]> www.infradead.org Git - users/jedix/linux-maple.git/commit
mac80211: prevent possible crypto tx tailroom corruption
authorMichal Kazior <michal.kazior@tieto.com>
Fri, 22 May 2015 08:22:40 +0000 (10:22 +0200)
committerGreg Kroah-Hartman <gregkh@linuxfoundation.org>
Mon, 3 Aug 2015 16:29:16 +0000 (09:29 -0700)
commit063c47a030bf17eac2d9010f0a899be5f7f4db87
treee7b1f2066660eef3bee5eea3ca253d0786d22f22
parent7b53ca5c69fd9054c0fc3af97717176a8dacdb4f
mac80211: prevent possible crypto tx tailroom corruption

commit ab499db80fcf07c18e4053f91a619500f663e90e upstream.

There was a possible race between
ieee80211_reconfig() and
ieee80211_delayed_tailroom_dec(). This could
result in inability to transmit data if driver
crashed during roaming or rekeying and subsequent
skbs with insufficient tailroom appeared.

This race was probably never seen in the wild
because a device driver would have to crash AND
recover within 0.5s which is very unlikely.

I was able to prove this race exists after
changing the delay to 10s locally and crashing
ath10k via debugfs immediately after GTK
rekeying. In case of ath10k the counter went below
0. This was harmless but other drivers which
actually require tailroom (e.g. for WEP ICV or
MMIC) could end up with the counter at 0 instead
of >0 and introduce insufficient skb tailroom
failures because mac80211 would not resize skbs
appropriately anymore.

Fixes: 8d1f7ecd2af5 ("mac80211: defer tailroom counter manipulation when roaming")
Signed-off-by: Michal Kazior <michal.kazior@tieto.com>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
net/mac80211/main.c