Re: Erroneous PPC spinlock code
Bruce Momjian <pgman@candle.pha.pa.us>
From: Bruce Momjian <pgman@candle.pha.pa.us>
To: Marcus Meissner <meissner@suse.de>
Cc: Reinhard Max <max@suse.de>, Tom Lane <tgl@sss.pgh.pa.us>, Peter Eisentraut <peter_e@gmx.net>, PostgreSQL Development <pgsql-hackers@postgresql.org>
Date: 2003-11-10T03:29:32Z
Lists: pgsql-hackers
Marcus Meissner wrote: -- Start of PGP signed section. > On Thu, Nov 06, 2003 at 12:08:56AM +0100, Reinhard Max wrote: > > On Wed, 5 Nov 2003 at 13:28, Tom Lane wrote: > > > > > Peter Eisentraut <peter_e@gmx.net> writes: > > > > > > > The SuSE PPC guru said that the PPC spinlock code we currently use > > > > may behave erroneously on multiprocessor systems. > > > > > > What's his evidence for that claim? > > > > Let's ask himself. > > > > > The code we have is based directly on the recommendations in the PPC > > > manuals, and has been tested on multi-CPU systems. > > > > Marcus, can you explain the details, please? > > I reviewed the documentation again (at: > http://www-1.ibm.com/servers/esdd/articles/powerpc.html > ) and it seems to agree with your opinion. > > I retract my comment, leave your code as-is. Cool. Thanks for checking. -- Bruce Momjian | http://candle.pha.pa.us pgman@candle.pha.pa.us | (610) 359-1001 + If your life is a hard drive, | 13 Roberts Road + Christ can be your backup. | Newtown Square, Pennsylvania 19073