/smsc95xx.c +1 -0
index eb529f1..16e24bd 100644
@@ -428,6 +428,8 @@ static void smsc95xx_set_multicast(struct ueth_data *dev)
 {
       /* No multicast in u-boot */
       dev->mac_cr &= ~(MAC_CR_PRMS_ | MAC_CR_MCPAS_ | MAC_CR_HPFILT_);
+

Quoted 8 lines

drivers/usb/eth/smsc95xx.c static void smsc95xx_set_multicast(struct ueth_data *dev) · line 432
Quoted 1 line
drivers/usb/eth/smsc95xx.c +1 -0
+       smsc95xx_write_reg(dev, MAC_CR, dev->mac_cr);
It seems a bit odd - what problem does your addition actually fix? At present both smsc95xx_start_tx_path() and smsc95xx_start_rx_path() write to this register, so it will be written three times in quick succession. If there are no other callers to smsc95xx_set_multicast() outside init, then perhaps we should just write it once in init, after calling the three functions?

The name "set_multicast" associated to me that the relevant register is really set. But from the functional poit of few, it is not necessary. Well, it's a minor issue and maybe not worth fixing. So, let's drop this patch.

Wolfgang.

Simon Glass

2011-11-14 20:06 UTC #13 lore raw

PATCH 2/3 smsc95xx: in smsc95xx_set_multicast write to reg

Hi Wolfgang,

Quoted 35 lines · Wolfgang Grandegger

Hi Simon,
Quoted 23 lines · Simon Glass
Hi Wolfgang,
Quoted 20 lines · Wolfgang Grandegger
The write to the mac_cr register was missing. This usually not cause an issue before, since the next function writing the register's shadow copy into the register would do it as a side effect.
Signed-off-by Wolfgang Grandegger <[email protected]>
Cc Simon Glass <[email protected]>
---
?drivers/usb/eth/smsc95xx.c | ? ?2 ++
?1 files changed, 2 insertions(+), 0 deletions(-)
drivers/usb/eth/smsc95xx.c +0 -0
index eb529f1..16e24bd 100644
@@ -428,6 +428,8 @@ static void smsc95xx_set_multicast(struct ueth_data *dev)
 ?{
? ? ? ?/* No multicast in u-boot */
? ? ? ?dev->mac_cr &= ~(MAC_CR_PRMS_ | MAC_CR_MCPAS_ | MAC_CR_HPFILT_); +
Quoted 8 lines
Quoted 1 line
+ ? ? ? smsc95xx_write_reg(dev, MAC_CR, dev->mac_cr);
It seems a bit odd - what problem does your addition actually fix? At present both smsc95xx_start_tx_path() and smsc95xx_start_rx_path() write to this register, so it will be written three times in quick succession. If there are no other callers to smsc95xx_set_multicast() outside init, then perhaps we should just write it once in init, after calling the three functions?

Quoted 2 lines

The name "set_multicast" associated to me that the relevant register is really set.

Yes, although it is a static function.

Quoted 3 lines

But from the functional poit of few, it is not necessary. Well, it's a minor issue and maybe not worth fixing. So, let's drop this patch.

If you like, although I agree a clean-up would be useful here.

Regards,
Simon

Quoted 3 lines

Wolfgang.

Wolfgang Grandegger

author

2011-11-11 10:59 UTC #14 lore raw

PATCH 3/3 smsc95xx: remove an unecessary debug messages

In reply to Wolfgang Grandegger

Signed-off-by Wolfgang Grandegger <[email protected]>

Cc Simon Glass <[email protected]>

---
 drivers/usb/eth/smsc95xx.c |    7 +++----
 1 files changed, 3 insertions(+), 4 deletions(-)

drivers/usb/eth/smsc95xx.c +3 -4

index 16e24bd..40f7f5b 100644
@@ -380,15 +380,14 @@ static int smsc95xx_write_hwaddr(struct eth_device *eth)
 	/* set hardware address */
 	debug("** %s()\n", __func__);
 
-	ret = smsc95xx_write_reg(dev, ADDRL, addr_lo);
-	if (ret < 0) {
-		debug("Failed to write ADDRL: %d\n", ret);
+	ret = smsc95xx_write_reg(dev, ADDRL, adddr_lo);
+	if (ret < 0)
 		return ret;
-	}
 
 	ret = smsc95xx_write_reg(dev, ADDRH, addr_hi);
 	if (ret < 0)
 		return ret;
+
 	debug("MAC %02x:%02x:%02x:%02x:%02x:%02x\n",
 		eth->enetaddr[0], eth->enetaddr[1],
 		eth->enetaddr[2], eth->enetaddr[3],

Signature · 1 line

1.7.4.1

Simon Glass

2011-11-11 15:26 UTC #15 lore raw

PATCH 3/3 smsc95xx: remove an unecessary debug messages

Hi Wolfgang,

Quoted 14 lines · Wolfgang Grandegger

Signed-off-by Wolfgang Grandegger <[email protected]>
Cc Simon Glass <[email protected]>
---
?drivers/usb/eth/smsc95xx.c | ? ?7 +++----
?1 files changed, 3 insertions(+), 4 deletions(-)
drivers/usb/eth/smsc95xx.c +0 -0
index 16e24bd..40f7f5b 100644
@@ -380,15 +380,14 @@ static int smsc95xx_write_hwaddr(struct eth_device *eth)
? ? ? ?/* set hardware address */
? ? ? ?debug("** %s()\n", __func__);

Quoted 4 lines

- ? ? ? ret = smsc95xx_write_reg(dev, ADDRL, addr_lo);
- ? ? ? if (ret < 0) {
- ? ? ? ? ? ? ? debug("Failed to write ADDRL: %d\n", ret);
+ ? ? ? ret = smsc95xx_write_reg(dev, ADDRL, adddr_lo);

addr_lo?

Quoted 8 lines

+ ? ? ? if (ret < 0)
? ? ? ? ? ? ? ?return ret;
- ? ? ? }
? ? ? ?ret = smsc95xx_write_reg(dev, ADDRH, addr_hi); ? ? ? ?if (ret < 0)
? ? ? ? ? ? ? ?return ret;
+

Since you are adding a blank line here, you need to fix the line below :-) See Mike's email...

Quoted 7 lines

? ? ? ?debug("MAC %02x:%02x:%02x:%02x:%02x:%02x\n", ? ? ? ? ? ? ? ?eth->enetaddr[0], eth->enetaddr[1], ? ? ? ? ? ? ? ?eth->enetaddr[2], eth->enetaddr[3], --
1.7.4.1

Regards,
Simon

Wolfgang Grandegger

2011-11-14 08:33 UTC #16 lore raw

PATCH 3/3 smsc95xx: remove an unecessary debug messages

Quoted 17 lines · Simon Glass

Hi Wolfgang,
Quoted 14 lines · Wolfgang Grandegger
Signed-off-by Wolfgang Grandegger <[email protected]>
Cc Simon Glass <[email protected]>
---
 drivers/usb/eth/smsc95xx.c |    7 +++----
 1 files changed, 3 insertions(+), 4 deletions(-)
drivers/usb/eth/smsc95xx.c +0 -0
index 16e24bd..40f7f5b 100644
@@ -380,15 +380,14 @@ static int smsc95xx_write_hwaddr(struct eth_device *eth)
       /* set hardware address */
       debug("** %s()\n", __func__);
 

Quoted 6 lines

drivers/usb/eth/smsc95xx.c static int smsc95xx_write_hwaddr(struct eth_device *eth) · line 383
Quoted 4 lines
drivers/usb/eth/smsc95xx.c +1 -3
-       ret = smsc95xx_write_reg(dev, ADDRL, addr_lo);
-       if (ret < 0) {
-               debug("Failed to write ADDRL: %d\n", ret);
+       ret = smsc95xx_write_reg(dev, ADDRL, adddr_lo);
addr_lo?

Oh yeah, sneaked in somehow, sorry ... I'm now focusing on Mike's patch fixing the same issue. Hope to have some feedback later today.

Wolfgang.