[Question] Dead code in xt_register_matches/targets()?

[Question] Dead code in xt_register_matches/targets()?

[Ingyu Jang]

Hi,

I noticed that in net/netfilter/x_tables.c, the functions
xt_register_match() and xt_register_target() always return 0.

Both functions simply perform:
  - mutex_lock()
  - list_add()
  - mutex_unlock()
  - return 0

However, xt_register_matches() and xt_register_targets() check
the return values and have error handling paths:

  for (i = 0; i < n; i++) {
      err = xt_register_match(&match[i]);
      if (err)
          goto err;
  }

Since xt_register_match/target() never fail, these error checks
appear to be dead code.

I found multiple callers that check these return values:
  - net/netfilter/xt_set.c
  - net/netfilter/xt_MASQUERADE.c
  - and others

Is this intentional defensive coding for potential future changes,
or could this be cleaned up by making xt_register_match/target()
return void?

Thanks,
Ingyu Jang

Re: [Question] Dead code in xt_register_matches/targets()?

[Jan Engelhardt]

On Saturday 2026-01-31 15:55, Ingyu Jang wrote:

>I noticed that in net/netfilter/x_tables.c, the functions
>xt_register_match() and xt_register_target() always return 0.

See commit 7926dbfa4bc14e27f4e18a6184a031a1c1e077dc
and the state prior.

Re: [Question] Dead code in xt_register_matches/targets()?

[Florian Westphal]

Ingyu Jang <ingyujang25@korea.ac.kr> wrote:
> I noticed that in net/netfilter/x_tables.c, the functions
> xt_register_match() and xt_register_target() always return 0.
> 
> Both functions simply perform:
>   - mutex_lock()
>   - list_add()
>   - mutex_unlock()
>   - return 0
> 
> However, xt_register_matches() and xt_register_targets() check
> the return values and have error handling paths:
> 
>   for (i = 0; i < n; i++) {
>       err = xt_register_match(&match[i]);
>       if (err)
>           goto err;
>   }
> 
> Since xt_register_match/target() never fail, these error checks
> appear to be dead code.
> 
> I found multiple callers that check these return values:
>   - net/netfilter/xt_set.c
>   - net/netfilter/xt_MASQUERADE.c
>   - and others
> 
> Is this intentional defensive coding for potential future changes,

No, leftover.  It could be cleaned up but I would prefer to keep it
as-is, x_tables is in maintenance mode.