Skip to contentSkip to conversation

scsi: hpsa: Replace deprecated strncpy() with strscpy()

PATCH In review Thorsten Blum 2 participants 3 messages Feb 12, 2025 – Feb 13, 2025

Subsystems (2)
lore
Share & tools
lore
#0 lore raw

PATCH 1/1 scsi: hpsa: Replace deprecated strncpy() with strscpy()

strncpy() is deprecated for NUL-terminated destination buffers [1]. Use strscpy() instead and remove the manual NUL-termination.

Use min() to simplify the size calculation.

Compile-tested only.

---
 drivers/scsi/hpsa.c | 10 ++++------
 1 file changed, 4 insertions(+), 6 deletions(-)
drivers/scsi/hpsa.c +4 -6
index 84d8de07b7ae..9399e101f150 100644
@@ -460,9 +460,8 @@ static ssize_t host_store_hp_ssd_smart_path_status(struct device *dev,
 
 	if (!capable(CAP_SYS_ADMIN) || !capable(CAP_SYS_RAWIO))
 		return -EACCES;
-	len = count > sizeof(tmpbuf) - 1 ? sizeof(tmpbuf) - 1 : count;
-	strncpy(tmpbuf, buf, len);
-	tmpbuf[len] = '\0';
+	len = min(count, sizeof(tmpbuf) - 1);
+	strscpy(tmpbuf, buf, len);
 	if (sscanf(tmpbuf, "%d", &status) != 1)
 		return -EINVAL;
 	h = shost_to_hba(shost);
@@ -484,9 +483,8 @@ static ssize_t host_store_raid_offload_debug(struct device *dev,
 
 	if (!capable(CAP_SYS_ADMIN) || !capable(CAP_SYS_RAWIO))
 		return -EACCES;
-	len = count > sizeof(tmpbuf) - 1 ? sizeof(tmpbuf) - 1 : count;
-	strncpy(tmpbuf, buf, len);
-	tmpbuf[len] = '\0';
+	len = min(count, sizeof(tmpbuf) - 1);
+	strscpy(tmpbuf, buf, len);
 	if (sscanf(tmpbuf, "%d", &debug_level) != 1)
 		return -EINVAL;
 	if (debug_level < 0)
Signature · 1 line
2.48.1
#1 lore raw
Quoted 20 lines · Thorsten Blum

strncpy() is deprecated for NUL-terminated destination buffers [1]. Use strscpy() instead and remove the manual NUL-termination.

Use min() to simplify the size calculation.

Compile-tested only.

---
drivers/scsi/hpsa.c | 10 ++++------

1 file changed, 4 insertions(+), 6 deletions(-)
drivers/scsi/hpsa.c +0 -0
index 84d8de07b7ae..9399e101f150 100644

drivers/scsi/hpsa.c static ssize_t host_store_hp_ssd_smart_path_status(struct device *dev, · line 464

Quoted 8 lines
drivers/scsi/hpsa.c +2 -3
@@ -460,9 +460,8 @@ static ssize_t host_store_hp_ssd_smart_path_status(struct device *dev,
 
 	if (!capable(CAP_SYS_ADMIN) || !capable(CAP_SYS_RAWIO))
 		return -EACCES;
-	len = count > sizeof(tmpbuf) - 1 ? sizeof(tmpbuf) - 1 : count;
-	strncpy(tmpbuf, buf, len);
-	tmpbuf[len] = '\0';
+	len = min(count, sizeof(tmpbuf) - 1);
+	strscpy(tmpbuf, buf, len);

With strscpy() it should probably just be sizeof(tmpbuf) without -1, and then add +1 to count for the number of copied bytes to be the same as with strncpy().

Like this:

	len = min(count + 1, sizeof(tmpbuf));

This subtle difference between strncpy() and strscpy() regarding the number of bytes copied isn't really documented anywhere, is it? The documentation I came across so far seems to focus mostly on the different return values of the two functions.

drivers/scsi/hpsa.c static ssize_t host_store_raid_offload_debug(struct device *dev, · line 487

Quoted 12 lines
drivers/scsi/hpsa.c +2 -3
 	if (sscanf(tmpbuf, "%d", &status) != 1)
 		return -EINVAL;
 	h = shost_to_hba(shost);
@@ -484,9 +483,8 @@ static ssize_t host_store_raid_offload_debug(struct device *dev,
 
 	if (!capable(CAP_SYS_ADMIN) || !capable(CAP_SYS_RAWIO))
 		return -EACCES;
-	len = count > sizeof(tmpbuf) - 1 ? sizeof(tmpbuf) - 1 : count;
-	strncpy(tmpbuf, buf, len);
-	tmpbuf[len] = '\0';
+	len = min(count, sizeof(tmpbuf) - 1);
+	strscpy(tmpbuf, buf, len);

Same here.

drivers/scsi/hpsa.c static ssize_t host_store_raid_offload_debug(struct device *dev, · line 490

Quoted 3 lines
drivers/scsi/hpsa.c +0 -0
 	if (sscanf(tmpbuf, "%d", &debug_level) != 1)
 		return -EINVAL;
 	if (debug_level < 0)

Maybe someone can confirm my reasoning before I submit a v2?

Thanks,
Thorsten

#2 lore raw
Quoted 4 lines · Thorsten Blum

This subtle difference between strncpy() and strscpy() regarding the number of bytes copied isn't really documented anywhere, is it? The documentation I came across so far seems to focus mostly on the different return values of the two functions.

From the description of commit 9022ed0e7e65 ("strscpy: write destination buffer only once"):

So strscpy not only guarantees NUL-termination (unlike strncpy), it
also
doesn't do unnecessary padding at the destination. But at the same
time
also avoids byte-at-a-time reads and writes by _allowing_ some
extra NUL
writes - within the size, of course - so that the whole copy can be
done
with word operations.

It is also stable in the face of a mutable source string: it explicitly does not read the source buffer multiple times (so an implementation using "strnlen()+memcpy()" would be wrong), and does not read the
source
buffer past the size (like the mis-design that is strlcpy does).

Finally, the return value is designed to be simple and unambiguous: if the string cannot be copied fully, it returns an actual negative error, making error handling clearer and simpler (and the caller already knows the size of the buffer). Otherwise it returns the string length of the result.

More information is available in the description of commit 30c44659f4a3
("Merge branch 'strscpy' of git://git.kernel.org/pub/scm/linux/kernel/git/cmetcalf/linux-tile").

Thanks,

Bart.