mac80211: always account for A-MSDU header changes
authorJohannes Berg <johannes.berg@intel.com>
Thu, 30 Aug 2018 08:55:49 +0000 (10:55 +0200)
committerJohannes Berg <johannes.berg@intel.com>
Thu, 30 Aug 2018 09:04:19 +0000 (11:04 +0200)
In the error path of changing the SKB headroom of the second
A-MSDU subframe, we would not account for the already-changed
length of the first frame that just got converted to be in
A-MSDU format and thus is a bit longer now.

Fix this by doing the necessary accounting.

It would be possible to reorder the operations, but that would
make the code more complex (to calculate the necessary pad),
and the headroom expansion should not fail frequently enough
to make that worthwhile.

Fixes: 6e0456b54545 ("mac80211: add A-MSDU tx support")
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Acked-by: Lorenzo Bianconi <lorenzo.bianconi@redhat.com>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
net/mac80211/tx.c

index 1aac5e3c7eeeadf4060f93c247d456941d659ea7..6ca0865de94518cb3f313657d37beda16dc39e08 100644 (file)
@@ -3239,7 +3239,7 @@ static bool ieee80211_amsdu_aggregate(struct ieee80211_sub_if_data *sdata,
 
        if (!ieee80211_amsdu_realloc_pad(local, skb, sizeof(rfc1042_header) +
                                                     2 + pad))
-               goto out;
+               goto out_recalc;
 
        ret = true;
        data = skb_push(skb, ETH_ALEN + 2);
@@ -3256,11 +3256,13 @@ static bool ieee80211_amsdu_aggregate(struct ieee80211_sub_if_data *sdata,
        head->data_len += skb->len;
        *frag_tail = skb;
 
-       flow->backlog += head->len - orig_len;
-       tin->backlog_bytes += head->len - orig_len;
-
-       fq_recalc_backlog(fq, tin, flow);
+out_recalc:
+       if (head->len != orig_len) {
+               flow->backlog += head->len - orig_len;
+               tin->backlog_bytes += head->len - orig_len;
 
+               fq_recalc_backlog(fq, tin, flow);
+       }
 out:
        spin_unlock_bh(&fq->lock);