2 <Title>Release Notes</Title>
7 Should include the migration notes from <FileName>migration/</FileName>.
13 The release notes have not yet been integrated into the new documentation.
14 Check for plain text files in the top of the distribution directory tree
15 and in the <FileName>migration/</FileName> directory for current information.
18 <Title>Release 6.3</Title>
27 <Title>Release 6.2.1</Title>
32 v6.2.1 was a bug-fix and usability release on v6.2. Needs only a few notes.
40 <Title>Release 6.2</Title>
45 This should include information based on Bruce's release summary.
53 <Title>Release 6.1</Title>
58 This should include information based on Bruce's release summary.
64 The regression tests have been adapted and extensively modified for the
65 v6.1 release of PostgreSQL.
69 Three new data types (datetime, timespan, and circle) have been added to
70 the native set of PostgreSQL types. Points, boxes, paths, and polygons
71 have had their output formats made consistant across the data types.
72 The polygon output in misc.out has only been spot-checked for correctness
73 relative to the original regression output.
77 PostgreSQL v6.1 introduces a new, alternate optimizer which uses <FirstTerm>genetic</FirstTerm>
78 algorithms. These algorithms introduce a random behavior in the ordering
79 of query results when the query contains multiple qualifiers or multiple
80 tables (giving the optimizer a choice on order of evaluation). Several
81 regression tests have been modified to explicitly order the results, and
82 hence are insensitive to optimizer choices. A few regression tests are
83 for data types which are inherently unordered (e.g. points and time
84 intervals) and tests involving those types are explicitly bracketed with
85 <Command>set geqo to 'off'</Command> and <Command>reset geqo</Command>.
89 The interpretation of array specifiers (the curly braces around atomic
90 values) appears to have changed sometime after the original regression
91 tests were generated. The current <FileName>./expected/*.out</FileName> files reflect this
92 new interpretation, which may not be correct!
96 The float8 regression test fails on at least some platforms. This is due
97 to differences in implementations of pow() and exp() and the signaling
98 mechanisms used for overflow and underflow conditions.
102 The "random" results in the random test should cause the "random" test
103 to be "failed", since the regression tests are evaluated using a simple
104 diff. However, "random" does not seem to produce random results on my
105 test machine (Linux/gcc/i686).
109 <Title>Timing Results</Title>
112 These timing results are from running the regression test with the command
119 Timing under Linux 2.0.27 seems to have a roughly 5% variation from run
120 to run, presumably due to the timing vagaries of multitasking systems.
128 02:30 Dual Pentium Pro 180, 96MB, UW-SCSI, Linux 2.0.30, gcc 2.7.2.1 -O2 -m486
129 04:12 Dual Pentium Pro 180, 96MB, EIDE, Linux 2.0.30, gcc 2.7.2.1 -O2 -m486
138 06:12 Pentium Pro 180, 32MB, Linux 2.0.30, gcc 2.7.2 -O2 -m486
139 12:06 P-100, 48MB, Linux 2.0.29, gcc
140 39:58 Sparc IPC 32MB, Solaris 2.5, gcc 2.7.2.1 -O -g