linux-wireless.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Johannes Stezenbach <js@sig21.net>
To: Johannes Berg <johannes@sipsolutions.net>
Cc: linux-wireless@vger.kernel.org, Ivo van Doorn <IvDoorn@gmail.com>,
	Gertjan van Wingerde <gwingerde@gmail.com>,
	"John W. Linville" <linville@tuxdriver.com>,
	Helmut Schaa <helmut.schaa@googlemail.com>
Subject: [PATCH v4] mac80211/rt2x00: add ieee80211_tx_status_ni()
Date: Tue, 30 Nov 2010 16:49:23 +0100	[thread overview]
Message-ID: <20101130154923.GA4618@sig21.net> (raw)
In-Reply-To: <20101129161316.GA32760@sig21.net>

All rt2x00 drivers except rt2800pci call ieee80211_tx_status() from
a workqueue, which causes "NOHZ: local_softirq_pending 08" messages.

To fix it, add ieee80211_tx_status_ni() similar to ieee80211_rx_ni()
which can be called from process context, and call it from
rt2x00lib_txdone().  For the rt2800pci special case a driver
flag is introduced.

Signed-off-by: Johannes Stezenbach <js@sig21.net>
---
v2: add ieee80211_tx_status_ni()
v3: simplify accorindg to review
v4: add rt2x00 as first user of ieee80211_tx_status_ni()

The DRIVER_REQUIRE_TASKLET_CONTEXT flag may seem ugly, but there is
a pending rewrite of the rt2800pci irq handling code by
Helmut, so we decided to use this easy solution.

Note: Only tested with rt2800usb.

 drivers/net/wireless/rt2x00/rt2800pci.c |    1 +
 drivers/net/wireless/rt2x00/rt2x00.h    |    1 +
 drivers/net/wireless/rt2x00/rt2x00dev.c |    9 ++++++---
 include/net/mac80211.h                  |   28 ++++++++++++++++++++++++----
 4 files changed, 32 insertions(+), 7 deletions(-)

diff --git a/drivers/net/wireless/rt2x00/rt2800pci.c b/drivers/net/wireless/rt2x00/rt2800pci.c
index b48d2d3..3ef0228 100644
--- a/drivers/net/wireless/rt2x00/rt2800pci.c
+++ b/drivers/net/wireless/rt2x00/rt2800pci.c
@@ -932,6 +932,7 @@ static int rt2800pci_probe_hw(struct rt2x00_dev *rt2x00dev)
 	__set_bit(DRIVER_REQUIRE_DMA, &rt2x00dev->flags);
 	__set_bit(DRIVER_REQUIRE_L2PAD, &rt2x00dev->flags);
 	__set_bit(DRIVER_REQUIRE_TXSTATUS_FIFO, &rt2x00dev->flags);
+	__set_bit(DRIVER_REQUIRE_TASKLET_CONTEXT, &rt2x00dev->flags);
 	if (!modparam_nohwcrypt)
 		__set_bit(CONFIG_SUPPORT_HW_CRYPTO, &rt2x00dev->flags);
 	__set_bit(DRIVER_SUPPORT_LINK_TUNING, &rt2x00dev->flags);
diff --git a/drivers/net/wireless/rt2x00/rt2x00.h b/drivers/net/wireless/rt2x00/rt2x00.h
index 556b744..28ea59a 100644
--- a/drivers/net/wireless/rt2x00/rt2x00.h
+++ b/drivers/net/wireless/rt2x00/rt2x00.h
@@ -670,6 +670,7 @@ enum rt2x00_flags {
 	DRIVER_REQUIRE_COPY_IV,
 	DRIVER_REQUIRE_L2PAD,
 	DRIVER_REQUIRE_TXSTATUS_FIFO,
+	DRIVER_REQUIRE_TASKLET_CONTEXT,
 
 	/*
 	 * Driver features
diff --git a/drivers/net/wireless/rt2x00/rt2x00dev.c b/drivers/net/wireless/rt2x00/rt2x00dev.c
index 87a421b..fa74acd 100644
--- a/drivers/net/wireless/rt2x00/rt2x00dev.c
+++ b/drivers/net/wireless/rt2x00/rt2x00dev.c
@@ -376,9 +376,12 @@ void rt2x00lib_txdone(struct queue_entry *entry,
 	 * through a mac80211 library call (RTS/CTS) then we should not
 	 * send the status report back.
 	 */
-	if (!(skbdesc_flags & SKBDESC_NOT_MAC80211))
-		ieee80211_tx_status(rt2x00dev->hw, entry->skb);
-	else
+	if (!(skbdesc_flags & SKBDESC_NOT_MAC80211)) {
+		if (test_bit(DRIVER_REQUIRE_TASKLET_CONTEXT, &rt2x00dev->flags))
+			ieee80211_tx_status(rt2x00dev->hw, entry->skb);
+		else
+			ieee80211_tx_status_ni(rt2x00dev->hw, entry->skb);
+	} else
 		dev_kfree_skb_any(entry->skb);
 
 	/*
diff --git a/include/net/mac80211.h b/include/net/mac80211.h
index eaa4aff..e411cf8 100644
--- a/include/net/mac80211.h
+++ b/include/net/mac80211.h
@@ -2055,8 +2055,8 @@ static inline void ieee80211_rx_ni(struct ieee80211_hw *hw,
  *
  * This function may not be called in IRQ context. Calls to this function
  * for a single hardware must be synchronized against each other. Calls
- * to this function and ieee80211_tx_status_irqsafe() may not be mixed
- * for a single hardware.
+ * to this function, ieee80211_tx_status_ni() and ieee80211_tx_status_irqsafe()
+ * may not be mixed for a single hardware.
  *
  * @hw: the hardware the frame was transmitted by
  * @skb: the frame that was transmitted, owned by mac80211 after this call
@@ -2065,13 +2065,33 @@ void ieee80211_tx_status(struct ieee80211_hw *hw,
 			 struct sk_buff *skb);
 
 /**
+ * ieee80211_tx_status_ni - transmit status callback (in process context)
+ *
+ * Like ieee80211_tx_status() but can be called in process context.
+ *
+ * Calls to this function, ieee80211_tx_status() and
+ * ieee80211_tx_status_irqsafe() may not be mixed
+ * for a single hardware.
+ *
+ * @hw: the hardware the frame was transmitted by
+ * @skb: the frame that was transmitted, owned by mac80211 after this call
+ */
+static inline void ieee80211_tx_status_ni(struct ieee80211_hw *hw,
+					  struct sk_buff *skb)
+{
+	local_bh_disable();
+	ieee80211_tx_status(hw, skb);
+	local_bh_enable();
+}
+
+/**
  * ieee80211_tx_status_irqsafe - IRQ-safe transmit status callback
  *
  * Like ieee80211_tx_status() but can be called in IRQ context
  * (internally defers to a tasklet.)
  *
- * Calls to this function and ieee80211_tx_status() may not be mixed for a
- * single hardware.
+ * Calls to this function, ieee80211_tx_status() and
+ * ieee80211_tx_status_ni() may not be mixed for a single hardware.
  *
  * @hw: the hardware the frame was transmitted by
  * @skb: the frame that was transmitted, owned by mac80211 after this call
-- 
1.7.2.3


  reply	other threads:[~2010-11-30 15:49 UTC|newest]

Thread overview: 12+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2010-11-29 14:54 [PATCH RFC] mac80211: fix "NOHZ: local_softirq_pending 08" Johannes Stezenbach
2010-11-29 15:09 ` Johannes Berg
2010-11-29 15:27   ` Johannes Stezenbach
2010-11-29 15:37     ` Johannes Berg
2010-11-29 15:51       ` [PATCH v2 " Johannes Stezenbach
2010-11-29 15:58         ` Johannes Berg
2010-11-29 16:13           ` [PATCH v3 RFC] mac80211: add ieee80211_tx_status_ni() Johannes Stezenbach
2010-11-30 15:49             ` Johannes Stezenbach [this message]
2010-11-30 15:53               ` [PATCH v4] mac80211/rt2x00: " Johannes Berg
2010-11-30 16:34                 ` Johannes Stezenbach
2010-11-30 16:35                   ` Johannes Berg
2010-11-30 18:38                     ` John W. Linville

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=20101130154923.GA4618@sig21.net \
    --to=js@sig21.net \
    --cc=IvDoorn@gmail.com \
    --cc=gwingerde@gmail.com \
    --cc=helmut.schaa@googlemail.com \
    --cc=johannes@sipsolutions.net \
    --cc=linux-wireless@vger.kernel.org \
    --cc=linville@tuxdriver.com \
    /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).