Subject:
|
Re: Stack Overflow
|
Newsgroups:
|
lugnet.robotics.handyboard
|
Date:
|
Thu, 11 Feb 1999 02:23:39 GMT
|
Original-From:
|
William Sitch <wsitch@engsoc.carleton#IHateSpam#.ca>
|
Viewed:
|
927 times
|
| |
|
|
Hi everyone:
Sorry for the reply latency, I was buried under an avalanche of work and
assignments. Changing the format of the code (as detailed below) helped
immensely. I re-wrote two of my drivers to use the increased size, and
everything is working perfectly.
I believe this is definitely something for the FAQ, as the solution is
fairly obscure, and the "yacc" keyword is best searched for. Thanks to Max,
Regan, and Fred, for all your kind help.
----- Original Message -----
From: Max Davies <mdavies@CAM.ORG>
To: William Sitch <wsitch@engsoc.carleton.ca>
Cc: <handyboard@media.mit.edu>
Sent: Sunday, January 24, 1999 11:16 AM
Subject: Re: Stack Overflow
> William Sitch wrote:
>
> > ...I tried to increase the stack size of the process I spawn, but the • error
> > still exists. I think I might be getting a compile-time stack overflow - • or
> > something like that. The procedure it crashes for is a leg controller, • and
> > consists of many "if .. else if" statements.
>
> If your code looks something like this...
>
> if (i == 1) do001(); else
> if (i == 2) do002(); else
> if (i == 3) do003(); else
> .
> .
> .
> if (i == 100) do100();
>
> ...then try changing it to this:
>
> /* Not really a loop, just so we can break */
>
> for (;;) {
> if (i == 1) {
> do001();
> break;
> }
> if (i == 2) {
> do002();
> break;
> }
> if (i == 3) {
> do003();
> break;
> }
> .
> .
> .
> if (i == 100) {
> do100();
> break;
> }
>
> break; /* Always break... this isn't really a loop */
> }
>
>
> ...this won't be any less efficient and should resolve your compile-time • stack
> overflow as it doesn't nest the IF statements.
>
> /Max
>
>
|
|
1 Message in This Thread:
- Entire Thread on One Page:
- Nested:
All | Brief | Compact | Dots
Linear:
All | Brief | Compact
|
|
|
|
