[PATCH] bsd/ntpq: Use the correct address length

[Kinsey Moore]

Yes, sorry about that. I apparently haven't added the subject tweak to this
repo yet.

Thanks,
Kinsey

On Thu, Sep 21, 2023 at 7:02 PM Chris Johns <chrisj at rtems.org> wrote:

> I assume this is in the network services repo?
>
> OK and thanks
>
> Chris
>
> On 22/9/2023 4:38 am, Kinsey Moore wrote:
> > lwIP includes more than the bare address structures in its accounting of
> > the total addrinfo struct size. Ensure that lwIP gets the correct
> > address size.
> > ---
> >  bsd/freebsd/contrib/ntp/ntpq/ntpq.c | 8 ++++++++
> >  1 file changed, 8 insertions(+)
> >
> > diff --git a/bsd/freebsd/contrib/ntp/ntpq/ntpq.c
> b/bsd/freebsd/contrib/ntp/ntpq/ntpq.c
> > index 1cea922..0ea6c08 100644
> > --- a/bsd/freebsd/contrib/ntp/ntpq/ntpq.c
> > +++ b/bsd/freebsd/contrib/ntp/ntpq/ntpq.c
> > @@ -828,6 +828,14 @@ openhost(
> >  #ifdef SYS_VXWORKS
> >          (connect(sockfd, (struct sockaddr *)&hostaddr,
> >                   sizeof(hostaddr)) == -1)
> > +#elif defined(__rtems__)
> > +        /*
> > +         * lwIP's ai_addrlen covers the entire struct which includes
> > +         * padding for extra data and is not accurate for calls which
> > +         * validate the size of the address structure. Use sa_len
> instead
> > +         */
> > +        (connect(sockfd, (struct sockaddr *)ai->ai_addr,
> > +             ai->ai_addr->sa_len) == -1)
> >  #else
> >          (connect(sockfd, (struct sockaddr *)ai->ai_addr,
> >               ai->ai_addrlen) == -1)
> _______________________________________________
> devel mailing list
> devel at rtems.org
> http://lists.rtems.org/mailman/listinfo/devel
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.rtems.org/pipermail/devel/attachments/20230921/9d000fb3/attachment.htm>