[bpf,v2] libbpf: sanitise map names before pinning

When we added sanitising of map names before loading programs to libbpf, we
still allowed periods in the name. While the kernel will accept these for
the map names themselves, they are not allowed in file names when pinning
maps. This means that bpf_object__pin_maps() will fail if called on an
object that contains internal maps (such as sections .rodata).

Fix this by replacing periods with underscores when constructing map pin
paths. This only affects the paths generated by libbpf when
bpf_object__ping_maps() is called with a path argument. Any pin paths set
by bpf_map__set_pin_path() are unaffected, and it will still be up to the
caller to avoid invalid characters in those.

Fixes: 113e6b7e15e2 ("libbpf: Sanitise internal map names so they are not rejected by the kernel")
Signed-off-by: Toke Høiland-Jørgensen <toke@redhat.com>
---
v2:
  - Move string munging to helper function

 tools/lib/bpf/libbpf.c | 18 ++++++++++++++++--
 1 file changed, 16 insertions(+), 2 deletions(-)

On Wed, Dec 2, 2020 at 3:19 AM Toke Høiland-Jørgensen <toke@redhat.com> wrote:
>
> When we added sanitising of map names before loading programs to libbpf, we
> still allowed periods in the name. While the kernel will accept these for
> the map names themselves, they are not allowed in file names when pinning
> maps. This means that bpf_object__pin_maps() will fail if called on an
> object that contains internal maps (such as sections .rodata).
>
> Fix this by replacing periods with underscores when constructing map pin
> paths. This only affects the paths generated by libbpf when
> bpf_object__ping_maps() is called with a path argument. Any pin paths set
> by bpf_map__set_pin_path() are unaffected, and it will still be up to the
> caller to avoid invalid characters in those.
>
> Fixes: 113e6b7e15e2 ("libbpf: Sanitise internal map names so they are not rejected by the kernel")
> Signed-off-by: Toke Høiland-Jørgensen <toke@redhat.com>
> ---
> v2:
>   - Move string munging to helper function
>
>  tools/lib/bpf/libbpf.c | 18 ++++++++++++++++--
>  1 file changed, 16 insertions(+), 2 deletions(-)
>
> diff --git a/tools/lib/bpf/libbpf.c b/tools/lib/bpf/libbpf.c
> index 8d05132e1945..08ff7783fb93 100644
> --- a/tools/lib/bpf/libbpf.c
> +++ b/tools/lib/bpf/libbpf.c
> @@ -7651,6 +7651,20 @@ bool bpf_map__is_pinned(const struct bpf_map *map)
>         return map->pinned;
>  }
>
> +static char *sanitize_pin_path(char *str)

don't want to be unnecessarily nitpicky, but the return of char *
suggests that this function might be allocating new string, so it's a
bit misleading. doing void function and having non-const char *str
feels most appropriate for this. Nice side-benefit: the implementation
will be even shorter :)


> +{
> +       char *s = str;
> +
> +       /* bpffs disallows periods in path names */
> +       while (*s) {
> +               if (*s == '.')
> +                       *s = '_';
> +               s++;
> +       }
> +
> +       return str;
> +}
> +
>  int bpf_object__pin_maps(struct bpf_object *obj, const char *path)
>  {
>         struct bpf_map *map;
> @@ -7680,7 +7694,7 @@ int bpf_object__pin_maps(struct bpf_object *obj, const char *path)
>                                 err = -ENAMETOOLONG;
>                                 goto err_unpin_maps;
>                         }
> -                       pin_path = buf;
> +                       pin_path = sanitize_pin_path(buf);
>                 } else if (!map->pin_path) {
>                         continue;
>                 }
> @@ -7724,7 +7738,7 @@ int bpf_object__unpin_maps(struct bpf_object *obj, const char *path)
>                                 return -EINVAL;
>                         else if (len >= PATH_MAX)
>                                 return -ENAMETOOLONG;
> -                       pin_path = buf;
> +                       pin_path = sanitize_pin_path(buf);
>                 } else if (!map->pin_path) {
>                         continue;
>                 }
> --
> 2.29.2
>

Andrii Nakryiko <andrii.nakryiko@gmail.com> writes:

> On Wed, Dec 2, 2020 at 3:19 AM Toke Høiland-Jørgensen <toke@redhat.com> wrote:
>>
>> When we added sanitising of map names before loading programs to libbpf, we
>> still allowed periods in the name. While the kernel will accept these for
>> the map names themselves, they are not allowed in file names when pinning
>> maps. This means that bpf_object__pin_maps() will fail if called on an
>> object that contains internal maps (such as sections .rodata).
>>
>> Fix this by replacing periods with underscores when constructing map pin
>> paths. This only affects the paths generated by libbpf when
>> bpf_object__ping_maps() is called with a path argument. Any pin paths set
>> by bpf_map__set_pin_path() are unaffected, and it will still be up to the
>> caller to avoid invalid characters in those.
>>
>> Fixes: 113e6b7e15e2 ("libbpf: Sanitise internal map names so they are not rejected by the kernel")
>> Signed-off-by: Toke Høiland-Jørgensen <toke@redhat.com>
>> ---
>> v2:
>>   - Move string munging to helper function
>>
>>  tools/lib/bpf/libbpf.c | 18 ++++++++++++++++--
>>  1 file changed, 16 insertions(+), 2 deletions(-)
>>
>> diff --git a/tools/lib/bpf/libbpf.c b/tools/lib/bpf/libbpf.c
>> index 8d05132e1945..08ff7783fb93 100644
>> --- a/tools/lib/bpf/libbpf.c
>> +++ b/tools/lib/bpf/libbpf.c
>> @@ -7651,6 +7651,20 @@ bool bpf_map__is_pinned(const struct bpf_map *map)
>>         return map->pinned;
>>  }
>>
>> +static char *sanitize_pin_path(char *str)
>
> don't want to be unnecessarily nitpicky, but the return of char *
> suggests that this function might be allocating new string, so it's a
> bit misleading. doing void function and having non-const char *str
> feels most appropriate for this. Nice side-benefit: the implementation
> will be even shorter :)

Hmm, fair enough. I added the return because I figured it was convenient
to fold the call into the assignment of pin_path, but I don't have a
strong preference either way; will send a v3 :)

-Toke