| Message ID | 1636070227-15909-1-git-send-email-quic_subbaram@quicinc.com (mailing list archive) |
|---|---|
| State | Mainlined, archived |
| Delegated to: | Rafael Wysocki |
| Headers | show |
| Series | [RESEND,v2] thermal: Fix a NULL pointer dereference | expand |
On Thu, Nov 04, 2021 at 04:57:07PM -0700, Subbaraman Narayanamurthy wrote: > of_parse_thermal_zones() parses the thermal-zones node and registers a > thermal_zone device for each subnode. However, if a thermal zone is > consuming a thermal sensor and that thermal sensor device hasn't probed > yet, an attempt to set trip_point_*_temp for that thermal zone device > can cause a NULL pointer dereference. Fix it. > > console:/sys/class/thermal/thermal_zone87 # echo 120000 > trip_point_0_temp > ... > Unable to handle kernel NULL pointer dereference at virtual address 0000000000000020 > ... > Call trace: > of_thermal_set_trip_temp+0x40/0xc4 > trip_point_temp_store+0xc0/0x1dc > dev_attr_store+0x38/0x88 > sysfs_kf_write+0x64/0xc0 > kernfs_fop_write_iter+0x108/0x1d0 > vfs_write+0x2f4/0x368 > ksys_write+0x7c/0xec > __arm64_sys_write+0x20/0x30 > el0_svc_common.llvm.7279915941325364641+0xbc/0x1bc > do_el0_svc+0x28/0xa0 > el0_svc+0x14/0x24 > el0_sync_handler+0x88/0xec > el0_sync+0x1c0/0x200 > > While at it, fix the possible NULL pointer dereference in other > functions as well: of_thermal_get_temp(), of_thermal_set_emul_temp(), > of_thermal_get_trend(). > > Suggested-by: David Collins <quic_collinsd@quicinc.com> > Signed-off-by: Subbaraman Narayanamurthy <quic_subbaram@quicinc.com> > --- > drivers/thermal/thermal_of.c | 9 ++++++--- > 1 file changed, 6 insertions(+), 3 deletions(-) > <formletter> This is not the correct way to submit patches for inclusion in the stable kernel tree. Please read: https://www.kernel.org/doc/html/latest/process/stable-kernel-rules.html for how to do this properly. </formletter>
On Fri, Nov 5, 2021 at 12:57 AM Subbaraman Narayanamurthy <quic_subbaram@quicinc.com> wrote: > > of_parse_thermal_zones() parses the thermal-zones node and registers a > thermal_zone device for each subnode. However, if a thermal zone is > consuming a thermal sensor and that thermal sensor device hasn't probed > yet, an attempt to set trip_point_*_temp for that thermal zone device > can cause a NULL pointer dereference. Fix it. > > console:/sys/class/thermal/thermal_zone87 # echo 120000 > trip_point_0_temp > ... > Unable to handle kernel NULL pointer dereference at virtual address 0000000000000020 > ... > Call trace: > of_thermal_set_trip_temp+0x40/0xc4 > trip_point_temp_store+0xc0/0x1dc > dev_attr_store+0x38/0x88 > sysfs_kf_write+0x64/0xc0 > kernfs_fop_write_iter+0x108/0x1d0 > vfs_write+0x2f4/0x368 > ksys_write+0x7c/0xec > __arm64_sys_write+0x20/0x30 > el0_svc_common.llvm.7279915941325364641+0xbc/0x1bc > do_el0_svc+0x28/0xa0 > el0_svc+0x14/0x24 > el0_sync_handler+0x88/0xec > el0_sync+0x1c0/0x200 > > While at it, fix the possible NULL pointer dereference in other > functions as well: of_thermal_get_temp(), of_thermal_set_emul_temp(), > of_thermal_get_trend(). Can the subject be more specific, please? The issue appears to be limited to the of_thermal_ family of functions, but the subject doesn't reflect that at all. > Suggested-by: David Collins <quic_collinsd@quicinc.com> > Signed-off-by: Subbaraman Narayanamurthy <quic_subbaram@quicinc.com> Daniel, any concerns regarding the code changes below? > --- > drivers/thermal/thermal_of.c | 9 ++++++--- > 1 file changed, 6 insertions(+), 3 deletions(-) > > diff --git a/drivers/thermal/thermal_of.c b/drivers/thermal/thermal_of.c > index 6379f26..9233f7e 100644 > --- a/drivers/thermal/thermal_of.c > +++ b/drivers/thermal/thermal_of.c > @@ -89,7 +89,7 @@ static int of_thermal_get_temp(struct thermal_zone_device *tz, > { > struct __thermal_zone *data = tz->devdata; > > - if (!data->ops->get_temp) > + if (!data->ops || !data->ops->get_temp) > return -EINVAL; > > return data->ops->get_temp(data->sensor_data, temp); > @@ -186,6 +186,9 @@ static int of_thermal_set_emul_temp(struct thermal_zone_device *tz, > { > struct __thermal_zone *data = tz->devdata; > > + if (!data->ops || !data->ops->set_emul_temp) > + return -EINVAL; > + > return data->ops->set_emul_temp(data->sensor_data, temp); > } > > @@ -194,7 +197,7 @@ static int of_thermal_get_trend(struct thermal_zone_device *tz, int trip, > { > struct __thermal_zone *data = tz->devdata; > > - if (!data->ops->get_trend) > + if (!data->ops || !data->ops->get_trend) > return -EINVAL; > > return data->ops->get_trend(data->sensor_data, trip, trend); > @@ -301,7 +304,7 @@ static int of_thermal_set_trip_temp(struct thermal_zone_device *tz, int trip, > if (trip >= data->ntrips || trip < 0) > return -EDOM; > > - if (data->ops->set_trip_temp) { > + if (data->ops && data->ops->set_trip_temp) { > int ret; > > ret = data->ops->set_trip_temp(data->sensor_data, trip, temp); > -- > 2.7.4 >
On 05/11/2021 16:14, Rafael J. Wysocki wrote: > On Fri, Nov 5, 2021 at 12:57 AM Subbaraman Narayanamurthy > <quic_subbaram@quicinc.com> wrote: >> >> of_parse_thermal_zones() parses the thermal-zones node and registers a >> thermal_zone device for each subnode. However, if a thermal zone is >> consuming a thermal sensor and that thermal sensor device hasn't probed >> yet, an attempt to set trip_point_*_temp for that thermal zone device >> can cause a NULL pointer dereference. Fix it. >> >> console:/sys/class/thermal/thermal_zone87 # echo 120000 > trip_point_0_temp >> ... >> Unable to handle kernel NULL pointer dereference at virtual address 0000000000000020 >> ... >> Call trace: >> of_thermal_set_trip_temp+0x40/0xc4 >> trip_point_temp_store+0xc0/0x1dc >> dev_attr_store+0x38/0x88 >> sysfs_kf_write+0x64/0xc0 >> kernfs_fop_write_iter+0x108/0x1d0 >> vfs_write+0x2f4/0x368 >> ksys_write+0x7c/0xec >> __arm64_sys_write+0x20/0x30 >> el0_svc_common.llvm.7279915941325364641+0xbc/0x1bc >> do_el0_svc+0x28/0xa0 >> el0_svc+0x14/0x24 >> el0_sync_handler+0x88/0xec >> el0_sync+0x1c0/0x200 >> >> While at it, fix the possible NULL pointer dereference in other >> functions as well: of_thermal_get_temp(), of_thermal_set_emul_temp(), >> of_thermal_get_trend(). > > Can the subject be more specific, please? > > The issue appears to be limited to the of_thermal_ family of > functions, but the subject doesn't reflect that at all. > >> Suggested-by: David Collins <quic_collinsd@quicinc.com> >> Signed-off-by: Subbaraman Narayanamurthy <quic_subbaram@quicinc.com> > > Daniel, any concerns regarding the code changes below? I've a concern about the root cause but I did not have time to investigate how to fix it nicely. thermal_of is responsible of introducing itself between the thermal core code and the backend. So it defines the ops which in turn call the sensor ops leading us to this problem. So, without a better solution, this fix can be applied until we rethink the thermal_of approach. Acked-by: Daniel Lezcano <daniel.lezcano@linaro.org>
On Fri, Nov 5, 2021 at 5:19 PM Daniel Lezcano <daniel.lezcano@linaro.org> wrote: > > On 05/11/2021 16:14, Rafael J. Wysocki wrote: > > On Fri, Nov 5, 2021 at 12:57 AM Subbaraman Narayanamurthy > > <quic_subbaram@quicinc.com> wrote: > >> > >> of_parse_thermal_zones() parses the thermal-zones node and registers a > >> thermal_zone device for each subnode. However, if a thermal zone is > >> consuming a thermal sensor and that thermal sensor device hasn't probed > >> yet, an attempt to set trip_point_*_temp for that thermal zone device > >> can cause a NULL pointer dereference. Fix it. > >> > >> console:/sys/class/thermal/thermal_zone87 # echo 120000 > trip_point_0_temp > >> ... > >> Unable to handle kernel NULL pointer dereference at virtual address 0000000000000020 > >> ... > >> Call trace: > >> of_thermal_set_trip_temp+0x40/0xc4 > >> trip_point_temp_store+0xc0/0x1dc > >> dev_attr_store+0x38/0x88 > >> sysfs_kf_write+0x64/0xc0 > >> kernfs_fop_write_iter+0x108/0x1d0 > >> vfs_write+0x2f4/0x368 > >> ksys_write+0x7c/0xec > >> __arm64_sys_write+0x20/0x30 > >> el0_svc_common.llvm.7279915941325364641+0xbc/0x1bc > >> do_el0_svc+0x28/0xa0 > >> el0_svc+0x14/0x24 > >> el0_sync_handler+0x88/0xec > >> el0_sync+0x1c0/0x200 > >> > >> While at it, fix the possible NULL pointer dereference in other > >> functions as well: of_thermal_get_temp(), of_thermal_set_emul_temp(), > >> of_thermal_get_trend(). > > > > Can the subject be more specific, please? > > > > The issue appears to be limited to the of_thermal_ family of > > functions, but the subject doesn't reflect that at all. > > > >> Suggested-by: David Collins <quic_collinsd@quicinc.com> > >> Signed-off-by: Subbaraman Narayanamurthy <quic_subbaram@quicinc.com> > > > > Daniel, any concerns regarding the code changes below? > > I've a concern about the root cause but I did not have time to > investigate how to fix it nicely. > > thermal_of is responsible of introducing itself between the thermal core > code and the backend. So it defines the ops which in turn call the > sensor ops leading us to this problem. > > So, without a better solution, this fix can be applied until we rethink > the thermal_of approach. > > Acked-by: Daniel Lezcano <daniel.lezcano@linaro.org> Thanks! I've queued it up for 5.16-rc as "thermal: Fix NULL pointer dereferences in of_thermal_ functions".
On 11/4/21 11:50 PM, Greg KH wrote: > On Thu, Nov 04, 2021 at 04:57:07PM -0700, Subbaraman Narayanamurthy wrote: >> of_parse_thermal_zones() parses the thermal-zones node and registers a >> thermal_zone device for each subnode. However, if a thermal zone is >> consuming a thermal sensor and that thermal sensor device hasn't probed >> yet, an attempt to set trip_point_*_temp for that thermal zone device >> can cause a NULL pointer dereference. Fix it. >> >> console:/sys/class/thermal/thermal_zone87 # echo 120000 > trip_point_0_temp >> ... >> Unable to handle kernel NULL pointer dereference at virtual address 0000000000000020 >> ... >> Call trace: >> of_thermal_set_trip_temp+0x40/0xc4 >> trip_point_temp_store+0xc0/0x1dc >> dev_attr_store+0x38/0x88 >> sysfs_kf_write+0x64/0xc0 >> kernfs_fop_write_iter+0x108/0x1d0 >> vfs_write+0x2f4/0x368 >> ksys_write+0x7c/0xec >> __arm64_sys_write+0x20/0x30 >> el0_svc_common.llvm.7279915941325364641+0xbc/0x1bc >> do_el0_svc+0x28/0xa0 >> el0_svc+0x14/0x24 >> el0_sync_handler+0x88/0xec >> el0_sync+0x1c0/0x200 >> >> While at it, fix the possible NULL pointer dereference in other >> functions as well: of_thermal_get_temp(), of_thermal_set_emul_temp(), >> of_thermal_get_trend(). >> >> Suggested-by: David Collins <quic_collinsd@quicinc.com> >> Signed-off-by: Subbaraman Narayanamurthy <quic_subbaram@quicinc.com> >> --- >> drivers/thermal/thermal_of.c | 9 ++++++--- >> 1 file changed, 6 insertions(+), 3 deletions(-) >> > <formletter> > > This is not the correct way to submit patches for inclusion in the > stable kernel tree. Please read: > https://www.kernel.org/doc/html/latest/process/stable-kernel-rules.html > for how to do this properly. > > </formletter> Hi Greg, For this case, is it because I've missed adding "Cc:stable@vger.kernel.org" in commit text itself and cc-ed stable@vger.kernel.org directly? Thanks, Subbaraman
On 11/5/21 9:37 AM, Rafael J. Wysocki wrote: > On Fri, Nov 5, 2021 at 5:19 PM Daniel Lezcano <daniel.lezcano@linaro.org> wrote: >> On 05/11/2021 16:14, Rafael J. Wysocki wrote: >>> On Fri, Nov 5, 2021 at 12:57 AM Subbaraman Narayanamurthy >>> <quic_subbaram@quicinc.com> wrote: >>>> of_parse_thermal_zones() parses the thermal-zones node and registers a >>>> thermal_zone device for each subnode. However, if a thermal zone is >>>> consuming a thermal sensor and that thermal sensor device hasn't probed >>>> yet, an attempt to set trip_point_*_temp for that thermal zone device >>>> can cause a NULL pointer dereference. Fix it. >>>> >>>> console:/sys/class/thermal/thermal_zone87 # echo 120000 > trip_point_0_temp >>>> ... >>>> Unable to handle kernel NULL pointer dereference at virtual address 0000000000000020 >>>> ... >>>> Call trace: >>>> of_thermal_set_trip_temp+0x40/0xc4 >>>> trip_point_temp_store+0xc0/0x1dc >>>> dev_attr_store+0x38/0x88 >>>> sysfs_kf_write+0x64/0xc0 >>>> kernfs_fop_write_iter+0x108/0x1d0 >>>> vfs_write+0x2f4/0x368 >>>> ksys_write+0x7c/0xec >>>> __arm64_sys_write+0x20/0x30 >>>> el0_svc_common.llvm.7279915941325364641+0xbc/0x1bc >>>> do_el0_svc+0x28/0xa0 >>>> el0_svc+0x14/0x24 >>>> el0_sync_handler+0x88/0xec >>>> el0_sync+0x1c0/0x200 >>>> >>>> While at it, fix the possible NULL pointer dereference in other >>>> functions as well: of_thermal_get_temp(), of_thermal_set_emul_temp(), >>>> of_thermal_get_trend(). >>> Can the subject be more specific, please? >>> >>> The issue appears to be limited to the of_thermal_ family of >>> functions, but the subject doesn't reflect that at all. >>> >>>> Suggested-by: David Collins <quic_collinsd@quicinc.com> >>>> Signed-off-by: Subbaraman Narayanamurthy <quic_subbaram@quicinc.com> >>> Daniel, any concerns regarding the code changes below? >> I've a concern about the root cause but I did not have time to >> investigate how to fix it nicely. >> >> thermal_of is responsible of introducing itself between the thermal core >> code and the backend. So it defines the ops which in turn call the >> sensor ops leading us to this problem. >> >> So, without a better solution, this fix can be applied until we rethink >> the thermal_of approach. >> >> Acked-by: Daniel Lezcano <daniel.lezcano@linaro.org> > Thanks! > > I've queued it up for 5.16-rc as "thermal: Fix NULL pointer > dereferences in of_thermal_ functions". Thanks, Daniel and Rafael. So, I guess I don't need to send v3 with fixing commit subject right? -Subbaraman
On Fri, Nov 5, 2021 at 9:08 PM Subbaraman Narayanamurthy <quic_subbaram@quicinc.com> wrote: > > On 11/5/21 9:37 AM, Rafael J. Wysocki wrote: > > On Fri, Nov 5, 2021 at 5:19 PM Daniel Lezcano <daniel.lezcano@linaro.org> wrote: > >> On 05/11/2021 16:14, Rafael J. Wysocki wrote: > >>> On Fri, Nov 5, 2021 at 12:57 AM Subbaraman Narayanamurthy > >>> <quic_subbaram@quicinc.com> wrote: > >>>> of_parse_thermal_zones() parses the thermal-zones node and registers a > >>>> thermal_zone device for each subnode. However, if a thermal zone is > >>>> consuming a thermal sensor and that thermal sensor device hasn't probed > >>>> yet, an attempt to set trip_point_*_temp for that thermal zone device > >>>> can cause a NULL pointer dereference. Fix it. > >>>> > >>>> console:/sys/class/thermal/thermal_zone87 # echo 120000 > trip_point_0_temp > >>>> ... > >>>> Unable to handle kernel NULL pointer dereference at virtual address 0000000000000020 > >>>> ... > >>>> Call trace: > >>>> of_thermal_set_trip_temp+0x40/0xc4 > >>>> trip_point_temp_store+0xc0/0x1dc > >>>> dev_attr_store+0x38/0x88 > >>>> sysfs_kf_write+0x64/0xc0 > >>>> kernfs_fop_write_iter+0x108/0x1d0 > >>>> vfs_write+0x2f4/0x368 > >>>> ksys_write+0x7c/0xec > >>>> __arm64_sys_write+0x20/0x30 > >>>> el0_svc_common.llvm.7279915941325364641+0xbc/0x1bc > >>>> do_el0_svc+0x28/0xa0 > >>>> el0_svc+0x14/0x24 > >>>> el0_sync_handler+0x88/0xec > >>>> el0_sync+0x1c0/0x200 > >>>> > >>>> While at it, fix the possible NULL pointer dereference in other > >>>> functions as well: of_thermal_get_temp(), of_thermal_set_emul_temp(), > >>>> of_thermal_get_trend(). > >>> Can the subject be more specific, please? > >>> > >>> The issue appears to be limited to the of_thermal_ family of > >>> functions, but the subject doesn't reflect that at all. > >>> > >>>> Suggested-by: David Collins <quic_collinsd@quicinc.com> > >>>> Signed-off-by: Subbaraman Narayanamurthy <quic_subbaram@quicinc.com> > >>> Daniel, any concerns regarding the code changes below? > >> I've a concern about the root cause but I did not have time to > >> investigate how to fix it nicely. > >> > >> thermal_of is responsible of introducing itself between the thermal core > >> code and the backend. So it defines the ops which in turn call the > >> sensor ops leading us to this problem. > >> > >> So, without a better solution, this fix can be applied until we rethink > >> the thermal_of approach. > >> > >> Acked-by: Daniel Lezcano <daniel.lezcano@linaro.org> > > Thanks! > > > > I've queued it up for 5.16-rc as "thermal: Fix NULL pointer > > dereferences in of_thermal_ functions". > > Thanks, Daniel and Rafael. So, I guess I don't need to send v3 with fixing commit subject right? Right.
diff --git a/drivers/thermal/thermal_of.c b/drivers/thermal/thermal_of.c index 6379f26..9233f7e 100644 --- a/drivers/thermal/thermal_of.c +++ b/drivers/thermal/thermal_of.c @@ -89,7 +89,7 @@ static int of_thermal_get_temp(struct thermal_zone_device *tz, { struct __thermal_zone *data = tz->devdata; - if (!data->ops->get_temp) + if (!data->ops || !data->ops->get_temp) return -EINVAL; return data->ops->get_temp(data->sensor_data, temp); @@ -186,6 +186,9 @@ static int of_thermal_set_emul_temp(struct thermal_zone_device *tz, { struct __thermal_zone *data = tz->devdata; + if (!data->ops || !data->ops->set_emul_temp) + return -EINVAL; + return data->ops->set_emul_temp(data->sensor_data, temp); } @@ -194,7 +197,7 @@ static int of_thermal_get_trend(struct thermal_zone_device *tz, int trip, { struct __thermal_zone *data = tz->devdata; - if (!data->ops->get_trend) + if (!data->ops || !data->ops->get_trend) return -EINVAL; return data->ops->get_trend(data->sensor_data, trip, trend); @@ -301,7 +304,7 @@ static int of_thermal_set_trip_temp(struct thermal_zone_device *tz, int trip, if (trip >= data->ntrips || trip < 0) return -EDOM; - if (data->ops->set_trip_temp) { + if (data->ops && data->ops->set_trip_temp) { int ret; ret = data->ops->set_trip_temp(data->sensor_data, trip, temp);
of_parse_thermal_zones() parses the thermal-zones node and registers a thermal_zone device for each subnode. However, if a thermal zone is consuming a thermal sensor and that thermal sensor device hasn't probed yet, an attempt to set trip_point_*_temp for that thermal zone device can cause a NULL pointer dereference. Fix it. console:/sys/class/thermal/thermal_zone87 # echo 120000 > trip_point_0_temp ... Unable to handle kernel NULL pointer dereference at virtual address 0000000000000020 ... Call trace: of_thermal_set_trip_temp+0x40/0xc4 trip_point_temp_store+0xc0/0x1dc dev_attr_store+0x38/0x88 sysfs_kf_write+0x64/0xc0 kernfs_fop_write_iter+0x108/0x1d0 vfs_write+0x2f4/0x368 ksys_write+0x7c/0xec __arm64_sys_write+0x20/0x30 el0_svc_common.llvm.7279915941325364641+0xbc/0x1bc do_el0_svc+0x28/0xa0 el0_svc+0x14/0x24 el0_sync_handler+0x88/0xec el0_sync+0x1c0/0x200 While at it, fix the possible NULL pointer dereference in other functions as well: of_thermal_get_temp(), of_thermal_set_emul_temp(), of_thermal_get_trend(). Suggested-by: David Collins <quic_collinsd@quicinc.com> Signed-off-by: Subbaraman Narayanamurthy <quic_subbaram@quicinc.com> --- drivers/thermal/thermal_of.c | 9 ++++++--- 1 file changed, 6 insertions(+), 3 deletions(-)