ath9k_hw: Fix AR9003 MPDU delimeter CRC check for middle subframes

An A-MPDU may contain several subframes each containing its own
CRC for the data. Each subframe also has a respective CRC for the
MPDU length and 4 reserved bits (aka delimeter CRC). AR9003 will
ACK frames that have a valid data CRC but have failed to pass the
CRC for the MPDU length, if and only if the subframe is not the
last subframe in an A-MPDU and if an OFDM phy OFDM reset error has
been caught. Discarding those subframes results in packet loss under
heavy stress conditions, an example being UDP video. Since the
frames are ACK'd by hardware we need to let these frames through
and process them as valid frames.

Cc: Tushit Jain <tushit.jain@atheros.com>
Cc: Kyungwan Nam <kyungwan.nam@atheros.com>
Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>

1 files changed, 29 insertions, 2 deletions

diff --git a/drivers/net/wireless/ath/ath9k/ar9003_mac.c b/drivers/net/wireless/ath/ath9k/ar9003_mac.c
index 06ef71019c12..5b995bee70ae 100644
--- a/drivers/net/wireless/ath/ath9k/ar9003_mac.c
+++ b/drivers/net/wireless/ath/ath9k/ar9003_mac.c
@@ -579,12 +579,39 @@ int ath9k_hw_process_rxdesc_edma(struct ath_hw *ah, struct ath_rx_status *rxs,
                 rxs->rs_flags |= ATH9K_RX_DECRYPT_BUSY;
 
         if ((rxsp->status11 & AR_RxFrameOK) == 0) {
+                /*
+                 * AR_CRCErr will bet set to true if we're on the last
+                 * subframe and the AR_PostDelimCRCErr is caught.
+                 * In a way this also gives us a guarantee that when
+                 * (!(AR_CRCErr) && (AR_PostDelimCRCErr)) we cannot
+                 * possibly be reviewing the last subframe. AR_CRCErr
+                 * is the CRC of the actual data.
+                 */
                 if (rxsp->status11 & AR_CRCErr) {
                         rxs->rs_status |= ATH9K_RXERR_CRC;
                 } else if (rxsp->status11 & AR_PHYErr) {
-                        rxs->rs_status |= ATH9K_RXERR_PHY;
                         phyerr = MS(rxsp->status11, AR_PHYErrCode);
-                        rxs->rs_phyerr = phyerr;
+                        /*
+                         * If we reach a point here where AR_PostDelimCRCErr is
+                         * true it implies we're *not* on the last subframe. In
+                         * in that case that we know already that the CRC of
+                         * the frame was OK, and MAC would send an ACK for that
+                         * subframe, even if we did get a phy error of type
+                         * ATH9K_PHYERR_OFDM_RESTART. This is only applicable
+                         * to frame that are prior to the last subframe.
+                         * The AR_PostDelimCRCErr is the CRC for the MPDU
+                         * delimiter, which contains the 4 reserved bits,
+                         * the MPDU length (12 bits), and follows the MPDU
+                         * delimiter for an A-MPDU subframe (0x4E = 'N' ASCII).
+                         */
+                        if ((phyerr == ATH9K_PHYERR_OFDM_RESTART) &&
+                            (rxsp->status11 & AR_PostDelimCRCErr)) {
+                                rxs->rs_phyerr = 0;
+                        } else {
+                                rxs->rs_status |= ATH9K_RXERR_PHY;
+                                rxs->rs_phyerr = phyerr;
+                        }
+
                 } else if (rxsp->status11 & AR_DecryptCRCErr) {
                         rxs->rs_status |= ATH9K_RXERR_DECRYPT;
                 } else if (rxsp->status11 & AR_MichaelErr) {