Message ID | 20180516084836.GB23043@mwanda (mailing list archive) |
---|---|
State | Not Applicable |
Delegated to: | Johannes Berg |
Headers | show |
Dan Carpenter <dan.carpenter@oracle.com> writes: > Smatch complains that we should use the passed in "gfp" instead of hard > coding GFP_KERNEL. I looked at some of the callers and this would > probably be a bug for rtw_cfg80211_indicate_sta_disassoc() which uses > GFP_ATOMIC and a NULL "sinfo". > > Fixes: 52539ca89f36 ("cfg80211: Expose TXQ stats and parameters to userspace") > Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com> Ah, right, didn't notice that that was supposed to be a gfp value. Thanks for the fix :) Acked-by: Toke Høiland-Jørgensen <toke@toke.dk> -Toke
diff --git a/net/wireless/nl80211.c b/net/wireless/nl80211.c index afbe5105bf7f..3eb645b81777 100644 --- a/net/wireless/nl80211.c +++ b/net/wireless/nl80211.c @@ -14644,7 +14644,7 @@ void cfg80211_del_sta_sinfo(struct net_device *dev, const u8 *mac_addr, struct station_info *empty_sinfo = NULL; if (!sinfo) { - empty_sinfo = kzalloc(sizeof(*empty_sinfo), GFP_KERNEL); + empty_sinfo = kzalloc(sizeof(*empty_sinfo), gfp); if (!empty_sinfo) return; sinfo = empty_sinfo;
Smatch complains that we should use the passed in "gfp" instead of hard coding GFP_KERNEL. I looked at some of the callers and this would probably be a bug for rtw_cfg80211_indicate_sta_disassoc() which uses GFP_ATOMIC and a NULL "sinfo". Fixes: 52539ca89f36 ("cfg80211: Expose TXQ stats and parameters to userspace") Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>