2018-02-14 21:39:47

by SF Markus Elfring

Subject: [PATCH 0/2] crypto/bcm: Adjustments for do_shash()

From: Markus Elfring <[email protected]>
Date: Wed, 14 Feb 2018 22:30:07 +0100

Two update suggestions were taken into account
from static source code analysis.

Markus Elfring (2):
Delete an error message for a failed memory allocation
One function call less after error detection

drivers/crypto/bcm/util.c | 7 +++----
1 file changed, 3 insertions(+), 4 deletions(-)

--
2.16.1

2018-02-14 21:41:16

by SF Markus Elfring

Subject: [PATCH 1/2] crypto: bcm: Delete an error message for a failed memory allocation in do_shash()

From: Markus Elfring <[email protected]>
Date: Wed, 14 Feb 2018 22:05:11 +0100

Omit an extra message for a memory allocation failure in this function.

This issue was detected by using the Coccinelle software.

Signed-off-by: Markus Elfring <[email protected]>
---
drivers/crypto/bcm/util.c | 1 -
1 file changed, 1 deletion(-)

diff --git a/drivers/crypto/bcm/util.c b/drivers/crypto/bcm/util.c
index d543c010ccd9..a912c6ad3e85 100644
--- a/drivers/crypto/bcm/util.c
+++ b/drivers/crypto/bcm/util.c
@@ -279,7 +279,6 @@ int do_shash(unsigned char *name, unsigned char *result,
sdesc = kmalloc(size, GFP_KERNEL);
if (!sdesc) {
rc = -ENOMEM;
- pr_err("%s: Memory allocation failure\n", __func__);
goto do_shash_err;
}
sdesc->shash.tfm = hash;
--
2.16.1

2018-02-14 21:42:06

by SF Markus Elfring

Subject: [PATCH 2/2] crypto: bcm: One function call less in do_shash() after error detection

From: Markus Elfring <[email protected]>
Date: Wed, 14 Feb 2018 22:22:20 +0100

The kfree() function was called in one case by the do_shash() function
during error handling even if the passed variable contained a null pointer.

* Reorder two function calls at the end.

* Add a jump target.

Signed-off-by: Markus Elfring <[email protected]>
---
drivers/crypto/bcm/util.c | 6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/crypto/bcm/util.c b/drivers/crypto/bcm/util.c
index a912c6ad3e85..c141a0242223 100644
--- a/drivers/crypto/bcm/util.c
+++ b/drivers/crypto/bcm/util.c
@@ -279,7 +279,7 @@ int do_shash(unsigned char *name, unsigned char *result,
sdesc = kmalloc(size, GFP_KERNEL);
if (!sdesc) {
rc = -ENOMEM;
- goto do_shash_err;
+ goto free_shash;
}
sdesc->shash.tfm = hash;
sdesc->shash.flags = 0x0;
@@ -314,9 +314,9 @@ int do_shash(unsigned char *name, unsigned char *result,
pr_err("%s: Could not generate %s hash\n", __func__, name);

do_shash_err:
- crypto_free_shash(hash);
kfree(sdesc);
-
+free_shash:
+ crypto_free_shash(hash);
return rc;
}

--
2.16.1

2018-02-22 13:43:36

by Herbert Xu

Subject: Re: [PATCH 2/2] crypto: bcm: One function call less in do_shash() after error detection

On Wed, Feb 14, 2018 at 10:40:26PM +0100, SF Markus Elfring wrote:
> From: Markus Elfring <[email protected]>
> Date: Wed, 14 Feb 2018 22:22:20 +0100
>
> The kfree() function was called in one case by the do_shash() function
> during error handling even if the passed variable contained a null pointer.
>
> * Reorder two function calls at the end.
>
> * Add a jump target.

This patch is pointless as kfree on NULL is a no-op.

Cheers,
--
Email: Herbert Xu <[email protected]>
Home Page: http://gondor.apana.org.au/~herbert/
PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txt

2018-02-22 15:15:24

by Herbert Xu

Subject: Re: [PATCH 1/2] crypto: bcm: Delete an error message for a failed memory allocation in do_shash()

On Wed, Feb 14, 2018 at 10:39:25PM +0100, SF Markus Elfring wrote:
> From: Markus Elfring <[email protected]>
> Date: Wed, 14 Feb 2018 22:05:11 +0100
>
> Omit an extra message for a memory allocation failure in this function.
>
> This issue was detected by using the Coccinelle software.
>
> Signed-off-by: Markus Elfring <[email protected]>

Patch applied. Thanks.
--
Email: Herbert Xu <[email protected]>
Home Page: http://gondor.apana.org.au/~herbert/
PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txt

2018-02-23 08:24:18

by SF Markus Elfring

Subject: Re: [2/2] crypto: bcm: One function call less in do_shash() after error detection

> This patch is pointless as kfree on NULL is a no-op.

I prefer to avoid unnecessary function calls generally.

Regards,
Markus