svn commit: r334077 - in head/sbin/devd: . tests

[Rodney W. Grimes]

> On 23 May 2018 at 13:09, Warner Losh <imp at bsdimp.com> wrote:
> >
> >
> > On Wed, May 23, 2018 at 1:39 AM, Eitan Adler <eadler at freebsd.org> wrote:
> >>
> >> Author: eadler
> >> Date: Wed May 23 07:39:02 2018
> >> New Revision: 334077
> >> URL: https://svnweb.freebsd.org/changeset/base/334077
> >>
> >> Log:
> >>   devd: correct two warnings
> >>
> >>   - catching a polymorphic type by value
> >>   - "output between 16 and 95 bytes into a destination of size 80"
> >>
> >> Modified:
> >>   head/sbin/devd/devd.cc
> >>   head/sbin/devd/tests/client_test.c
> >>
> >> Modified: head/sbin/devd/devd.cc
> >>
> >> ==============================================================================
> >> --- head/sbin/devd/devd.cc      Wed May 23 07:39:00 2018        (r334076)
> >> +++ head/sbin/devd/devd.cc      Wed May 23 07:39:02 2018        (r334077)
> >> @@ -1087,7 +1087,7 @@ event_loop(void)
> >>                                 try {
> >>                                         process_event(buffer);
> >>                                 }
> >> -                               catch (std::length_error e) {
> >> +                               catch (const std::length_error& e) {
> >>                                         devdlog(LOG_ERR, "Dropping event
> >> %s "
> >>                                             "due to low memory", buffer);
> >>                                 }
> >>
> >> Modified: head/sbin/devd/tests/client_test.c
> >>
> >> ==============================================================================
> >> --- head/sbin/devd/tests/client_test.c  Wed May 23 07:39:00 2018
> >> (r334076)
> >> +++ head/sbin/devd/tests/client_test.c  Wed May 23 07:39:02 2018
> >> (r334077)
> >> @@ -50,7 +50,7 @@ create_two_events(void)
> >>         FILE *create_stdout;
> >>         FILE *destroy_stdout;
> >>         char mdname[80];
> >> -       char destroy_cmd[80];
> >> +       char destroy_cmd[95];
> >>         char *error;
> >
> >
> > I know it's just a test, but 95 seems equally as magical as 80...
> 
> Agreed. I stole this number from the warning without too much thought.
> Tested that it still passed, and didn't do anything  funky.

It is extremly dangerous to be driven by compiler warnings,
it has driven some developers write off the major bug cliff
causing serious security issues.


-- 
Rod Grimes                                                 rgrimes at freebsd.org