From owner-freebsd-scsi Thu Dec 4 17:39:46 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id RAA24611 for freebsd-scsi-outgoing; Thu, 4 Dec 1997 17:39:46 -0800 (PST) (envelope-from owner-freebsd-scsi) Received: from gaia.coppe.ufrj.br (cisigw.coppe.ufrj.br [146.164.5.200]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id RAA24605 for ; Thu, 4 Dec 1997 17:39:43 -0800 (PST) (envelope-from jonny@coppe.ufrj.br) Received: (from jonny@localhost) by gaia.coppe.ufrj.br (8.8.8/8.8.8) id XAA11560 for freebsd-scsi@freebsd.org; Thu, 4 Dec 1997 23:39:40 -0200 (EDT) (envelope-from jonny) From: Joao Carlos Mendes Luis Message-Id: <199712050139.XAA11560@gaia.coppe.ufrj.br> Subject: AHC optimizations To: freebsd-scsi@freebsd.org Date: Thu, 4 Dec 1997 23:39:39 -0200 (EDT) X-Mailer: ELM [version 2.4ME+ PL32 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-freebsd-scsi@freebsd.org X-Loop: FreeBSD.org Precedence: bulk Hi, In 2.2.5 LINT there are the following lines: ... # enable tagged command queueing, which is a major performance win on # devices that support it (and controllers with enough SCB's) # # Note that some drives claim to grok tagged commands, but actually # don't. The HP C3725S is a known offender. options AHC_TAGENABLE # enable SCB paging - See the ahc.4 man page for usage warnings. # (IOW: better don't use it by now.) options AHC_SCBPAGING_ENABLE # The aic7xxx driver will attempt to use memory mapped I/O for all PCI # controllers that have it configured only if this option is set. Unfortunately, # this doesn't work on some motherboards, which prevents it from being the # default. options AHC_ALLOW_MEMIO ... I've read the manual for ahc(4), but could not find anything that could harm my configuration: ... ahc0 rev 0 int a irq 15 on pci0:6 ahc0: aic7880 Wide Channel, SCSI Id=7, 16/255 SCBs ahc0: target 0 Tagged Queuing Device (ahc0:0:0): "IBM XP34300W !z 81K1" type 0 fixed SCSI 2 sd0(ahc0:0:0): Direct-Access 4303MB (8813870 512 byte sectors) sd0(ahc0:0:0): with 3907 cyls, 20 heads, and an average 112 sectors/track (ahc0:5:0): "EXABYTE EXB-82058VQANXR1 0781" type 1 removable SCSI 2 st0(ahc0:5:0): Sequential-Access density code 0x0, drive empty ... But, everytime I enable these options, it panics with a page fault in kernel mode while squid is starting (lots of disk access). The process referred in the panic message is squid, if this matters. Isn't it yet safe to use these options ? Reading ahc(4) I could imagine a disk stopping to answer requests or simply not starting to work, but I can not imagine a kernel panic. Jonny -- Joao Carlos Mendes Luis jonny@gta.ufrj.br +55 21 290-4698 jonny@coppe.ufrj.br Universidade Federal do Rio de Janeiro UFRJ/COPPE/CISI PGP fingerprint: 29 C0 50 B9 B6 3E 58 F2 83 5F E3 26 BF 0F EA 67