Re: [PATCH -next] libbpf: Add additional null-pointer checking in make_parent_dir

[cuigaosheng <cuigaosheng1@huawei.com>]

This email adjusts the code format.

I don't understand why we don't check path for NULL, bpf_link__pin is an 
external
interface, It will be called by external functions and provide input 
parameters,
for example in samples/bpf/hbm.c:

> 201         link = bpf_program__attach_cgroup(bpf_prog, cg1);
> 202         if (libbpf_get_error(link)) {
> 203                 fprintf(stderr, "ERROR: bpf_program__attach_cgroup 
> failed\n");
> 204                 goto err;
> 205         }
> 206
> 207         sprintf(cg_pin_path, "/sys/fs/bpf/hbm%d", cg_id);
> 208         rc = bpf_link__pin(link, cg_pin_path);
> 209         if (rc < 0) {
> 210                 printf("ERROR: bpf_link__pin failed: %d\n", rc);
> 211                 goto err;
> 212         }

if cg_pin_path is NULL, strdup(NULL) will trigger a segmentation fault in
make_parent_dir, I think we should avoid this and add null-pointer checking
for path, just like check_path:
>  7673 static int check_path(const char *path)
>  7674 {
>  7675         char *cp, errmsg[STRERR_BUFSIZE];
>  7676         struct statfs st_fs;
>  7677         char *dname, *dir;
>  7678         int err = 0;
>  7679
>  7680         if (path == NULL)
>  7681                 return -EINVAL;
>  7682
>  7683         dname = strdup(path);
>  7684         if (dname == NULL)
>  7685                 return -ENOMEM;
>  7686
>  7687         dir = dirname(dname);
>  7688         if (statfs(dir, &st_fs)) {
>  7689                 cp = libbpf_strerror_r(errno, errmsg, 
> sizeof(errmsg));
>  7690                 pr_warn("failed to statfs %s: %s\n", dir, cp);
>  7691                 err = -errno;
>  7692         }
>  7693         free(dname);
>  7694
>  7695         if (!err && st_fs.f_type != BPF_FS_MAGIC) {
>  7696                 pr_warn("specified path %s is not on BPF FS\n", 
> path);
>  7697                 err = -EINVAL;
>  7698         }
>  7699
>  7700         return err;
>  7701 }

Thanks.


在 2022/4/22 0:55, Andrii Nakryiko 写道:
> On Thu, Apr 21, 2022 at 6:01 AM Gaosheng Cui <cuigaosheng1@huawei.com> wrote:
>> The make_parent_dir is called without null-pointer checking for path,
>> such as bpf_link__pin. To ensure there is no null-pointer dereference
>> in make_parent_dir, so make_parent_dir requires additional null-pointer
>> checking for path.
>>
>> Signed-off-by: Gaosheng Cui <cuigaosheng1@huawei.com>
>> ---
>>   tools/lib/bpf/libbpf.c | 3 +++
>>   1 file changed, 3 insertions(+)
>>
>> diff --git a/tools/lib/bpf/libbpf.c b/tools/lib/bpf/libbpf.c
>> index b53e51884f9e..5786e6184bf5 100644
>> --- a/tools/lib/bpf/libbpf.c
>> +++ b/tools/lib/bpf/libbpf.c
>> @@ -7634,6 +7634,9 @@ static int make_parent_dir(const char *path)
>>          char *dname, *dir;
>>          int err = 0;
>>
>> +       if (path == NULL)
>> +               return -EINVAL;
>> +
> API contract is that path shouldn't be NULL. Just like we don't check
> link, obj, prog for NULL in every single API, I don't think we need to
> do it here, unless I'm missing something?
>
>>          dname = strdup(path);
>>          if (dname == NULL)
>>                  return -ENOMEM;
>> --
>> 2.25.1
>>
> .