Re: [PATCH bpf-next] selftests/bpf: Use _REGION1_SIZE in test_snprintf_btf on s390

[Yonghong Song <yhs@fb.com>]

On 2/26/21 5:59 AM, Ilya Leoshkevich wrote:
> test_snprintf_btf fails on s390, because NULL points to a readable
> struct lowcore there. Fix by using _REGION1_SIZE instead.
> 
> Signed-off-by: Ilya Leoshkevich <iii@linux.ibm.com>

Ack with a minor nit below.

Acked-by: Yonghong Song <yhs@fb.com>

> ---
>   tools/testing/selftests/bpf/progs/netif_receive_skb.c | 11 +++++++++--
>   1 file changed, 9 insertions(+), 2 deletions(-)
> 
> diff --git a/tools/testing/selftests/bpf/progs/netif_receive_skb.c b/tools/testing/selftests/bpf/progs/netif_receive_skb.c
> index 6b670039ea67..fa54d2abc41e 100644
> --- a/tools/testing/selftests/bpf/progs/netif_receive_skb.c
> +++ b/tools/testing/selftests/bpf/progs/netif_receive_skb.c
> @@ -16,6 +16,13 @@ bool skip = false;
>   #define STRSIZE			2048
>   #define EXPECTED_STRSIZE	256
>   
> +#if defined(bpf_target_s390)
> +/* NULL points to a readable struct lowcore on s390, so take _REGION1_SIZE */
> +#define BADPTR			((void *)(1ULL << 53))
> +#else
> +#define BADPTR			0
> +#endif
> +
>   #ifndef ARRAY_SIZE
>   #define ARRAY_SIZE(x)	(sizeof(x) / sizeof((x)[0]))
>   #endif
> @@ -114,9 +121,9 @@ int BPF_PROG(trace_netif_receive_skb, struct sk_buff *skb)
>   
>   	/* Check invalid ptr value */
>   	p.ptr = 0;
> -	__ret = bpf_snprintf_btf(str, STRSIZE, &p, sizeof(p), 0);
> +	__ret = bpf_snprintf_btf(str, STRSIZE, &p, sizeof(p), BADPTR);
>   	if (__ret >= 0) {
> -		bpf_printk("printing NULL should generate error, got (%d)",
> +		bpf_printk("printing BADPTR should generate error, got (%d)",
>   			   __ret);

The previous error message NULL implies address 0. The new error message
did not print out the badptr address. Do you want to print them out so
in case of errors, you don't need to search source code?

>   		ret = -ERANGE;
>   	}
>