From: Erik Stromdahl <erik.stromdahl@gmail.com> To: johannes@sipsolutions.net, kvalo@codeaurora.org, davem@davemloft.net, linux-wireless@vger.kernel.org, ath10k@lists.infradead.org, linux-kernel@vger.kernel.org Cc: Erik Stromdahl <erik.stromdahl@gmail.com> Subject: [PATCH 1/2] mac80211: add tx dequeue function for process context Date: Mon, 17 Jun 2019 22:01:39 +0200 [thread overview] Message-ID: <20190617200140.6189-1-erik.stromdahl@gmail.com> (raw) Since ieee80211_tx_dequeue() must not be called with softirqs enabled (i.e. from process context without proper disable of bottom halves), we add a wrapper that disables bottom halves before calling ieee80211_tx_dequeue() The new function is named ieee80211_tx_dequeue_ni() just as all other from-process-context versions found in mac80211. The documentation of ieee80211_tx_dequeue() is also updated so it mentions that the function should not be called from process context. Signed-off-by: Erik Stromdahl <erik.stromdahl@gmail.com> --- include/net/mac80211.h | 26 ++++++++++++++++++++++++++ net/mac80211/tx.c | 2 ++ 2 files changed, 28 insertions(+) diff --git a/include/net/mac80211.h b/include/net/mac80211.h index 72080d9d617e..c47990d8db79 100644 --- a/include/net/mac80211.h +++ b/include/net/mac80211.h @@ -6251,10 +6251,36 @@ void ieee80211_unreserve_tid(struct ieee80211_sta *sta, u8 tid); * but for the duration of the frame handling. * However, also note that while in the wake_tx_queue() method, * rcu_read_lock() is already held. + * + * softirqs must also be disabled when this function is called. + * In process context, use ieee80211_tx_dequeue_ni() instead. */ struct sk_buff *ieee80211_tx_dequeue(struct ieee80211_hw *hw, struct ieee80211_txq *txq); +/** + * ieee80211_tx_dequeue_ni - dequeue a packet from a software tx queue + * (in process context) + * + * Like ieee80211_tx_dequeue() but can be called in process context + * (internally disables bottom halves). + * + * @hw: pointer as obtained from ieee80211_alloc_hw() + * @txq: pointer obtained from station or virtual interface, or from + * ieee80211_next_txq() + */ +static inline struct sk_buff *ieee80211_tx_dequeue_ni(struct ieee80211_hw *hw, + struct ieee80211_txq *txq) +{ + struct sk_buff *skb; + + local_bh_disable(); + skb = ieee80211_tx_dequeue(hw, txq); + local_bh_enable(); + + return skb; +} + /** * ieee80211_next_txq - get next tx queue to pull packets from * diff --git a/net/mac80211/tx.c b/net/mac80211/tx.c index dd220b977025..4bd0066ea7cd 100644 --- a/net/mac80211/tx.c +++ b/net/mac80211/tx.c @@ -3550,6 +3550,8 @@ struct sk_buff *ieee80211_tx_dequeue(struct ieee80211_hw *hw, ieee80211_tx_result r; struct ieee80211_vif *vif = txq->vif; + WARN_ON_ONCE(softirq_count() == 0); + begin: spin_lock_bh(&fq->lock); -- 2.22.0
WARNING: multiple messages have this Message-ID (diff)
From: Erik Stromdahl <erik.stromdahl@gmail.com> To: johannes@sipsolutions.net, kvalo@codeaurora.org, davem@davemloft.net, linux-wireless@vger.kernel.org, ath10k@lists.infradead.org, linux-kernel@vger.kernel.org Cc: Erik Stromdahl <erik.stromdahl@gmail.com> Subject: [PATCH 1/2] mac80211: add tx dequeue function for process context Date: Mon, 17 Jun 2019 22:01:39 +0200 [thread overview] Message-ID: <20190617200140.6189-1-erik.stromdahl@gmail.com> (raw) Since ieee80211_tx_dequeue() must not be called with softirqs enabled (i.e. from process context without proper disable of bottom halves), we add a wrapper that disables bottom halves before calling ieee80211_tx_dequeue() The new function is named ieee80211_tx_dequeue_ni() just as all other from-process-context versions found in mac80211. The documentation of ieee80211_tx_dequeue() is also updated so it mentions that the function should not be called from process context. Signed-off-by: Erik Stromdahl <erik.stromdahl@gmail.com> --- include/net/mac80211.h | 26 ++++++++++++++++++++++++++ net/mac80211/tx.c | 2 ++ 2 files changed, 28 insertions(+) diff --git a/include/net/mac80211.h b/include/net/mac80211.h index 72080d9d617e..c47990d8db79 100644 --- a/include/net/mac80211.h +++ b/include/net/mac80211.h @@ -6251,10 +6251,36 @@ void ieee80211_unreserve_tid(struct ieee80211_sta *sta, u8 tid); * but for the duration of the frame handling. * However, also note that while in the wake_tx_queue() method, * rcu_read_lock() is already held. + * + * softirqs must also be disabled when this function is called. + * In process context, use ieee80211_tx_dequeue_ni() instead. */ struct sk_buff *ieee80211_tx_dequeue(struct ieee80211_hw *hw, struct ieee80211_txq *txq); +/** + * ieee80211_tx_dequeue_ni - dequeue a packet from a software tx queue + * (in process context) + * + * Like ieee80211_tx_dequeue() but can be called in process context + * (internally disables bottom halves). + * + * @hw: pointer as obtained from ieee80211_alloc_hw() + * @txq: pointer obtained from station or virtual interface, or from + * ieee80211_next_txq() + */ +static inline struct sk_buff *ieee80211_tx_dequeue_ni(struct ieee80211_hw *hw, + struct ieee80211_txq *txq) +{ + struct sk_buff *skb; + + local_bh_disable(); + skb = ieee80211_tx_dequeue(hw, txq); + local_bh_enable(); + + return skb; +} + /** * ieee80211_next_txq - get next tx queue to pull packets from * diff --git a/net/mac80211/tx.c b/net/mac80211/tx.c index dd220b977025..4bd0066ea7cd 100644 --- a/net/mac80211/tx.c +++ b/net/mac80211/tx.c @@ -3550,6 +3550,8 @@ struct sk_buff *ieee80211_tx_dequeue(struct ieee80211_hw *hw, ieee80211_tx_result r; struct ieee80211_vif *vif = txq->vif; + WARN_ON_ONCE(softirq_count() == 0); + begin: spin_lock_bh(&fq->lock); -- 2.22.0 _______________________________________________ ath10k mailing list ath10k@lists.infradead.org http://lists.infradead.org/mailman/listinfo/ath10k
next reply other threads:[~2019-06-17 20:02 UTC|newest] Thread overview: 14+ messages / expand[flat|nested] mbox.gz Atom feed top 2019-06-17 20:01 Erik Stromdahl [this message] 2019-06-17 20:01 ` [PATCH 1/2] mac80211: add tx dequeue function for process context Erik Stromdahl 2019-06-17 20:01 ` [PATCH 2/2] ath10k: switch to ieee80211_tx_dequeue_ni Erik Stromdahl 2019-06-17 20:01 ` Erik Stromdahl 2019-10-01 11:16 ` Kalle Valo 2019-10-01 11:16 ` Kalle Valo 2019-10-01 11:48 ` Kalle Valo 2019-10-01 11:48 ` Kalle Valo 2019-10-01 17:13 ` Peter Oh 2019-10-01 17:13 ` Peter Oh 2019-10-09 19:23 ` Erik Stromdahl 2019-10-09 19:23 ` Erik Stromdahl 2019-10-09 19:45 ` Erik Stromdahl 2019-10-09 19:45 ` Erik Stromdahl
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=20190617200140.6189-1-erik.stromdahl@gmail.com \ --to=erik.stromdahl@gmail.com \ --cc=ath10k@lists.infradead.org \ --cc=davem@davemloft.net \ --cc=johannes@sipsolutions.net \ --cc=kvalo@codeaurora.org \ --cc=linux-kernel@vger.kernel.org \ --cc=linux-wireless@vger.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: linkBe sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes, see mirroring instructions on how to clone and mirror all data and code used by this external index.