Dev:ETL make check - Revision history

Zelgadis: moved Source:ETL make check to Dev:ETL make check

2010-02-22T05:08:47Z

moved Source:ETL make check to Dev:ETL make check

Zelgadis: Text replace - '{{l|Category:' to '{{Category|'

2010-01-13T16:11:55Z

Text replace - '{{l|Category:' to '{{Category|'

Zelgadis: Text replace - ']]' to '}}'

2009-09-08T07:25:23Z

Text replace - ']]' to '}}'

Zelgadis: Text replace - '[[' to '{{l|'

2009-09-08T05:56:12Z

Text replace - '[[' to '{{l|'

Rore: Added to the Code category

2008-04-21T10:54:09Z

Added to the Code category

Dooglus at 02:43, 8 October 2007

2007-10-08T02:43:34Z

Dooglus: explain the failure of the 'make check' tests for 'fixed'

2007-10-08T02:27:43Z

explain the failure of the 'make check' tests for 'fixed'

New page

If we 'make check' in the ETL directory, one of the tests fails.

The failing test is one which multiplies a fixed point number by PI. The result is outside the expected accuracy range.

The problem is that it's working with 12 bits of decimal.

PI is around 12867.9635 * 1/(2^12)

The code in ETL/_fixed.h rounds that down to 12867, and so gets the sum quite wrong.

If it rounded to the nearest integer instead, it would do better.

There are 3 lines in _fixed.h with commented "+0.5" code, like:

fixed_base<T,FIXED_BITS>::fixed_base(const float &f):_data(static_cast<value_type>(f*_ONE()/*+0.5f*/))

If those 3 bits are uncommented, the test passes. I guess it was commented to make the code faster.

Even once that's fixed, however, the 'fixed' check still crashes, when doing:

a=3+i;
b=40+i;
b/=a;

on all values of i from 1 to 10 million. When i gets to 1048573, 3+i is 2^20, which is represented by shifting it left 12 bits, giving 2^32, which on a machine with 32 bit integers is 0.

So the b/=a line crashes, with:
Floating point exception(core dumped)

← Older revision		Revision as of 16:11, 13 January 2010
Line 1:		Line 1:
−	{{l\|Category:Code}}	+	{{Category\|Code}}
	If we 'make check' in the ETL directory, one of the tests fails.		If we 'make check' in the ETL directory, one of the tests fails.

← Older revision		Revision as of 07:25, 8 September 2009
Line 1:		Line 1:
−	{{l\|Category:Code]]	+	{{l\|Category:Code}}
	If we 'make check' in the ETL directory, one of the tests fails.		If we 'make check' in the ETL directory, one of the tests fails.

← Older revision		Revision as of 05:56, 8 September 2009
Line 1:		Line 1:
−	[[Category:Code]]	+	{{l\|Category:Code]]
	If we 'make check' in the ETL directory, one of the tests fails.		If we 'make check' in the ETL directory, one of the tests fails.

← Older revision		Revision as of 10:54, 21 April 2008
Line 1:		Line 1:
		+	[[Category:Code]]
	If we 'make check' in the ETL directory, one of the tests fails.		If we 'make check' in the ETL directory, one of the tests fails.

← Older revision	Revision as of 05:08, 22 February 2010
(No difference)

@@ Line 1: / Line 1: @@
 If we 'make check' in the ETL directory, one of the tests fails.
-The failing test is one which multiplies a fixed point number by PI.  The result is outside the expected accuracy range.
+The failing test is one which multiplies a fixed point number by pi.  The result is outside the expected accuracy range.
-The problem is that it's working with 12 bits of decimal.
+The problem is that it's working with 12 bits of decimal, and pi is around 12867.9635 * 1/(2^12).  The code in ETL/_fixed.h rounds that down to 12867, and so gets the sum quite wrong.  If it rounded to the nearest integer instead, it would do better.
-−
-PI is around 12867.9635 * 1/(2^12)
-−
 The code in ETL/_fixed.h rounds that down to 12867, and so gets the sum quite wrong.
-−
 If it rounded to the nearest integer instead, it would do better.
 There are 3 lines in _fixed.h with commented "+0.5" code, like:
-fixed_base<T,FIXED_BITS>::fixed_base(const float &f):_data(static_cast<value_type>(f*_ONE()/*+0.5f*/))
+ fixed_base<T,FIXED_BITS>::fixed_base(const float &f):
 If those 3 bits are uncommented, the test passes.  I guess it was commented to make the code faster.
@@ Line 19: / Line 14: @@
 Even once that's fixed, however, the 'fixed' check still crashes, when doing:
-			a=3+i;
+ a=3+i;
-			b=40+i;
+ b=40+i;
-			b/=a;
+ b/=a;
 on all values of i from 1 to 10 million.  When i gets to 1048573, 3+i is 2^20, which is represented by shifting it left 12 bits, giving 2^32, which on a machine with 32 bit integers is 0.