Re: [gpsd-dev] tests hanging on netsbd-6 i386

[Eric S. Raymond]

Greg Troxel <address@hidden>:
> 
> "Eric S. Raymond" <address@hidden> writes:
> 
> > Yup.  The failure commit is the one where that's actually implemented.
> > My guess is that BSD doesn't implement VTIME right.  (You must be getting
> > VMIN right or the old code wouldn't work).
> >
> > No big deal if that's so, we'll just condition out that code for BSD
> > and all you'll lose is the anti-buzz protection.
> 
> The following goes back to 1/0 from n/1, and tests are passing.  I will
> add a better comment and commit.  Or this could be in calling
> gpsd_optimize_io, but to me the point is that the optimize attempt
> doesn't work, so this feels like the right place.
> 
> diff --git a/serial.c b/serial.c
> index 74e339b..0db14d4 100644
> --- a/serial.c
> +++ b/serial.c
> @@ -610,13 +610,17 @@ ssize_t gpsd_serial_write(struct gps_device_t * session,
>  }
>  
>  void gpsd_optimize_io(struct gps_device_t *session, 
> -                  const int minlength, const bool textual UNUSED)
> +                   /* const */ int minlength, const bool textual UNUSED)
>  /* optimize I/O mode depending on the minimum packet size */
>  {
>      /* bail out if this is not actually a tty */
>      if (isatty(session->gpsdata.gps_fd) == 0)
>       return;
>  
> +#ifdef __NetBSD__
> +    minlength = 1;
> +#endif
> +
>      /*
>       * This is an optimization hack.  The idea is to get away from
>       * character-at-a-time I/O in order to slow down the rate at

That's more complex than necessary.  When the device is opened it's set
up to do character-at-a-time blocking I/O.  If you just condition out
the whole body of gpsd_optimize_io() the right thing will happen.
-- 
                <a href="http://www.catb.org/~esr/";>Eric S. Raymond</a>

signature.asc
Description: Digital signature