[PATCH bpf] libbpf: Prevent overriding errno when logging errors

[PATCH bpf] libbpf: Prevent overriding errno when logging errors

[Toke Høiland-Jørgensen]

Turns out there were a few more instances where libbpf didn't save the
errno before writing an error message, causing errno to be overridden by
the printf() return and the error disappearing if logging is enabled.

Signed-off-by: Toke Høiland-Jørgensen <toke@redhat.com>
---
 tools/lib/bpf/libbpf.c | 12 +++++++-----
 1 file changed, 7 insertions(+), 5 deletions(-)

diff --git a/tools/lib/bpf/libbpf.c b/tools/lib/bpf/libbpf.c
index 0a06124f7999..fd256440e233 100644
--- a/tools/lib/bpf/libbpf.c
+++ b/tools/lib/bpf/libbpf.c
@@ -3478,10 +3478,11 @@ bpf_object__probe_global_data(struct bpf_object *obj)
 
 	map = bpf_create_map_xattr(&map_attr);
 	if (map < 0) {
-		cp = libbpf_strerror_r(errno, errmsg, sizeof(errmsg));
+		ret = -errno;
+		cp = libbpf_strerror_r(-ret, errmsg, sizeof(errmsg));
 		pr_warn("Error in %s():%s(%d). Couldn't create simple array map.\n",
-			__func__, cp, errno);
-		return -errno;
+			__func__, cp, -ret);
+		return ret;
 	}
 
 	insns[0].imm = map;
@@ -6012,9 +6013,10 @@ int bpf_program__pin_instance(struct bpf_program *prog, const char *path,
 	}
 
 	if (bpf_obj_pin(prog->instances.fds[instance], path)) {
-		cp = libbpf_strerror_r(errno, errmsg, sizeof(errmsg));
+		err = -errno;
+		cp = libbpf_strerror_r(-err, errmsg, sizeof(errmsg));
 		pr_warn("failed to pin program: %s\n", cp);
-		return -errno;
+		return err;
 	}
 	pr_debug("pinned program '%s'\n", path);
 
-- 
2.28.0

Re: [PATCH bpf] libbpf: Prevent overriding errno when logging errors

[Andrii Nakryiko]

On Thu, Aug 13, 2020 at 7:29 AM Toke Høiland-Jørgensen <toke@redhat.com> wrote:
>
> Turns out there were a few more instances where libbpf didn't save the
> errno before writing an error message, causing errno to be overridden by
> the printf() return and the error disappearing if logging is enabled.
>
> Signed-off-by: Toke Høiland-Jørgensen <toke@redhat.com>
> ---

Acked-by: Andrii Nakryiko <andriin@fb.com>

>  tools/lib/bpf/libbpf.c | 12 +++++++-----
>  1 file changed, 7 insertions(+), 5 deletions(-)
>
> diff --git a/tools/lib/bpf/libbpf.c b/tools/lib/bpf/libbpf.c
> index 0a06124f7999..fd256440e233 100644
> --- a/tools/lib/bpf/libbpf.c
> +++ b/tools/lib/bpf/libbpf.c
> @@ -3478,10 +3478,11 @@ bpf_object__probe_global_data(struct bpf_object *obj)
>
>         map = bpf_create_map_xattr(&map_attr);
>         if (map < 0) {
> -               cp = libbpf_strerror_r(errno, errmsg, sizeof(errmsg));
> +               ret = -errno;
> +               cp = libbpf_strerror_r(-ret, errmsg, sizeof(errmsg));

fyi, libbpf_strerror_r() is smart enough to work with both negative
and positive error numbers (it basically takes abs(err)), so no need
to ensure it's positive here and below.

>                 pr_warn("Error in %s():%s(%d). Couldn't create simple array map.\n",
> -                       __func__, cp, errno);
> -               return -errno;
> +                       __func__, cp, -ret);
> +               return ret;
>         }
>
>         insns[0].imm = map;
> @@ -6012,9 +6013,10 @@ int bpf_program__pin_instance(struct bpf_program *prog, const char *path,
>         }
>
>         if (bpf_obj_pin(prog->instances.fds[instance], path)) {
> -               cp = libbpf_strerror_r(errno, errmsg, sizeof(errmsg));
> +               err = -errno;
> +               cp = libbpf_strerror_r(-err, errmsg, sizeof(errmsg));
>                 pr_warn("failed to pin program: %s\n", cp);
> -               return -errno;
> +               return err;
>         }
>         pr_debug("pinned program '%s'\n", path);
>
> --
> 2.28.0
>

Re: [PATCH bpf] libbpf: Prevent overriding errno when logging errors

[Toke Høiland-Jørgensen]

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

> On Thu, Aug 13, 2020 at 7:29 AM Toke Høiland-Jørgensen <toke@redhat.com> wrote:
>>
>> Turns out there were a few more instances where libbpf didn't save the
>> errno before writing an error message, causing errno to be overridden by
>> the printf() return and the error disappearing if logging is enabled.
>>
>> Signed-off-by: Toke Høiland-Jørgensen <toke@redhat.com>
>> ---
>
> Acked-by: Andrii Nakryiko <andriin@fb.com>
>
>>  tools/lib/bpf/libbpf.c | 12 +++++++-----
>>  1 file changed, 7 insertions(+), 5 deletions(-)
>>
>> diff --git a/tools/lib/bpf/libbpf.c b/tools/lib/bpf/libbpf.c
>> index 0a06124f7999..fd256440e233 100644
>> --- a/tools/lib/bpf/libbpf.c
>> +++ b/tools/lib/bpf/libbpf.c
>> @@ -3478,10 +3478,11 @@ bpf_object__probe_global_data(struct bpf_object *obj)
>>
>>         map = bpf_create_map_xattr(&map_attr);
>>         if (map < 0) {
>> -               cp = libbpf_strerror_r(errno, errmsg, sizeof(errmsg));
>> +               ret = -errno;
>> +               cp = libbpf_strerror_r(-ret, errmsg, sizeof(errmsg));
>
> fyi, libbpf_strerror_r() is smart enough to work with both negative
> and positive error numbers (it basically takes abs(err)), so no need
> to ensure it's positive here and below.

Noted. Although that also means it doesn't hurt either, I suppose; so
not going to bother respinning this unless someone insists :)

-Toke

Re: [PATCH bpf] libbpf: Prevent overriding errno when logging errors

[Daniel Borkmann]

On 8/13/20 9:52 PM, Toke Høiland-Jørgensen wrote:
> Andrii Nakryiko <andrii.nakryiko@gmail.com> writes:
>> On Thu, Aug 13, 2020 at 7:29 AM Toke Høiland-Jørgensen <toke@redhat.com> wrote:
>>>
>>> Turns out there were a few more instances where libbpf didn't save the
>>> errno before writing an error message, causing errno to be overridden by
>>> the printf() return and the error disappearing if logging is enabled.
>>>
>>> Signed-off-by: Toke Høiland-Jørgensen <toke@redhat.com>
>>> ---
>>
>> Acked-by: Andrii Nakryiko <andriin@fb.com>
>>
>>>   tools/lib/bpf/libbpf.c | 12 +++++++-----
>>>   1 file changed, 7 insertions(+), 5 deletions(-)
>>>
>>> diff --git a/tools/lib/bpf/libbpf.c b/tools/lib/bpf/libbpf.c
>>> index 0a06124f7999..fd256440e233 100644
>>> --- a/tools/lib/bpf/libbpf.c
>>> +++ b/tools/lib/bpf/libbpf.c
>>> @@ -3478,10 +3478,11 @@ bpf_object__probe_global_data(struct bpf_object *obj)
>>>
>>>          map = bpf_create_map_xattr(&map_attr);
>>>          if (map < 0) {
>>> -               cp = libbpf_strerror_r(errno, errmsg, sizeof(errmsg));
>>> +               ret = -errno;
>>> +               cp = libbpf_strerror_r(-ret, errmsg, sizeof(errmsg));
>>
>> fyi, libbpf_strerror_r() is smart enough to work with both negative
>> and positive error numbers (it basically takes abs(err)), so no need
>> to ensure it's positive here and below.
> 
> Noted. Although that also means it doesn't hurt either, I suppose; so
> not going to bother respinning this unless someone insists :)

Fixed up while applying, thanks!

Re: [PATCH bpf] libbpf: Prevent overriding errno when logging errors

[Toke Høiland-Jørgensen]

Daniel Borkmann <daniel@iogearbox.net> writes:

> On 8/13/20 9:52 PM, Toke Høiland-Jørgensen wrote:
>> Andrii Nakryiko <andrii.nakryiko@gmail.com> writes:
>>> On Thu, Aug 13, 2020 at 7:29 AM Toke Høiland-Jørgensen <toke@redhat.com> wrote:
>>>>
>>>> Turns out there were a few more instances where libbpf didn't save the
>>>> errno before writing an error message, causing errno to be overridden by
>>>> the printf() return and the error disappearing if logging is enabled.
>>>>
>>>> Signed-off-by: Toke Høiland-Jørgensen <toke@redhat.com>
>>>> ---
>>>
>>> Acked-by: Andrii Nakryiko <andriin@fb.com>
>>>
>>>>   tools/lib/bpf/libbpf.c | 12 +++++++-----
>>>>   1 file changed, 7 insertions(+), 5 deletions(-)
>>>>
>>>> diff --git a/tools/lib/bpf/libbpf.c b/tools/lib/bpf/libbpf.c
>>>> index 0a06124f7999..fd256440e233 100644
>>>> --- a/tools/lib/bpf/libbpf.c
>>>> +++ b/tools/lib/bpf/libbpf.c
>>>> @@ -3478,10 +3478,11 @@ bpf_object__probe_global_data(struct bpf_object *obj)
>>>>
>>>>          map = bpf_create_map_xattr(&map_attr);
>>>>          if (map < 0) {
>>>> -               cp = libbpf_strerror_r(errno, errmsg, sizeof(errmsg));
>>>> +               ret = -errno;
>>>> +               cp = libbpf_strerror_r(-ret, errmsg, sizeof(errmsg));
>>>
>>> fyi, libbpf_strerror_r() is smart enough to work with both negative
>>> and positive error numbers (it basically takes abs(err)), so no need
>>> to ensure it's positive here and below.
>> 
>> Noted. Although that also means it doesn't hurt either, I suppose; so
>> not going to bother respinning this unless someone insists :)
>
> Fixed up while applying, thanks!

Great, thanks!

-Toke