Re: BUG #3362: xlog corruption just after initdb on irix - Mailing list pgsql-bugs

From Herve Boulouis
Subject Re: BUG #3362: xlog corruption just after initdb on irix
Date
Msg-id 20070606214446.GH592@ra.aabs
Whole thread Raw
In response to Re: BUG #3362: xlog corruption just after initdb on irix  (Tom Lane <tgl@sss.pgh.pa.us>)
List pgsql-bugs
Le 05/06/2007  12:06, Tom Lane a écrit:
>
> I'm willing to take a look, but first please see if you can eliminate
> the compiler-bug theory.  Check to see if there's a newer compiler
> version available; check whether building with CFLAGS=-O0 makes the
> problem go away.

I tried -O0 with same results but I've found something more interesting :

It seems that my compiler version is not 7.4.4m but 7.4.1m, I got lost
in the sgi compiler upgrades. As the folks at nekochan.net provide a binary
package of postgresql for irix that works correctly, I tried compiling
with their build flags but the problem is still there. I asked the
package maintainer which compiler version they use and it's 7.4.4m. So
our problem could very well be a compiler bug.

The mipspro compiler is a proprietary product and I cannot get the
necessary updates so for now I am unable to verify this theory. Anyway,
maybe you should a add a note in the pg irix faq to use the 7.4.4m version
of mipspro when building the sources ?

--
Herve Boulouis

pgsql-bugs by date:

Previous
From: Tom Lane
Date:
Subject: Re: BUG #3362: xlog corruption just after initdb on irix
Next
From: Alvaro Herrera
Date:
Subject: Re: BUG #3326: Invalid lower bound of autovacuum_cost_limit