RE: [HACKERS] Re: [PORTS] v6.3 release ToDo list and supported po rts
I just tried this on Debian with glibc 2.0.6 and it works fine. That is
my 6.3 postgresql release outputs 4 hours.
Michael
--
Dr. Michael Meskes, Projekt-Manager | topystem Systemhaus GmbH
meskes@topsystem.de | Europark A2, Adenauerstr. 20
meskes@debian.org | 52146 Wuerselen
Go SF49ers! Use Debian GNU/Linux! | Tel: (+49) 2405/4670-44
Show quoted text
----------
From: Thomas G. Lockhart[SMTP:lockhart@alumni.caltech.edu]
Sent: Freitag, 13. Februar 1998 17:57
To: Fred.Reimer@ctg.hboc.com; Bruce Momjian
Cc: hackers@postgreSQL.org
Subject: [HACKERS] Re: [PORTS] v6.3 release ToDo list and
supported portsIs the ix86/linux "confirmed" port using libc 5 or glibc 2.x (aka
libc6)? As most major distributions are going to release all new
versions with glibc I think it's prudent to test on both "platforms"
independently.Yes, that is a good point. There is a known bug in the glibc2 math
library which breaks the date/time routines:select '4 hours'::timespan;
---------
@ 4 hours
(1 row)comes out instead as
tgl=> select '4 hours'::timespan;
----------------------------
@ 3 hours 59 mins 60.00 secs
(1 row)Oliver was working on patches. Lost his e-mail message; is Oliver
still
here? I've got a RH5.0 linux system at work now, and the shipped
Postgres
installation has this problem. I'd like to see it fixed...Bruce, can you add this to the v6.3 ToDo (assuming you agree to take
it
on :)- Tom
I have a "custom" slackware with at least the following updates:
gcc-2.8.0
glibc-2.0.6
As per the glibc upgrade notes out there on the web, I've cleared
(actually moved) the /usr/include header files before installing
glibc, and have reinstalled some "standard" extra libraries (such as
ncurses) after that.
I get the 3 hours 59 minutes 60.00 seconds.
I also get a bunch of other problems in other tests. Some are just
differences in the error message, but others are actual differences
in the float/double numbers produced. Others don't show an error and
give a number where one "should" have been produced. This is all
vauge now, but I will recompile the latest snapshot and give a
complete update maybe tonight, if not tomorrow.
Could some of the other problems be due to gcc-2.8.0, which I'm not
sure is being used by anyone else for postgress (yet)? Should we
include the compiler used in the porting updates (I think so). For
instance, I like to use gcc on the Sparc/Solaris2.5.1 and HP-UX/10.20
boxes I have because it makes porting things easier between those
boxes and my Linux systems. Isn't it possible that postgres compiles
fine using the vendor-supplied compilers, but would fail on gcc, or
visa versa?
- Fred
Show quoted text
I just tried this on Debian with glibc 2.0.6 and it works fine. That
is my 6.3 postgresql release outputs 4 hours.Michael
--
Dr. Michael Meskes, Projekt-Manager | topystem Systemhaus GmbH
meskes@topsystem.de | Europark A2, Adenauerstr.
20 meskes@debian.org | 52146 Wuerselen Go
SF49ers! Use Debian GNU/Linux! | Tel: (+49) 2405/4670-44----------
From: Thomas G. Lockhart[SMTP:lockhart@alumni.caltech.edu]
Sent: Freitag, 13. Februar 1998 17:57
To: Fred.Reimer@ctg.hboc.com; Bruce Momjian
Cc: hackers@postgreSQL.org
Subject: [HACKERS] Re: [PORTS] v6.3 release ToDo list and
supported portsIs the ix86/linux "confirmed" port using libc 5 or glibc 2.x (aka
libc6)? As most major distributions are going to release all new
versions with glibc I think it's prudent to test on both "platforms"
independently.Yes, that is a good point. There is a known bug in the glibc2 math
library which breaks the date/time routines:select '4 hours'::timespan;
---------
@ 4 hours
(1 row)comes out instead as
tgl=> select '4 hours'::timespan;
----------------------------
@ 3 hours 59 mins 60.00 secs
(1 row)Oliver was working on patches. Lost his e-mail message; is Oliver
still
here? I've got a RH5.0 linux system at work now, and the shipped
Postgres
installation has this problem. I'd like to see it fixed...Bruce, can you add this to the v6.3 ToDo (assuming you agree to take
it
on :)- Tom
gcc-2.8.0 is know to be buggy. But that shouldn't bring you the glibc
bug, should it?
Michael
--
Dr. Michael Meskes, Project-Manager | topsystem Systemhaus GmbH
meskes@topsystem.de | Europark A2, Adenauerstr. 20
meskes@debian.org | 52146 Wuerselen
Go SF49ers! Go Rhein Fire! | Tel: (+49) 2405/4670-44
Use Debian GNU/Linux! | Fax: (+49) 2405/4670-10
Show quoted text
-----Original Message-----
From: Frederick W. Reimer [SMTP:Fred.Reimer@ctg.hboc.com]
Sent: Monday, February 16, 1998 12:00 PM
To: Meskes, Michael; hackers@postgreSQL.org;
Fred.Reimer@ctg.hboc.com
Subject: RE: [HACKERS] Re: [PORTS] v6.3 release ToDo list and
supported pI have a "custom" slackware with at least the following updates:
gcc-2.8.0
glibc-2.0.6As per the glibc upgrade notes out there on the web, I've cleared
(actually moved) the /usr/include header files before installing
glibc, and have reinstalled some "standard" extra libraries (such as
ncurses) after that.I get the 3 hours 59 minutes 60.00 seconds.
I also get a bunch of other problems in other tests. Some are just
differences in the error message, but others are actual differences
in the float/double numbers produced. Others don't show an error and
give a number where one "should" have been produced. This is all
vauge now, but I will recompile the latest snapshot and give a
complete update maybe tonight, if not tomorrow.Could some of the other problems be due to gcc-2.8.0, which I'm not
sure is being used by anyone else for postgress (yet)? Should we
include the compiler used in the porting updates (I think so). For
instance, I like to use gcc on the Sparc/Solaris2.5.1 and HP-UX/10.20
boxes I have because it makes porting things easier between those
boxes and my Linux systems. Isn't it possible that postgres compiles
fine using the vendor-supplied compilers, but would fail on gcc, or
visa versa?- Fred
I just tried this on Debian with glibc 2.0.6 and it works fine. That
is my 6.3 postgresql release outputs 4 hours.Michael
--
Dr. Michael Meskes, Projekt-Manager | topystem Systemhaus GmbH
meskes@topsystem.de | Europark A2, Adenauerstr.
20 meskes@debian.org | 52146 Wuerselen Go
SF49ers! Use Debian GNU/Linux! | Tel: (+49) 2405/4670-44----------
From: Thomas G. Lockhart[SMTP:lockhart@alumni.caltech.edu]
Sent: Freitag, 13. Februar 1998 17:57
To: Fred.Reimer@ctg.hboc.com; Bruce Momjian
Cc: hackers@postgreSQL.org
Subject: [HACKERS] Re: [PORTS] v6.3 release ToDo list and
supported portsIs the ix86/linux "confirmed" port using libc 5 or glibc 2.x
(aka
libc6)? As most major distributions are going to release all
new
versions with glibc I think it's prudent to test on both
"platforms"
independently.
Yes, that is a good point. There is a known bug in the glibc2 math
library which breaks the date/time routines:select '4 hours'::timespan;
---------
@ 4 hours
(1 row)comes out instead as
tgl=> select '4 hours'::timespan;
----------------------------
@ 3 hours 59 mins 60.00 secs
(1 row)Oliver was working on patches. Lost his e-mail message; is Oliver
still
here? I've got a RH5.0 linux system at work now, and the shipped
Postgres
installation has this problem. I'd like to see it fixed...Bruce, can you add this to the v6.3 ToDo (assuming you agree to
take
it
on :)- Tom
Import Notes
Resolved by subject fallback
On Mon, 16 Feb 1998, Frederick W. Reimer wrote:
Could some of the other problems be due to gcc-2.8.0, which I'm not
I'm using v2.8.0 on the Solaris Sparc tests I do here...but
nothing else yet...
I just checked the changelogs and Debian has integrated lots of upstream
patches into glibc-2.0.6. So maybe this wasn't fixed before 2.0.6. I'm
currently running 2.0.7pre1 and that one is fine.
Michael
--
Dr. Michael Meskes, Project-Manager | topsystem Systemhaus GmbH
meskes@topsystem.de | Europark A2, Adenauerstr. 20
meskes@debian.org | 52146 Wuerselen
Go SF49ers! Go Rhein Fire! | Tel: (+49) 2405/4670-44
Use Debian GNU/Linux! | Fax: (+49) 2405/4670-10
Show quoted text
-----Original Message-----
From: The Hermit Hacker [SMTP:scrappy@hub.org]
Sent: Monday, February 16, 1998 5:48 PM
To: Frederick W. Reimer
Cc: Meskes, Michael; hackers@postgreSQL.org
Subject: RE: [HACKERS] Re: [PORTS] v6.3 release ToDo list and
supported pOn Mon, 16 Feb 1998, Frederick W. Reimer wrote:
Could some of the other problems be due to gcc-2.8.0, which I'm not
I'm using v2.8.0 on the Solaris Sparc tests I do here...but
nothing else yet...
Import Notes
Resolved by subject fallback
I just checked the changelogs and Debian has integrated lots of upstream
patches into glibc-2.0.6. So maybe this wasn't fixed before 2.0.6. I'm
currently running 2.0.7pre1 and that one is fine.
Andrew, could you add this to the Linux FAQ? The problem statement is:
Q:
Why do I get funny rounding results in some date/time arithmetic, such as
select '4 hours'::timespan;
returning '3 hours 59 minutes 60 seconds'?
A:
You are running the new glibc2 libraries and have a version earlier than
2.0.7. It is a math rounding problem in the library. Upgrade your library.
And, it seems to me that we should avoid those platform-specific patches
which were discussed to try getting this to work with the buggy versions of
these libraries...
- Tom
I just checked the changelogs and Debian has integrated lots of upstream
patches into glibc-2.0.6. So maybe this wasn't fixed before 2.0.6. I'm
currently running 2.0.7pre1 and that one is fine.Andrew, could you add this to the Linux FAQ? The problem statement is:
Q:
Why do I get funny rounding results in some date/time arithmetic, such as
select '4 hours'::timespan;
returning '3 hours 59 minutes 60 seconds'?
Have done....
Andrew
p.s. BRUCE! Could I remind you to make sure the latest FAQs get into the
distribution :-)
----------------------------------------------------------------------------
Dr. Andrew C.R. Martin University College London
EMAIL: (Work) martin@biochem.ucl.ac.uk (Home) andrew@stagleys.demon.co.uk
URL: http://www.biochem.ucl.ac.uk/~martin
Tel: (Work) +44(0)171 419 3890 (Home) +44(0)1372 275775
Import Notes
Resolved by subject fallback
I just checked the changelogs and Debian has integrated lots of upstream
patches into glibc-2.0.6. So maybe this wasn't fixed before 2.0.6. I'm
currently running 2.0.7pre1 and that one is fine.Andrew, could you add this to the Linux FAQ? The problem statement is:
Q:
Why do I get funny rounding results in some date/time arithmetic, such as
select '4 hours'::timespan;
returning '3 hours 59 minutes 60 seconds'?Have done....
Andrew
p.s. BRUCE! Could I remind you to make sure the latest FAQs get into the
distribution :-)
Yep, I need that reminder. I am terrible with that.
--
Bruce Momjian
maillist@candle.pha.pa.us