Re: time_t size: The year 2038 bug?

From: Matti Aarnio (matti.aarnio@sonera.fi)
Date: Thu Jan 06 2000 - 16:55:26 EST


On Thu, Jan 06, 2000 at 12:06:40PM -0800, David Schwartz wrote:
...
> I realize that this is common sense, but it's just not true.
> The defintion of "the job" is depedent upon what chips are available.
> When the scope of the embedded system is defined, it's based upon
> the current level of technology.
>
> Why did the Hubble, and embedded system, just get upgraded to
> a 486-50? Obviously the system that was there before did the job, right?

        Aside of getting more of house-keeping data processing,
        and memory storage capacity ? (Than with the original
        16 bit 1172A processor.)

> No, what happened was the availability of more powerful embedded
> processors changed the _scope_ of what the job was. This is what has
> _always_ happened.

        The case with HUBBLE is *very* special, even though it is
        (sort of) embedded system. NASA is flying Linux on space-
        missions where it 1) isn't life critical system 2) where the
        computing hardware is in human rated environment.
        (Note: I didn't say *mission critical* -- mission (science)
         goals may well depend on functionality of the Linux based
         laptop, but that is "just" a big embarrasment, not dead
         astronauts.)

        The HUBBLE is *not* Human rated. In fact it flies thru so
        called South Atlantic Magnetic Anomaly at every orbit -- every
        about 1.5 hours. (Well, not *every* orbit, only every 3rd-4th,
        or so..) That area in near-earth space happens to be low-lying
        extension of Van Allen radiation belts around our planet.
        (The service missions don't have the astronauts doing EVA while
         the flightpath enters that area of space.)

        So why 486-50 ? It is available in radiation hardened model.
        (I didn't know that - but its been a few years since I checked
         what is available in that esoteric hardware field.)

        A friend of mine used to be writing software for embedded systems.
        He said roughly: Embedded software is where PROM change needs
        digger crew to get to to box buried under a busy street. (If not
        a cable-ship to drag it out from the bottom of an ocean ... or
        need a shuttle mission worth around USD 1 billion..)

        Those cadgets below the streets and in the ocean do tend stay in
        service for *long* times. Tens of years, very least. Mainly
        because the installation costs are so high... (The cadget costs
        next to nothing compared to the work.)

....
> DS

/Matti Aarnio

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/



This archive was generated by hypermail 2b29 : Fri Jan 07 2000 - 21:00:07 EST