Issue with 7.1.3 - regressions and such

From: Chris Browne <cbbrowne(at)acm(dot)org>
To: pgsql-hackers(at)postgresql(dot)org
Subject: Issue with 7.1.3 - regressions and such
Date: 2004-03-04 18:01:11
Message-ID: 60d67sh4pk.fsf@dev6.int.libertyrms.info
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

I recently had the 'joy' of needing to compile a copy of 7.1, to
support a fairly crusty application where we'd have to do more testing
than we can justify in order to upgrade to some (vastly) newer
generation.

Ran into a couple of things worth mentioning:

1. Had a whole lot of gory problems due to 'less than wonderful'
handling of GCC versioning.

In older versions of GCC, there was just 1 one of output. Now there
are many. In new versions of PG configure, it strips out just the
first line.

# Create compiler version string
if test x"$GCC" = x"yes" ; then
#cc_string="GCC `${CC} --version`"
cc_string="GCC `${CC} --version | sed q`"
else
cc_string=$CC
fi
cat >> confdefs.h <<EOF
#define PG_VERSION_STR "PostgreSQL $VERSION on $host, compiled by $cc_string"
EOF

It took quite some looking around to figure out the root of this one,
as wildly unusual things were breaking.

2. Regression tests broke a bit in the area of timestamps.

Apparently as of Red Hat 8.0, there are "entertaining changes" in how
GLIBC deals with dates back before the start-of-UNIX epoch, so that
any dates back in the 1960s and earlier will have some timing
discrepancies.

I don't see that being terribly much worth fixing, but it should
remain 'expected.'

*** ./expected/timestamp.out Thu May 3 15:00:37 2001
--- ./results/timestamp.out Thu Mar 4 12:26:22 2004
***************
*** 572,578 ****
| invalid
| -infinity
| infinity
! | Tue Dec 31 16:00:00 1968 PST
| Sat Feb 10 17:32:01 1996 PST
| invalid
| invalid
--- 572,578 ----
| invalid
| -infinity
| infinity
! | Tue Dec 31 08:00:00 1968 PST
| Sat Feb 10 17:32:01 1996 PST
| invalid
| invalid

======================================================================

*** ./expected/abstime.out Thu Aug 16 14:36:45 2001
--- ./results/abstime.out Thu Mar 4 12:26:21 2004
***************
*** 36,42 ****
| current
| infinity
| -infinity
! | Sat May 10 23:59:12 1947 PST
| invalid
(8 rows)

--- 36,42 ----
| current
| infinity
| -infinity
! | Sat May 10 15:59:12 1947 PST
| invalid
(8 rows)
--
output = ("cbbrowne" "@" "acm.org")
http://cbbrowne.com/info/spreadsheets.html
Rules of the Evil Overlord #17. "When I employ people as advisors, I
will occasionally listen to their advice."
<http://www.eviloverlord.com/>

Browse pgsql-hackers by date

  From Date Subject
Next Message Bruce Momjian 2004-03-04 18:18:47 Re: [HACKERS] Tablespaces
Previous Message tswan 2004-03-04 17:50:11 Re: [HACKERS] Tablespaces