2024-02-01 18:37:27

by Markus Elfring

[permalink] [raw]
Subject: [RFC] perf: Reconsider an error code selection in bpf_map__fprintf()

Hello,

A null pointer check is performed for the input parameter “map”.
It looks suspicious that the function “PTR_ERR” is applied then for
a corresponding return statement.
https://elixir.bootlin.com/linux/v6.8-rc2/source/tools/perf/util/bpf_map.c#L29
https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/tools/perf/util/bpf_map.c?h=v6.8-rc2&id=41bccc98fb7931d63d03f326a746ac4d429c1dd3#n29

This issue was detected by using the Coccinelle software.

Would any other error code determination be more appropriate here?

Regards,
Markus


2024-02-02 06:50:26

by Markus Elfring

[permalink] [raw]
Subject: Re: [RFC] perf: Reconsider an error code selection in bpf_map__fprintf()

> A null pointer check is performed for the input parameter “map”.
> It looks suspicious that the function “PTR_ERR” is applied then for
> a corresponding return statement.

Are contributions also by YueHaibing still waiting on further development considerations?

[PATCH -next] perf: Fix pass 0 to PTR_ERR
https://lore.kernel.org/lkml/[email protected]/
https://lkml.org/lkml/2022/6/11/3


Regards,
Markus

2024-02-06 02:13:29

by Namhyung Kim

[permalink] [raw]
Subject: Re: [RFC] perf: Reconsider an error code selection in bpf_map__fprintf()

Hello,

On Thu, Feb 1, 2024 at 10:49 PM Markus Elfring <[email protected]> wrote:
>
> > A null pointer check is performed for the input parameter “map”.
> > It looks suspicious that the function “PTR_ERR” is applied then for
> > a corresponding return statement.
>
> Are contributions also by YueHaibing still waiting on further development considerations?
>
> [PATCH -next] perf: Fix pass 0 to PTR_ERR
> https://lore.kernel.org/lkml/[email protected]/
> https://lkml.org/lkml/2022/6/11/3

I think we dropped the bpf-loader and it seems bpf_map.[ch] is
leftover. I don't see any users of bpf_map__fprintf() in the tree.
Maybe we can drop it too.

Thanks,
Namhyung