Hi Johannes,
We are noticing a issue in updating the multicast frame statistics.
Though mac80211 explicitly sets the IEEE80211_TX_CTL_NO_ACK, it checks
IEEE80211_TX_STAT_ACK to increment the frame count. So, for multicast
frames, is the driver expected to set this STAT_ACK if the frame is
successfully transmitted over the air?
The naming convention seems to be a bit confusing from unicast and
multicast frames perspectives:
For multicast, IEEE80211_TX_STAT_ACK can denote if a frame is
successfully transmitted over the air or failed to be transmitted.
In case of normal frames, IEEE80211_TX_STAT_ACK decides if the frame
is acked or not-acked/not-transmitted.
Do you think renaming this to something like STAT_SUCCESS would be a
better thing to do?
Vivek.
mac80211/status.c:
if (info->flags & IEEE80211_TX_STAT_ACK) {
if (ieee80211_is_first_frag(hdr->seq_ctrl)) {
local->dot11TransmittedFrameCount++;
if (is_multicast_ether_addr(hdr->addr1))
local->dot11MulticastTransmittedFrameCount++;
if (retry_count > 0)
local->dot11RetryCount++;
if (retry_count > 1)
local->dot11MultipleRetryCount++;
}
/* This counter shall be incremented for an acknowledged MPDU
* with an individual address in the address 1 field or an MPDU
* with a multicast address in the address 1 field of type Data
* or Management. */
if (!is_multicast_ether_addr(hdr->addr1) ||
ieee80211_is_data(fc) ||
ieee80211_is_mgmt(fc))
local->dot11TransmittedFragmentCount++;
} else {
if (ieee80211_is_first_frag(hdr->seq_ctrl))
local->dot11FailedCount++;
}
--
To unsubscribe from this list: send the line "unsubscribe linux-wireless" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
