Re: 7.1 vs. 7.2 on AIX 5L - Mailing list pgsql-hackers

From Tom Lane
Subject Re: 7.1 vs. 7.2 on AIX 5L
Date
Msg-id 25762.1011116064@sss.pgh.pa.us
Whole thread Raw
In response to Re: 7.1 vs. 7.2 on AIX 5L  ("Zeugswetter Andreas SB SD" <ZeugswetterA@spardat.at>)
List pgsql-hackers
"Zeugswetter Andreas SB SD" <ZeugswetterA@spardat.at> writes:
> I think that the culprit may also be the S_UNLOCK (use _clear_lock) ?

If isync is anything like the MB instruction on Alpha (ie, force memory
updates to occur before proceeding), then it must be used in S_UNLOCK
as well as S_LOCK.  All updates done within the spinlocked section must
reach memory before another processor is allowed to acquire the lock.
        regards, tom lane


pgsql-hackers by date:

Previous
From: "Zeugswetter Andreas SB SD"
Date:
Subject: Re: 7.1 vs. 7.2 on AIX 5L
Next
From: "Hiroshi Inoue"
Date:
Subject: Re: Theory about XLogFlush startup failures