Skip to contentSkip to conversation

iio: proximity: as3935: fix major drivers

PATCH Applied Matt Ranostay 2 participants 8 messages May 22, 2016 IIO

lore
Share & tools
lore
#0 lore raw

cover iio: proximity: as3935: fix major drivers

This series has some major ABI changes, and bugfixes related to stack corruption.

Seems these partly are to blame from the driver's author being in one the least active thunderstorm parts of the world (Oregon has almost none).

Matt Ranostay (3):
iio: proximity: as3935: correct IIO_CHAN_INFO_RAW output
iio: proximity: as3935: remove triggered buffer processing
iio: proximity: as3935: fix buffer stack trashing

 .../ABI/testing/sysfs-bus-iio-proximity-as3935          |  2 +-
 drivers/iio/proximity/as3935.c                          | 17 ++++++++++++-----
 2 files changed, 13 insertions(+), 6 deletions(-)
Signature · 1 line
2.7.4
#1 lore raw

PATCH 1/3 iio: proximity: as3935: correct IIO_CHAN_INFO_RAW output

IIO_CHAN_INFO_RAW was returning processed data which was incorrect. This also adds the IIO_CHAN_INFO_SCALE value to convert to a processed value.

---
 Documentation/ABI/testing/sysfs-bus-iio-proximity-as3935 |  2 +-
 drivers/iio/proximity/as3935.c                           | 10 ++++++++--
 2 files changed, 9 insertions(+), 3 deletions(-)
Documentation/ABI/testing/sysfs-bus-iio-proximity-as3935 +1 -1
index 6708c5e..33e96f7 100644
@@ -1,4 +1,4 @@
-What		/sys/bus/iio/devices/iio:deviceX/in_proximity_raw
+What		/sys/bus/iio/devices/iio:deviceX/in_proximity_input
 Date:		March 2014
 KernelVersion:	3.15
 Contact:	Matt Ranostay <[email protected]>
drivers/iio/proximity/as3935.c +8 -2
index f4d29d5..f0a0def 100644
@@ -72,7 +72,8 @@ static const struct iio_chan_spec as3935_channels[] = {
 		.type           = IIO_PROXIMITY,
 		.info_mask_separate =
 			BIT(IIO_CHAN_INFO_RAW) |
-			BIT(IIO_CHAN_INFO_PROCESSED),
+			BIT(IIO_CHAN_INFO_PROCESSED) |
+			BIT(IIO_CHAN_INFO_SCALE),
 		.scan_index     = 0,
 		.scan_type = {
 			.sign           = 'u',
@@ -181,7 +182,12 @@ static int as3935_read_raw(struct iio_dev *indio_dev,
 		/* storm out of range */
 		if (*val == AS3935_DATA_MASK)
 			return -EINVAL;
-		*val *= 1000;
+
+		if (m == IIO_CHAN_INFO_PROCESSED)
+			*val *= 1000;
+		break;
+	case IIO_CHAN_INFO_SCALE:
+		*val = 1000;
 		break;
 	default:
 		return -EINVAL;
Signature · 1 line
2.7.4
#3 lore raw

PATCH 2/3 iio: proximity: as3935: remove triggered buffer processing

In reply to Matt Ranostay

Triggered buffers shouldn't return processed data, and the respective conversion was overflowing the defined .realbits for the channel.

---
 drivers/iio/proximity/as3935.c | 1 -
 1 file changed, 1 deletion(-)
drivers/iio/proximity/as3935.c +0 -1
index f0a0def..6aed024 100644
@@ -213,7 +213,6 @@ static irqreturn_t as3935_trigger_handler(int irq, void *private)
 	if (ret)
 		goto err_read;
 	val &= AS3935_DATA_MASK;
-	val *= 1000;
 
 	iio_push_to_buffers_with_timestamp(indio_dev, &val, pf->timestamp);
 err_read:
Signature · 1 line
2.7.4
#5 lore raw

PATCH 3/3 iio: proximity: as3935: fix buffer stack trashing

In reply to Matt Ranostay

Buffer wasn't of a valid size to allow the timestamp, and correct padding. This patchset also moves the buffer off the stack, and onto the heap.

---
 drivers/iio/proximity/as3935.c | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)
drivers/iio/proximity/as3935.c +4 -2
index 6aed024..8730451 100644
@@ -64,6 +64,7 @@ struct as3935_state {
 	struct delayed_work work;
 
 	u32 tune_cap;
+	u8 buffer[16]; /* 8-bit data + 54-bit padding + 64-bit timestamp */
 	u8 buf[2] ____cacheline_aligned;
 };
 
@@ -212,9 +213,10 @@ static irqreturn_t as3935_trigger_handler(int irq, void *private)
 	ret = as3935_read(st, AS3935_DATA, &val);
 	if (ret)
 		goto err_read;
-	val &= AS3935_DATA_MASK;
 
-	iio_push_to_buffers_with_timestamp(indio_dev, &val, pf->timestamp);
+	st->buffer[0] = val & AS3935_DATA_MASK;
+	iio_push_to_buffers_with_timestamp(indio_dev, &st->buffer,
+					   pf->timestamp);
 err_read:
 	iio_trigger_notify_done(indio_dev->trig);
 
Signature · 1 line
2.7.4
#6 lore raw

Re: [PATCH 3/3] iio: proximity: as3935: fix buffer stack trashing

Quoted 17 lines · Matt Ranostay

Buffer wasn't of a valid size to allow the timestamp, and correct padding. This patchset also moves the buffer off the stack, and onto the heap.

---
drivers/iio/proximity/as3935.c | 6 ++++--

1 file changed, 4 insertions(+), 2 deletions(-)
drivers/iio/proximity/as3935.c +0 -0
index 6aed024..8730451 100644

drivers/iio/proximity/as3935.c struct as3935_state { · line 67

Quoted 4 lines
drivers/iio/proximity/as3935.c +1 -0
@@ -64,6 +64,7 @@ struct as3935_state {
   struct delayed_work work;
 
   u32 tune_cap;
+    u8 buffer[16]; /* 8-bit data + 54-bit padding + 64-bit timestamp */

Just noticed that should be noted as 56-bit

Quoted 19 lines
drivers/iio/proximity/as3935.c +3 -2
   u8 buf[2] ____cacheline_aligned;
 };
 
@@ -212,9 +213,10 @@ static irqreturn_t as3935_trigger_handler(int irq, void *private)
   ret = as3935_read(st, AS3935_DATA, &val);
   if (ret)
       goto err_read;
-    val &= AS3935_DATA_MASK;
 
-    iio_push_to_buffers_with_timestamp(indio_dev, &val, pf->timestamp);
+    st->buffer[0] = val & AS3935_DATA_MASK;
+    iio_push_to_buffers_with_timestamp(indio_dev, &st->buffer,
+                       pf->timestamp);

err_read:

   iio_trigger_notify_done(indio_dev->trig);
Signature · 1 line
2.7.4
maintainer Applied
#2 lore raw

Re: [PATCH 1/3] iio: proximity: as3935: correct IIO_CHAN_INFO_RAW output

In reply to Matt Ranostay
Quoted 4 lines · Matt Ranostay

IIO_CHAN_INFO_RAW was returning processed data which was incorrect. This also adds the IIO_CHAN_INFO_SCALE value to convert to a processed value.

Applied to the fixes-togreg-post-rc1 branch of iio.git and marked for stable.

Thanks,
Jonathan

Quoted 44 lines
---
 Documentation/ABI/testing/sysfs-bus-iio-proximity-as3935 |  2 +-
 drivers/iio/proximity/as3935.c                           | 10 ++++++++--
 2 files changed, 9 insertions(+), 3 deletions(-)
Documentation/ABI/testing/sysfs-bus-iio-proximity-as3935 +1 -1
index 6708c5e..33e96f7 100644
@@ -1,4 +1,4 @@
-What		/sys/bus/iio/devices/iio:deviceX/in_proximity_raw
+What		/sys/bus/iio/devices/iio:deviceX/in_proximity_input
 Date:		March 2014
 KernelVersion:	3.15
 Contact:	Matt Ranostay <[email protected]>
drivers/iio/proximity/as3935.c +8 -2
index f4d29d5..f0a0def 100644
@@ -72,7 +72,8 @@ static const struct iio_chan_spec as3935_channels[] = {
 		.type           = IIO_PROXIMITY,
 		.info_mask_separate =
 			BIT(IIO_CHAN_INFO_RAW) |
-			BIT(IIO_CHAN_INFO_PROCESSED),
+			BIT(IIO_CHAN_INFO_PROCESSED) |
+			BIT(IIO_CHAN_INFO_SCALE),
 		.scan_index     = 0,
 		.scan_type = {
 			.sign           = 'u',
@@ -181,7 +182,12 @@ static int as3935_read_raw(struct iio_dev *indio_dev,
 		/* storm out of range */
 		if (*val == AS3935_DATA_MASK)
 			return -EINVAL;
-		*val *= 1000;
+
+		if (m == IIO_CHAN_INFO_PROCESSED)
+			*val *= 1000;
+		break;
+	case IIO_CHAN_INFO_SCALE:
+		*val = 1000;
 		break;
 	default:
 		return -EINVAL;
maintainer Applied
#4 lore raw

Re: [PATCH 2/3] iio: proximity: as3935: remove triggered buffer processing

In reply to Matt Ranostay
Quoted 5 lines · Matt Ranostay

Triggered buffers shouldn't return processed data, and the respective conversion was overflowing the defined .realbits for the channel.

Applied to the fixes-togreg-post-rc1 branch of iio.git (because it was handy) and marked for stable.

Quoted 17 lines
---
 drivers/iio/proximity/as3935.c | 1 -
 1 file changed, 1 deletion(-)
drivers/iio/proximity/as3935.c +0 -1
index f0a0def..6aed024 100644
@@ -213,7 +213,6 @@ static irqreturn_t as3935_trigger_handler(int irq, void *private)
 	if (ret)
 		goto err_read;
 	val &= AS3935_DATA_MASK;
-	val *= 1000;
 
 	iio_push_to_buffers_with_timestamp(indio_dev, &val, pf->timestamp);
 err_read:
maintainer Applied
#7 lore raw

Re: [PATCH 3/3] iio: proximity: as3935: fix buffer stack trashing

In reply to Matt Ranostay
Quoted 8 lines · Matt Ranostay
Quoted 8 lines · Matt Ranostay

Buffer wasn't of a valid size to allow the timestamp, and correct padding. This patchset also moves the buffer off the stack, and onto the heap.

Applied as per the other 2.

Quoted 11 lines
Quoted 11 lines

---
drivers/iio/proximity/as3935.c | 6 ++++--

1 file changed, 4 insertions(+), 2 deletions(-)
drivers/iio/proximity/as3935.c +0 -0
index 6aed024..8730451 100644
@@ -64,6 +64,7 @@ struct as3935_state {
   struct delayed_work work;
 
Quoted 4 lines

drivers/iio/proximity/as3935.c struct as3935_state { · line 67

Quoted 2 lines
drivers/iio/proximity/as3935.c +1 -0
   u32 tune_cap;
+    u8 buffer[16]; /* 8-bit data + 54-bit padding + 64-bit timestamp */

Just noticed that should be noted as 56-bit

Fixed up.

Quoted 25 lines
Quoted 15 lines
drivers/iio/proximity/as3935.c +3 -2
   u8 buf[2] ____cacheline_aligned;
 };
 
@@ -212,9 +213,10 @@ static irqreturn_t as3935_trigger_handler(int irq, void *private)
   ret = as3935_read(st, AS3935_DATA, &val);
   if (ret)
       goto err_read;
-    val &= AS3935_DATA_MASK;
 
-    iio_push_to_buffers_with_timestamp(indio_dev, &val, pf->timestamp);
+    st->buffer[0] = val & AS3935_DATA_MASK;
+    iio_push_to_buffers_with_timestamp(indio_dev, &st->buffer,
+                       pf->timestamp);

err_read:

Quoted 5 lines
   iio_trigger_notify_done(indio_dev->trig);
Signature · 1 line
2.7.4

--
To unsubscribe from this list: send the line "unsubscribe linux-iio" in the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html