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
next prev 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).