hzy15610046011 <houzhanyi61@xxxxxxxxxxx> wrote: > > From: H-Mole <houzhanyi61@xxxxxxxxxxx> > > Since commit https://github.com/torvalds/linux/commit/f503ae90c7355e8506e68498fe84c1357894cd5b The pattern to reference commit is "commit + 12digits + subject", such as: commit f503ae90c735 ("wifi: mt76: mt7996: fix NULL pointer dereference in mt7996_mcu_sta_bfer_he") > showed that drivers/net/wireless/mediatek/mt76/mt7996/mcu.c misses a non-null judgement for the > return value of mt76_connac_get_he_phy_cap on variable vc, this null-pointer problem may also happen > on mt7915. This commit added such non-null judgement like what was added to function > mt7996_mcu_sta_bfer_he() > in commit f503ae90c7355e8506e68498fe84c1357894cd5b. Have you run checkpatch before sending out? > > Signed-off-by: H-Mole <houzhanyi61@xxxxxxxxxxx> > --- > drivers/net/wireless/mediatek/mt76/mt7915/mcu.c | 3 +++ > 1 file changed, 3 insertions(+) > > diff --git a/drivers/net/wireless/mediatek/mt76/mt7915/mcu.c > b/drivers/net/wireless/mediatek/mt76/mt7915/mcu.c > index 427542777abc..5dca50d40080 100644 > --- a/drivers/net/wireless/mediatek/mt76/mt7915/mcu.c > +++ b/drivers/net/wireless/mediatek/mt76/mt7915/mcu.c > @@ -1184,6 +1184,9 @@ mt7915_mcu_sta_bfer_he(struct ieee80211_sta *sta, struct ieee80211_vif *vif, > u8 nss_mcs = mt7915_mcu_get_sta_nss(mcs_map); > u8 snd_dim, sts; > > + if (!vc) > + return; > + Before checking vc, ve is induced by vc already. const struct ieee80211_he_cap_elem *ve = &vc->he_cap_elem; Though pointer doesn't really access the data, it would be better to get the pointer after checking vc. > bf->tx_mode = MT_PHY_TYPE_HE_SU; > > mt7915_mcu_sta_sounding_rate(bf); > -- > 2.39.5 >
