Skip to contentSkip to conversation

net: asix: fix uninit value in asix_mdio_read

PATCH Superseded Pavel Skripkin 4 participants 12 messages Aug 13, 2021 – Aug 18, 2021

Subsystems (4)
lore
Share & tools
lore
#0 lore raw

PATCH 1/1 net: asix: fix uninit value in asix_mdio_read

Syzbot reported uninit-value in asix_mdio_read(). The problem was in missing error handling. asix_read_cmd() should initialize passed stack variable smsr, but it can fail in some cases. Then while condidition checks possibly uninit smsr variable.

Since smsr is uninitialized stack variable, driver can misbehave, because smsr will be random in case of asix_read_cmd() failure. Fix it by adding error cheking and just continue the loop instead of checking uninit value.

Fixes 8a46f665833a ("net: asix: Avoid looping when the device is disconnected")
Reported-and-tested-by [email protected]
---
 drivers/net/usb/asix_common.c | 8 +++++++-
 1 file changed, 7 insertions(+), 1 deletion(-)
drivers/net/usb/asix_common.c +7 -1
index ac92bc52a85e..572ca3077f8f 100644
@@ -479,7 +479,13 @@ int asix_mdio_read(struct net_device *netdev, int phy_id, int loc)
 		usleep_range(1000, 1100);
 		ret = asix_read_cmd(dev, AX_CMD_STATMNGSTS_REG,
 				    0, 0, 1, &smsr, 0);
-	} while (!(smsr & AX_HOST_EN) && (i++ < 30) && (ret != -ENODEV));
+		if (ret == -ENODEV) {
+			break;
+		} else if (ret < 0) {
+			++i;
+			continue;
+		}
+	} while (!(smsr & AX_HOST_EN) && (i++ < 30));
 	if (ret == -ENODEV || ret == -ETIMEDOUT) {
 		mutex_unlock(&dev->phy_mutex);
 		return ret;
Signature · 1 line
2.32.0
#1 lore raw
Quoted 23 lines · Pavel Skripkin

Syzbot reported uninit-value in asix_mdio_read(). The problem was in missing error handling. asix_read_cmd() should initialize passed stack variable smsr, but it can fail in some cases. Then while condidition checks possibly uninit smsr variable.

Since smsr is uninitialized stack variable, driver can misbehave, because smsr will be random in case of asix_read_cmd() failure. Fix it by adding error cheking and just continue the loop instead of checking uninit value.

Fixes 8a46f665833a ("net: asix: Avoid looping when the device is disconnected")
Reported-and-tested-by [email protected]
---
 drivers/net/usb/asix_common.c | 8 +++++++-
 1 file changed, 7 insertions(+), 1 deletion(-)
drivers/net/usb/asix_common.c +0 -0
index ac92bc52a85e..572ca3077f8f 100644
@@ -479,7 +479,13 @@ int asix_mdio_read(struct net_device *netdev, int phy_id, int loc)
 		usleep_range(1000, 1100);

drivers/net/usb/asix_common.c int asix_mdio_read(struct net_device *netdev, int phy_id, int loc) · line 488

Quoted 10 lines
drivers/net/usb/asix_common.c +7 -1
 		ret = asix_read_cmd(dev, AX_CMD_STATMNGSTS_REG,
 				    0, 0, 1, &smsr, 0);
-	} while (!(smsr & AX_HOST_EN) && (i++ < 30) && (ret != -ENODEV));
+		if (ret == -ENODEV) {
+			break;
+		} else if (ret < 0) {
+			++i;
+			continue;
+		}
+	} while (!(smsr & AX_HOST_EN) && (i++ < 30));

No ret < 0, don't you end up with a double increment of i? So it will only retry 15 times, not 30?

Humm.

If ret < 0 is true, smsr is uninitialized? The continue statement causes a jump into the condition expression, where we evaluate smsr & AX_HOST_EN. Isn't this just as broken as the original version?

Andrew

#2 lore raw
Quoted 38 lines · Andrew Lunn
Quoted 23 lines · Pavel Skripkin

Syzbot reported uninit-value in asix_mdio_read(). The problem was in missing error handling. asix_read_cmd() should initialize passed stack variable smsr, but it can fail in some cases. Then while condidition checks possibly uninit smsr variable.

Since smsr is uninitialized stack variable, driver can misbehave, because smsr will be random in case of asix_read_cmd() failure. Fix it by adding error cheking and just continue the loop instead of checking uninit value.

Fixes 8a46f665833a ("net: asix: Avoid looping when the device is disconnected")
Reported-and-tested-by [email protected]
---
 drivers/net/usb/asix_common.c | 8 +++++++-
 1 file changed, 7 insertions(+), 1 deletion(-)
drivers/net/usb/asix_common.c +0 -0
index ac92bc52a85e..572ca3077f8f 100644
@@ -479,7 +479,13 @@ int asix_mdio_read(struct net_device *netdev, int phy_id, int loc)
 		usleep_range(1000, 1100);

drivers/net/usb/asix_common.c int asix_mdio_read(struct net_device *netdev, int phy_id, int loc) · line 488

Quoted 10 lines
drivers/net/usb/asix_common.c +7 -1
 		ret = asix_read_cmd(dev, AX_CMD_STATMNGSTS_REG,
 				    0, 0, 1, &smsr, 0);
-	} while (!(smsr & AX_HOST_EN) && (i++ < 30) && (ret != -ENODEV));
+		if (ret == -ENODEV) {
+			break;
+		} else if (ret < 0) {
+			++i;
+			continue;
+		}
+	} while (!(smsr & AX_HOST_EN) && (i++ < 30));

No ret < 0, don't you end up with a double increment of i? So it will only retry 15 times, not 30?

Quoted 8 lines

Humm.

If ret < 0 is true, smsr is uninitialized? The continue statement causes a jump into the condition expression, where we evaluate smsr & AX_HOST_EN. Isn't this just as broken as the original version?

Andrew

Yes, you are right, I missed that, sorry. I will rewrote this loop into for loop in v2.

Im wondering why this wrong patch passed KMSAN testing...

With regards,
Pavel Skripkin

#3 lore raw

PATCH v2 1/1 net: asix: fix uninit value in asix_mdio_read

In reply to Andrew Lunn

Syzbot reported uninit-value in asix_mdio_read(). The problem was in missing error handling. asix_read_cmd() should initialize passed stack variable smsr, but it can fail in some cases. Then while condidition checks possibly uninit smsr variable.

Since smsr is uninitialized stack variable, driver can misbehave, because smsr will be random in case of asix_read_cmd() failure. Fix it by adding error cheking and just continue the loop instead of checking uninit value.

Fixes 8a46f665833a ("net: asix: Avoid looping when the device is disconnected")
Reported-by [email protected]

---

Changes in v2:
1. Fixed previous wrong approach and changed while loop to for loop
2. Reported-and-tested-by: tag removed, since KMSAN tests can be false positive. Used Reported-by instead.

---
 drivers/net/usb/asix_common.c | 13 ++++++++++---
 1 file changed, 10 insertions(+), 3 deletions(-)
drivers/net/usb/asix_common.c +10 -3
index ac92bc52a85e..7019c25e591c 100644
@@ -468,18 +468,25 @@ int asix_mdio_read(struct net_device *netdev, int phy_id, int loc)
 	struct usbnet *dev = netdev_priv(netdev);
 	__le16 res;
 	u8 smsr;
-	int i = 0;
+	int i;
 	int ret;
 
 	mutex_lock(&dev->phy_mutex);
-	do {
+	for (i = 0; i < 30; ++i) {
 		ret = asix_set_sw_mii(dev, 0);
 		if (ret == -ENODEV || ret == -ETIMEDOUT)
 			break;
 		usleep_range(1000, 1100);
 		ret = asix_read_cmd(dev, AX_CMD_STATMNGSTS_REG,
 				    0, 0, 1, &smsr, 0);
-	} while (!(smsr & AX_HOST_EN) && (i++ < 30) && (ret != -ENODEV));
+		if (ret == -ENODEV)
+			break;
+		else if (ret < 0)
+			continue;
+		else if (smsr & AX_HOST_EN)
+			break;
+	}
+
 	if (ret == -ENODEV || ret == -ETIMEDOUT) {
 		mutex_unlock(&dev->phy_mutex);
 		return ret;
Signature · 1 line
2.32.0
maintainer
#4 lore raw
Quoted 11 lines · On Sat, 14 Aug 2021 01:42:19 +0300 Pavel Skripkin wrote:

Syzbot reported uninit-value in asix_mdio_read(). The problem was in missing error handling. asix_read_cmd() should initialize passed stack variable smsr, but it can fail in some cases. Then while condidition checks possibly uninit smsr variable.

Since smsr is uninitialized stack variable, driver can misbehave, because smsr will be random in case of asix_read_cmd() failure. Fix it by adding error cheking and just continue the loop instead of checking uninit value.

Fixes 8a46f665833a ("net: asix: Avoid looping when the device is disconnected")

This is not the right tag, the Fixes tag should point to the commit where the problem is introduced. Robert/Vincent only added some error checking, the issue was there before, right?

Once you locate the right starting point for the fix please make sure to add the author to CC.

Thanks!

Quoted 11 lines
 drivers/net/usb/asix_common.c | 13 ++++++++++---
 1 file changed, 10 insertions(+), 3 deletions(-)
drivers/net/usb/asix_common.c +0 -0
index ac92bc52a85e..7019c25e591c 100644

drivers/net/usb/asix_common.c int asix_mdio_read(struct net_device *netdev, int phy_id, int loc) · line 472

Quoted 6 lines
drivers/net/usb/asix_common.c +1 -1
@@ -468,18 +468,25 @@ int asix_mdio_read(struct net_device *netdev, int phy_id, int loc)
 	struct usbnet *dev = netdev_priv(netdev);
 	__le16 res;
 	u8 smsr;
-	int i = 0;
+	int i;
 	int ret;

nit: move i after ret, reverse xmas tree style

Quoted 11 lines
drivers/net/usb/asix_common.c +2 -2
 	mutex_lock(&dev->phy_mutex);
-	do {
+	for (i = 0; i < 30; ++i) {
 		ret = asix_set_sw_mii(dev, 0);
 		if (ret == -ENODEV || ret == -ETIMEDOUT)
 			break;
 		usleep_range(1000, 1100);
 		ret = asix_read_cmd(dev, AX_CMD_STATMNGSTS_REG,
 				    0, 0, 1, &smsr, 0);
-	} while (!(smsr & AX_HOST_EN) && (i++ < 30) && (ret != -ENODEV));
+		if (ret == -ENODEV)

drivers/net/usb/asix_common.c int asix_mdio_read(struct net_device *netdev, int phy_id, int loc) · line 492

Quoted 10 lines
drivers/net/usb/asix_common.c +7 -0
+			break;
+		else if (ret < 0)
+			continue;
+		else if (smsr & AX_HOST_EN)
+			break;
+	}
+
 	if (ret == -ENODEV || ret == -ETIMEDOUT) {
 		mutex_unlock(&dev->phy_mutex);
 		return ret;

Code LGTM, do other functions which Robert/Vincent touched not need the same treatment tho?

#5 lore raw

PATCH v3 1/1 net: asix: fix uninit value bugs

Syzbot reported uninit-value in asix_mdio_read(). The problem was in missing error handling. asix_read_cmd() should initialize passed stack variable smsr, but it can fail in some cases. Then while condidition checks possibly uninit smsr variable.

Since smsr is uninitialized stack variable, driver can misbehave, because smsr will be random in case of asix_read_cmd() failure. Fix it by adding error handling and just continue the loop instead of checking uninit value.

Also, same loop was used in 3 other functions. Fixed uninit value bug in them too.

Cc Robert Foss <[email protected]>
Fixes d9fe64e51114 ("net: asix: Add in_pm parameter")
Reported-by [email protected]

---

Changes in v2:
1. Fixed previous wrong approach and changed while loop to for loop
2. Reported-and-tested-by: tag removed, since KMSAN tests can be false positive. Used Reported-by instead.

Changes in v3:
1. Addressed uninit value bugs in asix_mdio_write(), asix_mdio_read_nopm() and asix_mdio_write_nopm()
2. Moved i after ret to reverse xmas tree style
3. Fixed Fixes: tag

---
 drivers/net/usb/asix_common.c | 51 ++++++++++++++++++++++++++---------
 1 file changed, 39 insertions(+), 12 deletions(-)
drivers/net/usb/asix_common.c +39 -12
index ac92bc52a85e..f0ee35edb746 100644
@@ -468,18 +468,25 @@ int asix_mdio_read(struct net_device *netdev, int phy_id, int loc)
 	struct usbnet *dev = netdev_priv(netdev);
 	__le16 res;
 	u8 smsr;
-	int i = 0;
 	int ret;
+	int i;
 
 	mutex_lock(&dev->phy_mutex);
-	do {
+	for (i = 0; i < 30; ++i) {
 		ret = asix_set_sw_mii(dev, 0);
 		if (ret == -ENODEV || ret == -ETIMEDOUT)
 			break;
 		usleep_range(1000, 1100);
 		ret = asix_read_cmd(dev, AX_CMD_STATMNGSTS_REG,
 				    0, 0, 1, &smsr, 0);
-	} while (!(smsr & AX_HOST_EN) && (i++ < 30) && (ret != -ENODEV));
+		if (ret == -ENODEV)
+			break;
+		else if (ret < 0)
+			continue;
+		else if (smsr & AX_HOST_EN)
+			break;
+	}
+
 	if (ret == -ENODEV || ret == -ETIMEDOUT) {
 		mutex_unlock(&dev->phy_mutex);
 		return ret;
@@ -506,21 +513,27 @@ static int __asix_mdio_write(struct net_device *netdev, int phy_id, int loc,
 	struct usbnet *dev = netdev_priv(netdev);
 	__le16 res = cpu_to_le16(val);
 	u8 smsr;
-	int i = 0;
 	int ret;
+	int i;
 
 	netdev_dbg(dev->net, "asix_mdio_write() phy_id=0x%02x, loc=0x%02x, val=0x%04x\n",
 			phy_id, loc, val);
 
 	mutex_lock(&dev->phy_mutex);
-	do {
+	for (i = 0; i < 30; ++i) {
 		ret = asix_set_sw_mii(dev, 0);
 		if (ret == -ENODEV)
 			break;
 		usleep_range(1000, 1100);
 		ret = asix_read_cmd(dev, AX_CMD_STATMNGSTS_REG,
 				    0, 0, 1, &smsr, 0);
-	} while (!(smsr & AX_HOST_EN) && (i++ < 30) && (ret != -ENODEV));
+		if (ret == -ENODEV)
+			break;
+		else if (ret < 0)
+			continue;
+		else if (smsr & AX_HOST_EN)
+			break;
+	}
 
 	if (ret == -ENODEV)
 		goto out;
@@ -562,18 +575,25 @@ int asix_mdio_read_nopm(struct net_device *netdev, int phy_id, int loc)
 	struct usbnet *dev = netdev_priv(netdev);
 	__le16 res;
 	u8 smsr;
-	int i = 0;
 	int ret;
+	int i;
 
 	mutex_lock(&dev->phy_mutex);
-	do {
+	for (i = 0; i < 30; ++i) {
 		ret = asix_set_sw_mii(dev, 1);
 		if (ret == -ENODEV || ret == -ETIMEDOUT)
 			break;
 		usleep_range(1000, 1100);
 		ret = asix_read_cmd(dev, AX_CMD_STATMNGSTS_REG,
 				    0, 0, 1, &smsr, 1);
-	} while (!(smsr & AX_HOST_EN) && (i++ < 30) && (ret != -ENODEV));
+		if (ret == -ENODEV)
+			break;
+		else if (ret < 0)
+			continue;
+		else if (smsr & AX_HOST_EN)
+			break;
+	}
+
 	if (ret == -ENODEV || ret == -ETIMEDOUT) {
 		mutex_unlock(&dev->phy_mutex);
 		return ret;
@@ -596,21 +616,28 @@ asix_mdio_write_nopm(struct net_device *netdev, int phy_id, int loc, int val)
 	struct usbnet *dev = netdev_priv(netdev);
 	__le16 res = cpu_to_le16(val);
 	u8 smsr;
-	int i = 0;
 	int ret;
+	int i;
 
 	netdev_dbg(dev->net, "asix_mdio_write() phy_id=0x%02x, loc=0x%02x, val=0x%04x\n",
 			phy_id, loc, val);
 
 	mutex_lock(&dev->phy_mutex);
-	do {
+	for (i = 0; i < 30; ++i) {
 		ret = asix_set_sw_mii(dev, 1);
 		if (ret == -ENODEV)
 			break;
 		usleep_range(1000, 1100);
 		ret = asix_read_cmd(dev, AX_CMD_STATMNGSTS_REG,
 				    0, 0, 1, &smsr, 1);
-	} while (!(smsr & AX_HOST_EN) && (i++ < 30) && (ret != -ENODEV));
+		if (ret == -ENODEV)
+			break;
+		else if (ret < 0)
+			continue;
+		else if (smsr & AX_HOST_EN)
+			break;
+	}
+
 	if (ret == -ENODEV) {
 		mutex_unlock(&dev->phy_mutex);
 		return;
Signature · 1 line
2.32.0
#6 lore raw

Re: [PATCH v3] net: asix: fix uninit value bugs

Quoted 5 lines · Pavel Skripkin

Syzbot reported uninit-value in asix_mdio_read(). The problem was in missing error handling. asix_read_cmd() should initialize passed stack variable smsr, but it can fail in some cases. Then while condidition checks possibly uninit smsr variable.

Quoted 7 lines

Since smsr is uninitialized stack variable, driver can misbehave, because smsr will be random in case of asix_read_cmd() failure. Fix it by adding error handling and just continue the loop instead of checking uninit value.

Also, same loop was used in 3 other functions. Fixed uninit value bug in them too.

Hi Pavel

Which suggests it might make sense to refactor the code to make a helper? I will leave you to decide if you want to do that.

The code does looks correct now.

Andrew

#7 lore raw

Re: [PATCH v3] net: asix: fix uninit value bugs

Quoted 11 lines · Andrew Lunn
Quoted 10 lines · Pavel Skripkin

Syzbot reported uninit-value in asix_mdio_read(). The problem was in missing error handling. asix_read_cmd() should initialize passed stack variable smsr, but it can fail in some cases. Then while condidition checks possibly uninit smsr variable.

Since smsr is uninitialized stack variable, driver can misbehave, because smsr will be random in case of asix_read_cmd() failure. Fix it by adding error handling and just continue the loop instead of checking uninit value.

Quoted 7 lines
Quoted 2 lines

Also, same loop was used in 3 other functions. Fixed uninit value bug in them too.

Hi Pavel

Which suggests it might make sense to refactor the code to make a helper? I will leave you to decide if you want to do that.

It makes sense. Will add a helper function in v4. Thank you for suggestion and review!

Quoted 3 lines

The code does looks correct now.

With regards,
Pavel Skripkin

#8 lore raw

Re: [PATCH v3] net: asix: fix uninit value bugs

In reply to Andrew Lunn
Quoted 14 lines · Andrew Lunn
Quoted 5 lines · Pavel Skripkin

Syzbot reported uninit-value in asix_mdio_read(). The problem was in missing error handling. asix_read_cmd() should initialize passed stack variable smsr, but it can fail in some cases. Then while condidition checks possibly uninit smsr variable.

Quoted 7 lines

Since smsr is uninitialized stack variable, driver can misbehave, because smsr will be random in case of asix_read_cmd() failure. Fix it by adding error handling and just continue the loop instead of checking uninit value.

Also, same loop was used in 3 other functions. Fixed uninit value bug in them too.

Quoted 9 lines

Hi Pavel

Which suggests it might make sense to refactor the code to make a helper? I will leave you to decide if you want to do that.

The code does looks correct now.

Andrew

I noticed strange thing. For example: driver looped 30 times, there wasn't any errors with usb transfer, but Host_En bit is not set. Datasheet says, that if Host_En is not set, that means software access will be ignored. Driver code doesn't handle this situation. We only check if ret is -ENODEV or -ETIMEOUT.

I guess, next register access will fail, but anyway, does it make sense to return when Host_En bit is not set?

With regards,
Pavel Skripkin

#9 lore raw

PATCH v4 1/1 net: asix: fix uninit value bugs

In reply to Andrew Lunn

Syzbot reported uninit-value in asix_mdio_read(). The problem was in missing error handling. asix_read_cmd() should initialize passed stack variable smsr, but it can fail in some cases. Then while condidition checks possibly uninit smsr variable.

Since smsr is uninitialized stack variable, driver can misbehave, because smsr will be random in case of asix_read_cmd() failure. Fix it by adding error handling and just continue the loop instead of checking uninit value.

Added helper function for checking Host_En bit, since wrong loop was used in 4 functions and there is no need in copy-pasting code parts.

Cc Robert Foss <[email protected]>
Fixes d9fe64e51114 ("net: asix: Add in_pm parameter")
Reported-by [email protected]

---

Changes in v2:
1. Fixed previous wrong approach and changed while loop to for loop
2. Reported-and-tested-by: tag removed, since KMSAN tests can be false positive. Used Reported-by instead.

Changes in v3:
1. Addressed uninit value bugs in asix_mdio_write(), asix_mdio_read_nopm() and asix_mdio_write_nopm()
2. Moved i after ret to reverse xmas tree style
3. Fixed Fixes: tag

Changes in v4:
1. Added helper for checking Host_En bit, since wrong loop was used in 4 functions. (Suggested by Andrew Lunn)

---
 drivers/net/usb/asix_common.c | 70 +++++++++++++++--------------------
 1 file changed, 30 insertions(+), 40 deletions(-)
drivers/net/usb/asix_common.c +30 -40
index ac92bc52a85e..38cda590895c 100644
@@ -63,6 +63,29 @@ void asix_write_cmd_async(struct usbnet *dev, u8 cmd, u16 value, u16 index,
 			       value, index, data, size);
 }
 
+static int asix_check_host_enable(struct usbnet *dev, int in_pm)
+{
+	int i, ret;
+	u8 smsr;
+
+	for (i = 0; i < 30; ++i) {
+		ret = asix_set_sw_mii(dev, in_pm);
+		if (ret == -ENODEV || ret == -ETIMEDOUT)
+			break;
+		usleep_range(1000, 1100);
+		ret = asix_read_cmd(dev, AX_CMD_STATMNGSTS_REG,
+				    0, 0, 1, &smsr, in_pm);
+		if (ret == -ENODEV)
+			break;
+		else if (ret < 0)
+			continue;
+		else if (smsr & AX_HOST_EN)
+			break;
+	}
+
+	return ret;
+}
+
 static void reset_asix_rx_fixup_info(struct asix_rx_fixup_info *rx)
 {
 	/* Reset the variables that have a lifetime outside of
@@ -467,19 +490,11 @@ int asix_mdio_read(struct net_device *netdev, int phy_id, int loc)
 {
 	struct usbnet *dev = netdev_priv(netdev);
 	__le16 res;
-	u8 smsr;
-	int i = 0;
 	int ret;
 
 	mutex_lock(&dev->phy_mutex);
-	do {
-		ret = asix_set_sw_mii(dev, 0);
-		if (ret == -ENODEV || ret == -ETIMEDOUT)
-			break;
-		usleep_range(1000, 1100);
-		ret = asix_read_cmd(dev, AX_CMD_STATMNGSTS_REG,
-				    0, 0, 1, &smsr, 0);
-	} while (!(smsr & AX_HOST_EN) && (i++ < 30) && (ret != -ENODEV));
+
+	ret = asix_check_host_enable(dev, 0);
 	if (ret == -ENODEV || ret == -ETIMEDOUT) {
 		mutex_unlock(&dev->phy_mutex);
 		return ret;
@@ -505,23 +520,14 @@ static int __asix_mdio_write(struct net_device *netdev, int phy_id, int loc,
 {
 	struct usbnet *dev = netdev_priv(netdev);
 	__le16 res = cpu_to_le16(val);
-	u8 smsr;
-	int i = 0;
 	int ret;
 
 	netdev_dbg(dev->net, "asix_mdio_write() phy_id=0x%02x, loc=0x%02x, val=0x%04x\n",
 			phy_id, loc, val);
 
 	mutex_lock(&dev->phy_mutex);
-	do {
-		ret = asix_set_sw_mii(dev, 0);
-		if (ret == -ENODEV)
-			break;
-		usleep_range(1000, 1100);
-		ret = asix_read_cmd(dev, AX_CMD_STATMNGSTS_REG,
-				    0, 0, 1, &smsr, 0);
-	} while (!(smsr & AX_HOST_EN) && (i++ < 30) && (ret != -ENODEV));
 
+	ret = asix_check_host_enable(dev, 0);
 	if (ret == -ENODEV)
 		goto out;
 
@@ -561,19 +567,11 @@ int asix_mdio_read_nopm(struct net_device *netdev, int phy_id, int loc)
 {
 	struct usbnet *dev = netdev_priv(netdev);
 	__le16 res;
-	u8 smsr;
-	int i = 0;
 	int ret;
 
 	mutex_lock(&dev->phy_mutex);
-	do {
-		ret = asix_set_sw_mii(dev, 1);
-		if (ret == -ENODEV || ret == -ETIMEDOUT)
-			break;
-		usleep_range(1000, 1100);
-		ret = asix_read_cmd(dev, AX_CMD_STATMNGSTS_REG,
-				    0, 0, 1, &smsr, 1);
-	} while (!(smsr & AX_HOST_EN) && (i++ < 30) && (ret != -ENODEV));
+
+	ret = asix_check_host_enable(dev, 1);
 	if (ret == -ENODEV || ret == -ETIMEDOUT) {
 		mutex_unlock(&dev->phy_mutex);
 		return ret;
@@ -595,22 +593,14 @@ asix_mdio_write_nopm(struct net_device *netdev, int phy_id, int loc, int val)
 {
 	struct usbnet *dev = netdev_priv(netdev);
 	__le16 res = cpu_to_le16(val);
-	u8 smsr;
-	int i = 0;
 	int ret;
 
 	netdev_dbg(dev->net, "asix_mdio_write() phy_id=0x%02x, loc=0x%02x, val=0x%04x\n",
 			phy_id, loc, val);
 
 	mutex_lock(&dev->phy_mutex);
-	do {
-		ret = asix_set_sw_mii(dev, 1);
-		if (ret == -ENODEV)
-			break;
-		usleep_range(1000, 1100);
-		ret = asix_read_cmd(dev, AX_CMD_STATMNGSTS_REG,
-				    0, 0, 1, &smsr, 1);
-	} while (!(smsr & AX_HOST_EN) && (i++ < 30) && (ret != -ENODEV));
+
+	ret = asix_check_host_enable(dev, 1);
 	if (ret == -ENODEV) {
 		mutex_unlock(&dev->phy_mutex);
 		return;
Signature · 1 line
2.32.0
#10 lore raw

Re: [PATCH v4] net: asix: fix uninit value bugs

This patch was applied to netdev/net.git (refs/heads/master):

Hello:

This patch was applied to netdev/net.git (refs/heads/master):

Quoted 11 lines · On Tue, 17 Aug 2021 19:37:23 +0300 you wrote:

Syzbot reported uninit-value in asix_mdio_read(). The problem was in missing error handling. asix_read_cmd() should initialize passed stack variable smsr, but it can fail in some cases. Then while condidition checks possibly uninit smsr variable.

Since smsr is uninitialized stack variable, driver can misbehave, because smsr will be random in case of asix_read_cmd() failure. Fix it by adding error handling and just continue the loop instead of checking uninit value.

[...]

Here is the summary with links:
- [v4] net: asix: fix uninit value bugs

    https://git.kernel.org/netdev/net/c/a786e3195d6a

You are awesome, thank you!
--
Deet-doot-dot, I am a bot.
https://korg.docs.kernel.org/patchwork/pwbot.html

#11 lore raw
In reply to Pavel Skripkin
Quoted 20 lines
drivers/net/usb/asix_common.c +2 -2
index ac92bc52a85e..7019c25e591c 100644
@@ -468,18 +468,25 @@ int asix_mdio_read(struct net_device *netdev, int phy_id, int loc)
 	struct usbnet *dev = netdev_priv(netdev);
 	__le16 res;
 	u8 smsr;
-	int i = 0;
+	int i;
 	int ret;
 
 	mutex_lock(&dev->phy_mutex);
-	do {
+	for (i = 0; i < 30; ++i) {
 		ret = asix_set_sw_mii(dev, 0);
 		if (ret == -ENODEV || ret == -ETIMEDOUT)
 			break;
 		usleep_range(1000, 1100);
 		ret = asix_read_cmd(dev, AX_CMD_STATMNGSTS_REG,

drivers/net/usb/asix_common.c int asix_mdio_read(struct net_device *netdev, int phy_id, int loc) · line 489

Quoted 10 lines
drivers/net/usb/asix_common.c +8 -1
 				    0, 0, 1, &smsr, 0);
-	} while (!(smsr & AX_HOST_EN) && (i++ < 30) && (ret != -ENODEV));
+		if (ret == -ENODEV)
+			break;
+		else if (ret < 0)
+			continue;
+		else if (smsr & AX_HOST_EN)
+			break;
+	}
+

Yes, this looks good. And Jakub is correct, there are 3 other bits of similar code you should look at.

Andrew