linux-wireless.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Felix Fietkau <nbd@nbd.name>
To: linux-wireless@vger.kernel.org
Cc: toke@kernel.org, johannes@sipsolutions.net
Subject: [PATCH 3/7] mac80211: consider aql_tx_pending when checking airtime deficit
Date: Sat, 25 Jun 2022 23:24:07 +0200	[thread overview]
Message-ID: <20220625212411.36675-3-nbd@nbd.name> (raw)
In-Reply-To: <20220625212411.36675-1-nbd@nbd.name>

When queueing packets for a station, deficit only gets added once the packets
have been transmitted, which could be much later. During that time, a lot of
temporary unfairness could happen, which could lead to bursty behavior.
Fix this by subtracting the aql_tx_pending when checking the deficit in tx
scheduling.

Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 net/mac80211/tx.c | 11 +++++++++--
 1 file changed, 9 insertions(+), 2 deletions(-)

diff --git a/net/mac80211/tx.c b/net/mac80211/tx.c
index a615cadf7728..0509486ac40a 100644
--- a/net/mac80211/tx.c
+++ b/net/mac80211/tx.c
@@ -3800,6 +3800,13 @@ struct sk_buff *ieee80211_tx_dequeue(struct ieee80211_hw *hw,
 }
 EXPORT_SYMBOL(ieee80211_tx_dequeue);
 
+static inline s32 ieee80211_sta_deficit(struct sta_info *sta, u8 ac)
+{
+	struct airtime_info *air_info = &sta->airtime[ac];
+
+	return air_info->deficit - atomic_read(&air_info->aql_tx_pending);
+}
+
 struct ieee80211_txq *ieee80211_next_txq(struct ieee80211_hw *hw, u8 ac)
 {
 	struct ieee80211_local *local = hw_to_local(hw);
@@ -3830,7 +3837,7 @@ struct ieee80211_txq *ieee80211_next_txq(struct ieee80211_hw *hw, u8 ac)
 		struct sta_info *sta = container_of(txqi->txq.sta,
 						    struct sta_info, sta);
 		bool aql_check = ieee80211_txq_airtime_check(hw, &txqi->txq);
-		s32 deficit = sta->airtime[txqi->txq.ac].deficit;
+		s32 deficit = ieee80211_sta_deficit(sta, txqi->txq.ac);
 
 		if (aql_check)
 			found_eligible_txq = true;
@@ -3955,7 +3962,7 @@ bool ieee80211_txq_may_transmit(struct ieee80211_hw *hw,
 			continue;
 		}
 		sta = container_of(iter->txq.sta, struct sta_info, sta);
-		if (sta->airtime[ac].deficit < 0)
+		if (ieee80211_sta_deficit(sta, ac) < 0)
 			sta->airtime[ac].deficit += sta->airtime_weight;
 		list_move_tail(&iter->schedule_order, &local->active_txqs[ac]);
 	}
-- 
2.36.1


  parent reply	other threads:[~2022-06-25 21:24 UTC|newest]

Thread overview: 14+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-06-25 21:24 [PATCH 1/7] mac80211: switch airtime fairness back to deficit round-robin scheduling Felix Fietkau
2022-06-25 21:24 ` [PATCH 2/7] mac80211: make sta airtime deficit field s32 instead of s64 Felix Fietkau
2022-06-27 15:02   ` Toke Høiland-Jørgensen
2022-06-25 21:24 ` Felix Fietkau [this message]
2022-06-27 15:02   ` [PATCH 3/7] mac80211: consider aql_tx_pending when checking airtime deficit Toke Høiland-Jørgensen
2022-06-25 21:24 ` [PATCH 4/7] mac80211: keep recently active tx queues in scheduling list Felix Fietkau
2022-06-27 15:03   ` Toke Høiland-Jørgensen
2022-06-25 21:24 ` [PATCH 5/7] mac80211: add a per-PHY AQL limit to improve fairness Felix Fietkau
2022-06-27 15:04   ` Toke Høiland-Jørgensen
2022-06-25 21:24 ` [PATCH 6/7] mac80211: add debugfs file to display per-phy AQL pending airtime Felix Fietkau
2022-06-27 15:04   ` Toke Høiland-Jørgensen
2022-06-25 21:24 ` [PATCH 7/7] mac80211: only accumulate airtime deficit for active clients Felix Fietkau
2022-06-27 15:04   ` Toke Høiland-Jørgensen
2022-06-27 15:02 ` [PATCH 1/7] mac80211: switch airtime fairness back to deficit round-robin scheduling Toke Høiland-Jørgensen

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20220625212411.36675-3-nbd@nbd.name \
    --to=nbd@nbd.name \
    --cc=johannes@sipsolutions.net \
    --cc=linux-wireless@vger.kernel.org \
    --cc=toke@kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).