crypto: qat - use strscpy_pad to simplify buffer initialization

PATCH In review Thorsten Blum 3 participants 10 messages Oct 22, 2025 – Oct 27, 2025

Subsystems (2)

Crypto API QAT driver

lore

Share & tools

lore

Thorsten Blum

author

2025-10-22 12:36 UTC #0 lore raw

PATCH 1/1 crypto: qat - use strscpy_pad to simplify buffer initialization

Use strscpy_pad() to copy the string and zero-pad the destination buffer in a single step instead of zero-initializing the buffer first and then immediately overwriting it using strscpy().

Replace the magic number 16 with sizeof(buf) and remove the redundant parentheses around kstrtoul() while we're at it.

Signed-off-by Thorsten Blum <[email protected]>

---
 drivers/crypto/intel/qat/qat_common/qat_uclo.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

drivers/crypto/intel/qat/qat_common/qat_uclo.c +4 -4

index 18c3e4416dc5..41a7bd434e97 100644
@@ -200,18 +200,18 @@ qat_uclo_cleanup_batch_init_list(struct icp_qat_fw_loader_handle *handle,
 
 static int qat_uclo_parse_num(char *str, unsigned int *num)
 {
-	char buf[16] = {0};
+	char buf[16] = {};
 	unsigned long ae = 0;
 	int i;
 
-	strscpy(buf, str, sizeof(buf));
-	for (i = 0; i < 16; i++) {
+	strscpy_pad(buf, str);
+	for (i = 0; i < sizeof(buf); i++) {
 		if (!isdigit(buf[i])) {
 			buf[i] = '\0';
 			break;
 		}
 	}
-	if ((kstrtoul(buf, 10, &ae)))
+	if (kstrtoul(buf, 10, &ae))
 		return -EFAULT;
 
 	*num = (unsigned int)ae;

Signature · 1 line

2.51.0

Andy Shevchenko

2025-10-22 18:17 UTC #1 lore raw

Quoted 6 lines · Thorsten Blum

Use strscpy_pad() to copy the string and zero-pad the destination buffer in a single step instead of zero-initializing the buffer first and then immediately overwriting it using strscpy().
Replace the magic number 16 with sizeof(buf) and remove the redundant parentheses around kstrtoul() while we're at it.

I understand that you focused on strscpy*() conversions, but the below I think needs a bigger refactoring, see my remarks.

...

drivers/crypto/intel/qat/qat_common/qat_uclo.c qat_uclo_cleanup_batch_init_list(struct icp_qat_fw_loader_handle *handle, · line 207

Quoted 8 lines

drivers/crypto/intel/qat/qat_common/qat_uclo.c +2 -3

-	char buf[16] = {0};
+	char buf[16] = {};
 	unsigned long ae = 0;
 	int i;
 
-	strscpy(buf, str, sizeof(buf));
-	for (i = 0; i < 16; i++) {
+	strscpy_pad(buf, str);

First of all, why do we need a _pad() version here? Is the data somehow being used as a whole?

drivers/crypto/intel/qat/qat_common/qat_uclo.c qat_uclo_cleanup_batch_init_list(struct icp_qat_fw_loader_handle *handle, · line 215

Quoted 9 lines

drivers/crypto/intel/qat/qat_common/qat_uclo.c +2 -1

+	for (i = 0; i < sizeof(buf); i++) {
 		if (!isdigit(buf[i])) {
 			buf[i] = '\0';
 			break;
 		}
 	}
-	if ((kstrtoul(buf, 10, &ae)))
+	if (kstrtoul(buf, 10, &ae))
 		return -EFAULT;

Looking at this, it tries to work around the kstrtoul() inability to perform partial parses. Instead, this should do something like

	unsigned long long x;
	const char *end;

	simple_strtoull(...);
	if (x > UINT_MAX || end == buf)
		return $ERR; // wrong input / overflow

Signature · 2 lines

With Best Regards,
Andy Shevchenko

Andy Shevchenko

2025-10-22 18:29 UTC #2 lore raw

Quoted 10 lines · Andy Shevchenko

Quoted 6 lines · Thorsten Blum
Use strscpy_pad() to copy the string and zero-pad the destination buffer in a single step instead of zero-initializing the buffer first and then immediately overwriting it using strscpy().
Replace the magic number 16 with sizeof(buf) and remove the redundant parentheses around kstrtoul() while we're at it.
I understand that you focused on strscpy*() conversions, but the below I think needs a bigger refactoring, see my remarks.

...

Quoted 12 lines

drivers/crypto/intel/qat/qat_common/qat_uclo.c qat_uclo_cleanup_batch_init_list(struct icp_qat_fw_loader_handle *handle, · line 207
Quoted 8 lines
drivers/crypto/intel/qat/qat_common/qat_uclo.c +2 -3
-	char buf[16] = {0};
+	char buf[16] = {};
 	unsigned long ae = 0;
 	int i;
 
-	strscpy(buf, str, sizeof(buf));
-	for (i = 0; i < 16; i++) {
+	strscpy_pad(buf, str);
First of all, why do we need a _pad() version here? Is the data somehow being used as a whole?

Quoted 9 lines

drivers/crypto/intel/qat/qat_common/qat_uclo.c qat_uclo_cleanup_batch_init_list(struct icp_qat_fw_loader_handle *handle, · line 215
Quoted 9 lines
drivers/crypto/intel/qat/qat_common/qat_uclo.c +2 -1
+	for (i = 0; i < sizeof(buf); i++) {
 		if (!isdigit(buf[i])) {
 			buf[i] = '\0';
 			break;
 		}
 	}
-	if ((kstrtoul(buf, 10, &ae)))
+	if (kstrtoul(buf, 10, &ae))
 		return -EFAULT;

On top of that the function is called only from one place and returns different error code, instead it would have returned what kstrtoul() gives...

Quoted 9 lines

Looking at this, it tries to work around the kstrtoul() inability to perform partial parses. Instead, this should do something like
	unsigned long long x;
	const char *end;

	simple_strtoull(...);
	if (x > UINT_MAX || end == buf)
		return $ERR; // wrong input / overflow

Yeah, the overflow check here is not comprehensive, it won't catch the overflow (wrap around) of 64-bit value. But we can add a check for the end not to be farther than ~19 characters from the start, which would correspond the initial copy of 16 characters.

Signature · 2 lines

With Best Regards,
Andy Shevchenko

Thorsten Blum

author

2025-10-23 15:35 UTC #3 lore raw

In reply to Andy Shevchenko

Quoted 8 lines · Andy Shevchenko

Quoted 6 lines · Thorsten Blum
Use strscpy_pad() to copy the string and zero-pad the destination buffer in a single step instead of zero-initializing the buffer first and then immediately overwriting it using strscpy().
Replace the magic number 16 with sizeof(buf) and remove the redundant parentheses around kstrtoul() while we're at it.

Quoted 7 lines

I understand that you focused on strscpy*() conversions, but the below I think needs a bigger refactoring, see my remarks.
...
Review
Quoted 2 lines
(quoted diff) +1 -1
-	char buf[16] = {0};
+	char buf[16] = {};

Sorry, this should have been just 'char buf[16];' since {} and {0} are equivalent and both zero-initialize the array.

Quoted 9 lines

Review
Quoted 6 lines
	unsigned long ae = 0;
	int i;
(quoted diff) +1 -2
-	strscpy(buf, str, sizeof(buf));
-	for (i = 0; i < 16; i++) {
+	strscpy_pad(buf, str);
First of all, why do we need a _pad() version here? Is the data somehow being used as a whole?

I honestly didn't question this, but it looks like strscpy() would be sufficient (with this approach at least).

Quoted 13 lines

Quoted 9 lines
(quoted diff) +1 -0
+	for (i = 0; i < sizeof(buf); i++) {
		if (!isdigit(buf[i])) {
			buf[i] = '\0';
			break;
		}
	}
(quoted diff) +1 -1
-	if ((kstrtoul(buf, 10, &ae)))
+	if (kstrtoul(buf, 10, &ae))
		return -EFAULT;
Looking at this, it tries to work around the kstrtoul() inability to perform partial parses. Instead, this should do something like

Quoted 6 lines

	unsigned long long x;
	const char *end;

	simple_strtoull(...);
	if (x > UINT_MAX || end == buf)
		return $ERR; // wrong input / overflow

How about this?

drivers/crypto/intel/qat/qat_common/qat_uclo.c +5 ~~-13~~

index 18c3e4416dc5..04628dc01456 100644
@@ -200,20 +200,12 @@ qat_uclo_cleanup_batch_init_list(struct icp_qat_fw_loader_handle *handle,
 
 static int qat_uclo_parse_num(char *str, unsigned int *num)
 {
-	char buf[16] = {0};
-	unsigned long ae = 0;
-	int i;
-
-	strscpy(buf, str, sizeof(buf));
-	for (i = 0; i < 16; i++) {
-		if (!isdigit(buf[i])) {
-			buf[i] = '\0';
-			break;
-		}
-	}
-	if ((kstrtoul(buf, 10, &ae)))
-		return -EFAULT;
+	unsigned long long ae;
+	char *end;
 
+	ae = simple_strtoull(str, &end, 10);
+	if (ae > UINT_MAX || str == end || (end - str) > 20)
+		return -EINVAL;
 	*num = (unsigned int)ae;
 	return 0;
 }

Andy Shevchenko

2025-10-23 18:44 UTC #4 lore raw

Quoted 2 lines · Thorsten Blum

Quoted 1 line · Andy Shevchenko
On Wed, Oct 22, 2025 at 02:36:19PM +0200, Thorsten Blum wrote:

...

Quoted 1 line

How about this?

LGTM, and that's what I had in mind, but please double check the behaviour of kstrtox() on an empty strings.

Quoted 16 lines

drivers/crypto/intel/qat/qat_common/qat_uclo.c +0 -8

index 18c3e4416dc5..04628dc01456 100644
@@ -200,20 +200,12 @@ qat_uclo_cleanup_batch_init_list(struct icp_qat_fw_loader_handle *handle,
 
 static int qat_uclo_parse_num(char *str, unsigned int *num)
 {
-	char buf[16] = {0};
-	unsigned long ae = 0;
-	int i;
-
-	strscpy(buf, str, sizeof(buf));
-	for (i = 0; i < 16; i++) {
-		if (!isdigit(buf[i])) {
-			buf[i] = '\0';

drivers/crypto/intel/qat/qat_common/qat_uclo.c qat_uclo_cleanup_batch_init_list(struct icp_qat_fw_loader_handle *handle, · line 207

Quoted 10 lines

drivers/crypto/intel/qat/qat_common/qat_uclo.c +4 -5

-			break;
-		}
-	}
-	if ((kstrtoul(buf, 10, &ae)))
-		return -EFAULT;
+	unsigned long long ae;
+	char *end;
 
+	ae = simple_strtoull(str, &end, 10);
+	if (ae > UINT_MAX || str == end || (end - str) > 20)

I would go with >= 20, the 64-bit value is approx. 1 * 10^19.

Quoted 4 lines

drivers/crypto/intel/qat/qat_common/qat_uclo.c +1 -0
+		return -EINVAL;
 	*num = (unsigned int)ae;
 	return 0;
 }

Signature · 2 lines

With Best Regards,
Andy Shevchenko

Giovanni Cabiddu

maintainer

2025-10-24 08:50 UTC #5 lore raw

Quoted 10 lines · Andy Shevchenko

Quoted 2 lines · Thorsten Blum
Quoted 1 line · Andy Shevchenko
On Wed, Oct 22, 2025 at 02:36:19PM +0200, Thorsten Blum wrote:
...
Quoted 1 line
How about this?
LGTM, and that's what I had in mind, but please double check the behaviour of kstrtox() on an empty strings.

LGTM as well.

I checked the behaviour of kstrtoul() when given an empty string. It returns -EINVAL (-22). The result variable (the third parameter) is only modified if the conversion is succesful.

Anyway, the caller will not provide a NULL string to this function [1].

Quoted 24 lines

drivers/crypto/intel/qat/qat_common/qat_uclo.c +2 ~~-13~~

index 18c3e4416dc5..04628dc01456 100644
@@ -200,20 +200,12 @@ qat_uclo_cleanup_batch_init_list(struct icp_qat_fw_loader_handle *handle,
 
 static int qat_uclo_parse_num(char *str, unsigned int *num)
 {
-	char buf[16] = {0};
-	unsigned long ae = 0;
-	int i;
-
-	strscpy(buf, str, sizeof(buf));
-	for (i = 0; i < 16; i++) {
-		if (!isdigit(buf[i])) {
-			buf[i] = '\0';
-			break;
-		}
-	}
-	if ((kstrtoul(buf, 10, &ae)))
-		return -EFAULT;
+	unsigned long long ae;
+	char *end;

Quoted 4 lines

drivers/crypto/intel/qat/qat_common/qat_uclo.c qat_uclo_cleanup_batch_init_list(struct icp_qat_fw_loader_handle *handle, · line 207
Quoted 2 lines
drivers/crypto/intel/qat/qat_common/qat_uclo.c +2 -0
+	ae = simple_strtoull(str, &end, 10);
+	if (ae > UINT_MAX || str == end || (end - str) > 20)
I would go with >= 20, the 64-bit value is approx. 1 * 10^19.

Just an insight into the type of strings being parsed here. If they are well-formed, the format looks like:

    <AE_NUMBER>!<...>

Example:

    11!l0000!lm_thread_ctrl_struct_base

This function just extract the first number. Currently, that is 2 digits [2], and I believe it is unlikely to exceed 3 in future gens.

[1] https://elixir.bootlin.com/linux/v6.17.4/source/drivers/crypto/intel/qat/qat_common/qat_uclo.c#L237
[2] https://elixir.bootlin.com/linux/v6.17.4/source/drivers/crypto/intel/qat/qat_common/icp_qat_uclo.h#L11

Regards,

Signature · 1 line

Giovanni

Andy Shevchenko

2025-10-24 09:49 UTC #6 lore raw

Quoted 4 lines · Giovanni Cabiddu

Quoted 3 lines · Andy Shevchenko
Quoted 2 lines · Thorsten Blum
Quoted 1 line · Andy Shevchenko
On Wed, Oct 22, 2025 at 02:36:19PM +0200, Thorsten Blum wrote:

...

Quoted 10 lines

Quoted 4 lines
Quoted 1 line
How about this?
LGTM, and that's what I had in mind, but please double check the behaviour of kstrtox() on an empty strings.
LGTM as well.
I checked the behaviour of kstrtoul() when given an empty string. It returns -EINVAL (-22). The result variable (the third parameter) is only modified if the conversion is succesful.

Thanks for confirming!

Quoted 13 lines

Anyway, the caller will not provide a NULL string to this function [1].
Quoted 7 lines
drivers/crypto/intel/qat/qat_common/qat_uclo.c qat_uclo_cleanup_batch_init_list(struct icp_qat_fw_loader_handle *handle, · line 207
Quoted 5 lines
drivers/crypto/intel/qat/qat_common/qat_uclo.c +4 -0
+	unsigned long long ae;
+	char *end;
 
+	ae = simple_strtoull(str, &end, 10);
+	if (ae > UINT_MAX || str == end || (end - str) > 20)
I would go with >= 20, the 64-bit value is approx. 1 * 10^19.
Just an insight into the type of strings being parsed here. If they are well-formed, the format looks like:

Quoted 8 lines

    <AE_NUMBER>!<...>
Example:
    11!l0000!lm_thread_ctrl_struct_base
This function just extract the first number. Currently, that is 2 digits [2], and I believe it is unlikely to exceed 3 in future gens.

Yep, but it is better to make code more robust (and esp. in case somebody copies'n'pastes to somewhere else).

Quoted 2 lines

[1] https://elixir.bootlin.com/linux/v6.17.4/source/drivers/crypto/intel/qat/qat_common/qat_uclo.c#L237
[2] https://elixir.bootlin.com/linux/v6.17.4/source/drivers/crypto/intel/qat/qat_common/icp_qat_uclo.h#L11

Signature · 2 lines

With Best Regards,
Andy Shevchenko

Thorsten Blum

author

2025-10-24 18:47 UTC #7 lore raw

In reply to Andy Shevchenko

Hi Andy,

Quoted 9 lines · Andy Shevchenko

Quoted 2 lines · Thorsten Blum
Quoted 1 line · Andy Shevchenko
On Wed, Oct 22, 2025 at 02:36:19PM +0200, Thorsten Blum wrote:
...
Quoted 1 line
How about this?
LGTM, and that's what I had in mind, ...

I was about to submit v2, but checkpatch warns me about simple_strtoull:

WARNING: simple_strtoull is obsolete, use kstrtoull instead

Any recommendations?

Thanks,
Thorsten

Andy Shevchenko

2025-10-27 08:19 UTC #8 lore raw

Quoted 4 lines · Thorsten Blum

Quoted 3 lines · Andy Shevchenko
Quoted 2 lines · Thorsten Blum
Quoted 1 line · Andy Shevchenko
On Wed, Oct 22, 2025 at 02:36:19PM +0200, Thorsten Blum wrote:

...

Quoted 9 lines

Quoted 3 lines
Quoted 1 line
How about this?
LGTM, and that's what I had in mind, ...
I was about to submit v2, but checkpatch warns me about simple_strtoull:
WARNING: simple_strtoull is obsolete, use kstrtoull instead
Any recommendations?

Yes, fix checkpatch as per commit 885e68e8b7b1 ("kernel.h: update comment about simple_strto<foo>() functions") or ignore this false positive.

Signature · 2 lines

With Best Regards,
Andy Shevchenko

Thorsten Blum

author

2025-10-27 08:43 UTC #9 lore raw

Quoted 6 lines · Andy Shevchenko

Quoted 2 lines · Thorsten Blum
[...]
Any recommendations?
Yes, fix checkpatch as per commit 885e68e8b7b1 ("kernel.h: update comment about simple_strto<foo>() functions") or ignore this false positive.

Thanks for the commit hash. I ignored it and submitted v2:

https://lore.kernel.org/lkml/[email protected]/