diff mbox series

[v2,-next] KEYS: trusted: remove set but not used variable 'keyhndl'

Message ID 1539949699-74650-1-git-send-email-yuehaibing@huawei.com (mailing list archive)
State New, archived
Headers show
Series [v2,-next] KEYS: trusted: remove set but not used variable 'keyhndl' | expand

Commit Message

Yue Haibing Oct. 19, 2018, 11:48 a.m. UTC
Fixes gcc '-Wunused-but-set-variable' warning:

security/keys/trusted.c: In function 'tpm_unseal':
security/keys/trusted.c:592:11: warning:
 variable 'keyhndl' set but not used [-Wunused-but-set-variable]
  uint32_t keyhndl;

It never used since introduction in 
commit d00a1c72f7f4 ("keys: add new trusted key-type")

Signed-off-by: YueHaibing <yuehaibing@huawei.com>
---
v2: fix patch title
---
 security/keys/trusted.c | 2 --
 1 file changed, 2 deletions(-)

Comments

James Bottomley Oct. 19, 2018, 4:30 p.m. UTC | #1
On Fri, 2018-10-19 at 11:48 +0000, YueHaibing wrote:
> Fixes gcc '-Wunused-but-set-variable' warning:
> 
> security/keys/trusted.c: In function 'tpm_unseal':
> security/keys/trusted.c:592:11: warning:
>  variable 'keyhndl' set but not used [-Wunused-but-set-variable]
>   uint32_t keyhndl;
> 
> It never used since introduction in 
> commit d00a1c72f7f4 ("keys: add new trusted key-type")

This triggered an "er, shouldn't we be using the SRK parent by
default?"  search through the code, but it turns out that's all taken
care of in trusted_options_alloc() where we'll set up the SRK as parent
with well known auth and unless overridden on the command line, that's
what we use, so I'm happy this is simply left over code from a
different implementation of that use case.

James
Yue Haibing Oct. 21, 2018, 9 a.m. UTC | #2
On 2018/10/20 0:30, James Bottomley wrote:
> On Fri, 2018-10-19 at 11:48 +0000, YueHaibing wrote:
>> Fixes gcc '-Wunused-but-set-variable' warning:
>>
>> security/keys/trusted.c: In function 'tpm_unseal':
>> security/keys/trusted.c:592:11: warning:
>>  variable 'keyhndl' set but not used [-Wunused-but-set-variable]
>>   uint32_t keyhndl;
>>
>> It never used since introduction in 
>> commit d00a1c72f7f4 ("keys: add new trusted key-type")
> 
> This triggered an "er, shouldn't we be using the SRK parent by
> default?"  search through the code, but it turns out that's all taken
> care of in trusted_options_alloc() where we'll set up the SRK as parent
> with well known auth and unless overridden on the command line, that's
> what we use, so I'm happy this is simply left over code from a
> different implementation of that use case.

Thank you for comment.

> 
> James
> 
> 
> .
>
diff mbox series

Patch

diff --git a/security/keys/trusted.c b/security/keys/trusted.c
index ff67893..169df79 100644
--- a/security/keys/trusted.c
+++ b/security/keys/trusted.c
@@ -589,7 +589,6 @@  static int tpm_unseal(struct tpm_buf *tb,
 	uint32_t authhandle2 = 0;
 	unsigned char cont = 0;
 	uint32_t ordinal;
-	uint32_t keyhndl;
 	int ret;
 
 	/* sessions for unsealing key and data */
@@ -605,7 +604,6 @@  static int tpm_unseal(struct tpm_buf *tb,
 	}
 
 	ordinal = htonl(TPM_ORD_UNSEAL);
-	keyhndl = htonl(SRKHANDLE);
 	ret = tpm_get_random(NULL, nonceodd, TPM_NONCE_SIZE);
 	if (ret != TPM_NONCE_SIZE) {
 		pr_info("trusted_key: tpm_get_random failed (%d)\n", ret);