Skip to contentSkip to conversation

iio: dac: ad3552r-hs: fix uninitialized data ni ad3552r_hs_write_data_source()

PATCH Applied Dan Carpenter 4 participants 5 messages May 25 – May 26

Subsystems (3)
lore
Share & tools
lore
#0 lore raw

PATCH 1/1 iio: dac: ad3552r-hs: fix uninitialized data ni ad3552r_hs_write_data_source()

If the *ppos value is non-zero then the simple_write_to_buffer() function won't initialized the start of the buf[] buffer. Non-zero values for *ppos won't work here generally, so just test for them and return -EINVAL at the start of the function.

Fixes b1c5d68ea66e ("iio: dac: ad3552r-hs: add support for internal ramp")
---
 drivers/iio/dac/ad3552r-hs.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/iio/dac/ad3552r-hs.c +1 -1
index a9578afa7015..6bc64f53bce9 100644
@@ -549,7 +549,7 @@ static ssize_t ad3552r_hs_write_data_source(struct file *f,
 
 	guard(mutex)(&st->lock);
 
-	if (count >= sizeof(buf))
+	if (*ppos != 0 || count >= sizeof(buf))
 		return -ENOSPC;
 
 	ret = simple_write_to_buffer(buf, sizeof(buf) - 1, ppos, userbuf,
Signature · 1 line
2.53.0
Reviewed-by on v1 1/1
#1 lore raw

Hi Dan,

Quoted 16 lines · Dan Carpenter

If the *ppos value is non-zero then the simple_write_to_buffer() function won't initialized the start of the buf[] buffer. Non-zero values for *ppos won't work here generally, so just test for them and return -EINVAL at the start of the function.

Fixes b1c5d68ea66e ("iio: dac: ad3552r-hs: add support for internal ramp")
---
 drivers/iio/dac/ad3552r-hs.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/iio/dac/ad3552r-hs.c +0 -0
index a9578afa7015..6bc64f53bce9 100644

drivers/iio/dac/ad3552r-hs.c static ssize_t ad3552r_hs_write_data_source(struct file *f,

Quoted 9 lines
drivers/iio/dac/ad3552r-hs.c +1 -2
@@ -549,7 +549,7 @@ static ssize_t ad3552r_hs_write_data_source(struct file *f,
 
 	guard(mutex)(&st->lock);
 
-	if (count >= sizeof(buf))
+	if (*ppos != 0 || count >= sizeof(buf))
 		return -ENOSPC;
 
 	ret = simple_write_to_buffer(buf, sizeof(buf) - 1, ppos, userbuf,
--

thanks for the fix,

it looks like this may fix an additional overflow in case ppos is near the end of buf[].
But i am actually missing the use case when ppos is non zero, can this happen from shell "echo" ? Or some fseek/dd is needed to trigger it ?

Thanks,
regards,
angelo

Quoted 2 lines

2.53.0

#2 lore raw
Quoted 31 lines · Angelo Dureghello

Hi Dan,

Quoted 16 lines · Dan Carpenter

If the *ppos value is non-zero then the simple_write_to_buffer() function won't initialized the start of the buf[] buffer. Non-zero values for *ppos won't work here generally, so just test for them and return -EINVAL at the start of the function.

Fixes b1c5d68ea66e ("iio: dac: ad3552r-hs: add support for internal ramp")
---
 drivers/iio/dac/ad3552r-hs.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/iio/dac/ad3552r-hs.c +0 -0
index a9578afa7015..6bc64f53bce9 100644

drivers/iio/dac/ad3552r-hs.c static ssize_t ad3552r_hs_write_data_source(struct file *f,

Quoted 9 lines
drivers/iio/dac/ad3552r-hs.c +1 -2
@@ -549,7 +549,7 @@ static ssize_t ad3552r_hs_write_data_source(struct file *f,
 
 	guard(mutex)(&st->lock);
 
-	if (count >= sizeof(buf))
+	if (*ppos != 0 || count >= sizeof(buf))
 		return -ENOSPC;
 
 	ret = simple_write_to_buffer(buf, sizeof(buf) - 1, ppos, userbuf,
--

thanks for the fix,

Quoted 4 lines

it looks like this may fix an additional overflow in case ppos is near the end of buf[].

simple_write_to_buffer() won't overflow. It takes ppos into consideration properly.

Quoted 2 lines

But i am actually missing the use case when ppos is non zero, can this happen from shell "echo" ? Or some fseek/dd is needed to trigger it ?

I've never actually tried to do this. I think lseek() or fseek() would work. But also the simple_write_to_buffer() function itself updates *ppos so multiple writes would do the trick as well.

regards,
dan carpenter

#3 lore raw
In reply to Dan Carpenter

Hi Dan,

Quoted 6 lines · Dan Carpenter

If the *ppos value is non-zero then the simple_write_to_buffer() function won't initialized the start of the buf[] buffer. Non-zero values for *ppos won't work here generally, so just test for them and return -EINVAL at the start of the function.

Please see my other email, you're also returning -ENOSPC instead of -EINVAL here...

Quoted 12 lines
Fixes b1c5d68ea66e ("iio: dac: ad3552r-hs: add support for internal ramp")
---
 drivers/iio/dac/ad3552r-hs.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/iio/dac/ad3552r-hs.c +0 -0
index a9578afa7015..6bc64f53bce9 100644

drivers/iio/dac/ad3552r-hs.c static ssize_t ad3552r_hs_write_data_source(struct file *f, · line 554

Quoted 7 lines
drivers/iio/dac/ad3552r-hs.c +1 -1
@@ -549,7 +549,7 @@ static ssize_t ad3552r_hs_write_data_source(struct file *f,
 
        guard(mutex)(&st->lock);
 
-       if (count >= sizeof(buf))
+       if (*ppos != 0 || count >= sizeof(buf))
                return -ENOSPC;
 

See above comment.

best regards,
max

Quoted 5 lines
drivers/iio/dac/ad3552r-hs.c +0 -1
        ret = simple_write_to_buffer(buf, sizeof(buf) - 1, ppos, userbuf,
--

2.53.0

maintainer Applied
#4 lore raw
Quoted 4 lines · On Mon, 25 May 2026 08:21:51 -0500 Maxwell Doose <[email protected]> wrote:

Hi Dan,

Quoted 1 line · Dan Carpenter
Quoted 9 lines
Quoted 5 lines

If the *ppos value is non-zero then the simple_write_to_buffer() function won't initialized the start of the buf[] buffer. Non-zero values for *ppos won't work here generally, so just test for them and return -EINVAL at the start of the function.

Please see my other email, you're also returning -ENOSPC instead of -EINVAL here...

Good spot. Tweaked whilst applying.

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

Thanks

Jonathan

Quoted 32 lines
Quoted 12 lines
Fixes b1c5d68ea66e ("iio: dac: ad3552r-hs: add support for internal ramp")
---
 drivers/iio/dac/ad3552r-hs.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/iio/dac/ad3552r-hs.c +0 -0
index a9578afa7015..6bc64f53bce9 100644

drivers/iio/dac/ad3552r-hs.c static ssize_t ad3552r_hs_write_data_source(struct file *f, · line 554

Quoted 7 lines
drivers/iio/dac/ad3552r-hs.c +1 -1
@@ -549,7 +549,7 @@ static ssize_t ad3552r_hs_write_data_source(struct file *f,
 
        guard(mutex)(&st->lock);
 
-       if (count >= sizeof(buf))
+       if (*ppos != 0 || count >= sizeof(buf))
                return -ENOSPC;
 

See above comment.

best regards,
max

Quoted 5 lines
drivers/iio/dac/ad3552r-hs.c +0 -1
        ret = simple_write_to_buffer(buf, sizeof(buf) - 1, ppos, userbuf,
--

2.53.0