| Message ID | 1401357008-16259-1-git-send-email-michal.kazior@tieto.com (mailing list archive) |
|---|---|
| State | Not Applicable, archived |
| Headers | show |
Michal Kazior <michal.kazior@tieto.com> writes: > While fixing a bug reported by Avery I went ahead > and added a warning suspecting there might be > something more to the bug. This ended up with > people reporting they see warnings during heavy > traffic. This bought me some time and helped me > understand the problem better - apparently fw/hw > can report a chained msdus as follows: 1 msdu, 1 > chained, 1 msdu (0 length). > > The patch removes the extra check but leaves the > other change that fixed the original skb_push > panic bug (msdu_chaining was overwritten in an > unfortunate way which made the above example to > be treated as non-chained case). > > Reported-by: Yeoh Chun-Yeow <yeohchunyeow@gmail.com> > Reported-by: Tim Harvey <tharvey@gateworks.com> > Signed-off-by: Michal Kazior <michal.kazior@tieto.com> Thanks, applied to ath-current branch.
diff --git a/drivers/net/wireless/ath/ath10k/htt_rx.c b/drivers/net/wireless/ath/ath10k/htt_rx.c index 6c102b1..eebc860 100644 --- a/drivers/net/wireless/ath/ath10k/htt_rx.c +++ b/drivers/net/wireless/ath/ath10k/htt_rx.c @@ -312,7 +312,6 @@ static int ath10k_htt_rx_amsdu_pop(struct ath10k_htt *htt, int msdu_len, msdu_chaining = 0; struct sk_buff *msdu; struct htt_rx_desc *rx_desc; - bool corrupted = false; lockdep_assert_held(&htt->rx_ring.lock); @@ -439,9 +438,6 @@ static int ath10k_htt_rx_amsdu_pop(struct ath10k_htt *htt, last_msdu = __le32_to_cpu(rx_desc->msdu_end.info0) & RX_MSDU_END_INFO0_LAST_MSDU; - if (msdu_chaining && !last_msdu) - corrupted = true; - if (last_msdu) { msdu->next = NULL; break; @@ -457,20 +453,6 @@ static int ath10k_htt_rx_amsdu_pop(struct ath10k_htt *htt, msdu_chaining = -1; /* - * Apparently FW sometimes reports weird chained MSDU sequences with - * more than one rx descriptor. This seems like a bug but needs more - * analyzing. For the time being fix it by dropping such sequences to - * avoid blowing up the host system. - */ - if (corrupted) { - ath10k_warn("failed to pop chained msdus, dropping\n"); - ath10k_htt_rx_free_msdu_chain(*head_msdu); - *head_msdu = NULL; - *tail_msdu = NULL; - msdu_chaining = -EINVAL; - } - - /* * Don't refill the ring yet. * * First, the elements popped here are still in use - it is not
While fixing a bug reported by Avery I went ahead and added a warning suspecting there might be something more to the bug. This ended up with people reporting they see warnings during heavy traffic. This bought me some time and helped me understand the problem better - apparently fw/hw can report a chained msdus as follows: 1 msdu, 1 chained, 1 msdu (0 length). The patch removes the extra check but leaves the other change that fixed the original skb_push panic bug (msdu_chaining was overwritten in an unfortunate way which made the above example to be treated as non-chained case). Reported-by: Yeoh Chun-Yeow <yeohchunyeow@gmail.com> Reported-by: Tim Harvey <tharvey@gateworks.com> Signed-off-by: Michal Kazior <michal.kazior@tieto.com> --- drivers/net/wireless/ath/ath10k/htt_rx.c | 18 ------------------ 1 file changed, 18 deletions(-)