Skip to contentSkip to conversation

ath10k: fix different tx duration output

PATCH In review Lei Wang 3 participants 8 messages Apr 17, 2019 – May 7, 2019 Qualcomm Atheros ATH10K wireless driver

lore
Share & tools
lore

Lei Wang

author
#0 lore raw

PATCH v2 1/1 ath10k: fix different tx duration output

TX duration output of tx_stats in debugfs and station dump had big difference because they got tx duration value from different statistic data. We should use the same statistic data.

Tested: QCA988X with firmware ver 10.2.4-1.0-00043

---
 drivers/net/wireless/ath/ath10k/mac.c | 5 +++++
 1 file changed, 5 insertions(+)
drivers/net/wireless/ath/ath10k/mac.c +5 -0
index b73c23d..5414169 100644
@@ -7977,6 +7977,11 @@ static void ath10k_sta_statistics(struct ieee80211_hw *hw,
 	sinfo->rx_duration = arsta->rx_duration;
 	sinfo->filled |= BIT_ULL(NL80211_STA_INFO_RX_DURATION);
 
+	if (arsta->tx_stats && ath10k_debug_is_extd_tx_stats_enabled(ar)) {
+		sinfo->tx_duration = arsta->tx_stats->tx_duration;
+		sinfo->filled |= 1ULL << NL80211_STA_INFO_TX_DURATION;
+	}
+
 	if (!arsta->txrate.legacy && !arsta->txrate.nss)
 		return;
 
Signature · 1 line
1.9.1
#1 lore raw
Quoted 3 lines · Lei Wang

TX duration output of tx_stats in debugfs and station dump had big difference because they got tx duration value from different statistic data. We should use the same statistic data.

So are you sure you picked the most accurate one of the two? :)

-Toke

#2 lore raw
Quoted 9 lines · Toke Høiland-Jørgensen
Quoted 3 lines · Lei Wang

TX duration output of tx_stats in debugfs and station dump had big difference because they got tx duration value from different statistic data. We should use the same statistic data.

So are you sure you picked the most accurate one of the two? :)

-Toke

Hi Toke,

Yes.
Now for ath10k, there are two ways to get tx duration output.
One is got from tx_stats in debugfs reported by firmware. It is a total value including all the frames which created by host and firmware sent to the peer. And the second is calculated from ath10k_htt_rx_tx_compl_ind()-->ieee80211_sta_register_airtime(), here the tx duration just includes the data frames sent from host to the peer. So the first value is preferable for station dump.

Thanks.
Lei

_______________________________________________
ath10k mailing list
[email protected]
http://lists.infradead.org/mailman/listinfo/ath10k

#3 lore raw
Quoted 13 lines · [email protected] writes:
Quoted 9 lines · Toke Høiland-Jørgensen
Quoted 3 lines · Lei Wang

TX duration output of tx_stats in debugfs and station dump had big difference because they got tx duration value from different statistic data. We should use the same statistic data.

So are you sure you picked the most accurate one of the two? :)

-Toke

Hi Toke,

Quoted 9 lines

Yes.
Now for ath10k, there are two ways to get tx duration output. One is got from tx_stats in debugfs reported by firmware. It is a total value including all the frames which created by host and firmware sent to the peer.
And the second is calculated from ath10k_htt_rx_tx_compl_ind()-->ieee80211_sta_register_airtime(), here the tx duration just includes the data frames sent from host to the peer.

So the difference is that the former includes control frames as well? Is that the only difference? And what exactly is a "big difference" (from the commit message)?

Quoted 1 line

So the first value is preferable for station dump.

Hmm, I'm not sure if I agree with this. I specifically added the tx_duration to the station dump to be able to get the values used by the airtime scheduler. This breaks with this patch.

-Toke

_______________________________________________
ath10k mailing list
[email protected]
http://lists.infradead.org/mailman/listinfo/ath10k

#4 lore raw
Quoted 25 lines · Toke Høiland-Jørgensen
Quoted 13 lines · [email protected] writes:
Quoted 9 lines · Toke Høiland-Jørgensen
Quoted 3 lines · Lei Wang

TX duration output of tx_stats in debugfs and station dump had big difference because they got tx duration value from different statistic data. We should use the same statistic data.

So are you sure you picked the most accurate one of the two? :)

-Toke

Hi Toke,

Quoted 9 lines

Yes.
Now for ath10k, there are two ways to get tx duration output. One is got from tx_stats in debugfs reported by firmware. It is a total value including all the frames which created by host and firmware sent to the peer.
And the second is calculated from ath10k_htt_rx_tx_compl_ind()-->ieee80211_sta_register_airtime(), here the tx duration just includes the data frames sent from host to the peer.

Quoted 4 lines

So the difference is that the former includes control frames as well? Is that the only difference? And what exactly is a "big difference" (from the commit message)?

Yes,it adds the duration time of receiving ACK frames.
From my test,TX from AP to station with iperf UDP test in 10s,tx_stats->tx_duration:5496623us, and another value is 3934327us.

Quoted 7 lines
Quoted 1 line

So the first value is preferable for station dump.

Hmm, I'm not sure if I agree with this. I specifically added the tx_duration to the station dump to be able to get the values used by the airtime scheduler. This breaks with this patch.

-Toke

From our internal discussing, we will revert this change. Thanks.

Lei

_______________________________________________
ath10k mailing list
[email protected]
http://lists.infradead.org/mailman/listinfo/ath10k

#5 lore raw
Quoted 33 lines · [email protected] writes:
Quoted 27 lines · Toke Høiland-Jørgensen
Quoted 14 lines · [email protected] writes:
Quoted 10 lines · Toke Høiland-Jørgensen
Quoted 4 lines · Lei Wang

TX duration output of tx_stats in debugfs and station dump had big difference because they got tx duration value from different statistic
data. We should use the same statistic data.

So are you sure you picked the most accurate one of the two? :)

-Toke

Hi Toke,

Quoted 10 lines

Yes.
Now for ath10k, there are two ways to get tx duration output. One is got from tx_stats in debugfs reported by firmware. It is a total
value including all the frames which created by host and firmware sent to the peer.
And the second is calculated from ath10k_htt_rx_tx_compl_ind()-->ieee80211_sta_register_airtime(), here the tx duration just includes the data frames sent from host to the peer.

Quoted 5 lines

So the difference is that the former includes control frames as well? Is
that the only difference? And what exactly is a "big difference" (from the commit message)?

Quoted 4 lines

Yes,it adds the duration time of receiving ACK frames.
From my test,TX from AP to station with iperf UDP test in
10s,tx_stats->tx_duration:5496623us,
and another value is 3934327us.

Hmm, that's quite a big difference. Is this really only ACKs, or is it also a question of whether retries are accounted? If so, it may actually be that what we should do is change which value is passed to ieee80211_sta_register_airtime()?

Quoted 9 lines
Quoted 8 lines
Quoted 1 line

So the first value is preferable for station dump.

Hmm, I'm not sure if I agree with this. I specifically added the tx_duration to the station dump to be able to get the values used by the
airtime scheduler. This breaks with this patch.

-Toke

From our internal discussing, we will revert this change.

Cool, but see above :)

-Toke

#6 lore raw
In reply to Lei Wang
Quoted 8 lines · Lei Wang

TX duration output of tx_stats in debugfs and station dump had big difference because they got tx duration value from different statistic data. We should use the same statistic data.

Tested: QCA988X with firmware ver 10.2.4-1.0-00043

Dropping per discussion, please resend once everything is clarified.

Patch set to Changes Requested.

Signature · 3 lines
https://patchwork.kernel.org/patch/10904909/

https://wireless.wiki.kernel.org/en/developers/documentation/submittingpatches
#7 lore raw
In reply to Lei Wang
Quoted 8 lines · Lei Wang

TX duration output of tx_stats in debugfs and station dump had big difference because they got tx duration value from different statistic data. We should use the same statistic data.

Tested: QCA988X with firmware ver 10.2.4-1.0-00043

Dropping per discussion, please resend once everything is clarified.

Patch set to Changes Requested.

Signature · 9 lines
https://patchwork.kernel.org/patch/10904909/

https://wireless.wiki.kernel.org/en/developers/documentation/submittingpatches


_______________________________________________
ath10k mailing list
[email protected]
http://lists.infradead.org/mailman/listinfo/ath10k