Re: [LTP] [PATCH] hugeshmget02: remove the existent key before testing

[Wanlong G. <gao...@cn...>]

On 04/11/2012 06:03 PM, Caspar Zhang wrote:

> On 04/10/2012 10:54 AM, Wanlong Gao wrote:
> 
>>  testcases/kernel/mem/hugetlb/hugeshmget/hugeshmget02.c |   14 +++++++++++---
>>  1 file changed, 11 insertions(+), 3 deletions(-)
>>
>> diff --git a/testcases/kernel/mem/hugetlb/hugeshmget/hugeshmget02.c b/testcases/kernel/mem/hugetlb/hugeshmget/hugeshmget02.c
>> index 0f5f7d2..51740a7 100644
>> --- a/testcases/kernel/mem/hugetlb/hugeshmget/hugeshmget02.c
>> +++ b/testcases/kernel/mem/hugetlb/hugeshmget/hugeshmget02.c
>> @@ -60,7 +60,7 @@ int TST_TOTAL = 4;
>>  
>>  static size_t shm_size;
>>  static int shm_id_1 = -1;
>> -static int shm_id_2 = -1;
>> +static int shm_nonexistent_key = -1;
> 
> The rename looks good, if only there is no another `shm_id_2`. The name
> `shm_id_2` below caused a little confusion when reviewing the patch (I
> thought it was replaced by `shm_nonexistent_key`.. hmmm). But overall,
> the patch looks good to me.


yeah, pushed, thanks,

Wanlong Gao

> 
> Reviewed-by: Caspar Zhang <ca...@ca...>
> 
>>  static key_t shmkey2;
> [snip]
>>  	char *msg;
>> +	int shm_id_2 = -1;
>>  
>>  	msg = parse_opts(ac, av, options, &help);
>>  	if (msg != NULL)
>> @@ -103,6 +104,13 @@ int main(int ac, char **av)
>>  		Tst_count = 0;
>>  
>>  		for (i = 0; i < TST_TOTAL; i++) {
>> +			/* If this key is existent, just remove it */
>> +			if (*TC[i].skey == -1) {
>> +				shm_id_2 = shmget(*(TC[i].skey), 0, 0);
>> +				if (shm_id_2 != -1)
>> +					shmctl(shm_id_2, IPC_RMID, NULL);
>> +			}
>> +
>>  			TEST(shmget(*(TC[i].skey), TC[i].size_coe*shm_size,
>>  				    TC[i].flags));
>>  			if (TEST_RETURN != -1) {
> 
>