[LTP] [PATCH v2] block/ltp_block_dev.c: Update tests for the newer kernel

[Petr Vorel]

Hi Xiao,

> The maximum major number is set to 511 since kernel commit 133d55c,
> so register_blkdev() with a major number grater than the maximum(511)
> will get failure instead of pass as expected.

> 1) Update tests for the newer behavior of register_blkdev().
> 2) Add more major numbers(i.e. 511 and 512) to test.

> Note:
> It is possible that specified major numbers have been used
> by system, so we skip them if register_blkdev() returns EBUSY.

> Signed-off-by: Xiao Yang <yangx.jy@cn.fujitsu.com>
> Reviewed-by: Petr Vorel <pvorel@suse.cz>

Acked-by: Petr Vorel <pvorel@suse.cz>
Thanks for your updates.

> ---
>  .../block/block_dev_kernel/ltp_block_dev.c         | 64 +++++++++++-----------
>  1 file changed, 32 insertions(+), 32 deletions(-)

> diff --git a/testcases/kernel/device-drivers/block/block_dev_kernel/ltp_block_dev.c b/testcases/kernel/device-drivers/block/block_dev_kernel/ltp_block_dev.c
...
> @@ -225,18 +221,22 @@ static int tc04(void)
>  static int tc05(void)
...

> -	major = register_blkdev(UINT_MAX, BLK_DEV_NAME);
> -	prk_debug("major = %i\n", major);
> +	for (i = 0; i < sizeof(test_major) / sizeof(unsigned int); i++) {
> +		major = register_blkdev(test_major[i], BLK_DEV_NAME);
> +		prk_debug("major = %i\n", major);

> -	if (major == 0) {
> -		unregister_blkdev(UINT_MAX, BLK_DEV_NAME);
> -	} else {
> -		prk_debug("reg blkdev with major %d failed with error %i\n",
> -			UINT_MAX, major);
> -		pass = 0;
> +		if (major == 0) {
In v1 this was major >= 0.
It looks to me as well that major == 0 is correct.

> +			unregister_blkdev(test_major[i], BLK_DEV_NAME);
> +			pass = 0;
> +		} else {
> +			prk_debug("register_blkdev() with major %u got error %i\n",
> +				  test_major[i], major);
> +		}
>  	}

>  	prk_info("Test Case Result: %s\n", result_str(pass));


Kind regards,
Petr