From: Daniel Drake <dsd@laptop.org>
To: linville@tuxdriver.com
Cc: dcbw@redhat.com
Cc: linux-wireless@vger.kernel.org
Cc: libertas-dev@lists.infradead.org
Subject: [PATCH v2] libertas: detect TX lockups and reset hardware
Date: Mon, 3 Oct 2011 11:33:02 +0100 (BST) [thread overview]
Message-ID: <20111003103302.644D79D401D@zog.reactivated.net> (raw)
Recent patches added support for resetting the SD8686 hardware when
commands time out, which seems to happen quite frequently soon after
resuming the system from a Wake-on-WLAN-triggered resume.
At http://dev.laptop.org/ticket/10969 we see the same thing happen
with transmits. In this case, the hardware will fail to respond to
a frame passed for transmission, and libertas (correctly) will block
all further commands and transmissions as the hardware can only
deal with one thing at a time. This results in a lockup while the
system waits indefinitely for the dead card to respond.
Hook up a TX lockup timer to detect this and reset the hardware.
Signed-off-by: Daniel Drake <dsd@laptop.org>
---
drivers/net/wireless/libertas/dev.h | 1 +
drivers/net/wireless/libertas/main.c | 35 ++++++++++++++++++++++++++++++++++
2 files changed, 36 insertions(+), 0 deletions(-)
v2: regenerated against wireless-next.git. Compared to the first version,
minor changes were needed to the way the timer is stopped when the interface
is brought down (these routines changed as a result of other recent work).
diff --git a/drivers/net/wireless/libertas/dev.h b/drivers/net/wireless/libertas/dev.h
index fb3e40b..f3fd447 100644
--- a/drivers/net/wireless/libertas/dev.h
+++ b/drivers/net/wireless/libertas/dev.h
@@ -158,6 +158,7 @@ struct lbs_private {
/* protected by hard_start_xmit serialization */
u8 txretrycount;
struct sk_buff *currenttxskb;
+ struct timer_list tx_lockup_timer;
/* Locks */
struct mutex lock;
diff --git a/drivers/net/wireless/libertas/main.c b/drivers/net/wireless/libertas/main.c
index d62d1fb..6a32623 100644
--- a/drivers/net/wireless/libertas/main.c
+++ b/drivers/net/wireless/libertas/main.c
@@ -188,6 +188,7 @@ int lbs_stop_iface(struct lbs_private *priv)
spin_unlock_irqrestore(&priv->driver_lock, flags);
cancel_work_sync(&priv->mcast_work);
+ del_timer_sync(&priv->tx_lockup_timer);
/* Disable command processing, and wait for all commands to complete */
lbs_deb_main("waiting for commands to complete\n");
@@ -243,6 +244,7 @@ void lbs_host_to_card_done(struct lbs_private *priv)
lbs_deb_enter(LBS_DEB_THREAD);
spin_lock_irqsave(&priv->driver_lock, flags);
+ del_timer(&priv->tx_lockup_timer);
priv->dnld_sent = DNLD_RES_RECEIVED;
@@ -585,6 +587,9 @@ static int lbs_thread(void *data)
if (ret) {
lbs_deb_tx("host_to_card failed %d\n", ret);
priv->dnld_sent = DNLD_RES_RECEIVED;
+ } else {
+ mod_timer(&priv->tx_lockup_timer,
+ jiffies + (HZ * 5));
}
priv->tx_pending_len = 0;
if (!priv->currenttxskb) {
@@ -601,6 +606,7 @@ static int lbs_thread(void *data)
}
del_timer(&priv->command_timer);
+ del_timer(&priv->tx_lockup_timer);
del_timer(&priv->auto_deepsleep_timer);
lbs_deb_leave(LBS_DEB_THREAD);
@@ -735,6 +741,32 @@ out:
}
/**
+ * lbs_tx_lockup_handler - handles the timeout of the passing of TX frames
+ * to the hardware. This is known to frequently happen with SD8686 when
+ * waking up after a Wake-on-WLAN-triggered resume.
+ *
+ * @data: &struct lbs_private pointer
+ */
+static void lbs_tx_lockup_handler(unsigned long data)
+{
+ struct lbs_private *priv = (struct lbs_private *)data;
+ unsigned long flags;
+
+ lbs_deb_enter(LBS_DEB_TX);
+ spin_lock_irqsave(&priv->driver_lock, flags);
+
+ netdev_info(priv->dev, "TX lockup detected\n");
+ if (priv->reset_card)
+ priv->reset_card(priv);
+
+ priv->dnld_sent = DNLD_RES_RECEIVED;
+ wake_up_interruptible(&priv->waitq);
+
+ spin_unlock_irqrestore(&priv->driver_lock, flags);
+ lbs_deb_leave(LBS_DEB_TX);
+}
+
+/**
* auto_deepsleep_timer_fn - put the device back to deep sleep mode when
* timer expires and no activity (command, event, data etc.) is detected.
* @data: &struct lbs_private pointer
@@ -820,6 +852,8 @@ static int lbs_init_adapter(struct lbs_private *priv)
setup_timer(&priv->command_timer, lbs_cmd_timeout_handler,
(unsigned long)priv);
+ setup_timer(&priv->tx_lockup_timer, lbs_tx_lockup_handler,
+ (unsigned long)priv);
setup_timer(&priv->auto_deepsleep_timer, auto_deepsleep_timer_fn,
(unsigned long)priv);
@@ -857,6 +891,7 @@ static void lbs_free_adapter(struct lbs_private *priv)
lbs_free_cmd_buffer(priv);
kfifo_free(&priv->event_fifo);
del_timer(&priv->command_timer);
+ del_timer(&priv->tx_lockup_timer);
del_timer(&priv->auto_deepsleep_timer);
lbs_deb_leave(LBS_DEB_MAIN);
--
1.7.6.4
reply other threads:[~2011-10-03 11:21 UTC|newest]
Thread overview: [no followups] expand[flat|nested] mbox.gz Atom feed
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=20111003103302.644D79D401D@zog.reactivated.net \
--to=dsd@laptop.org \
--cc=dcbw@redhat.com \
--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