From owner-freebsd-smp Fri Sep 25 12:54:36 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id MAA29199 for freebsd-smp-outgoing; Fri, 25 Sep 1998 12:54:36 -0700 (PDT) (envelope-from owner-freebsd-smp@FreeBSD.ORG) Received: from magnet.geophysik.tu-freiberg.de (magnet.geophysik.tu-freiberg.de [139.20.128.6]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id MAA29184; Fri, 25 Sep 1998 12:54:23 -0700 (PDT) (envelope-from freebsd@magnet.geophysik.tu-freiberg.de) Received: (from freebsd@localhost) by magnet.geophysik.tu-freiberg.de (8.9.1/8.7.3) id VAA02617; Fri, 25 Sep 1998 21:50:02 +0200 (CEST) From: Holm Tiffe Message-Id: <199809251950.VAA02617@magnet.geophysik.tu-freiberg.de> Subject: AHA 2742T+CAM+SMP problems 2nd. try #2 To: freebsd-scsi@FreeBSD.ORG Date: Fri, 25 Sep 1998 21:50:01 +0200 (CEST) Cc: freebsd-smp@FreeBSD.ORG Reply-To: freebsd@magnet.geophysik.tu-freiberg.de X-Mailer: ELM [version 2.4ME+ PL26 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org More from my SMP problem: Today evening I have upgraded my src/sys to cvs-cur.4676 (cam_xpt.c version 1.13) and now I get the following with the new SMP kernel: SMP: AP CPU #1 Launched! ........................... here is a long pause with a stady lightening SCSI LED (probe9:ahc0:1:2:0): Sending SDTR! (what this means Scsi Device Target Reset ?) (probe9:ahc0:1:2:0): SCB 0x9 timed out in message in phase, SCSISIGI == 0xb6 SEQADDR == 0x15b SSTAT1 == 0x13 (probe9:ahc0:1:2:0): BDR message in message buffer Fatal trap 12: page fault while in kernel mode mp_lock = 01000002; cpuid = 1, lapicid = 01000000 fault virtual address = 0xb8 funkt code = supervisor read, page not present instruction pointer = 0x8:0xf0106a78 stack pointer = 0x10:0xff804cc4 frame pointer = 0x10:0xff805e04 code segment = base 0x0,limit 0xfffff, type 0x16 = DPL0; pres1, def32 1, gran 1 processor eflags = interrupt enabled, resume, IOPL=0 current process = Idle interrupt mask = cam <- SMP: XXX kernel type 12 trap, code=0 stopped at _xpt_set_transfer_settings+0xd8: -> testb $0x20,0xb8(%esi) trace: _xpt_set_transfer_settings(ff804e78,1) at _xpt_set_transfer_settings+0xd8 _xpt_async(200,f05acc40,ff804e78,ff804e78,f05acc40,1,f07b8000) at _xpt_async+0xb0 _ahc_set_syncrate(f07b8000,ff804f38,f05acc40,0,0,0,1,f07b8000,ff804f38,f05acc40,0,1) at ahc_set_syncrate+0x1de _ahc_handle_devreset() _ahc_handle_scsiint() _ahc_intr() .... this is shorted, pencil broken :-) this seams to be a NULL pointer reference in cam_xpt.c around line 5302 , maybe device->quirk is not jet initialized ? Holm PS: Changing the AHA 2742T to another (rev C) and changing the P5/90 CPU to a P5/100 has nothing changed. The SCSI bus hangs after the SMP: AP CPU #1 Launched line... -- ******************************************************************************* * Holm Tiffe holm@geophysik.tu-freiberg.de * * Freiberger Strasse 24 * * 09600 Kleinschirma, Germany Microsoft is not the Answer - * * Tel.: 49 3731 74233 Microsoft is the Question, * * UUCP: 49 3731 73719 unicorn!holm and the Answer is no ! * ******************************************************************************* To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message