From owner-freebsd-bugs Sat Feb 12 13:42:23 2000 Delivered-To: freebsd-bugs@freebsd.org Received: from mail1.pcom.de (mail1.pcom.de [194.25.152.10]) by builder.freebsd.org (Postfix) with ESMTP id BB8303FA0 for ; Sat, 12 Feb 2000 13:42:17 -0800 (PST) Received: from tkuehn@websurf.pcom.de (212.184.127.40) by mail1.pcom.de with MERCUR-SMTP/POP3/IMAP4-Server (v3.10.18 AS-0098309) for ; Sat, 12 Feb 2000 22:25:03 +0100 Date: Sat, 12 Feb 00 22:26:31 GMT From: tkuehn@websurf.pcom.de (Torsten Kuehn) Reply-To: tkuehn@websurf.pcom.de (Torsten Kuehn) To: freebsd-bugs@freebsd.org Cc: ken@plutotech.com, wilko@freebsd.org X-Mailer: Torsten Kuehn's PMMail v1.1 Subject: ahb Bug still not resolved in 3.2-release (ref. kern/12495) Message-Id: <000212222503224400@mail1.pcom.de> Sender: owner-freebsd-bugs@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org I've got a problem similar PR kern/12495 reported by Wilko Bulte. With a freshly installed system, 3.2-release's GENERIC kernel fails to attach a Nec 502 (6x Caddy) CDROM drive connected to AHA-1742. dmesg reveals (cd0:ahb0:0:3:0): got CAM status 0x4a " " " fatal error, failed to attach to device " " " lost device " " " removing device entry Unlike Wilko's experiences, this happens with an *installed* system. In fact, I am not completely sure if I probably used a different drive during installation. Any post install modifications can only be run over a network connection (just like Wilko did). The drive itself and whole SCSI subsystem run fine with Warp SMP, NT351SMP, Linux 2.2 (the latter has a horrible driver bug which only turns out in SMP mode, did 'fix' it by disabling EISA to PCI buffers in chipset setup). I have tried various modifications to GENERIC kernel, wired-down and unwired devices, got unbootable kernels and reverted - nothing helps. Another oberservation: if machine is restarted (issuing 'reboot') after successful SMP boot (without CDROM), the next reboot fails with SMP: AP CPU #1 launched! <- normally, SCSI probing starts here (probe0:ahb0:0:0:0) ECB 0xc2e83e80 - timed out (probe0:ahb0:0:0:0) Queuing BDR ahb0: no longer in timeout [...] (probe0:ahb0:0:0:0) ECB 0xc2e83e80 - timed out System only restarts successfully after reset button is pressed once. Torsten Kuehn To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-bugs" in the body of the message