Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > comp.lang.java.programmer > #4098

Re: Java puzzler

Show key headers only | View raw

>> On 5/14/2011 2:18 AM, Tom Anderson wrote:
>>> byte b = 100;
>>> byte c = (b + b) - b;

> On Sat, 14 May 2011, markspace wrote:
>> I have no opinion about that. I'm only interested in cases where the
>> actual result would be incorrect due to overflow at the machine level.

On 5/14/2011 4:41 PM, Tom Anderson wrote:
> So an expression which might give the right answer, or might blow up,
> depending on how the compiler felt, is okay, just as long as it never
> gives the wrong answer?


Right.  If c above in your example is 100, that's fine.  Or if the 
machine tries to compute it without any algebraeic optimization and gets 
an overflow after the first operation, that's fine too.  As long as it 
doesn't complete and tell me the answer is -20 or something, without 
throwing an exception.


> I'm only saying that the semantics would be defined in terms of infinite
> precision; the implementation could be anything the compiler can come up
> with which simulates the semantics. Since we're talking about fairly
> small expressions with integers, i would expect it to do a good job of
> finding efficient ways to compute things. For example, in the c = (b +
> b) - b case, it would be free to reduce that to c = b.


I think I understand what you are saying, but that's not what I'm after, 
personally.

What I'm saying is: at the machine level, the implementation is free to 
throw an excpetion if any intermediate calculation overflows.  I think 
that's the difference between "infinite precision" and saying the 
compiler and optimizer have some latitude.

What I'm really worried about is how do you simulate "infinite 
precision" with longs or something like that, where there is no larger 
width word?  It becomes a mess.  Then there's also really large numbers 
of small width integers, which could overflow quickly, but "infinite 
precision" means they don't overflow until the end.  Both of those 
scenarios are impractical to implement with "infinite precision" I think.


> The problem is that not all correct results are practical to compute
> with finite-width arithmetic. That means that as well as a rule that
> results have to be correct, we have to have a rule about when results
> that would be correct explode.


It's alright for a result to not be correct, as long the system throws 
an exception instead.  That's the rule: incorrect result means throw an 
exception.

Just to be 100% clear, what I'm really after is for Java to use the 
existing hardware detection that exists in most platforms for integer 
overflow/underflow.  It's practically cost free, and would really go a 
long ways to making programs error-free, imo.  That's what I'm gunning for.

Back to comp.lang.java.programmer | Previous | Next — Previous in thread | Next in thread | Find similar

Thread

Re: Java puzzler Gábor Bakos <aborgabor@gmail.com> - 2011-05-11 01:44 -0700
  Re: Java puzzler Patricia Shanahan <pats@acm.org> - 2011-05-11 05:24 -0700
    Re: Java puzzler Tom Anderson <twic@urchin.earth.li> - 2011-05-11 21:57 +0100
      Re: Java puzzler Patricia Shanahan <pats@acm.org> - 2011-05-11 16:07 -0700
        Re: Java puzzler Tom Anderson <twic@urchin.earth.li> - 2011-05-12 21:38 +0100
          Re: Java puzzler Patricia Shanahan <pats@acm.org> - 2011-05-12 13:45 -0700
            Re: Java puzzler Nancy 3 <n3@gmai1.c0m> - 2011-05-12 16:56 -0400
              Re: Java puzzler markspace <-@.> - 2011-05-12 14:32 -0700
                Re: Java puzzler Patricia Shanahan <pats@acm.org> - 2011-05-12 16:05 -0700
                Re: Java puzzler markspace <-@.> - 2011-05-12 17:40 -0700
                Re: Java puzzler Patricia Shanahan <pats@acm.org> - 2011-05-12 18:18 -0700
                Re: Java puzzler markspace <-@.> - 2011-05-12 19:26 -0700
                Re: Java puzzler Tom Anderson <twic@urchin.earth.li> - 2011-05-14 10:18 +0100
                Re: Java puzzler Patricia Shanahan <pats@acm.org> - 2011-05-14 08:09 -0700
                Re: Java puzzler Lew <noone@lewscanon.com> - 2011-05-14 13:00 -0400
                Re: Java puzzler Joshua Cranmer <Pidgeot18@verizon.invalid> - 2011-05-14 15:08 -0400
                Re: Java puzzler Patricia Shanahan <pats@acm.org> - 2011-05-14 13:25 -0700
                Re: Java puzzler markspace <-@.> - 2011-05-14 08:46 -0700
                Re: Java puzzler Spock <spock@starfleet.ufp> - 2011-05-14 19:33 -0400
                Re: Java puzzler Tom Anderson <twic@urchin.earth.li> - 2011-05-15 00:41 +0100
                Re: Java puzzler markspace <-@.> - 2011-05-14 18:35 -0700
                Re: Java puzzler Joshua Cranmer <Pidgeot18@verizon.invalid> - 2011-05-14 23:43 -0400
                Re: Java puzzler Tom Anderson <twic@urchin.earth.li> - 2011-05-15 11:16 +0100
                Re: Java puzzler markspace <-@.> - 2011-05-15 05:40 -0700
                Re: Java puzzler Tom Anderson <twic@urchin.earth.li> - 2011-05-15 15:23 +0100
                Re: Java puzzler markspace <-@.> - 2011-05-15 08:09 -0700
                Re: Java puzzler Patricia Shanahan <pats@acm.org> - 2011-05-15 07:37 -0700
                Re: Java puzzler markspace <-@.> - 2011-05-15 08:00 -0700
                Re: Java puzzler Patricia Shanahan <pats@acm.org> - 2011-05-15 08:16 -0700
                Re: Java puzzler Patricia Shanahan <pats@acm.org> - 2011-05-15 07:35 -0700
                Re: Java puzzler markspace <-@.> - 2011-05-15 08:11 -0700
                Re: Java puzzler Patricia Shanahan <pats@acm.org> - 2011-05-15 08:15 -0700
                Re: Java puzzler Lew <noone@lewscanon.com> - 2011-05-15 13:11 -0400
                Re: Java puzzler Nancy 3 <n3@gmai1.c0m> - 2011-05-12 20:47 -0400
                Re: Java puzzler Patricia Shanahan <pats@acm.org> - 2011-05-12 18:00 -0700
                Re: Java puzzler markspace <-@.> - 2011-05-12 18:01 -0700
                Re: Java puzzler Nancy 3 <n3@gmai1.c0m> - 2011-05-12 21:22 -0400
                Re: Java puzzler Patricia Shanahan <pats@acm.org> - 2011-05-12 18:32 -0700
                Re: Java puzzler Eric Sosman <esosman@ieee-dot-org.invalid> - 2011-05-12 22:34 -0400
              Re: Java puzzler Eric Sosman <esosman@ieee-dot-org.invalid> - 2011-05-12 22:27 -0400
      Re: Java puzzler "John B. Matthews" <nospam@nospam.invalid> - 2011-05-12 00:28 -0400
      Re: Java puzzler Joshua Cranmer <Pidgeot18@verizon.invalid> - 2011-05-12 10:33 -0400
  Re: Java puzzler Lew <noone@lewscanon.com> - 2011-05-11 09:12 -0400
  Re: Java puzzler Lew <noone@lewscanon.com> - 2011-05-11 09:13 -0400
  Re: Java puzzler markspace <-@.> - 2011-05-11 08:00 -0700

csiph-web