USB: dwc3: qcom: fix NULL-deref on suspend
cover USB: dwc3: qcom: fix NULL-deref on suspend
Some dwc3 glue drivers are currently accessing the driver data of the child core device directly, which is clearly a bad idea as the child may not have probed yet or may have been unbound from its driver.
This series fixes some of the symptoms of these layering violations as fixing the glue drivers properly is going to take a fair bit of work.
Johan
Johan Hovold (2):
USB: dwc3: qcom: fix NULL-deref on suspend
USB: dwc3: fix use-after-free on core driver unbind
drivers/usb/dwc3/core.c | 5 +++++ drivers/usb/dwc3/dwc3-qcom.c | 11 ++++++++++- 2 files changed, 15 insertions(+), 1 deletion(-)
Signature · 1 line
2.39.3
PATCH 1/2 USB: dwc3: qcom: fix NULL-deref on suspend
The Qualcomm dwc3 glue driver is currently accessing the driver data of the child core device during suspend and on wakeup interrupts. This is clearly a bad idea as the child may not have probed yet or could have been unbound from its driver.
The first such layering violation was part of the initial version of the driver, but this was later made worse when the hack that accesses the driver data of the grand child xhci device to configure the wakeup interrupts was added.
Fixing this properly is not that easily done, so add a sanity check to make sure that the child driver data is non-NULL before dereferencing it for now.
Note that this relies on subtleties like the fact that driver core is making sure that the parent is not suspended while the child is probing.
--- drivers/usb/dwc3/dwc3-qcom.c | 11 ++++++++++- 1 file changed, 10 insertions(+), 1 deletion(-)
drivers/usb/dwc3/dwc3-qcom.c +10 -1
@@ -308,7 +308,16 @@ static void dwc3_qcom_interconnect_exit(struct dwc3_qcom *qcom)/* Only usable in contexts where the role can not change. */ static bool dwc3_qcom_is_host (struct dwc3_qcom *qcom) { -struct dwc3 *dwc =platform_get_drvdata (qcom->dwc3); +struct dwc3 *dwc; + +/* +* FIXME: Fix this layering violation. +*/ + dwc =platform_get_drvdata (qcom->dwc3); + +/* Core driver may not have probed yet. */ +if (!dwc) +return false;return dwc->xhci; }
Signature · 1 line
2.39.3
Quoted 17 lines · Johan Hovold
The Qualcomm dwc3 glue driver is currently accessing the driver data of the child core device during suspend and on wakeup interrupts. This is clearly a bad idea as the child may not have probed yet or could have been unbound from its driver.
The first such layering violation was part of the initial version of the driver, but this was later made worse when the hack that accesses the driver data of the grand child xhci device to configure the wakeup interrupts was added.
Fixing this properly is not that easily done, so add a sanity check to make sure that the child driver data is non-NULL before dereferencing it for now.
Note that this relies on subtleties like the fact that driver core is making sure that the parent is not suspended while the child is probing.
Quoted 8 lines
Reported-by Manivannan Sadhasivam <[email protected]>Fixes d9152161b4bf ("usb: dwc3: Add Qualcomm DWC3 glue layer driver")Fixes 6895ea55c385 ("usb: dwc3: qcom: Configure wakeup interrupts during suspend")Cc [email protected] # 3.18: a872ab303d5d: "usb: dwc3: qcom: fix use-after-free on runtime-PM wakeup"Cc Sandeep Maheswaram <[email protected]>Cc Krishna Kurapati <[email protected]>Signed-off-by Johan Hovold <[email protected]>
- Mani
Quoted 29 lines
--- drivers/usb/dwc3/dwc3-qcom.c | 11 ++++++++++- 1 file changed, 10 insertions(+), 1 deletion(-)drivers/usb/dwc3/dwc3-qcom.c +10
-1@@ -308,7 +308,16 @@ static void dwc3_qcom_interconnect_exit(struct dwc3_qcom *qcom)/* Only usable in contexts where the role can not change. */ static bool dwc3_qcom_is_host (struct dwc3_qcom *qcom) { -struct dwc3 *dwc =platform_get_drvdata (qcom->dwc3); +struct dwc3 *dwc; + +/* +* FIXME: Fix this layering violation. +*/ + dwc =platform_get_drvdata (qcom->dwc3); + +/* Core driver may not have probed yet. */ +if (!dwc) +return false;return dwc->xhci; }Signature · 1 line
2.39.3
Signature · 1 line
மணிவண்ணன் சதாசிவம்
Quoted 54 lines · Johan Hovold
The Qualcomm dwc3 glue driver is currently accessing the driver data of the child core device during suspend and on wakeup interrupts. This is clearly a bad idea as the child may not have probed yet or could have been unbound from its driver.
The first such layering violation was part of the initial version of the driver, but this was later made worse when the hack that accesses the driver data of the grand child xhci device to configure the wakeup interrupts was added.
Fixing this properly is not that easily done, so add a sanity check to make sure that the child driver data is non-NULL before dereferencing it for now.
Note that this relies on subtleties like the fact that driver core is making sure that the parent is not suspended while the child is probing.
Reported-by Manivannan Sadhasivam <[email protected]>Fixes d9152161b4bf ("usb: dwc3: Add Qualcomm DWC3 glue layer driver")Fixes 6895ea55c385 ("usb: dwc3: qcom: Configure wakeup interrupts during suspend")Cc [email protected] # 3.18: a872ab303d5d: "usb: dwc3: qcom: fix use-after-free on runtime-PM wakeup"Cc Sandeep Maheswaram <[email protected]>Cc Krishna Kurapati <[email protected]>Signed-off-by Johan Hovold <[email protected]>--- drivers/usb/dwc3/dwc3-qcom.c | 11 ++++++++++- 1 file changed, 10 insertions(+), 1 deletion(-)drivers/usb/dwc3/dwc3-qcom.c +10
-1@@ -308,7 +308,16 @@ static void dwc3_qcom_interconnect_exit(struct dwc3_qcom *qcom)/* Only usable in contexts where the role can not change. */ static bool dwc3_qcom_is_host (struct dwc3_qcom *qcom) { -struct dwc3 *dwc =platform_get_drvdata (qcom->dwc3); +struct dwc3 *dwc; + +/* +* FIXME: Fix this layering violation. +*/ + dwc =platform_get_drvdata (qcom->dwc3); + +/* Core driver may not have probed yet. */ +if (!dwc) +return false;return dwc->xhci; }Signature · 1 line
2.39.3
Thanks for the catch.
BR,
Thinh
PATCH 2/2 USB: dwc3: fix use-after-free on core driver unbind
In reply to Johan HovoldSome dwc3 glue drivers are currently accessing the driver data of the child core device directly, which is clearly a bad idea as the child may not have probed yet or may have been unbound from its driver.
As a workaround until the glue drivers have been fixed, clear the driver data pointer before allowing the glue parent device to runtime suspend to prevent its driver from accessing data that has been freed during unbind.
--- drivers/usb/dwc3/core.c | 5 +++++ 1 file changed, 5 insertions(+)
drivers/usb/dwc3/core.c +5 -0
@@ -1929,6 +1929,11 @@ static int dwc3_remove(struct platform_device *pdev)pm_runtime_disable (&pdev->dev);pm_runtime_dont_use_autosuspend (&pdev->dev);pm_runtime_put_noidle (&pdev->dev); +/* +* HACK: Clear the driver data, which is currently accessed by parent +* glue drivers, before allowing the parent to suspend. +*/ +platform_set_drvdata (pdev,NULL );pm_runtime_set_suspended (&pdev->dev);dwc3_free_event_buffers (dwc);
Signature · 1 line
2.39.3
Re: [PATCH 2/2] USB: dwc3: fix use-after-free on core driver unbind
Quoted 25 lines · Johan Hovold
Some dwc3 glue drivers are currently accessing the driver data of the child core device directly, which is clearly a bad idea as the child may not have probed yet or may have been unbound from its driver.
As a workaround until the glue drivers have been fixed, clear the driver data pointer before allowing the glue parent device to runtime suspend to prevent its driver from accessing data that has been freed during unbind.
Fixes 6dd2565989b4 ("usb: dwc3: add imx8mp dwc3 glue layer driver")Fixes 6895ea55c385 ("usb: dwc3: qcom: Configure wakeup interrupts during suspend")Cc [email protected] # 5.12Cc Li Jun <[email protected]>Cc Sandeep Maheswaram <[email protected]>Cc Krishna Kurapati <[email protected]>Signed-off-by Johan Hovold <[email protected]>--- drivers/usb/dwc3/core.c | 5 +++++ 1 file changed, 5 insertions(+)drivers/usb/dwc3/core.c +0
-0
drivers/usb/dwc3/core.c static int dwc3_remove(struct platform_device *pdev) · line 1936
Quoted 8 lines
drivers/usb/dwc3/core.c +5
-0@@ -1929,6 +1929,11 @@ static int dwc3_remove(struct platform_device *pdev)pm_runtime_disable (&pdev->dev);pm_runtime_dont_use_autosuspend (&pdev->dev);pm_runtime_put_noidle (&pdev->dev); +/* +* HACK: Clear the driver data, which is currently accessed by parent +* glue drivers, before allowing the parent to suspend. +*/ +platform_set_drvdata (pdev,NULL );
This is required because you have seen the glue driver going to runtime suspend once the below pm_runtime_set_suspended() is completed?
- Mani
Quoted 6 lines
drivers/usb/dwc3/core.c +0
-0pm_runtime_set_suspended (&pdev->dev);dwc3_free_event_buffers (dwc);Signature · 1 line
2.39.3
Signature · 1 line
மணிவண்ணன் சதாசிவம்
Re: [PATCH 2/2] USB: dwc3: fix use-after-free on core driver unbind
Quoted 10 lines · Manivannan Sadhasivam
Quoted 8 lines · Johan Hovold
Some dwc3 glue drivers are currently accessing the driver data of the child core device directly, which is clearly a bad idea as the child may not have probed yet or may have been unbound from its driver.
As a workaround until the glue drivers have been fixed, clear the driver data pointer before allowing the glue parent device to runtime suspend to prevent its driver from accessing data that has been freed during unbind.
Quoted 12 lines
drivers/usb/dwc3/core.c static int dwc3_remove(struct platform_device *pdev) · line 1936
Quoted 9 lines
drivers/usb/dwc3/core.c +5
-0@@ -1929,6 +1929,11 @@ static int dwc3_remove(struct platform_device *pdev)pm_runtime_disable (&pdev->dev);pm_runtime_dont_use_autosuspend (&pdev->dev);pm_runtime_put_noidle (&pdev->dev); +/* +* HACK: Clear the driver data, which is currently accessed by parent +* glue drivers, before allowing the parent to suspend. +*/ +platform_set_drvdata (pdev,NULL );This is required because you have seen the glue driver going to runtime suspend once the below
pm_runtime_set_suspended()is completed?
This is based on analysis of the code. The parent (glue) can not suspend while the child (core) is in the active state, but once we set the suspended state that could happen.
Quoted 2 lines
drivers/usb/dwc3/core.c static int dwc3_remove(struct platform_device *pdev) · line 1939
Quoted 2 lines
drivers/usb/dwc3/core.c +0
-0dwc3_free_event_buffers (dwc);
Johan
Re: [PATCH 2/2] USB: dwc3: fix use-after-free on core driver unbind
Quoted 24 lines · Johan Hovold
Quoted 10 lines · Manivannan Sadhasivam
Quoted 8 lines · Johan Hovold
Some dwc3 glue drivers are currently accessing the driver data of the child core device directly, which is clearly a bad idea as the child may not have probed yet or may have been unbound from its driver.
As a workaround until the glue drivers have been fixed, clear the driver data pointer before allowing the glue parent device to runtime suspend to prevent its driver from accessing data that has been freed during unbind.
Quoted 12 lines
drivers/usb/dwc3/core.c static int dwc3_remove(struct platform_device *pdev) · line 1936
Quoted 9 lines
drivers/usb/dwc3/core.c +5
-0@@ -1929,6 +1929,11 @@ static int dwc3_remove(struct platform_device *pdev)pm_runtime_disable (&pdev->dev);pm_runtime_dont_use_autosuspend (&pdev->dev);pm_runtime_put_noidle (&pdev->dev); +/* +* HACK: Clear the driver data, which is currently accessed by parent +* glue drivers, before allowing the parent to suspend. +*/ +platform_set_drvdata (pdev,NULL );This is required because you have seen the glue driver going to runtime suspend once the below
pm_runtime_set_suspended()is completed?
Quoted 4 lines
This is based on analysis of the code. The parent (glue) can not suspend while the child (core) is in the active state, but once we set the suspended state that could happen.
I could see that the driver core is setting drvdata to NULL during device_unbind_cleanup(), so not sure if this scenario could be met otherwise it will be redundant.
- Mani
Quoted 4 lines
Quoted 2 lines
drivers/usb/dwc3/core.c static int dwc3_remove(struct platform_device *pdev) · line 1939
Quoted 2 lines
drivers/usb/dwc3/core.c +0
-0dwc3_free_event_buffers (dwc);Johan
Signature · 1 line
மணிவண்ணன் சதாசிவம்
Re: [PATCH 2/2] USB: dwc3: fix use-after-free on core driver unbind
Quoted 25 lines · Manivannan Sadhasivam
Quoted 24 lines · Johan Hovold
Quoted 22 lines · Manivannan Sadhasivam
Quoted 10 lines · Johan Hovold
Some dwc3 glue drivers are currently accessing the driver data of the child core device directly, which is clearly a bad idea as the child may not have probed yet or may have been unbound from its driver.
As a workaround until the glue drivers have been fixed, clear the driver data pointer before allowing the glue parent device to runtime suspend to prevent its driver from accessing data that has been freed during unbind.
drivers/usb/dwc3/core.c static int dwc3_remove(struct platform_device *pdev) · line 1936
Quoted 8 lines
drivers/usb/dwc3/core.c +5
-0@@ -1929,6 +1929,11 @@ static int dwc3_remove(struct platform_device *pdev)pm_runtime_disable (&pdev->dev);pm_runtime_dont_use_autosuspend (&pdev->dev);pm_runtime_put_noidle (&pdev->dev); +/* +* HACK: Clear the driver data, which is currently accessed by parent +* glue drivers, before allowing the parent to suspend. +*/ +platform_set_drvdata (pdev,NULL );This is required because you have seen the glue driver going to runtime suspend once the below
pm_runtime_set_suspended()is completed?
Quoted 7 lines
Quoted 3 lines
This is based on analysis of the code. The parent (glue) can not suspend while the child (core) is in the active state, but once we set the suspended state that could happen.
I could see that the driver core is setting drvdata to NULL during
device_unbind_cleanup(), so not sure if this scenario could be met otherwise it will be redundant.
If this was redundant I wouldn't have added it. ;)
The parent driver has no business accessing the driver data of the child in the first place, but it must absolutely not do so after the child has been unbound from its driver and the driver data is getting freed.
Relying on the clean up in driver core that resets this pointer does not work as that would still leave a window where the parent could access this stale data.
Johan
Manivannan Sadhasivam
Reviewed-byRe: [PATCH 2/2] USB: dwc3: fix use-after-free on core driver unbind
Quoted 36 lines · Johan Hovold
Quoted 25 lines · Manivannan Sadhasivam
Quoted 24 lines · Johan Hovold
Quoted 22 lines · Manivannan Sadhasivam
Quoted 10 lines · Johan Hovold
Some dwc3 glue drivers are currently accessing the driver data of the child core device directly, which is clearly a bad idea as the child may not have probed yet or may have been unbound from its driver.
As a workaround until the glue drivers have been fixed, clear the driver data pointer before allowing the glue parent device to runtime suspend to prevent its driver from accessing data that has been freed during unbind.
drivers/usb/dwc3/core.c static int dwc3_remove(struct platform_device *pdev) · line 1936
Quoted 8 lines
drivers/usb/dwc3/core.c +5
-0@@ -1929,6 +1929,11 @@ static int dwc3_remove(struct platform_device *pdev)pm_runtime_disable (&pdev->dev);pm_runtime_dont_use_autosuspend (&pdev->dev);pm_runtime_put_noidle (&pdev->dev); +/* +* HACK: Clear the driver data, which is currently accessed by parent +* glue drivers, before allowing the parent to suspend. +*/ +platform_set_drvdata (pdev,NULL );This is required because you have seen the glue driver going to runtime suspend once the below
pm_runtime_set_suspended()is completed?Quoted 7 lines
Quoted 3 lines
This is based on analysis of the code. The parent (glue) can not suspend while the child (core) is in the active state, but once we set the suspended state that could happen.
I could see that the driver core is setting drvdata to NULL during
device_unbind_cleanup(), so not sure if this scenario could be met otherwise it will be redundant.If this was redundant I wouldn't have added it. ;)
Quoted 8 lines
The parent driver has no business accessing the driver data of the child in the first place, but it must absolutely not do so after the child has been unbound from its driver and the driver data is getting freed.
Relying on the clean up in driver core that resets this pointer does not work as that would still leave a window where the parent could access this stale data.
Well, I agree with the small window here.
- Mani
Quoted 1 line
Johan
Signature · 1 line
மணிவண்ணன் சதாசிவம்
Re: [PATCH 2/2] USB: dwc3: fix use-after-free on core driver unbind
In reply to Johan HovoldQuoted 39 lines · Johan Hovold
Some dwc3 glue drivers are currently accessing the driver data of the child core device directly, which is clearly a bad idea as the child may not have probed yet or may have been unbound from its driver.
As a workaround until the glue drivers have been fixed, clear the driver data pointer before allowing the glue parent device to runtime suspend to prevent its driver from accessing data that has been freed during unbind.
Fixes 6dd2565989b4 ("usb: dwc3: add imx8mp dwc3 glue layer driver")Fixes 6895ea55c385 ("usb: dwc3: qcom: Configure wakeup interrupts during suspend")Cc [email protected] # 5.12Cc Li Jun <[email protected]>Cc Sandeep Maheswaram <[email protected]>Cc Krishna Kurapati <[email protected]>Signed-off-by Johan Hovold <[email protected]>--- drivers/usb/dwc3/core.c | 5 +++++ 1 file changed, 5 insertions(+)drivers/usb/dwc3/core.c +5
-0@@ -1929,6 +1929,11 @@ static int dwc3_remove(struct platform_device *pdev)pm_runtime_disable (&pdev->dev);pm_runtime_dont_use_autosuspend (&pdev->dev);pm_runtime_put_noidle (&pdev->dev); +/* +* HACK: Clear the driver data, which is currently accessed by parent +* glue drivers, before allowing the parent to suspend. +*/ +platform_set_drvdata (pdev,NULL );pm_runtime_set_suspended (&pdev->dev);dwc3_free_event_buffers (dwc);Signature · 1 line
2.39.3
Thanks,
Thinh