Skip to contentSkip to conversation

misc/mei: Add NULL check to component match callback functions

PATCH In review Won Chung 4 participants 7 messages Mar 31, 2022

Subsystems (3)
lore
Share & tools
lore

Won Chung

author
#0 lore raw

PATCH v3 1/1 misc/mei: Add NULL check to component match callback functions

Component match callback functions need to check if expected data is passed to them. Without this check, it can cause a NULL pointer dereference when another driver registers a component before i915 drivers have their component master fully bind.

Fixes 1e8d19d9b0dfc ("mei: hdcp: bind only with i915 on the same PCH")
Fixes c2004ce99ed73 ("mei: pxp: export pavp client to me client bus")

---
Changes from v2:
- Correctly add "Suggested-by" tag
- Add "Cc: [email protected]"

Changes from v1:
- Add "Fixes" tag
- Send to [email protected]

 drivers/misc/mei/hdcp/mei_hdcp.c | 2 +-
 drivers/misc/mei/pxp/mei_pxp.c   | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)
drivers/misc/mei/hdcp/mei_hdcp.c +1 -1
index ec2a4fce8581..843dbc2b21b1 100644
@@ -784,7 +784,7 @@ static int mei_hdcp_component_match(struct device *dev, int subcomponent,
 {
 	struct device *base = data;
 
-	if (strcmp(dev->driver->name, "i915") ||
+	if (!base || !dev->driver || strcmp(dev->driver->name, "i915") ||
 	    subcomponent != I915_COMPONENT_HDCP)
 		return 0;
 
drivers/misc/mei/pxp/mei_pxp.c +1 -1
index f7380d387bab..e32a81da8af6 100644
@@ -131,7 +131,7 @@ static int mei_pxp_component_match(struct device *dev, int subcomponent,
 {
 	struct device *base = data;
 
-	if (strcmp(dev->driver->name, "i915") ||
+	if (!base || !dev->driver || strcmp(dev->driver->name, "i915") ||
 	    subcomponent != I915_COMPONENT_PXP)
 		return 0;
 
Signature · 1 line
2.35.1.1021.g381101b075-goog
maintainer
#1 lore raw
Quoted 4 lines · Won Chung

Component match callback functions need to check if expected data is passed to them. Without this check, it can cause a NULL pointer dereference when another driver registers a component before i915 drivers have their component master fully bind.

How can that happen in a real system? Or does this just happen for when you are doing development and testing?

Quoted 7 lines
Fixes 1e8d19d9b0dfc ("mei: hdcp: bind only with i915 on the same PCH")
Fixes c2004ce99ed73 ("mei: pxp: export pavp client to me client bus")

Why does this need to go to stable? How can this be triggered in older kernels?

Quoted 18 lines

---
Changes from v2:
- Correctly add "Suggested-by" tag
- Add "Cc: [email protected]"

Changes from v1:
- Add "Fixes" tag
- Send to [email protected]

 drivers/misc/mei/hdcp/mei_hdcp.c | 2 +-
 drivers/misc/mei/pxp/mei_pxp.c   | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)
drivers/misc/mei/hdcp/mei_hdcp.c +0 -0
index ec2a4fce8581..843dbc2b21b1 100644

drivers/misc/mei/hdcp/mei_hdcp.c static int mei_hdcp_component_match(struct device *dev, int subcomponent, · line 787

Quoted 5 lines
drivers/misc/mei/hdcp/mei_hdcp.c +1 -1
@@ -784,7 +784,7 @@ static int mei_hdcp_component_match(struct device *dev, int subcomponent,
 {
 	struct device *base = data;
 
-	if (strcmp(dev->driver->name, "i915") ||
+	if (!base || !dev->driver || strcmp(dev->driver->name, "i915") ||

How can base be NULL?

drivers/misc/mei/pxp/mei_pxp.c static int mei_pxp_component_match(struct device *dev, int subcomponent, · line 134

Quoted 13 lines
drivers/misc/mei/hdcp/mei_hdcp.c +0 -0
 	    subcomponent != I915_COMPONENT_HDCP)
 		return 0;
 
drivers/misc/mei/pxp/mei_pxp.c +1 -1
index f7380d387bab..e32a81da8af6 100644
@@ -131,7 +131,7 @@ static int mei_pxp_component_match(struct device *dev, int subcomponent,
 {
 	struct device *base = data;
 
-	if (strcmp(dev->driver->name, "i915") ||
+	if (!base || !dev->driver || strcmp(dev->driver->name, "i915") ||

Same here, shouldn't this be caught by the driver core or bus and match should not be called?

Why not fix this in the component/driver core instead?

thanks,

greg k-h

#2 lore raw
Quoted 58 lines · Greg KH
Quoted 4 lines · Won Chung

Component match callback functions need to check if expected data is passed to them. Without this check, it can cause a NULL pointer dereference when another driver registers a component before i915 drivers have their component master fully bind.

How can that happen in a real system? Or does this just happen for when you are doing development and testing?

Quoted 7 lines
Fixes 1e8d19d9b0dfc ("mei: hdcp: bind only with i915 on the same PCH")
Fixes c2004ce99ed73 ("mei: pxp: export pavp client to me client bus")

Why does this need to go to stable? How can this be triggered in older kernels?

Quoted 18 lines

---
Changes from v2:
- Correctly add "Suggested-by" tag
- Add "Cc: [email protected]"

Changes from v1:
- Add "Fixes" tag
- Send to [email protected]

 drivers/misc/mei/hdcp/mei_hdcp.c | 2 +-
 drivers/misc/mei/pxp/mei_pxp.c   | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)
drivers/misc/mei/hdcp/mei_hdcp.c +0 -0
index ec2a4fce8581..843dbc2b21b1 100644

drivers/misc/mei/hdcp/mei_hdcp.c static int mei_hdcp_component_match(struct device *dev, int subcomponent, · line 787

Quoted 5 lines
drivers/misc/mei/hdcp/mei_hdcp.c +1 -1
@@ -784,7 +784,7 @@ static int mei_hdcp_component_match(struct device *dev, int subcomponent,
 {
 	struct device *base = data;
 
-	if (strcmp(dev->driver->name, "i915") ||
+	if (!base || !dev->driver || strcmp(dev->driver->name, "i915") ||

How can base be NULL?

Quoted 11 lines
drivers/misc/mei/hdcp/mei_hdcp.c +0 -0
 	    subcomponent != I915_COMPONENT_HDCP)
 		return 0;
 
drivers/misc/mei/pxp/mei_pxp.c +0 -0
index f7380d387bab..e32a81da8af6 100644
@@ -131,7 +131,7 @@ static int mei_pxp_component_match(struct device *dev, int subcomponent,
 {
 	struct device *base = data;
 
Quoted 7 lines

drivers/misc/mei/pxp/mei_pxp.c static int mei_pxp_component_match(struct device *dev, int subcomponent, · line 134

Quoted 2 lines
drivers/misc/mei/pxp/mei_pxp.c +1 -1
-	if (strcmp(dev->driver->name, "i915") ||
+	if (!base || !dev->driver || strcmp(dev->driver->name, "i915") ||

Same here, shouldn't this be caught by the driver core or bus and match should not be called?

Why not fix this in the component/driver core instead?

A component is just a device that is declared to be a "component", and the code that declares it as component does not have to be the driver of that device. You simply can't assume that it's bind to a driver like this function does.

In our case the "components" are USB ports, so devices that are never bind to drivers.

thanks,

Signature · 1 line
heikki
maintainer
#3 lore raw
Quoted 67 lines · Heikki Krogerus
Quoted 58 lines · Greg KH
Quoted 4 lines · Won Chung

Component match callback functions need to check if expected data is passed to them. Without this check, it can cause a NULL pointer dereference when another driver registers a component before i915 drivers have their component master fully bind.

How can that happen in a real system? Or does this just happen for when you are doing development and testing?

Quoted 7 lines
Fixes 1e8d19d9b0dfc ("mei: hdcp: bind only with i915 on the same PCH")
Fixes c2004ce99ed73 ("mei: pxp: export pavp client to me client bus")

Why does this need to go to stable? How can this be triggered in older kernels?

Quoted 18 lines

---
Changes from v2:
- Correctly add "Suggested-by" tag
- Add "Cc: [email protected]"

Changes from v1:
- Add "Fixes" tag
- Send to [email protected]

 drivers/misc/mei/hdcp/mei_hdcp.c | 2 +-
 drivers/misc/mei/pxp/mei_pxp.c   | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)
drivers/misc/mei/hdcp/mei_hdcp.c +0 -0
index ec2a4fce8581..843dbc2b21b1 100644

drivers/misc/mei/hdcp/mei_hdcp.c static int mei_hdcp_component_match(struct device *dev, int subcomponent, · line 787

Quoted 5 lines
drivers/misc/mei/hdcp/mei_hdcp.c +1 -1
@@ -784,7 +784,7 @@ static int mei_hdcp_component_match(struct device *dev, int subcomponent,
 {
 	struct device *base = data;
 
-	if (strcmp(dev->driver->name, "i915") ||
+	if (!base || !dev->driver || strcmp(dev->driver->name, "i915") ||

How can base be NULL?

Quoted 11 lines
drivers/misc/mei/hdcp/mei_hdcp.c +0 -0
 	    subcomponent != I915_COMPONENT_HDCP)
 		return 0;
 
drivers/misc/mei/pxp/mei_pxp.c +0 -0
index f7380d387bab..e32a81da8af6 100644
@@ -131,7 +131,7 @@ static int mei_pxp_component_match(struct device *dev, int subcomponent,
 {
 	struct device *base = data;
 
Quoted 7 lines

drivers/misc/mei/pxp/mei_pxp.c static int mei_pxp_component_match(struct device *dev, int subcomponent, · line 134

Quoted 2 lines
drivers/misc/mei/pxp/mei_pxp.c +1 -1
-	if (strcmp(dev->driver->name, "i915") ||
+	if (!base || !dev->driver || strcmp(dev->driver->name, "i915") ||

Same here, shouldn't this be caught by the driver core or bus and match should not be called?

Why not fix this in the component/driver core instead?

Quoted 7 lines

A component is just a device that is declared to be a "component", and the code that declares it as component does not have to be the driver of that device. You simply can't assume that it's bind to a driver like this function does.

In our case the "components" are USB ports, so devices that are never bind to drivers.

And going off of the driver name is sane? That feels ripe for bugs and problems in the future, but hey, I don't understand the need for this driver to care about another driver at all.

And why is a USB device being passed to something that it thinks is a PCI device? That too feels really wrong and ripe for problems.

thanks,

greg k-h

#4 lore raw

Hi Greg,

Quoted 23 lines · Greg KH
Quoted 23 lines
Quoted 10 lines
Quoted 10 lines
(quoted diff) +0 -0
 		return 0;
drivers/misc/mei/pxp/mei_pxp.c +0 -0
index f7380d387bab..e32a81da8af6 100644
@@ -131,7 +131,7 @@ static int mei_pxp_component_match(struct device *dev, int subcomponent,
 {
 	struct device *base = data;
 
Quoted 7 lines

drivers/misc/mei/pxp/mei_pxp.c static int mei_pxp_component_match(struct device *dev, int subcomponent, · line 134

Quoted 2 lines
drivers/misc/mei/pxp/mei_pxp.c +1 -1
-	if (strcmp(dev->driver->name, "i915") ||
+	if (!base || !dev->driver || strcmp(dev->driver->name, "i915") ||

Same here, shouldn't this be caught by the driver core or bus and match should not be called?

Why not fix this in the component/driver core instead?

A component is just a device that is declared to be a "component", and the code that declares it as component does not have to be the driver of that device. You simply can't assume that it's bind to a driver like this function does.

Quoted 6 lines
Quoted 2 lines

In our case the "components" are USB ports, so devices that are never bind to drivers.

And going off of the driver name is sane? That feels ripe for bugs and problems in the future, but hey, I don't understand the need for this driver to care about another driver at all.

I think the component framework is meant to be this loose confederation of devices, so going into component match, you don't know what the other device is yet.

The USB drivers and the i915 drivers 100% don't care about each other, but the framework doesn't know that yet until all the drivers try to match.

Quoted 4 lines

And why is a USB device being passed to something that it thinks is a PCI device? That too feels really wrong and ripe for problems.

The problematic device that's being passed through here is actually the usb4_port, not a usb device. My guess would be that's why it's getting past any checks for whether it's a PCI device.

The component framework currently being used by (hdac_i915, mei_hdcp, mei_pxp) to connect those three devices together, and completely separately, the component framework is being used by the typec connector class's port mapper.

These two clusters of devices are using the same component framework, but are not supposed to interact with each other. When we attempted to add the usb4_port and its retimer in order to link tbt/usb4 to the typec connector, we discovered this interaction because we happened to build the usb4_port built-in in our configs, so it does its component_add earlier.

I agree, by the way. This is all a bit ugly.

Thanks,
Benson

Quoted 3 lines

thanks,

greg k-h

Signature · 7 lines
Benson Leung
Staff Software Engineer
Chrome OS Kernel
Google Inc.
[email protected]
Chromium OS Project
[email protected]
maintainer
#5 lore raw
Quoted 41 lines · Benson Leung

Hi Greg,

Quoted 23 lines · Greg KH
Quoted 23 lines
Quoted 10 lines
Quoted 10 lines
(quoted diff) +0 -0
 		return 0;
drivers/misc/mei/pxp/mei_pxp.c +0 -0
index f7380d387bab..e32a81da8af6 100644
@@ -131,7 +131,7 @@ static int mei_pxp_component_match(struct device *dev, int subcomponent,
 {
 	struct device *base = data;
 
Quoted 7 lines

drivers/misc/mei/pxp/mei_pxp.c static int mei_pxp_component_match(struct device *dev, int subcomponent, · line 134

Quoted 2 lines
drivers/misc/mei/pxp/mei_pxp.c +1 -1
-	if (strcmp(dev->driver->name, "i915") ||
+	if (!base || !dev->driver || strcmp(dev->driver->name, "i915") ||

Same here, shouldn't this be caught by the driver core or bus and match should not be called?

Why not fix this in the component/driver core instead?

A component is just a device that is declared to be a "component", and the code that declares it as component does not have to be the driver of that device. You simply can't assume that it's bind to a driver like this function does.

Quoted 6 lines
Quoted 2 lines

In our case the "components" are USB ports, so devices that are never bind to drivers.

And going off of the driver name is sane? That feels ripe for bugs and problems in the future, but hey, I don't understand the need for this driver to care about another driver at all.

I think the component framework is meant to be this loose confederation of devices, so going into component match, you don't know what the other device is yet.

The USB drivers and the i915 drivers 100% don't care about each other, but the framework doesn't know that yet until all the drivers try to match.

Quoted 1 line
Quoted 7 lines
Quoted 3 lines

And why is a USB device being passed to something that it thinks is a PCI device? That too feels really wrong and ripe for problems.

The problematic device that's being passed through here is actually the usb4_port, not a usb device. My guess would be that's why it's getting past any checks for whether it's a PCI device.

a usb4 port should not be a pci device.

So fix up the checks, don't do a random "is this the driver name?" check, look for the real driver pointer or something like that that you KNOW is the correct match.

Quoted 11 lines

The component framework currently being used by (hdac_i915, mei_hdcp, mei_pxp) to connect those three devices together, and completely separately, the component framework is being used by the typec connector class's port mapper.

These two clusters of devices are using the same component framework, but are not supposed to interact with each other. When we attempted to add the usb4_port and its retimer in order to link tbt/usb4 to the typec connector, we discovered this interaction because we happened to build the usb4_port built-in in our configs, so it does its component_add earlier.

I agree, by the way. This is all a bit ugly.

Something is wrong with the component code here if this is happening, as that's not a solid interface that can actually work at all.

Perhaps do not use this framework until it is fixed?

And again, is this something that you see today on an unmodified 5.17.1 release? If not, why all of the stable backport requests?

confused,

greg k-h

#6 lore raw
In reply to Won Chung

Hi Won,

Quoted 11 lines · Won Chung

Component match callback functions need to check if expected data is passed to them. Without this check, it can cause a NULL pointer dereference when another driver registers a component before i915 drivers have their component master fully bind.

Fixes 1e8d19d9b0dfc ("mei: hdcp: bind only with i915 on the same PCH")
Fixes c2004ce99ed73 ("mei: pxp: export pavp client to me client bus")
Quoted 8 lines

---
Changes from v2:
- Correctly add "Suggested-by" tag
- Add "Cc: [email protected]"

Changes from v1:
- Add "Fixes" tag
- Send to [email protected]

You are sending these really fast. Before you send the next one, please wait for at least one day.

thanks,

Signature · 1 line
heikki