From owner-freebsd-mobile Thu Aug 16 10:53:41 2001 Delivered-To: freebsd-mobile@freebsd.org Received: from rover.village.org (rover.bsdimp.com [204.144.255.66]) by hub.freebsd.org (Postfix) with ESMTP id D525337B408 for ; Thu, 16 Aug 2001 10:53:36 -0700 (PDT) (envelope-from imp@harmony.village.org) Received: from harmony.village.org (harmony.village.org [10.0.0.6]) by rover.village.org (8.11.3/8.11.3) with ESMTP id f7GHrZq17427; Thu, 16 Aug 2001 11:53:35 -0600 (MDT) (envelope-from imp@harmony.village.org) Received: from harmony.village.org (localhost.village.org [127.0.0.1]) by harmony.village.org (8.11.3/8.11.4) with ESMTP id f7GHrZW34243; Thu, 16 Aug 2001 11:53:35 -0600 (MDT) (envelope-from imp@harmony.village.org) Message-Id: <200108161753.f7GHrZW34243@harmony.village.org> To: nate@yogotech.com (Nate Williams) Subject: Re: HEADS UP: pcic pci attachments merged from current Cc: mobile@FreeBSD.ORG In-reply-to: Your message of "Thu, 16 Aug 2001 11:45:02 MDT." <15228.1694.871200.901582@nomad.yogotech.com> References: <15228.1694.871200.901582@nomad.yogotech.com> <15227.21642.728459.378078@nomad.yogotech.com> <15226.52618.594233.704448@nomad.yogotech.com> <200108150433.f7F4X1W20487@harmony.village.org> <200108160317.f7G3HHW28812@harmony.village.org> <200108160533.f7G5XWW29552@harmony.village.org> <15227.65384.819476.423333@nomad.yogotech.com> Date: Thu, 16 Aug 2001 11:53:35 -0600 From: Warner Losh Sender: owner-freebsd-mobile@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.org In message <15228.1694.871200.901582@nomad.yogotech.com> Nate Williams writes: : I'm working on tracking down new hangs as to which commit is causing : this. Cool. I'm getting ready to spend some time out in the Big Blue Room. To be honest, I'd suspect one of two things is going on: First, one of the bits in the config registers that you sent me that are different is hozing you. The interrupt is routed in a funky way that the pcic interrupt isn't seeing since it is coming out a different IRQ than is attached to the PCI line. Likely due to one bit in one of the config registers saying "route this via comsat and map it" or maybe it might be the INTRTIE bit isn't needed on that card. Could be either one. Or, Maybe we do see it once, and fail to properly handle the interrupt, so when we exit our interrupt handler, we go right back in. I consider this less likely since you didn't say you saw any pcic0: Event mask 0xX in your failing dmesg. Warner To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-mobile" in the body of the message