re: CVS commit: src/usr.sbin/npf/npfctl

2018-09-21 Thread matthew green 
> >> Modified Files:
> >>src/usr.sbin/npf/npfctl: npf.conf.5
> >>
> >> Log Message:
> >> Switch back to tabs, it was nicer this way.
> >
> > now these examples *really* overflow 80 lines, vs only just.
> >
> > can you try fix this?  thanks.
> 
> Even with whitespaces there were overflows.
> 
> I've fixed it, not by replacing random tabs by whitespaces, but by cutting
> long lines.

thanks.   i can read it on a normal sized tty, and
better than the old too.


.mrg.

Re: CVS commit: src/usr.sbin/npf/npfctl

2018-09-21 Thread Maxime Villard 

Le 19/09/2018 à 20:44, matthew green a écrit :

"Maxime Villard" writes:

Module Name:src
Committed By:   maxv
Date:   Wed Sep 19 15:36:12 UTC 2018

Modified Files:
src/usr.sbin/npf/npfctl: npf.conf.5

Log Message:
Switch back to tabs, it was nicer this way.


now these examples *really* overflow 80 lines, vs only just.

can you try fix this?  thanks.


Even with whitespaces there were overflows.

I've fixed it, not by replacing random tabs by whitespaces, but by cutting
long lines.

re: CVS commit: src/usr.sbin/npf/npfctl

2018-09-19 Thread matthew green 
"Maxime Villard" writes:
> Module Name:  src
> Committed By: maxv
> Date: Wed Sep 19 15:36:12 UTC 2018
> 
> Modified Files:
>   src/usr.sbin/npf/npfctl: npf.conf.5
> 
> Log Message:
> Switch back to tabs, it was nicer this way.

now these examples *really* overflow 80 lines, vs only just.

can you try fix this?  thanks.


.mrg.

Re: CVS commit: src/usr.sbin/npf/npfctl

2018-09-01 Thread Maxime Villard 

Le 01/09/2018 à 18:28, Mindaugas Rasiukevicius a écrit :

Module Name:src
Committed By:   rmind
Date:   Sat Sep  1 16:28:57 UTC 2018

Modified Files:
src/usr.sbin/npf/npfctl: npf.conf.5

Log Message:
npf.conf(5): fix some of the previous incorrect or inaccurate changes.
The TCP flags option is not only for the stateful tracking.  Dynamic NAT
implies NAPT; algorithms, at least for now, are for static NAT mappings.
Mention that ALG ICMP is also for traceroute behind NAT; also mention
"MSS clamping" (some users might search for this term, so keeping the
terminology is helpful).


To generate a diff of this commit:
cvs rdiff -u -r1.70 -r1.71 src/usr.sbin/npf/npfctl/npf.conf.5

Please note that diffs are not public domain; they are subject to the
copyright notices on the relevant files.


Indeed the 'flags' option was not documented in the right section, thanks
for moving it in the correct one.

I believe it would be better to change

-   The protocol keyword can additionally have ...
+   The
+   .Cd proto
+   keyword can additionally have ...

To keep things consistent. Or even "This keyword" to avoid repeating.

And the end of the sentence should maybe be "protocol-specific options such
as 'flags'", to make it clearer that you just want to talk about 'flags' in
what follows.

Re: CVS commit: src/usr.sbin/npf/npfctl

2013-11-19 Thread Mindaugas Rasiukevicius 
Christos Zoulas chris...@netbsd.org wrote:
 Module Name:  src
 Committed By: christos
 Date: Tue Nov 19 17:01:45 UTC 2013
 
 Modified Files:
   src/usr.sbin/npf/npfctl: npf_show.c
 
 Log Message:
 CID 1129614: dereference after null
 

It should never be NULL.  Perhaps assert(tl != NULL)?

-- 
Mindaugas

Re: CVS commit: src/usr.sbin/npf/npfctl

2013-11-19 Thread Christos Zoulas 
In article 20131119213306.88c0114a...@mail.netbsd.org,
Mindaugas Rasiukevicius  rm...@netbsd.org wrote:
Christos Zoulas chris...@netbsd.org wrote:
 Module Name: src
 Committed By:christos
 Date:Tue Nov 19 17:01:45 UTC 2013
 
 Modified Files:
  src/usr.sbin/npf/npfctl: npf_show.c
 
 Log Message:
 CID 1129614: dereference after null
 

It should never be NULL.  Perhaps assert(tl != NULL)?

That would be fine too.

christos

Re: CVS commit: src/usr.sbin/npf/npfctl

2012-10-31 Thread Iain Hibbert 
On Wed, 31 Oct 2012, Martin Husemann wrote:

 Module Name:  src
 Committed By: martin
 Date: Wed Oct 31 08:54:39 UTC 2012

 Modified Files:
   src/usr.sbin/npf/npfctl: npfctl.c

 Log Message:
 gcc 4.1 is not smart enough to notice arg is only used when initialized
 correctly and produces a might be used unintialized warning.

how can it, when nct is passed to external function ioctl() before
potentially looping back? the structure contents may have changed..

regards,
iain

Re: CVS commit: src/usr.sbin/npf/npfctl

2012-10-31 Thread Martin Husemann 
On Wed, Oct 31, 2012 at 09:38:09AM +, Iain Hibbert wrote:
 how can it, when nct is passed to external function ioctl() before
 potentially looping back? the structure contents may have changed..

I see - and indeed moving again: down stops it from complaining.
I apologize, apparently it is smarter than me.

Martin