From Robert.Eckardt@robert-eckardt.de  Thu Mar  6 14:08:39 2003
Return-Path: <Robert.Eckardt@robert-eckardt.de>
Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125])
	by hub.freebsd.org (Postfix) with ESMTP id 46ED137B401
	for <FreeBSD-gnats-submit@freebsd.org>; Thu,  6 Mar 2003 14:08:39 -0800 (PST)
Received: from robert-eckardt.de (pD958A68F.dip.t-dialin.net [217.88.166.143])
	by mx1.FreeBSD.org (Postfix) with ESMTP id 7A85F43FEA
	for <FreeBSD-gnats-submit@freebsd.org>; Thu,  6 Mar 2003 14:08:37 -0800 (PST)
	(envelope-from Robert.Eckardt@robert-eckardt.de)
Received: (from roberte@localhost)
	by robert-eckardt.de (8.11.6/8.11.6) id h26M8b100947;
	Thu, 6 Mar 2003 23:08:37 +0100 (CET)
	(envelope-from roberte)
Message-Id: <200303062208.h26M8b100947@robert-eckardt.de>
Date: Thu, 6 Mar 2003 23:08:37 +0100 (CET)
From: Robert Eckardt <Robert.Eckardt@robert-eckardt.de>
To: FreeBSD-gnats-submit@freebsd.org
Cc:
Subject: 5.0-R/4.7R system hangs on HighPoint RAID w/ ata driver
X-Send-Pr-Version: 3.113
X-GNATS-Notify:

>Number:         48986
>Category:       kern
>Synopsis:       5.0-R/4.7R system hangs on HighPoint RAID w/ ata driver
>Confidential:   no
>Severity:       serious
>Priority:       high
>Responsible:    sos
>State:          closed
>Quarter:        
>Keywords:       
>Date-Required:  
>Class:          sw-bug
>Submitter-Id:   current-users
>Arrival-Date:   Thu Mar 06 14:10:07 PST 2003
>Closed-Date:    Mon Sep 08 12:04:02 PDT 2003
>Last-Modified:  Tue Sep  9 12:50:13 PDT 2003
>Originator:     Robert Eckardt
>Release:        FreeBSD 5.0-RELEASE i386
>Organization:
Home
>Environment:
System: FreeBSD quasar 5.0-RELEASE GENERIC-kernel
EPoX EP-4GEA+ board with IntelCeleron 2.0GHz using i845GE chipset and
AwardBIOS v6.00PG (12/04/2002 for Intel 845GE chipset) with
HighPoint HPT372 UDMA/ATA133 on-board RAID-controller (BIOS v2.32) and
2 IBM DTLA-307075 HDDs as RAID-1 attached to it.

>Description:
FreeBSD's ATA-driver recogizes the RAID-controller and can start installation.
Controlling the RAID with atacontrol (in the emergency shell) works.
On unpacking the distributions the system hangs (not always at the same point).
### This is also valid for 4.7-RELEASE. ###

>How-To-Repeat:
Install 4.7-RELEASE or 5.0-RELEASE on platform with EP-4GEA+ mainboard and
a RAID attached to the HighPoint controller and produce heavy disk I/O.

>Fix:
HighPoint (www.highpoint-tech.com) provides kernel modules for
FreeBSD 4.{1,3,4,5,6.2,7}-RELEASE for controller BIOS v2.34, which
works fine (even with BIOS v2.32). RAID is attached as SCSI device da:
pci1: <PCI bus> on pcib1
hpt37x20: <HPT372 UDMA/ATA133 RAID Controller> port 0xa000-0xa0ff,0x9c00-0x9c03,0x9800-0x9807,0x9400-0x9403,0x9000-0x9007 irq 11 at device 0.0 on pci1
 [...]
Device Driver for HPT37x2 ATA RAID Controller Version 1.31
 [...]
Waiting 5 seconds for SCSI devices to settle
da0 at hpt37x20 bus 0 target 0 lun 0
da0: <HPT Inc. HPT37x2   RAID 1 1.05> Fixed Direct Access SCSI-2 device 
da0: 73308MB (150136560 512 byte sectors: 255H 63S/T 9345C)

Though, for 5.0-RELEASE no workaround known.


>Release-Note:
>Audit-Trail:
Responsible-Changed-From-To: freebsd-bugs->sos 
Responsible-Changed-By: dwmalone 
Responsible-Changed-When: Thu Mar 13 13:42:39 PST 2003 
Responsible-Changed-Why:  
Highpoint ATA Raid PR for Soren. 

http://www.freebsd.org/cgi/query-pr.cgi?pr=48986 
State-Changed-From-To: open->feedback 
State-Changed-By: sos 
State-Changed-When: Mon Apr 28 11:35:44 PDT 2003 
State-Changed-Why:  
Does this still happen on a very resent -current ? 

http://www.freebsd.org/cgi/query-pr.cgi?pr=48986 

From: "marko" <marko@tartu.merit.ee>
To: <freebsd-gnats-submit@freebsd.org>,
	<Robert.Eckardt@robert-eckardt.de>
Cc:  
Subject: Re: kern/48986: 5.0-R/4.7R system hangs on HighPoint RAID w/ ata driver
Date: Sun, 11 May 2003 18:33:27 +0300

 This is a multi-part message in MIME format.
 
 ------=_NextPart_000_0003_01C317EB.D04A75C0
 Content-Type: text/plain;
 	charset="windows-1257"
 Content-Transfer-Encoding: quoted-printable
 
 Commentate agp off to kernel and works fine.
 
 ------=_NextPart_000_0003_01C317EB.D04A75C0
 Content-Type: text/html;
 	charset="windows-1257"
 Content-Transfer-Encoding: quoted-printable
 
 <!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">
 <HTML><HEAD>
 <META http-equiv=3DContent-Type content=3D"text/html; =
 charset=3Dwindows-1257">
 <META content=3D"MSHTML 6.00.2800.1170" name=3DGENERATOR>
 <STYLE></STYLE>
 </HEAD>
 <BODY bgColor=3D#ffffff>
 <DIV><FONT face=3DArial size=3D2>Commentate agp off to kernel and works=20
 fine.</FONT></DIV>
 <DIV><FONT face=3DArial size=3D2></FONT>&nbsp;</DIV></BODY></HTML>
 
 ------=_NextPart_000_0003_01C317EB.D04A75C0--
 
State-Changed-From-To: feedback->closed 
State-Changed-By: sos 
State-Changed-When: Mon Sep 8 12:03:04 PDT 2003 
State-Changed-Why:  
If i read this correctly disabling agp solved the problem. 

http://www.freebsd.org/cgi/query-pr.cgi?pr=48986 

From: "Robert Eckardt" <Robert.Eckardt@Robert-Eckardt.de>
To: freebsd-gnats-submit@FreeBSD.org
Cc:  
Subject: Re: kern/48986: 5.0-R/4.7R system hangs on HighPoint RAID w/ ata driver
Date: Tue, 9 Sep 2003 21:40:19 +0100

 On Mon, 8 Sep 2003 12:04:01 -0700 (PDT), Sren Schmidt wrote
 > Synopsis: 5.0-R/4.7R system hangs on HighPoint RAID w/ ata driver
 >
 > State-Changed-From-To: feedback->closed
 > State-Changed-By: sos
 > State-Changed-When: Mon Sep 8 12:03:04 PDT 2003
 > State-Changed-Why:
 > If i read this correctly disabling agp solved the problem.
 >
 > http://www.freebsd.org/cgi/query-pr.cgi?pr=48986
 
 Hi Sren,
 
 NO, it didn't solve the problem at all!
 
 Currently, I'm running FreeBSD-5.0-R with HighPoint's driver HPT372-1.32.
 Since HighPoint is always some time behind the recent RELEASE (esp. for
 FreeBSD), the latest release that runs on my hardware, and by carefully
 avoiding the RAID-functionality in FreeBSD, is 5.0-R.
 Despite the fact that 5.0 was not intended for "long-time" production
 systems, I won't grade-up my system any time soon now.
 
 As I wrote in private mail to you on 28 May 2003, "Someone [marko] on
 the list recommended to deactivate AGP, which didn't help however ...
 and which is very difficult when the installation itself already fails."
 Though, I never received a reply from you.
 The statement is still true. Commenting off "device agp" did not avoid
 the freeze.
 
 My systems's reaction on using the GENERIC kernel as on CD (latest:
 5.0-CURRENT-20030505-JPSNAP.iso) while creating RAID1 ar0 on ad4 and
 ad6 also was a panic:
 quasar# atacontrol create RAID1 ad4 ad6
 ar0: 73308MB  <ATA RAID1 array> [9345/255/63] status: READY subdisks:
 disk0 READY on ad4 at ata2-master
 disk1 READY on ad6 at ata3-master
 panic: mutex Giant not owned at /usr/src/sys/kern/kern_exit.c:122
 Debugger("panic")
 Stopped at      Debugger+0x54:  xchgl   %ebx,in_Debugger.0
 db>
 
 Another answer to my reply to a posting with title "Raid0 mit IDE-Raid"
 by Dominik Noetzel was that he was also using HighPoint's driver,
 which, however, recognized his controller as HPT372A whereas FreeBSD
 only saw HPT372.
 
 My controller is recognized by both (HighPoint and FreeBSD) as HPT372.
 
 This is, what I'm currently working with:
 (I modified the kernel-Makefiles to automatically copy hpt37x2.ko to
 the new kernel modules, since otherwise ...)
 
 FreeBSD 5.0-RELEASE #5: Mon Aug 25 22:41:44 CEST 2003
    root@quasar:/src/src/sys/i386/compile/QUASAR50v5
 Preloaded elf kernel "/boot/kernel/kernel" at 0xc05df000.
 Preloaded elf module "/boot/kernel/hpt37x2.ko" at 0xc05df0a8.
 Preloaded elf module "/boot/kernel/acpi.ko" at 0xc05df154.
 Timecounter "i8254"  frequency 1193182 Hz
 Timecounter "TSC"  frequency 2019893756 Hz
 CPU: Pentium 4 (2019.89-MHz 686-class CPU)
 <...>
 hpt37x20: <HPT372 UDMA/ATA133 RAID Controller> port
 0xb000-0xb0ff,0xac00-0xac03,0xa800-0xa807,0xa400-0xa403,0xa000-0xa007 irq 10
 at device 0.0 on pci1
 <...>
 Device Driver for HPT37x2 ATA RAID Controller Version 1.32  
 <...>
 Waiting 5 seconds for SCSI devices to settle
 sa0 at sym0 bus 0 target 2 lun 0
 sa0: <HP C1533A A708> Removable Sequential Access SCSI-2 device
 sa0: 10.000MB/s transfers (10.000MHz, offset 8)    
 da0 at hpt37x20 bus 0 target 0 lun 0
 da0: <HPT Inc. HPT37x2   RAID 1 1.05> Fixed Direct Access SCSI-2 device
 da0: 73308MB (150136560 512 byte sectors: 255H 63S/T 9345C)
 
 Therefore I plead for re-openinig the PR and fixing the ATA driver,
 although some people for now have a workaround provided by HighPoint
 Technologies.
 
 Regards,
 Robert
 
 --
 Dr. Robert Eckardt    ---    Robert.Eckardt@Robert-Eckardt.de
 
>Unformatted:
