Re: [PATCH bpf-next] libbpf: fix faccessat() usage on Android

[Jiri Olsa <olsajiri@gmail.com>]

On Fri, Jan 26, 2024 at 02:09:44PM -0800, Andrii Nakryiko wrote:
> Android implementation of libc errors out with -EINVAL in faccessat() if
> passed AT_EACCESS ([0]), this leads to ridiculous issue with libbpf
> refusing to load /sys/kernel/btf/vmlinux on Androids ([1]). Fix by
> detecting Android and redefining AT_EACCESS to 0, it's equivalent on
> Android.
> 
>   [0] https://android.googlesource.com/platform/bionic/+/refs/heads/android13-release/libc/bionic/faccessat.cpp#50
>   [1] https://github.com/libbpf/libbpf-bootstrap/issues/250#issuecomment-1911324250
> 
> Fixes: 6a4ab8869d0b ("libbpf: Fix the case of running as non-root with capabilities")
> Signed-off-by: Andrii Nakryiko <andrii@kernel.org>

Acked-by: Jiri Olsa <jolsa@kernel.org>

jirka

> ---
>  tools/lib/bpf/libbpf_internal.h | 14 ++++++++++++++
>  1 file changed, 14 insertions(+)
> 
> diff --git a/tools/lib/bpf/libbpf_internal.h b/tools/lib/bpf/libbpf_internal.h
> index 930cc9616527..5b30f3b67a02 100644
> --- a/tools/lib/bpf/libbpf_internal.h
> +++ b/tools/lib/bpf/libbpf_internal.h
> @@ -19,6 +19,20 @@
>  #include <libelf.h>
>  #include "relo_core.h"
>  
> +/* Android's libc doesn't support AT_EACCESS in faccessat() implementation
> + * ([0]), and just returns -EINVAL even if file exists and is accessible.
> + * See [1] for issues caused by this.
> + *
> + * So just redefine it to 0 on Android.
> + *
> + * [0] https://android.googlesource.com/platform/bionic/+/refs/heads/android13-release/libc/bionic/faccessat.cpp#50
> + * [1] https://github.com/libbpf/libbpf-bootstrap/issues/250#issuecomment-1911324250
> + */
> +#ifdef __ANDROID__
> +#undef AT_EACCESS
> +#define AT_EACCESS 0
> +#endif
> +
>  /* make sure libbpf doesn't use kernel-only integer typedefs */
>  #pragma GCC poison u8 u16 u32 u64 s8 s16 s32 s64
>  
> -- 
> 2.34.1
> 
>