lists.openwall.net   lists  /  announce  owl-users  owl-dev  john-users  john-dev  passwdqc-users  yescrypt  popa3d-users  /  oss-security  kernel-hardening  musl  sabotage  tlsify  passwords  /  crypt-dev  xvendor  /  Bugtraq  Full-Disclosure  linux-kernel  linux-netdev  linux-ext4  linux-hardening  linux-cve-announce  PHC 
Open Source and information security mailing list archives
 
Hash Suite for Android: free password hash cracker in your pocket
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Date:	Sun, 29 Sep 2013 12:28:22 -0700
From:	Greg Kroah-Hartman <gregkh@...uxfoundation.org>
To:	linux-kernel@...r.kernel.org
Cc:	Greg Kroah-Hartman <gregkh@...uxfoundation.org>,
	stable@...r.kernel.org, Solomon Peachy <pizza@...ftnet.org>,
	"John W. Linville" <linville@...driver.com>
Subject: [ 70/71] cw1200: Dont perform SPI transfers in interrupt context

3.11-stable review patch.  If anyone has any objections, please let me know.

------------------

From: Solomon Peachy <pizza@...ftnet.org>

commit aec8e88c947b7017e2b4bbcb68a4bfc4a1f8ad35 upstream.

When we get an interrupt from the hardware, the first thing the driver does
is tell the device to mask off the interrupt line.  Unfortunately this
involves a SPI transaction in interrupt context.  Some (most?) SPI
controllers perform the transfer asynchronously and try to sleep.
This is bad, and triggers a BUG().

So, work around this by using adding a hwbus hook for the cw1200 driver
core to call.  The cw1200_spi driver translates this into
irq_disable()/irq_enable() calls instead, which can safely be called in
interrupt context.

Apparently the platforms I used to develop the cw1200_spi driver used
synchronous spi_sync() implementations, which is why this didn't surface
until now.

Many thanks to Dave Sizeburns for the inital bug report and his services
as a tester.

Signed-off-by: Solomon Peachy <pizza@...ftnet.org>
Signed-off-by: John W. Linville <linville@...driver.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@...uxfoundation.org>

---
 drivers/net/wireless/cw1200/cw1200_spi.c |   19 ++++++++++++++++---
 drivers/net/wireless/cw1200/fwio.c       |    2 +-
 drivers/net/wireless/cw1200/hwbus.h      |    1 +
 drivers/net/wireless/cw1200/hwio.c       |   15 +++++++++++++++
 4 files changed, 33 insertions(+), 4 deletions(-)

--- a/drivers/net/wireless/cw1200/cw1200_spi.c
+++ b/drivers/net/wireless/cw1200/cw1200_spi.c
@@ -42,6 +42,7 @@ struct hwbus_priv {
 	spinlock_t		lock; /* Serialize all bus operations */
 	wait_queue_head_t       wq;
 	int claimed;
+	int irq_disabled;
 };
 
 #define SDIO_TO_SPI_ADDR(addr) ((addr & 0x1f)>>2)
@@ -237,6 +238,8 @@ static irqreturn_t cw1200_spi_irq_handle
 	struct hwbus_priv *self = dev_id;
 
 	if (self->core) {
+		disable_irq_nosync(self->func->irq);
+		self->irq_disabled = 1;
 		cw1200_irq_handler(self->core);
 		return IRQ_HANDLED;
 	} else {
@@ -270,13 +273,22 @@ exit:
 
 static int cw1200_spi_irq_unsubscribe(struct hwbus_priv *self)
 {
-	int ret = 0;
-
 	pr_debug("SW IRQ unsubscribe\n");
 	disable_irq_wake(self->func->irq);
 	free_irq(self->func->irq, self);
 
-	return ret;
+	return 0;
+}
+
+static int cw1200_spi_irq_enable(struct hwbus_priv *self, int enable)
+{
+	/* Disables are handled by the interrupt handler */
+	if (enable && self->irq_disabled) {
+		enable_irq(self->func->irq);
+		self->irq_disabled = 0;
+	}
+
+	return 0;
 }
 
 static int cw1200_spi_off(const struct cw1200_platform_data_spi *pdata)
@@ -356,6 +368,7 @@ static struct hwbus_ops cw1200_spi_hwbus
 	.unlock			= cw1200_spi_unlock,
 	.align_size		= cw1200_spi_align_size,
 	.power_mgmt		= cw1200_spi_pm,
+	.irq_enable             = cw1200_spi_irq_enable,
 };
 
 /* Probe Function to be called by SPI stack when device is discovered */
--- a/drivers/net/wireless/cw1200/fwio.c
+++ b/drivers/net/wireless/cw1200/fwio.c
@@ -485,7 +485,7 @@ int cw1200_load_firmware(struct cw1200_c
 
 	/* Enable interrupt signalling */
 	priv->hwbus_ops->lock(priv->hwbus_priv);
-	ret = __cw1200_irq_enable(priv, 1);
+	ret = __cw1200_irq_enable(priv, 2);
 	priv->hwbus_ops->unlock(priv->hwbus_priv);
 	if (ret < 0)
 		goto unsubscribe;
--- a/drivers/net/wireless/cw1200/hwbus.h
+++ b/drivers/net/wireless/cw1200/hwbus.h
@@ -28,6 +28,7 @@ struct hwbus_ops {
 	void (*unlock)(struct hwbus_priv *self);
 	size_t (*align_size)(struct hwbus_priv *self, size_t size);
 	int (*power_mgmt)(struct hwbus_priv *self, bool suspend);
+	int (*irq_enable)(struct hwbus_priv *self, int enable);
 };
 
 #endif /* CW1200_HWBUS_H */
--- a/drivers/net/wireless/cw1200/hwio.c
+++ b/drivers/net/wireless/cw1200/hwio.c
@@ -273,6 +273,21 @@ int __cw1200_irq_enable(struct cw1200_co
 	u16 val16;
 	int ret;
 
+	/* We need to do this hack because the SPI layer can sleep on I/O
+	   and the general path involves I/O to the device in interrupt
+	   context.
+
+	   However, the initial enable call needs to go to the hardware.
+
+	   We don't worry about shutdown because we do a full reset which
+	   clears the interrupt enabled bits.
+	*/
+	if (priv->hwbus_ops->irq_enable) {
+		ret = priv->hwbus_ops->irq_enable(priv->hwbus_priv, enable);
+		if (ret || enable < 2)
+			return ret;
+	}
+
 	if (HIF_8601_SILICON == priv->hw_type) {
 		ret = __cw1200_reg_read_32(priv, ST90TDS_CONFIG_REG_ID, &val32);
 		if (ret < 0) {


--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ