New CVSup static binaries for Linux

From: "Thomas G(dot) Lockhart" <lockhart(at)alumni(dot)caltech(dot)edu>
To: Postgres Hackers List <hackers(at)postgresql(dot)org>, pgsql-announce(at)postgresql(dot)org, John Polstra <jdp(at)polstra(dot)com>
Subject: New CVSup static binaries for Linux
Date: 1999-01-22 06:46:48
Message-ID: 36A81ED8.C2C5718E@alumni.caltech.edu
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

I've posted new static binaries for CVSup on Linux on the ftp site:

ftp://postgresql.org/pub/CVSup/cvsup-15.5-*linux*.tar.gz

There are binaries for both libc5 and glibc2 platforms, and I've
included binaries in separate tar files for cvsupd, the server-side
program just in case it helps someone.

The new versions fix "checksum error" problems with pre-v15.4 versions
of the clients (originating from whitespace changes in the output of
newer versions of diff or chksum, I can't remember which).

For reference, I built the glibc2 versions using the "PM3" version of
the Modula-3 compiler, available as RPMs, source, and other packages
from

http://m3.polymtl.ca/m3/

Once Modula-3 is installed, cvsup was built by unpacking a tarball into
/usr/local/src and then compiling using the following commands:

$ make M3TARGET=LINUXLIBC6 M3FLAGS="-D_pm3 -DSTATIC -DNOGUI"
$ make M3TARGET=LINUXLIBC6 M3FLAGS="-D_pm3 -DSTATIC -DNOGUI" install

I built the libc5 version using the older DEC/SRC version of the
Modula-3 compiler, installed from-source, with the following commands:

$ make M3TARGET=LINUXELF M3FLAGS="-D_pm3 -DSTATIC -DNOGUI"
$ make M3TARGET=LINUXELF M3FLAGS="-D_pm3 -DSTATIC -DNOGUI" install

Once again, John Polstra has been very helpful in giving us fresh
versions of source code and in helping to resolve minor porting issues.

Please let me know if you find any problems.

- Tom

Browse pgsql-hackers by date

  From Date Subject
Next Message Michael Meskes 1999-01-22 07:36:41 Re: [HACKERS] ecpg docs
Previous Message Hiroshi Inoue 1999-01-22 06:38:28 RE: [HACKERS] SPI_prepare() doesn't work well ?