[PATCH] uefirtvariable: fix the unchecked return value

[IvanHu]

On 05/08/2013 03:07 PM, Colin Ian King wrote:
> On 08/05/13 04:06, Ivan Hu wrote:
>> Coverity CID #997310: Unchecked return value
>>
>> Need to check the return value for calling function "ioctl".
>>
>> Signed-off-by: Ivan Hu <ivan.hu at canonical.com>
>> ---
>>   src/uefi/uefirtvariable/uefirtvariable.c |    5 +++--
>>   1 file changed, 3 insertions(+), 2 deletions(-)
>>
>> diff --git a/src/uefi/uefirtvariable/uefirtvariable.c b/src/uefi/uefirtvariable/uefirtvariable.c
>> index 16cd240..7c2fc1f 100644
>> --- a/src/uefi/uefirtvariable/uefirtvariable.c
>> +++ b/src/uefi/uefirtvariable/uefirtvariable.c
>> @@ -858,6 +858,7 @@ static int setvariable_invalidattr(
>>   	EFI_GUID *gtestguid,
>>   	const uint8_t datadiff)
>>   {
>> +	long ioret;
>>   	struct efi_setvariable setvariable;
>>   	uint64_t status;
>>   	uint64_t dataindex;
>> @@ -873,9 +874,9 @@ static int setvariable_invalidattr(
>>   	setvariable.Data = data;
>>   	setvariable.status = &status;
>>
>> -	ioctl(fd, EFI_RUNTIME_SET_VARIABLE, &setvariable);
>> +	ioret = ioctl(fd, EFI_RUNTIME_SET_VARIABLE, &setvariable);
>>
>> -	if (status == EFI_SUCCESS) {
>> +	if ((status == EFI_SUCCESS) && (ioret != -1)) {
>>   		fwts_warning(fw,
>>   			"After ExitBootServices() is performed, the "
>>   			"attributes %" PRIu32 ", "
>>
> No entirely sure about this.  What about the case when ioret == -1, is
> this an error/failure?  In other tests this is reported as a failure,
> where as in this case it seems to be silently ignored.
>
> Colin
>
>
>

In this test case, we do the setvariable with the invalid attributes.
We expects the firmware return a fail for the setvariable, ioret == -1. 
If not, will give an warnings.

Ivan