Re: [GENERAL] Probs with float8, geometry, abstime regression tests on Linux

O/S: Redhat 6.0 Linux 2.2.14
hardware:DELL POWEREDGE 8E. 500MHZ 512MB.
CPU0: Intel Pentium III (Katmai) stepping 03
(SMP support on kernel, but single CPU).

PG Version: Postgres 6.5.3
configure option: just used --prefix=<mypath>

Failed regression tests:
float8 .. failed
geometry .. failed
abstime .. failed

float8:
Don't get pow() and exp() out of range errors but get like 5 rows of
results instead. The README says that some platforms are different, however
I had the impression that the platform I'm using was similar to the
reference one. Should this be happening?
e.g.
QUERY: SELECT '' AS bad, f.f1 ^ '1e200' from FLOAT8_TBL f;

bad|?column?
---+--------
   |0
   |NaN
   |NaN
   |NaN
   |NaN
(5 rows)

QUERY: SELECT '' AS bad, : (f.f1) from FLOAT8_TBL f;

bad|            ?column?
---+--------------------
   |                   1
   |7.39912306090513e-16
   |                   0
   |                   0
   |                   1

geometry:
Quite a number of failures, probably the usual rounding stuff. However I
thought our Linux Pentium III system should give the same results as the
Postgres reference one (same goes for float8). Are there any
compiler/configuration options which I should be using so as to match the
reference results?

abstime:
The regression test not Y2K compliant?

SELECT '' AS four, ABSTIME_TBL.*
  WHERE ABSTIME_TBL.f1 <?>
   '["Apr 1 1950 00:00:00" "Dec 30 1999 23:00:00"]'::tinterval;
Expected:
four|f1
----+----------------------------
    |Sun Jan 14 03:14:21 1973 PST
    |Mon May 01 00:30:30 1995 PDT
    |epoch
    |current
(4 rows)

But nowadays current is not < Dec 30 1999 so should not appear in actual ;).

Best regards,

Link.