From owner-freebsd-bugs  Sun May 19 09:10:03 1996
Return-Path: owner-bugs
Received: (from root@localhost)
          by freefall.freebsd.org (8.7.3/8.7.3) id JAA03530
          for bugs-outgoing; Sun, 19 May 1996 09:10:03 -0700 (PDT)
Received: (from gnats@localhost)
          by freefall.freebsd.org (8.7.3/8.7.3) id JAA03524;
          Sun, 19 May 1996 09:10:02 -0700 (PDT)
Resent-Date: Sun, 19 May 1996 09:10:02 -0700 (PDT)
Resent-Message-Id: <199605191610.JAA03524@freefall.freebsd.org>
Resent-From: gnats (GNATS Management)
Resent-To: freebsd-bugs
Resent-Reply-To: FreeBSD-gnats@freefall.FreeBSD.org, mi@aldan.zipnet.net
Received: from guest.zipnet.net (guest.zipnet.net [199.232.255.43])
          by freefall.freebsd.org (8.7.3/8.7.3) with ESMTP id JAA03431
          for <FreeBSD-gnats-submit@freebsd.org>; Sun, 19 May 1996 09:08:18 -0700 (PDT)
Received: (from mi@localhost) by guest.zipnet.net (8.7.5/8.7.3) id MAA00384; Sun, 19 May 1996 12:10:08 GMT
Message-Id: <199605191210.MAA00384@guest.zipnet.net>
Date: Sun, 19 May 1996 12:10:08 GMT
From: Mikhail Teterin <mi@guest.zipnet.net>
Reply-To: mi@aldan.zipnet.net
To: FreeBSD-gnats-submit@freebsd.org
X-Send-Pr-Version: 3.2
Subject: kern/1217: second IDE-harddrive on the second IDE channel 
Sender: owner-bugs@freebsd.org
X-Loop: FreeBSD.org
Precedence: bulk


>Number:         1217
>Category:       kern
>Synopsis:       separating to hardrives to two IDE channels hangs the OS
>Confidential:   no
>Severity:       critical
>Priority:       high
>Responsible:    freebsd-bugs
>State:          open
>Class:          sw-bug
>Submitter-Id:   current-users
>Arrival-Date:   Sun May 19 09:10:01 PDT 1996
>Last-Modified:
>Originator:     Mikhail Teterin
>Organization:
>Release:        FreeBSD 2.2-960501-SNAP i386
>Environment:

	Embedded PCI dual-channel IDE controller (seen as wdc0 and
	wdc1) Neptune chip-set. EISA and PCI buses. Pentium-100.
	NEC Image-P100E machine.  Two Conner CFS1257A disks. Custom
	built kernel.

>Description:

	If I put every disk on it's own IDE channel (motherboard has
	two IDE plugs), mounting the partition on the second drive goes
	fine, but attempt to read a file from there hangs the entire machine,
	although I can do `ls' before reading, and see the correct output.
	Tried async and read-only mounting also. Same effect.

>How-To-Repeat:

	See description.
>Fix:
	
	Put the drives on the same IDE channel (as master and slave).
	But, I thought, I can gain a few percents of performance by
	splitting them.
>Audit-Trail:
>Unformatted: