| Message ID | 20190324231237.2246-1-kjlu@umn.edu (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
| Series | sound: codecs: fix a potential NULL pointer dereference | expand |
On 3/25/2019 4:42 AM, Kangjie Lu wrote: > In case devm_kzalloc fails, the patch returns -ENOMEM to avoid > potential NULL pointer dereference. > > Signed-off-by: Kangjie Lu <kjlu@umn.edu> > --- > sound/soc/codecs/rt5663.c | 2 ++ > 1 file changed, 2 insertions(+) > > diff --git a/sound/soc/codecs/rt5663.c b/sound/soc/codecs/rt5663.c > index da6647015708..909ab99a1995 100644 > --- a/sound/soc/codecs/rt5663.c > +++ b/sound/soc/codecs/rt5663.c > @@ -3480,6 +3480,8 @@ static int rt5663_parse_dp(struct rt5663_priv *rt5663, struct device *dev) > table_size = sizeof(struct impedance_mapping_table) * > rt5663->pdata.impedance_sensing_num; > rt5663->imp_table = devm_kzalloc(dev, table_size, GFP_KERNEL); > + if (!rt5663->imp_table) > + return -ENOMEM; add checks in rt5663_parse_dp as well. Thanks. Mukesh > device_property_read_u32_array(dev, > "realtek,impedance_sensing_table", > (u32 *)rt5663->imp_table, table_size);
Hi Kangjje/Aditya, Please do take care of the return value you are sending upstream whether still is checked or not otherwise NULL pointer dereference will still come. Also resource release properly otherwise your patch may looks simple but it can introduce memory leak as well in other path. Thanks, Mukesh On 3/25/2019 12:15 PM, Mukesh Ojha wrote: > > On 3/25/2019 4:42 AM, Kangjie Lu wrote: >> In case devm_kzalloc fails, the patch returns -ENOMEM to avoid >> potential NULL pointer dereference. >> >> Signed-off-by: Kangjie Lu <kjlu@umn.edu> >> --- >> sound/soc/codecs/rt5663.c | 2 ++ >> 1 file changed, 2 insertions(+) >> >> diff --git a/sound/soc/codecs/rt5663.c b/sound/soc/codecs/rt5663.c >> index da6647015708..909ab99a1995 100644 >> --- a/sound/soc/codecs/rt5663.c >> +++ b/sound/soc/codecs/rt5663.c >> @@ -3480,6 +3480,8 @@ static int rt5663_parse_dp(struct rt5663_priv >> *rt5663, struct device *dev) >> table_size = sizeof(struct impedance_mapping_table) * >> rt5663->pdata.impedance_sensing_num; >> rt5663->imp_table = devm_kzalloc(dev, table_size, GFP_KERNEL); >> + if (!rt5663->imp_table) >> + return -ENOMEM; > > add checks in rt5663_parse_dp as well. > > Thanks. > Mukesh > > >> device_property_read_u32_array(dev, >> "realtek,impedance_sensing_table", >> (u32 *)rt5663->imp_table, table_size);
diff --git a/sound/soc/codecs/rt5663.c b/sound/soc/codecs/rt5663.c index da6647015708..909ab99a1995 100644 --- a/sound/soc/codecs/rt5663.c +++ b/sound/soc/codecs/rt5663.c @@ -3480,6 +3480,8 @@ static int rt5663_parse_dp(struct rt5663_priv *rt5663, struct device *dev) table_size = sizeof(struct impedance_mapping_table) * rt5663->pdata.impedance_sensing_num; rt5663->imp_table = devm_kzalloc(dev, table_size, GFP_KERNEL); + if (!rt5663->imp_table) + return -ENOMEM; device_property_read_u32_array(dev, "realtek,impedance_sensing_table", (u32 *)rt5663->imp_table, table_size);
In case devm_kzalloc fails, the patch returns -ENOMEM to avoid potential NULL pointer dereference. Signed-off-by: Kangjie Lu <kjlu@umn.edu> --- sound/soc/codecs/rt5663.c | 2 ++ 1 file changed, 2 insertions(+)