Re: gcc 4.6 and hot standby - Mailing list pgsql-hackers

From Tom Lane
Subject Re: gcc 4.6 and hot standby
Date
Msg-id 2180.1307559909@sss.pgh.pa.us
Whole thread Raw
In response to Re: gcc 4.6 and hot standby  (Alex Hunsaker <badalex@gmail.com>)
List pgsql-hackers
Alex Hunsaker <badalex@gmail.com> writes:
> So I tracked it down to "-fgcse", that is CFLAGS="-O2 -fno-gcse"
> ./configure works. I then took a few guesses and compiled all of
> postgres with -O2, then manually recompiled xlog.c with -f-no-gcse.
> that combination seems to work.

Huh, interesting.  So the bug must be lurking somewhere around the logic
that deals with failedSources: somehow we're getting to the
ShutdownWalRcv call in XLogPageRead.
        regards, tom lane


pgsql-hackers by date:

Previous
From: Alex Hunsaker
Date:
Subject: Re: gcc 4.6 and hot standby
Next
From: Tom Lane
Date:
Subject: Re: contrib/citext versus collations