|
|
Hi Everybody,
I'm measuring timing jitter on my Smalltalk program. Every 1/2 hour or so I
get a divide by zero error. See the Smalltalk printout, the offending line,
#117, and the 1 second timing jitter data below.
1) I don't think the (direct) cause of the divide by zero error is from
line 117.
2) Could the 1000+ microSecond jump in deviation (red below) be due to
garbage collection?
3) 3.2.91 running on a 6 year old Macbook Air is close (ignore 1000 uSec
jump) to being suitable for high speed machine control. Frankly, I'm
surprised!
4) Is there a way errors such as this on can be trapped?
Just letting everyone know about the divide by zeo error. My program is
just for fun if anybody wants a copy of it.
Comments welcome!
Gary
Object: 1 error: The program attempted to divide a number by zero
ZeroDivide(Exception)>>signal (ExcHandling.st:254)
SmallInteger(Number)>>zeroDivide (SysExcept.st:1386)
Fraction>>setNumerator:setDenominator: (Fraction.st:485)
Fraction class>>numerator:denominator: (Fraction.st:66)
Fraction>>- (Fraction.st:151)
FloatD(Float)>>printOn:special: (Float.st:535)
FloatD(Float)>>printOn: (Float.st:436)
FloatD(Object)>>printString (Object.st:546)
FloatD(Object)>>printNl (Object.st:582)
optimized [] in RunStop>>init (polaris1a.st:117)
[] in Process>>onBlock:at:suspend: (Process.st:394)
BlockClosure>>on:do: (BlkClosure.st:196)
[] in Process>>onBlock:at:suspend: (Process.st:395)
BlockClosure>>ensure: (BlkClosure.st:271)
[] in Process>>onBlock:at:suspend: (Process.st:373)
[] in BlockClosure>>asContext: (BlkClosure.st:182)
BlockContext class>>fromClosure:parent: (BlkContext.st:68)
*Offending line, #117*
oldTime := Time nanosecondClock .
*Deviation from 1 Sec marker, microSeconds:*
32.7289
76.6629
58.759
20.789
66.858
20.0159
53.874
1423.52899999999
-189.252
-52.014
110.686
92.5799
26.289855149499
47.2049
62.557
|
Locked
