mirrors
/
postgres
mirror of https://github.com/postgres/postgres
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

Update discussion of floating-point problems in regression tests.

This commit is contained in:
Tom Lane 2002-11-08 20:26:12 +00:00
parent bea4792125
commit f2ef470196
1 changed files with 9 additions and 30 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

39
doc/src/sgml/regress.sgml
View File

@ -1,4 +1,4 @@
<!-- $Header: /cvsroot/pgsql/doc/src/sgml/regress.sgml,v 1.29 2002/10/12 16:31:55 petere Exp $ --> <!-- $Header: /cvsroot/pgsql/doc/src/sgml/regress.sgml,v 1.30 2002/11/08 20:26:12 tgl Exp $ -->
<chapter id="regress"> <chapter id="regress">
<title id="regress-title">Regression Tests</title> <title id="regress-title">Regression Tests</title>
@ -264,14 +264,19 @@ PGTZ='PST8PDT7,M04.01.0,M10.05.03'; export PGTZ
Some of the tests involve computing 64-bit (<type>double Some of the tests involve computing 64-bit (<type>double
precision</type>) numbers from table columns. Differences in precision</type>) numbers from table columns. Differences in
results involving mathematical functions of <type>double results involving mathematical functions of <type>double
precision</type> columns have been observed. The float8 and precision</type> columns have been observed. The <literal>float8</> and
geometry tests are particularly prone to small differences across <literal>geometry</> tests are particularly prone to small differences
platforms, or even with different compiler optimization options. across platforms, or even with different compiler optimization options.
Human eyeball comparison is needed to determine the real Human eyeball comparison is needed to determine the real
significance of these differences which are usually 10 places to significance of these differences which are usually 10 places to
the right of the decimal point. the right of the decimal point.
</para> </para>
<para>
Some systems display minus zero as <literal>-0</>, while others
just show <literal>0</>.
</para>
<para> <para>
Some systems signal errors from <function>pow()</function> and Some systems signal errors from <function>pow()</function> and
<function>exp()</function> differently from the mechanism <function>exp()</function> differently from the mechanism
@ -279,32 +284,6 @@ PGTZ='PST8PDT7,M04.01.0,M10.05.03'; export PGTZ
code. code.
</para> </para>
</sect2> </sect2>
<sect2>
<title>Polygon differences</title>
<para>
Several of the tests involve operations on geographic data about
the Oakland/Berkeley, California street map. The map data is expressed as
polygons whose vertices are represented as pairs of <type>double
precision</type> numbers (decimal latitude and
longitude). Initially, some tables are created and loaded with
geographic data, then some views are created that join two
tables using the polygon intersection operator
(<literal>##</literal>), then a select is done on the view.
</para>
<para>
When comparing the results from different platforms, differences
occur in the 2nd or 3rd place to the right of the decimal
point. The SQL statements where these problems occur are the
following:
<programlisting>
SELECT * from street;
SELECT * from iexit;
</programlisting>
</para>
</sect2>
<sect2> <sect2>
<title>Row ordering differences</title> <title>Row ordering differences</title>