NetBSD Problem Report #23529
Received: (qmail 28810 invoked by uid 605); 22 Nov 2003 00:23:05 -0000
Message-Id: <20031122002303.F3EB911153@narn.netbsd.org>
Date: Sat, 22 Nov 2003 00:23:03 +0000 (UTC)
From: m.hertrick@neovera.com
Sender: gnats-bugs-owner@NetBSD.org
Reply-To: m.hertrick@neovera.com
To: gnats-bugs@gnats.NetBSD.org
Subject: pciide0:0 and 0:1 bogus intr messages repeating forever on kernel load
X-Send-Pr-Version: www-1.0
>Number: 23529
>Category: kern
>Synopsis: pciide0:0 and 0:1 bogus intr messages repeating forever on kernel load
>Confidential: no
>Severity: critical
>Priority: high
>Responsible: kern-bug-people
>State: closed
>Class: sw-bug
>Submitter-Id: net
>Arrival-Date: Sat Nov 22 00:24:00 +0000 2003
>Closed-Date: Tue Nov 25 21:05:54 +0000 2003
>Last-Modified: Tue Nov 25 21:05:54 +0000 2003
>Originator: Michael Hertrick
>Release: -current
>Organization:
Neovera, Inc.
>Environment:
NetBSD squarepeg 1.6.1_STABLE NetBSD 1.6.1_STABLE (squarepeg) #0: Mon Sep 15 00:56:19 EDT 2003 root@squarepeg:/usr/src/sys/arch/i386/compile/squarepeg i386
>Description:
Boot of latest -current kernel goes into endless loop just before it would have printed "Kernelized RAIDframe activated". The kernel prints "pciide0:0 bogus intr" and "pciide0:1 bogus intr" indefinitely.
1.6.1-STABLE works fine with the same hardware configuration, however a single "pciide1:0 bogus intr" message is printed... but then finishes booting normally and functions properly.
Here's a dmesg from a good boot:
NetBSD 1.6.1_STABLE (squarepeg) #0: Mon Sep 15 00:56:19 EDT 2003
root@squarepeg:/usr/src/sys/arch/i386/compile/squarepeg
cpu0: AMD Athlon Model 7 (Morgan) (686-class), 995.80 MHz
cpu0: I-cache 64 KB 64b/line 2-way, D-cache 64 KB 64b/line 2-way
cpu0: L2 cache 64 KB 64b/line 16-way
cpu0: features 383f9ff<FPU,VME,DE,PSE,TSC,MSR,PAE,MCE,CX8,SEP,MTRR>
cpu0: features 383f9ff<PGE,MCA,CMOV,FGPAT,PSE36,MMX>
cpu0: features 383f9ff<FXSR,SSE>
total memory = 191 MB
avail memory = 174 MB
using 2478 buffers containing 9912 KB of memory
BIOS32 rev. 0 found at 0xfdae0
mainbus0 (root)
pci0 at mainbus0 bus 0: configuration mode 1
pci0: i/o space, memory space enabled, rd/line, rd/mult, wr/inv ok
pchb0 at pci0 dev 0 function 0
pchb0: vendor 0x1039 product 0x0735 (rev. 0x01)
agp0 at pchb0: aperture at 0xd0000000, size 0x4000000
ppb0 at pci0 dev 1 function 0: vendor 0x1039 product 0x0001 (rev. 0x00)
pci1 at ppb0 bus 1
pci1: i/o space, memory space enabled
vga0 at pci1 dev 0 function 0: vendor 0x10de product 0x0110 (rev. 0xa1)
wsdisplay0 at vga0 kbdmux 1: console (80x25, vt100 emulation)
wsmux1: connecting to wsdisplay0
pcib0 at pci0 dev 2 function 0
pcib0: vendor 0x1039 product 0x0008 (rev. 0x00)
ohci0 at pci0 dev 2 function 2: vendor 0x1039 product 0x7001 (rev. 0x07)
ohci0: interrupting at irq 11
ohci0: OHCI version 1.0, legacy support
usb0 at ohci0: USB revision 1.0
uhub0 at usb0
uhub0: vendor 0x1039 OHCI root hub, class 9/0, rev 1.00/1.00, addr 1
uhub0: 3 ports with 3 removable, self powered
ohci1 at pci0 dev 2 function 3: vendor 0x1039 product 0x7001 (rev. 0x07)
ohci1: interrupting at irq 11
ohci1: OHCI version 1.0, legacy support
usb1 at ohci1: USB revision 1.0
uhub1 at usb1
uhub1: vendor 0x1039 OHCI root hub, class 9/0, rev 1.00/1.00, addr 1
uhub1: 3 ports with 3 removable, self powered
pciide0 at pci0 dev 2 function 5: Silicon Integrated System 735 IDE controller (rev. 0xd0)
pciide0: bus-master DMA support present
pciide0: primary channel wired to compatibility mode
wd0 at pciide0 channel 0 drive 0: <QUANTUM FIREBALLlct08 08>
wd0: drive supports 16-sector PIO transfers, LBA addressing
wd0: 8063 MB, 16383 cyl, 16 head, 63 sec, 512 bytes/sect x 16514064 sectors
wd0: 32-bit data port
wd0: drive supports PIO mode 4, DMA mode 2, Ultra-DMA mode 4 (Ultra/66)
pciide0: primary channel interrupting at irq 14
wd0(pciide0:0:0): using PIO mode 4, Ultra-DMA mode 4 (Ultra/66) (using DMA data transfers)
pciide0: secondary channel wired to compatibility mode
atapibus0 at pciide0 channel 1: 2 targets
cd0 at atapibus0 drive 0: <SONY CD-RW CRX175A1, , 5YS2> type 5 cdrom removable
cd0: 32-bit data port
cd0: drive supports PIO mode 4, DMA mode 2, Ultra-DMA mode 2 (Ultra/33)
wd1 at pciide0 channel 1 drive 1: <Maxtor 6Y080P0>
wd1: drive supports 16-sector PIO transfers, LBA addressing
wd1: 78167 MB, 16383 cyl, 16 head, 63 sec, 512 bytes/sect x 160086528 sectors
wd1: 32-bit data port
wd1: drive supports PIO mode 4, DMA mode 2, Ultra-DMA mode 6 (Ultra/133)
pciide0: secondary channel interrupting at irq 15
cd0(pciide0:1:0): using PIO mode 4, Ultra-DMA mode 2 (Ultra/33) (using DMA data transfers)
wd1(pciide0:1:1): using PIO mode 4, Ultra-DMA mode 5 (Ultra/100) (using DMA data transfers)
vendor 0x1039 product 0x7013 (modem communications, revision 0xa0) at pci0 dev 2 function 6 not configured
vendor 0x1039 product 0x7012 (audio multimedia, revision 0xa0) at pci0 dev 2 function 7 not configured
sip0 at pci0 dev 3 function 0: SiS 900 10/100 Ethernet, rev 0x90
sip0: interrupting at irq 11
sip0: Ethernet address 00:07:95:a8:a6:45
ukphy0 at sip0 phy 1: Generic IEEE 802.3u media interface
ukphy0: OUI 0x000004, model 0x0020, rev. 1
ukphy0: 10baseT, 10baseT-FDX, 100baseTX, 100baseTX-FDX, auto
pciide1 at pci0 dev 9 function 0: Promise Ultra100/ATA Bus Master IDE Accelerator (rev. 0x02)
pciide1: bus-master DMA support present
pciide1: primary channel configured to native-PCI mode
pciide1: using irq 5 for native-PCI interrupt
wd2 at pciide1 channel 0 drive 0: <QUANTUM FIREBALLlct08 08>
wd2: drive supports 16-sector PIO transfers, LBA addressing
wd2: 8063 MB, 16383 cyl, 16 head, 63 sec, 512 bytes/sect x 16514064 sectors
wd2: 32-bit data port
wd2: drive supports PIO mode 4, DMA mode 2, Ultra-DMA mode 4 (Ultra/66)
wd2(pciide1:0:0): using PIO mode 4, Ultra-DMA mode 4 (Ultra/66) (using DMA data transfers)
pciide1: secondary channel configured to native-PCI mode
wd3 at pciide1 channel 1 drive 0: <Maxtor 6Y080P0>
wd3: drive supports 16-sector PIO transfers, LBA addressing
wd3: 78167 MB, 16383 cyl, 16 head, 63 sec, 512 bytes/sect x 160086528 sectors
wd3: 32-bit data port
wd3: drive supports PIO mode 4, DMA mode 2, Ultra-DMA mode 6 (Ultra/133)
wd3(pciide1:1:0): using PIO mode 4, Ultra-DMA mode 5 (Ultra/100) (using DMA data transfers)
vendor 0x104c product 0x8400 (miscellaneous network) at pci0 dev 11 function 0 not configured
ex0 at pci0 dev 13 function 0: 3Com 3c905C-TX 10/100 Ethernet with mngmt (rev. 0x74)
ex0: interrupting at irq 11
ex0: MAC address 00:50:da:b4:ac:4b
bmtphy0 at ex0 phy 24: Broadcom 3c905C internal PHY, rev. 6
bmtphy0: 10baseT, 10baseT-FDX, 100baseTX, 100baseTX-FDX, auto
bktr0 at pci0 dev 15 function 0
bktr0: interrupting at irq 11
bktr0: Hauppauge Model 38101 B410
bktr0: Hauppauge WinCast/TV, Philips NTSC tuner.
vendor 0x109e product 0x0878 (miscellaneous multimedia, revision 0x11) at pci0 dev 15 function 1 not configured
emuxki0 at pci0 dev 17 function 0: vendor 0x1102 product 0x0002 (audio multimedia, revision 0x05)
emuxki0: interrupting at irq 5
emuxki0: SigmaTel STAC9721/23 codec; 18 bit DAC, 18 bit ADC, SigmaTel 3D
audio0 at emuxki0: full duplex, mmap, independent
vendor 0x1102 product 0x7002 (miscellaneous input, revision 0x05) at pci0 dev 17 function 1 not configured
isa0 at pcib0
com0 at isa0 port 0x3f8-0x3ff irq 4: ns16550a, working fifo
com1 at isa0 port 0x2f8-0x2ff irq 3: ns16550a, working fifo
pckbc0 at isa0 port 0x60-0x64
pckbd0 at pckbc0 (kbd slot)
pckbc0: using irq 1 for kbd slot
wskbd0 at pckbd0: console keyboard, using wsdisplay0
pms0 at pckbc0 (aux slot)
pckbc0: using irq 12 for aux slot
wsmouse0 at pms0 mux 0
lpt0 at isa0 port 0x378-0x37b irq 7
pcppi0 at isa0 port 0x61
midi0 at pcppi0: PC speaker
spkr0 at pcppi0
sysbeep0 at pcppi0
isapnp0 at isa0 port 0x279: ISA Plug 'n Play device support
npx0 at isa0 port 0xf0-0xff: using exception 16
fdc0 at isa0 port 0x3f0-0x3f7 irq 6 drq 2
fd0 at fdc0 drive 0: 1.44MB, 80 cyl, 2 head, 18 sec
isapnp0: no ISA Plug 'n Play devices found
biomask ef65 netmask ef65 ttymask ffe7
pciide1:0: bogus intr
Kernelized RAIDframe activated
IPsec: Initialized Security Association Processing.
RAID autoconfigure
Configuring raid0:
RAIDFRAME: protectedSectors is 64
RAIDFRAME: Configure (RAID Level 1): total number of sectors is 15227776 (7435 MB)
RAIDFRAME(RAID Level 1): Using 6 floating recon bufs with no head sep limit
RAID autoconfigure
Configuring raid1:
RAIDFRAME: Configure (RAID Level 1): total number of sectors is 160086464 (78167 MB)
RAIDFRAME(RAID Level 1): Using 6 floating recon bufs with no head sep limit
boot device: raid0
root on raid0a dumps on raid0b
root file system type: ffs
raid0: Device already configured!
raid1: Device already configured!
wsdisplay0: screen 1 added (80x25, vt100 emulation)
wsdisplay0: screen 2 added (80x25, vt100 emulation)
wsdisplay0: screen 3 added (80x25, vt100 emulation)
wsdisplay0: screen 4 added (80x25, vt100 emulation)
IP Filter: v3.4.29 initialized. Default = pass all, Logging = enabled
As you can see, at pciide0 I have a Silicon Integrated System 735 IDE controller with QUANTUM FIREBALL, Sony CD-RW, and Maxtor drives connected. At pciide1 I have a Promise Ultra100/ATA Bus Master IDE Accelerator with QUANTUM FIREBALL and Maxtor drives connected.
My RAID configuration has the two QUANTUM drives mirrored and the two Maxtor drives mirrored.
>How-To-Repeat:
Begin with said hardware configuration.
Download the latest -current sources via anoncvs:
cvs checkout src
Build the tools and GENERIC kernel:
./build.sh tools kernel=GENERIC
Install the new kernel:
cp sys/arch/i386/compile/obj.i386/GENERIC/netbsd /netbsd
Reboot into the new kernel:
reboot
>Fix:
>Release-Note:
>Audit-Trail:
From: Manuel Bouyer <bouyer@antioche.eu.org>
To: m.hertrick@neovera.com
Cc: gnats-bugs@gnats.NetBSD.org, netbsd-bugs@netbsd.org
Subject: Re: kern/23529: pciide0:0 and 0:1 bogus intr messages repeating forever on kernel load
Date: Sat, 22 Nov 2003 14:26:06 +0100
On Sat, Nov 22, 2003 at 12:23:03AM +0000, m.hertrick@neovera.com wrote:
> >Description:
> Boot of latest -current kernel goes into endless loop just before it would have printed "Kernelized RAIDframe activated". The kernel prints "pciide0:0 bogus intr" and "pciide0:1 bogus intr" indefinitely.
Is it really pciide0 ? With current your IDE controllers should show up as
siside0 and pdcide0.
--
Manuel Bouyer <bouyer@antioche.eu.org>
NetBSD: 24 ans d'experience feront toujours la difference
--
From: Michael Hertrick <m.hertrick@neovera.com>
To: Manuel Bouyer <bouyer@antioche.eu.org>
Cc: gnats-bugs@gnats.NetBSD.org, netbsd-bugs@netbsd.org
Subject: Re: kern/23529: pdcide0:0 and 0:1 bogus intr messages repeating forever
on kernel load
Date: Sat, 22 Nov 2003 13:55:47 -0500
Manuel Bouyer wrote:
>On Sat, Nov 22, 2003 at 12:23:03AM +0000, m.hertrick@neovera.com wrote:
>
>
>>>Description:
>>>
>>>
>>Boot of latest -current kernel goes into endless loop just before it would have printed "Kernelized RAIDframe activated". The kernel prints "pciide0:0 bogus intr" and "pciide0:1 bogus intr" indefinitely.
>>
>>
>
>Is it really pciide0 ? With current your IDE controllers should show up as
>siside0 and pdcide0.
>
>
>
Pardon... 10 days ago when this was happening to me it said pciide, I'm
sure. Probably because of that I misread... it is in fact pdcide0.
Sorry for the confusion.
~Mike.
From: Manuel Bouyer <bouyer@antioche.eu.org>
To: Michael Hertrick <m.hertrick@neovera.com>
Cc: gnats-bugs@gnats.NetBSD.org, netbsd-bugs@netbsd.org
Subject: Re: kern/23529: pdcide0:0 and 0:1 bogus intr messages repeating forever on kernel load
Date: Sat, 22 Nov 2003 20:12:28 +0100
On Sat, Nov 22, 2003 at 01:55:47PM -0500, Michael Hertrick wrote:
> Pardon... 10 days ago when this was happening to me it said pciide, I'm
> sure. Probably because of that I misread... it is in fact pdcide0.
10 days ago ...
Did you try with a recent current kernel ?
pdcide.c rev 1.7, commited 2003/11/15 16:40:46 may fix the problem.
--
Manuel Bouyer <bouyer@antioche.eu.org>
NetBSD: 24 ans d'experience feront toujours la difference
--
From: Michael Hertrick <m.hertrick@neovera.com>
To: Manuel Bouyer <bouyer@antioche.eu.org>
Cc: gnats-bugs@gnats.NetBSD.org, netbsd-bugs@netbsd.org
Subject: Re: kern/23529: pdcide0:0 and 0:1 bogus intr messages repeating forever
on kernel load
Date: Sat, 22 Nov 2003 14:25:11 -0500
Manuel Bouyer wrote:
>On Sat, Nov 22, 2003 at 01:55:47PM -0500, Michael Hertrick wrote:
>
>
>>Pardon... 10 days ago when this was happening to me it said pciide, I'm
>>sure. Probably because of that I misread... it is in fact pdcide0.
>>
>>
>
>10 days ago ...
>Did you try with a recent current kernel ?
>pdcide.c rev 1.7, commited 2003/11/15 16:40:46 may fix the problem.
>
>
>
Just before I submitted the PR I tried again with the latest sources as
of 11/21, downloaded via CVS.
1.7 2003/11/15 is the version I had last night when I tried.
/* $NetBSD: pdcide.c,v 1.7 2003/11/15 16:40:46 bouyer Exp $ */
From: Michael Hertrick <m.hertrick@neovera.com>
To: Manuel Bouyer <bouyer@antioche.eu.org>
Cc: gnats-bugs@NetBSD.org, netbsd-bugs@NetBSD.org
Subject: Re: kern/23529: the attached message may be related.
Date: Sat, 22 Nov 2003 14:47:53 -0500
This is a multi-part message in MIME format.
--------------060107010901080001040406
Content-Type: text/plain; charset=us-ascii; format=flowed
Content-Transfer-Encoding: 7bit
see attached message.
nishio@hh.iij4u.or.jp says he has a QUANTUM fireball, too... and his
workaround was:
``do NOT attach "QUANTUM FIREBALL" to
Ultra100 or Ultra66''
Unfortunately, this workaround is no good for me. I need the QUANTUM fireball drives attached to each of my controllers.
Nevertheless, I thought this information may help diagnose the problem.
~Mike.
--------------060107010901080001040406
Content-Type: message/rfc822;
name="Re: i386 - Cannot boot from wd(Promise Ultra100) with infinite ``bogus inter''msg"
Content-Transfer-Encoding: 7bit
Content-Disposition: inline;
filename="Re: i386 - Cannot boot from wd(Promise Ultra100) with infinite ``bogus inter''msg"
Return-Path: <current-users-owner-m.hertrick=neovera.com@NetBSD.org>
Received: from mail.netbsd.org (mail.isc.netbsd.org [204.152.185.212])
by lsh131.siteprotect.com (8.11.6/8.11.6) with SMTP id h9MIv9d27402
for <m.hertrick@neovera.com>; Wed, 22 Oct 2003 13:57:09 -0500
Received: (qmail 3322 invoked by uid 605); 22 Oct 2003 18:56:35 -0000
Delivered-To: current-users@netbsd.org
Received: (qmail 3315 invoked from network); 22 Oct 2003 18:56:34 -0000
Received: from bouyer.net1.nerim.net (HELO chassiron.antioche.eu.org) (62.212.96.44)
by mail.netbsd.org with SMTP; 22 Oct 2003 18:56:34 -0000
Received: from rochebonne.antioche.eu.org (rochebonne [2001:7a8:242c:0:240:95ff:fe45:4702])
by chassiron.antioche.eu.org (8.11.6p3/8.11.6) with ESMTP id h9MIuVe17457;
Wed, 22 Oct 2003 20:56:31 +0200 (MEST)
Received: (from bouyer@localhost)
by rochebonne.antioche.eu.org (8.12.9p1/8.12.8/Submit) id h9MIuVdZ001512;
Wed, 22 Oct 2003 20:56:31 +0200 (MEST)
Date: Wed, 22 Oct 2003 20:56:31 +0200
From: Manuel Bouyer <bouyer@antioche.eu.org>
To: NISHIO Yasuhiro <nishio@hh.iij4u.or.jp>
Cc: current-users@NetBSD.org
Subject: Re: i386 - Cannot boot from wd(Promise Ultra100) with infinite ``bogus inter'' msg
Message-ID: <20031022185631.GA857@antioche.eu.org>
References: <20031016211744.GA327@antioche.eu.org> <20031017.122734.730563953.nishio@cerberus.yhn.jp> <20031019143258.GA5981@antioche.eu.org> <20031020.075611.730552568.nishio@cerberus.yhn.jp>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20031020.075611.730552568.nishio@cerberus.yhn.jp>
User-Agent: Mutt/1.4i
Sender: current-users-owner@NetBSD.org
Precedence: list
On Mon, Oct 20, 2003 at 07:56:11AM +0900, NISHIO Yasuhiro wrote:
>
> I tried with U66, and I got following message.
>
> :
> fd0 at fdc0 drive 0: 1.44MB, 80 cyl, 2 head, 18 sec
> Kernelized RAIDframe activated
> pdcide0:0: bogus intr (reg 0x1131520)
> pdcide0:0: bogus intr (reg 0x1131520)
> pdcide0:0: bogus intr (reg 0x1131520)
> :
>
> But I found workaround. That is ``do NOT attach "QUANTUM FIREBALL" to
> Ultra100 or Ultra66''. I think my QUANTUM FIREBALL has
> hardware/firmware bug.
Possibly, it asserts its interrupt line while reset is active.
Now that the probe is done with interrupts enabled, this fails.
>
> Here is the summary;
> --------------------
>
> Kernel: NetBSD-current 2003-09-19 21:40:00 UTC or later
>
> - The "bogus intr" message appears only when the "QUANTUM
> <FIREBALLlct15 30>" attached to Ultra100. (reappearance 100%)
>
> There is no effect if
> - remove Other(cbb, fxp, wm) PCI Card,
> - replace PCI card slot,
> - use new UDMA100 IDE cable.
> - disable/enable on board IDE,
> - disable/enable VGA IRQ,
> - disable/enable on board USB,
> - etc.
>
> - pdcide driver work well with
> - SEAGATE <ST340810ACE> on U100 Primary master
> Western Digital <WDC WD136AA> on U100 Secondary master
> - WDC on Primary, ST on Secondary
>
>
> - U66(PDC20262) & FIREBALL dose not work.
> ``pdcide0:0: bogus intr (reg 0x1131520)''
>
> only ``pdcide0:0:...'' messages, ``pdcide0:0:..'' and
> ``pdcide0:1:...'' when I used U100.
>
> ==> FIREBALL has hardware bug?
Maybe, but not handled the same way by different controllers.
--
Manuel Bouyer <bouyer@antioche.eu.org>
NetBSD: 24 ans d'experience feront toujours la difference
--
--------------060107010901080001040406--
From: Michael Hertrick <m.hertrick@neovera.com>
To: Manuel Bouyer <bouyer@antioche.eu.org>
Cc: gnats-bugs@NetBSD.org, netbsd-bugs@NetBSD.org
Subject: Re: kern/23529 pdcide0:0 and 0:1 bogus intr messages repeating forever
on kernel load
Date: Sun, 23 Nov 2003 00:47:49 -0500
I attempted the following in kernel config file:
wd* at atabus? drive ? flags 0x0fac
cd* at atapibus? drive ? flags 0x0fac
To force PIO mode 4 and DMA mode 2 (disabling UltraDMA). This had no
effect... still bogus intr problem.
--------------
Then I tried disabling DMA altogether, with "flags 0x0ffc" on wd and
cd. This also had no effect... still bogus intr problem.
--------------
By breaking into gdb (ctrl-alt-esc) as soon as the problem starts and
entering "dmesg", I could see it never seems to get to attaching wd*. I
suppose that's what it's trying to do when the "bogus intr" occurs.
What can I do to help troubleshoot this problem?
~Mike.
From: Manuel Bouyer <bouyer@antioche.eu.org>
To: Michael Hertrick <m.hertrick@neovera.com>
Cc: gnats-bugs@NetBSD.org, netbsd-bugs@NetBSD.org
Subject: Re: kern/23529 pdcide0:0 and 0:1 bogus intr messages repeating forever on kernel load
Date: Sun, 23 Nov 2003 22:14:41 +0100
On Sun, Nov 23, 2003 at 12:47:49AM -0500, Michael Hertrick wrote:
> I attempted the following in kernel config file:
>
> wd* at atabus? drive ? flags 0x0fac
>
> cd* at atapibus? drive ? flags 0x0fac
>
> To force PIO mode 4 and DMA mode 2 (disabling UltraDMA). This had no
> effect... still bogus intr problem.
>
> --------------
>
> Then I tried disabling DMA altogether, with "flags 0x0ffc" on wd and
> cd. This also had no effect... still bogus intr problem.
>
> --------------
>
> By breaking into gdb (ctrl-alt-esc) as soon as the problem starts and
> entering "dmesg", I could see it never seems to get to attaching wd*. I
> suppose that's what it's trying to do when the "bogus intr" occurs.
>
> What can I do to help troubleshoot this problem?
I think this happens while the channel is being reset.
First I'll see if I can find a quantun drive at work, as I have
both U100 and U66 promise IDE controllers to test. What is the exact model
you have ?
--
Manuel Bouyer <bouyer@antioche.eu.org>
NetBSD: 24 ans d'experience feront toujours la difference
--
From: Michael Hertrick <m.hertrick@neovera.com>
To: Manuel Bouyer <bouyer@antioche.eu.org>
Cc: gnats-bugs@NetBSD.org, netbsd-bugs@NetBSD.org
Subject: Re: kern/23529 pdcide0:0 and 0:1 bogus intr messages repeating forever
on kernel load
Date: Sun, 23 Nov 2003 17:13:34 -0500
Manuel Bouyer wrote:
>I think this happens while the channel is being reset.
>First I'll see if I can find a quantun drive at work, as I have
>both U100 and U66 promise IDE controllers to test. What is the exact model
>you have ?
>
>
>
The drives are:
QUANTUM FIREBALLct08 08, 8063MB -- it's old, at least 3 years, if not 4
or more.
Maxtor 6Y080P0, 78167 MB -- this one's new, <1 year old.
If the information above does not suffice, then I'll pull my case apart
later and get more specific.
And the card, just in case, is a Maxtor UltraATA/100 PCI Adapter Card.
The model number from a sticker on the box is K01PCAT100
From: Manuel Bouyer <bouyer@antioche.eu.org>
To: Michael Hertrick <m.hertrick@neovera.com>
Cc: gnats-bugs@NetBSD.org, netbsd-bugs@NetBSD.org
Subject: Re: kern/23529 pdcide0:0 and 0:1 bogus intr messages repeating forever on kernel load
Date: Sun, 23 Nov 2003 23:22:41 +0100
On Sun, Nov 23, 2003 at 05:13:34PM -0500, Michael Hertrick wrote:
> The drives are:
> QUANTUM FIREBALLct08 08, 8063MB -- it's old, at least 3 years, if not 4
> or more.
> Maxtor 6Y080P0, 78167 MB -- this one's new, <1 year old.
You mean, both drives cause priblems on the promise, or only the QUANTUM ?
--
Manuel Bouyer <bouyer@antioche.eu.org>
NetBSD: 24 ans d'experience feront toujours la difference
--
From: Michael Hertrick <m.hertrick@neovera.com>
To: Manuel Bouyer <bouyer@antioche.eu.org>
Cc: gnats-bugs@NetBSD.org, netbsd-bugs@NetBSD.org
Subject: Re: kern/23529 pdcide0:0 and 0:1 bogus intr messages repeating forever
on kernel load
Date: Sun, 23 Nov 2003 17:46:38 -0500
The messages for "bogus intr" say pdcide0:0 and pdcide0:1, and I have
the Quantum on 0:0 and the Maxtor on 0:1. So, I'm tempted to say it's
both...but what do I know?
I honestly don't know where the problem lies. I've been hesitant to
mess around with the hardware setup for numerous reasons. I'll tell ya
what, though. I'm going to take the machine down after I send this
e-mail, take my mirrored drives out and see if I can determine whether
it's just the Quantum or if it's both.
Hang on and I'll get back to you within thirty minutes...
~Mike.
Manuel Bouyer wrote:
>On Sun, Nov 23, 2003 at 05:13:34PM -0500, Michael Hertrick wrote:
>
>
>>The drives are:
>>QUANTUM FIREBALLct08 08, 8063MB -- it's old, at least 3 years, if not 4
>>or more.
>>Maxtor 6Y080P0, 78167 MB -- this one's new, <1 year old.
>>
>>
>
>You mean, both drives cause priblems on the promise, or only the QUANTUM ?
>
>
>
From: Michael Hertrick <m.hertrick@neovera.com>
To: Manuel Bouyer <bouyer@antioche.eu.org>
Cc: gnats-bugs@NetBSD.org, netbsd-bugs@NetBSD.org
Subject: Re: kern/23529 pdcide0:0 and 0:1 bogus intr messages repeating forever
on kernel load
Date: Sun, 23 Nov 2003 17:54:04 -0500
Alright, I disconnected the Quantum drive from the Promise controller
and didn't get the "bogus intr" messages with -current kernel.
So, it's safe to say now that the problem only occurs with the Quantum
Fireball on Promise controller.
~Mike.
Michael Hertrick wrote:
> The messages for "bogus intr" say pdcide0:0 and pdcide0:1, and I have
> the Quantum on 0:0 and the Maxtor on 0:1. So, I'm tempted to say it's
> both...but what do I know?
>
> I honestly don't know where the problem lies. I've been hesitant to
> mess around with the hardware setup for numerous reasons. I'll tell
> ya what, though. I'm going to take the machine down after I send this
> e-mail, take my mirrored drives out and see if I can determine whether
> it's just the Quantum or if it's both.
>
> Hang on and I'll get back to you within thirty minutes...
>
> ~Mike.
>
> Manuel Bouyer wrote:
>
>> On Sun, Nov 23, 2003 at 05:13:34PM -0500, Michael Hertrick wrote:
>>
>>
>>> The drives are:
>>> QUANTUM FIREBALLct08 08, 8063MB -- it's old, at least 3 years, if
>>> not 4 or more.
>>> Maxtor 6Y080P0, 78167 MB -- this one's new, <1 year old.
>>>
>>
>>
>> You mean, both drives cause priblems on the promise, or only the
>> QUANTUM ?
>>
>>
>>
>
>
From: Manuel Bouyer <bouyer@antioche.eu.org>
To: Michael Hertrick <m.hertrick@neovera.com>
Cc: gnats-bugs@NetBSD.org, netbsd-bugs@NetBSD.org
Subject: Re: kern/23529 pdcide0:0 and 0:1 bogus intr messages repeating forever on kernel load
Date: Mon, 24 Nov 2003 00:24:23 +0100
On Sun, Nov 23, 2003 at 05:54:04PM -0500, Michael Hertrick wrote:
> Alright, I disconnected the Quantum drive from the Promise controller
> and didn't get the "bogus intr" messages with -current kernel.
>
> So, it's safe to say now that the problem only occurs with the Quantum
> Fireball on Promise controller.
OK, thanks. I'll see if I can find a similar drive at work tomorow.
Time to go to bed now :)
--
Manuel Bouyer <bouyer@antioche.eu.org>
NetBSD: 24 ans d'experience feront toujours la difference
--
From: Manuel Bouyer <bouyer@antioche.eu.org>
To: Michael Hertrick <m.hertrick@neovera.com>
Cc: gnats-bugs@NetBSD.org
Subject: Re: kern/23529 pdcide0:0 and 0:1 bogus intr messages repeating forever on kernel load
Date: Mon, 24 Nov 2003 20:36:03 +0100
--17pEHd4RhPHOinZp
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
On Sun, Nov 23, 2003 at 05:13:34PM -0500, Michael Hertrick wrote:
> The drives are:
> QUANTUM FIREBALLct08 08, 8063MB -- it's old, at least 3 years, if not 4
> or more.
Well, I tried with
wd5 at atabus7 drive 0: <QUANTUM FIREBALL CR8.4A>
wd5: drive supports 16-sector PIO transfers, LBA addressing
wd5: 8063 MB, 16383 cyl, 16 head, 63 sec, 512 bytes/sect x 16514064 sectors
but couldn't reproduce the problem, on all of the promise IDE I have:
Ultra/66, 100, 100Tx2, and 133Tx2.
Can you try the attached patch ? I'd like to know if it has any effect at all.
--
Manuel Bouyer <bouyer@antioche.eu.org>
NetBSD: 24 ans d'experience feront toujours la difference
--
--17pEHd4RhPHOinZp
Content-Type: text/plain; charset=us-ascii
Content-Disposition: attachment; filename=diff
Index: wdc.c
===================================================================
RCS file: /cvsroot/src/sys/dev/ic/wdc.c,v
retrieving revision 1.155
diff -u -r1.155 wdc.c
--- wdc.c 2003/11/17 20:01:35 1.155
+++ wdc.c 2003/11/24 19:35:14
@@ -614,6 +614,7 @@
u_int8_t st0, st1, sc, sn, cl, ch;
u_int8_t ret_value = 0x03;
u_int8_t drive;
+ int s;
/*
* Sanity check to see if the wdc channel responds at all.
@@ -710,6 +711,8 @@
return 0;
}
+ s=splbio();
+
if (chp->wdc && (chp->wdc->cap & WDC_CAPABILITY_SELECT))
chp->wdc->select(chp,0);
/* assert SRST, wait for reset to complete */
@@ -727,6 +730,10 @@
WDCDEBUG_PRINT(("%s:%d: after reset, ret_value=0x%d\n",
chp->wdc ? chp->wdc->sc_dev.dv_xname : "wdcprobe", chp->channel,
ret_value), DEBUG_PROBE);
+ /* ACK interrupt in case there is one pending left (Promise ATA100) */
+ if (chp->wdc->cap & WDC_CAPABILITY_IRQACK)
+ chp->wdc->irqack(chp);
+ splx(s);
/* if reset failed, there's nothing here */
if (ret_value == 0)
@@ -1109,9 +1116,12 @@
int poll;
{
int drv_mask1, drv_mask2;
+ int s;
if (chp->wdc->cap & WDC_CAPABILITY_SELECT)
chp->wdc->select(chp,0);
+ if (poll != RESET_SLEEP)
+ s = splbio();
bus_space_write_1(chp->cmd_iot, chp->cmd_ioh, wd_sdh,
WDSD_IBM); /* master */
delay(10); /* 400ns delay */
@@ -1127,6 +1137,11 @@
drv_mask1 |= (chp->ch_drive[1].drive_flags & DRIVE) ? 0x02:0x00;
drv_mask2 = __wdcwait_reset(chp, drv_mask1,
(poll == RESET_SLEEP) ? 0 : 1);
+ if (poll != RESET_SLEEP) {
+ if (chp->wdc->cap & WDC_CAPABILITY_IRQACK)
+ chp->wdc->irqack(chp);
+ splx(s);
+ }
if (drv_mask2 != drv_mask1) {
printf("%s channel %d: reset failed for",
chp->wdc->sc_dev.dv_xname, chp->channel);
--17pEHd4RhPHOinZp--
From: Michael Hertrick <m.hertrick@neovera.com>
To: Manuel Bouyer <bouyer@antioche.eu.org>
Cc: gnats-bugs@NetBSD.org
Subject: Re: kern/23529 pdcide0:0 and 0:1 bogus intr messages repeating forever
on kernel load
Date: Mon, 24 Nov 2003 16:34:51 -0500
Manuel Bouyer wrote:
>On Sun, Nov 23, 2003 at 05:13:34PM -0500, Michael Hertrick wrote:
>
>
>>The drives are:
>>QUANTUM FIREBALLct08 08, 8063MB -- it's old, at least 3 years, if not 4
>>or more.
>>
>>
>
>Well, I tried with
>wd5 at atabus7 drive 0: <QUANTUM FIREBALL CR8.4A>
>wd5: drive supports 16-sector PIO transfers, LBA addressing
>wd5: 8063 MB, 16383 cyl, 16 head, 63 sec, 512 bytes/sect x 16514064 sectors
>
>but couldn't reproduce the problem, on all of the promise IDE I have:
>Ultra/66, 100, 100Tx2, and 133Tx2.
>
The CR8.4A must be the firmware revision. I don't suppose firmware is
upgradable on these drives....... I'll check and see, though.
Did you even get a single "bogus intr" message?
>Can you try the attached patch ? I'd like to know if it has any effect at all.
>
>
FYI: The compiler complained "s may be used uninitialized in this
function" at line 1119, so I changed that "int s" to "int s = 0" and it
compiled without warning. Anyway.......
Yes, it had an effect. Aside from the lone comment in the patch, I
don't know what it does different, but it's much better...... the kernel
finished booting this time, but I still got 3 "pdcide0:0: bogus intr"
messages. I don't perceive any danger in continuing to run -current at
this point. Should I be worried about data integrity? Or even
performance issues?
Are you going to look into the erroneous "bogus intr" messages any
further, or are they really nothing to worry about?
Thank you very much.
~Mike.
From: Manuel Bouyer <bouyer@antioche.eu.org>
To: Michael Hertrick <m.hertrick@neovera.com>
Cc: gnats-bugs@NetBSD.org
Subject: Re: kern/23529 pdcide0:0 and 0:1 bogus intr messages repeating forever on kernel load
Date: Mon, 24 Nov 2003 23:02:17 +0100
--oyUTqETQ0mS9luUI
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
On Mon, Nov 24, 2003 at 04:34:51PM -0500, Michael Hertrick wrote:
> Manuel Bouyer wrote:
>
> >On Sun, Nov 23, 2003 at 05:13:34PM -0500, Michael Hertrick wrote:
> >
> >
> >>The drives are:
> >>QUANTUM FIREBALLct08 08, 8063MB -- it's old, at least 3 years, if not 4
> >>or more.
> >>
> >>
> >
> >Well, I tried with
> >wd5 at atabus7 drive 0: <QUANTUM FIREBALL CR8.4A>
> >wd5: drive supports 16-sector PIO transfers, LBA addressing
> >wd5: 8063 MB, 16383 cyl, 16 head, 63 sec, 512 bytes/sect x 16514064 sectors
> >
> >but couldn't reproduce the problem, on all of the promise IDE I have:
> >Ultra/66, 100, 100Tx2, and 133Tx2.
> >
>
> The CR8.4A must be the firmware revision. I don't suppose firmware is
> upgradable on these drives....... I'll check and see, though.
>
> Did you even get a single "bogus intr" message?
Not since I commited pdcide.c 1.7.
>
> >Can you try the attached patch ? I'd like to know if it has any effect at
> >all.
> >
> >
> FYI: The compiler complained "s may be used uninitialized in this
> function" at line 1119, so I changed that "int s" to "int s = 0" and it
> compiled without warning. Anyway.......
This warning is bogus, the compiler should know that poll won't change
between lines 1123 and 1140.
anyway initialising s doesn't hurt here.
>
> Yes, it had an effect. Aside from the lone comment in the patch, I
> don't know what it does different, but it's much better...... the kernel
> finished booting this time, but I still got 3 "pdcide0:0: bogus intr"
> messages.
Did it probe the drives ?
> I don't perceive any danger in continuing to run -current at
> this point. Should I be worried about data integrity? Or even
> performance issues?
No it shouldn't. The issue really only affects the channel reset.
>
> Are you going to look into the erroneous "bogus intr" messages any
> further, or are they really nothing to worry about?
I'll look at it eventually. But as it only affects a particular combination
of controller/drive, and is purely cosmetic, it's low priority :)
But first, I'd like to try to narrow down the period which needs interrupts
blocked. A reset can take up to 31s, I don't like the idea of BIO interrupts
being blocked for 31s.
The attached patch (against my previous patch, not unmodified wdc.c) moves the
splx() to a more sensible place. Can you check if it works ?
If it doesn't please try to bump the delay(10) before the 2 splx(s) to
delay(2000)
--
Manuel Bouyer <bouyer@antioche.eu.org>
NetBSD: 24 ans d'experience feront toujours la difference
--
--oyUTqETQ0mS9luUI
Content-Type: text/plain; charset=us-ascii
Content-Disposition: attachment; filename=diff2
--- wdc.c.orig 2003-11-24 22:58:43.000000000 +0100
+++ wdc.c 2003-11-24 22:59:14.000000000 +0100
@@ -725,15 +725,15 @@
(void) bus_space_read_1(chp->cmd_iot, chp->cmd_ioh, wd_error);
bus_space_write_1(chp->ctl_iot, chp->ctl_ioh, wd_aux_ctlr, WDCTL_4BIT);
delay(10); /* 400ns delay */
+ /* ACK interrupt in case there is one pending left (Promise ATA100) */
+ if (chp->wdc->cap & WDC_CAPABILITY_IRQACK)
+ chp->wdc->irqack(chp);
+ splx(s);
ret_value = __wdcwait_reset(chp, ret_value, poll);
WDCDEBUG_PRINT(("%s:%d: after reset, ret_value=0x%d\n",
chp->wdc ? chp->wdc->sc_dev.dv_xname : "wdcprobe", chp->channel,
ret_value), DEBUG_PROBE);
- /* ACK interrupt in case there is one pending left (Promise ATA100) */
- if (chp->wdc->cap & WDC_CAPABILITY_IRQACK)
- chp->wdc->irqack(chp);
- splx(s);
/* if reset failed, there's nothing here */
if (ret_value == 0)
@@ -1116,7 +1116,7 @@
int poll;
{
int drv_mask1, drv_mask2;
- int s;
+ int s = 0;
if (chp->wdc->cap & WDC_CAPABILITY_SELECT)
chp->wdc->select(chp,0);
@@ -1132,16 +1132,16 @@
bus_space_write_1(chp->ctl_iot, chp->ctl_ioh, wd_aux_ctlr,
WDCTL_4BIT | WDCTL_IDS);
delay(10); /* 400ns delay */
-
- drv_mask1 = (chp->ch_drive[0].drive_flags & DRIVE) ? 0x01:0x00;
- drv_mask1 |= (chp->ch_drive[1].drive_flags & DRIVE) ? 0x02:0x00;
- drv_mask2 = __wdcwait_reset(chp, drv_mask1,
- (poll == RESET_SLEEP) ? 0 : 1);
if (poll != RESET_SLEEP) {
if (chp->wdc->cap & WDC_CAPABILITY_IRQACK)
chp->wdc->irqack(chp);
splx(s);
}
+
+ drv_mask1 = (chp->ch_drive[0].drive_flags & DRIVE) ? 0x01:0x00;
+ drv_mask1 |= (chp->ch_drive[1].drive_flags & DRIVE) ? 0x02:0x00;
+ drv_mask2 = __wdcwait_reset(chp, drv_mask1,
+ (poll == RESET_SLEEP) ? 0 : 1);
if (drv_mask2 != drv_mask1) {
printf("%s channel %d: reset failed for",
chp->wdc->sc_dev.dv_xname, chp->channel);
--oyUTqETQ0mS9luUI--
From: Michael Hertrick <m.hertrick@neovera.com>
To: Manuel Bouyer <bouyer@antioche.eu.org>
Cc: gnats-bugs@NetBSD.org
Subject: Re: kern/23529 pdcide0:0 and 0:1 bogus intr messages repeating forever
on kernel load
Date: Mon, 24 Nov 2003 17:33:49 -0500
Manuel Bouyer wrote:
>Did it probe the drives ?
>
>
wd2 at atabus2 drive 0: <QUANTUM FIREBALLlct08 08>
wd2: drive supports 16-sector PIO transfers, LBA addressing
wd2: 8063 MB, 16383 cyl, 16 head, 63 sec, 512 bytes/sect x 16514064 sectors
wd2: 32-bit data port
wd2: drive supports PIO mode 4, DMA mode 2, Ultra-DMA mode 4 (Ultra/66)
wd2(pdcide0:0:0): using PIO mode 4
wd3 at atabus3 drive 0: <Maxtor 6Y080P0>
wd3: drive supports 16-sector PIO transfers, LBA addressing
wd3: 78167 MB, 158816 cyl, 16 head, 63 sec, 512 bytes/sect x 160086528
sectors
wd3: 32-bit data port
wd3: drive supports PIO mode 4, DMA mode 2, Ultra-DMA mode 6 (Ultra/133)
wd3(pdcide0:1:0): using PIO mode 4
I forgot I still had the modes hardcoded in the kernel config file, but
based on the dmesg it did probe the drives. If it hadn't, it wouldn't
have been able to mount, right?
I will try the new patch shortly...
From: Manuel Bouyer <bouyer@antioche.eu.org>
To: Michael Hertrick <m.hertrick@neovera.com>
Cc: gnats-bugs@NetBSD.org
Subject: Re: kern/23529 pdcide0:0 and 0:1 bogus intr messages repeating forever on kernel load
Date: Mon, 24 Nov 2003 23:41:42 +0100
On Mon, Nov 24, 2003 at 05:33:49PM -0500, Michael Hertrick wrote:
> Manuel Bouyer wrote:
>
> >Did it probe the drives ?
> >
> >
>
> wd2 at atabus2 drive 0: <QUANTUM FIREBALLlct08 08>
> wd2: drive supports 16-sector PIO transfers, LBA addressing
> wd2: 8063 MB, 16383 cyl, 16 head, 63 sec, 512 bytes/sect x 16514064 sectors
> wd2: 32-bit data port
> wd2: drive supports PIO mode 4, DMA mode 2, Ultra-DMA mode 4 (Ultra/66)
> wd2(pdcide0:0:0): using PIO mode 4
> wd3 at atabus3 drive 0: <Maxtor 6Y080P0>
> wd3: drive supports 16-sector PIO transfers, LBA addressing
> wd3: 78167 MB, 158816 cyl, 16 head, 63 sec, 512 bytes/sect x 160086528
> sectors
> wd3: 32-bit data port
> wd3: drive supports PIO mode 4, DMA mode 2, Ultra-DMA mode 6 (Ultra/133)
> wd3(pdcide0:1:0): using PIO mode 4
>
>
>
> I forgot I still had the modes hardcoded in the kernel config file, but
You have them:
wd2(pdcide0:0:0): using PIO mode 4
wd3(pdcide0:1:0): using PIO mode 4
> based on the dmesg it did probe the drives. If it hadn't, it wouldn't
> have been able to mount, right?
Yes :)
--
Manuel Bouyer <bouyer@antioche.eu.org>
NetBSD: 24 ans d'experience feront toujours la difference
--
From: Michael Hertrick <m.hertrick@neovera.com>
To: Manuel Bouyer <bouyer@antioche.eu.org>
Cc: gnats-bugs@NetBSD.org
Subject: Re: kern/23529 pdcide0:0 and 0:1 bogus intr messages repeating forever
on kernel load
Date: Mon, 24 Nov 2003 19:13:06 -0500
Manuel Bouyer wrote:
>But first, I'd like to try to narrow down the period which needs interrupts
>blocked. A reset can take up to 31s, I don't like the idea of BIO interrupts
>being blocked for 31s.
>The attached patch (against my previous patch, not unmodified wdc.c) moves the
>splx() to a more sensible place. Can you check if it works ?
>
>If it doesn't please try to bump the delay(10) before the 2 splx(s) to
>delay(2000)
>
>
>
You're quite the beautician ;-). With the latest patch to wdc.c I don't
receive any more "bogus intr" messages at all. Was that expected? Does
it make sense? BTW, I did not change the delay.
Everything was detected, probed, set, reset, plotted, arranged,
allocated, built, loaded, read, written, and illuminated as it should
be...... I think.
If there're any more patches you'd like to experiment with, send them
along and I'll try them out.
~Mike.
From: Manuel Bouyer <bouyer@antioche.eu.org>
To: Michael Hertrick <m.hertrick@neovera.com>
Cc: gnats-bugs@NetBSD.org
Subject: Re: kern/23529 pdcide0:0 and 0:1 bogus intr messages repeating forever on kernel load
Date: Tue, 25 Nov 2003 22:00:41 +0100
On Mon, Nov 24, 2003 at 07:13:06PM -0500, Michael Hertrick wrote:
> You're quite the beautician ;-). With the latest patch to wdc.c I don't
> receive any more "bogus intr" messages at all. Was that expected?
No, but it's hard to predic behavior with such hardware that doens't behave
the expected way :)
> Does
> it make sense? BTW, I did not change the delay.
>
> Everything was detected, probed, set, reset, plotted, arranged,
> allocated, built, loaded, read, written, and illuminated as it should
> be...... I think.
Cool, I'll commit this patch then.
>
>
> If there're any more patches you'd like to experiment with, send them
> along and I'll try them out.
Well, now it works without blocking interrupts for too long. I'm
happy with it :)
--
Manuel Bouyer <bouyer@antioche.eu.org>
NetBSD: 24 ans d'experience feront toujours la difference
--
State-Changed-From-To: open->closed
State-Changed-By: bouyer
State-Changed-When: Tue Nov 25 21:05:26 UTC 2003
State-Changed-Why:
Fixed with wdc.c 1.156
>Unformatted:
(Contact us)
$NetBSD: query-full-pr,v 1.51 2026/08/10 02:28:17 riastradh Exp $
$NetBSD: gnats_config.sh,v 1.10 2026/05/13 22:00:09 riastradh Exp $
Copyright © 1994-2026
The NetBSD Foundation, Inc. ALL RIGHTS RESERVED.