hwmon: couple of fixes on HWMON_C_REGISTER_TZ
v2 cover hwmon: couple of fixes on HWMON_C_REGISTER_TZ
Hello Guenter,
I found these bugs in the error path of hwmon_device_register(). One related to calling of-thermal when no dev->of_node is present. And another in the error path handling of it.
Only difference from V1 is that I changed patch 2/2 to remove the device_unregister() before jumping into the new label.
Eduardo Valentin (2):
hwmon: core: add thermal sensors only if dev->of_node is present
hwmon: core: fix potential memory leak in *hwmon_device_register*
drivers/hwmon/hwmon.c | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-)
Signature · 1 line
2.21.0
PATCH v2 1/2 hwmon: core: add thermal sensors only if dev->of_node is present
Drivers may register to hwmon and request for also registering with the thermal subsystem (HWMON_C_REGISTER_TZ). However, some of these driver, e.g. marvell phy, may be probed from Device Tree or being dynamically allocated, and in the later case, it will not have a dev->of_node entry.
Registering with hwmon without the dev->of_node may result in different outcomes depending on the device tree, which may be a bit misleading. If the device tree blob has no 'thermal-zones' node, the *hwmon_device_register*() family functions are going to gracefully succeed, because of-thermal, *thermal_zone_of_sensor_register() return -ENODEV in this case, and the hwmon error path handles this error code as success to cover for the case where CONFIG_THERMAL_OF is not set. However, if the device tree blob has the 'thermal-zones' entry, the *hwmon_device_register*() will always fail on callers with no dev->of_node, propagating -EINVAL.
If dev->of_node is not present, calling of-thermal does not make sense. For this reason, this patch checks first if the device has a of_node before going over the process of registering with the thermal subsystem of-thermal interface. And in this case, when a caller of *hwmon_device_register*() with HWMON_C_REGISTER_TZ and no dev->of_node will still register with hwmon, but not with the thermal subsystem. If all the hwmon part bits are in place, the registration will succeed.
---V1->V2: no change
drivers/hwmon/hwmon.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/hwmon/hwmon.c +1 -1
@@ -636,7 +636,7 @@ __hwmon_device_register(struct device *dev, const char *name, void *drvdata,if (err) goto free_hwmon; -if (dev && chip && chip->ops->read && +if (dev && dev->of_node && chip && chip->ops->read && chip->info[0 ]->type == hwmon_chip && (chip->info[0 ]->config[0 ] &HWMON_C_REGISTER_TZ )) {const struct hwmon_channel_info **info = chip->info;
Signature · 1 line
2.21.0
PATCH v2 2/2 hwmon: core: fix potential memory leak in *hwmon_device_register*
In reply to Eduardo ValentinWhen registering a hwmon device with HWMON_C_REGISTER_TZ flag in place, the hwmon subsystem will attempt to register the device also with the thermal subsystem. When the of-thermal registration fails, __hwmon_device_register jumps to ida_remove, leaving the locally allocated hwdev pointer.
This patch fixes the leak by jumping to a new label that will first unregister hdev and then fall into the kfree of hwdev to finally remove the idas and propagate the error code.
---V1->V2: removed the device_unregister() before jumping into the new label, as suggested in the first review round.
drivers/hwmon/hwmon.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-)
drivers/hwmon/hwmon.c +4 -4
@@ -652,10 +652,8 @@ __hwmon_device_register(struct device *dev, const char *name, void *drvdata,if (info[i]->config[j] &HWMON_T_INPUT ) { err =hwmon_thermal_add_sensor (dev, hwdev, j); -if (err) { -device_unregister (hdev); - goto ida_remove; - } +if (err) + goto device_unregister; } } }
@@ -663,6 +661,8 @@ __hwmon_device_register(struct device *dev, const char *name, void *drvdata,return hdev; +device_unregister: +device_unregister (hdev); free_hwmon:kfree (hwdev); ida_remove:
Signature · 1 line
2.21.0
Re: [PATCHv2 2/2] hwmon: core: fix potential memory leak in *hwmon_device_register*
Quoted 10 lines · Eduardo Valentin
When registering a hwmon device with
HWMON_C_REGISTER_TZflag in place, the hwmon subsystem will attempt to register the device also with the thermal subsystem. When the of-thermal registration fails,__hwmon_device_registerjumps toida_remove, leaving the locally allocated hwdev pointer.This patch fixes the leak by jumping to a new label that will first unregister hdev and then fall into the kfree of hwdev to finally remove the idas and propagate the error code.
Quoted 5 lines
Cc Jean Delvare <[email protected]>Cc Guenter Roeck <[email protected]>Signed-off-by Eduardo Valentin <[email protected]>
Applied.
Thanks,
Guenter
Quoted 33 lines
---
V1->V2: removed thedevice_unregister()before jumping into the new label, as suggested in the first review round.drivers/hwmon/hwmon.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-)drivers/hwmon/hwmon.c +4
-4@@ -652,10 +652,8 @@ __hwmon_device_register(struct device *dev, const char *name, void *drvdata,if (info[i]->config[j] &HWMON_T_INPUT ) { err =hwmon_thermal_add_sensor (dev, hwdev, j); -if (err) { -device_unregister (hdev); - goto ida_remove; - } +if (err) + goto device_unregister; } } }@@ -663,6 +661,8 @@ __hwmon_device_register(struct device *dev, const char *name, void *drvdata,return hdev; +device_unregister: +device_unregister (hdev); free_hwmon:kfree (hwdev); ida_remove:
Guenter Roeck
maintainerRe: [PATCHv2 2/2] hwmon: core: fix potential memory leak in *hwmon_device_register*
In reply to Eduardo ValentinQuoted 10 lines · Eduardo Valentin
When registering a hwmon device with
HWMON_C_REGISTER_TZflag in place, the hwmon subsystem will attempt to register the device also with the thermal subsystem. When the of-thermal registration fails,__hwmon_device_registerjumps toida_remove, leaving the locally allocated hwdev pointer.This patch fixes the leak by jumping to a new label that will first unregister hdev and then fall into the kfree of hwdev to finally remove the idas and propagate the error code.
Hah, actually this is wrong. hwdev is freed indirectly with the device_unregister() call. See commit 74e3512731bd ("hwmon: (core) Fix double-free in __hwmon_device_register()").
It may make sense to add a respective comment to the code, though.
Guenter
Quoted 38 lines
Cc Jean Delvare <[email protected]>Cc Guenter Roeck <[email protected]>Signed-off-by Eduardo Valentin <[email protected]>---
V1->V2: removed thedevice_unregister()before jumping into the new label, as suggested in the first review round.drivers/hwmon/hwmon.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-)drivers/hwmon/hwmon.c +4
-4@@ -652,10 +652,8 @@ __hwmon_device_register(struct device *dev, const char *name, void *drvdata,if (info[i]->config[j] &HWMON_T_INPUT ) { err =hwmon_thermal_add_sensor (dev, hwdev, j); -if (err) { -device_unregister (hdev); - goto ida_remove; - } +if (err) + goto device_unregister; } } }@@ -663,6 +661,8 @@ __hwmon_device_register(struct device *dev, const char *name, void *drvdata,return hdev; +device_unregister: +device_unregister (hdev); free_hwmon:kfree (hwdev); ida_remove:
Re: [PATCHv2 2/2] hwmon: core: fix potential memory leak in *hwmon_device_register*
Quoted 7 lines · Guenter Roeck
Quoted 6 lines · Eduardo Valentin
When registering a hwmon device with
HWMON_C_REGISTER_TZflag in place, the hwmon subsystem will attempt to register the device also with the thermal subsystem. When the of-thermal registration fails,__hwmon_device_registerjumps toida_remove, leaving the locally allocated hwdev pointer.
Quoted 8 lines
Quoted 4 lines
This patch fixes the leak by jumping to a new label that will first unregister hdev and then fall into the kfree of hwdev to finally remove the idas and propagate the error code.
Hah, actually this is wrong. hwdev is freed indirectly with the
device_unregister()call. See commit74e3512731bd("hwmon: (core) Fix double-free in__hwmon_device_register()").
heh.. I see it now. Well, it is not a straight catch though.
Quoted 3 lines
It may make sense to add a respective comment to the code, though.
I agree. Or a simple comment saying "dont worry about freeing hwdev because hwmon_dev_release() takes care of it".
Are you patching it ?
Quoted 40 lines
Guenter
Quoted 38 lines
Cc Jean Delvare <[email protected]>Cc Guenter Roeck <[email protected]>Signed-off-by Eduardo Valentin <[email protected]>---
V1->V2: removed thedevice_unregister()before jumping into the new label, as suggested in the first review round.drivers/hwmon/hwmon.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-)drivers/hwmon/hwmon.c +4
-4@@ -652,10 +652,8 @@ __hwmon_device_register(struct device *dev, const char *name, void *drvdata,if (info[i]->config[j] &HWMON_T_INPUT ) { err =hwmon_thermal_add_sensor (dev, hwdev, j); -if (err) { -device_unregister (hdev); - goto ida_remove; - } +if (err) + goto device_unregister; } } }@@ -663,6 +661,8 @@ __hwmon_device_register(struct device *dev, const char *name, void *drvdata,return hdev; +device_unregister: +device_unregister (hdev); free_hwmon:kfree (hwdev); ida_remove:
Signature · 2 lines
All the best, Eduardo Valentin
Guenter Roeck
maintainerRe: [PATCHv2 2/2] hwmon: core: fix potential memory leak in *hwmon_device_register*
Quoted 20 lines · Eduardo Valentin
Quoted 7 lines · Guenter Roeck
Quoted 6 lines · Eduardo Valentin
When registering a hwmon device with
HWMON_C_REGISTER_TZflag in place, the hwmon subsystem will attempt to register the device also with the thermal subsystem. When the of-thermal registration fails,__hwmon_device_registerjumps toida_remove, leaving the locally allocated hwdev pointer.Quoted 8 lines
Quoted 4 lines
This patch fixes the leak by jumping to a new label that will first unregister hdev and then fall into the kfree of hwdev to finally remove the idas and propagate the error code.
Hah, actually this is wrong. hwdev is freed indirectly with the
device_unregister()call. See commit74e3512731bd("hwmon: (core) Fix double-free in__hwmon_device_register()").heh.. I see it now. Well, it is not a straight catch though.
Quoted 1 line
Quoted 8 lines
Quoted 2 lines
It may make sense to add a respective comment to the code, though.
I agree. Or a simple comment saying "dont worry about freeing hwdev because
hwmon_dev_release()takes care of it".Are you patching it ?
Will do. I'll send a patch in a minute.
Thanks,
Guenter