Received: from sd.tpf.co.jp (sd.tpf.co.jp [210.161.239.34]) by hub.org (8.9.3/8.9.3) with ESMTP id WAA65140 for ; Wed, 7 Jul 1999 22:05:16 -0400 (EDT) (envelope-from Inoue@tpf.co.jp) Received: from cadzone ([126.0.1.40] (may be forged)) by sd.tpf.co.jp (2.5 Build 2640 (Berkeley 8.8.6)/8.8.4) with SMTP id LAA02147; Thu, 08 Jul 1999 11:04:53 +0900 From: "Hiroshi Inoue" To: "Bruce Momjian" Cc: "pgsql-hackers" Subject: RE: [HACKERS] spinlock freeze again Date: Thu, 8 Jul 1999 11:07:10 +0900 Message-ID: <000901bec8e6$967c14a0$2801007e@cadzone.tpf.co.jp> MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit X-Priority: 3 (Normal) X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook 8.5, Build 4.71.2173.0 In-Reply-To: <199907080109.VAA18621@candle.pha.pa.us> X-MimeOLE: Produced By Microsoft MimeOLE V4.72.2106.4 Importance: Normal > -----Original Message----- > From: Bruce Momjian [mailto:maillist@candle.pha.pa.us] > Sent: Thursday, July 08, 1999 10:09 AM > To: Hiroshi Inoue > Cc: pgsql-hackers > Subject: Re: [HACKERS] spinlock freeze again > > > > > Are their any locks acquired before InitProcess()? > > > > > > > OidGenLockId spinlock is acquired in InitTransactionLog(). > > > > Regards. > > > > Well, seems we have a Proc queue that holds locks, but for these other > cases, we don't. We could use the on_shmexit queue to add an cleanup > handler once we get the lock, and remove it from the queue once we > release the lock. We don't currently have the ability to remove > specific queue entries, but we could easily do that. > > Is the lock failure a problem that happens a lot? > It doesn't happen oridinarily. I don't remember well how it happend. Probably it was caused by other spinlock(io_in_progress_lock ?) freeze while testing 6.5-beta. Regards. Hiroshi Inoue Inoue@tpf.co.jp