System crashes with via82cxxx ide driver

System crashes with via82cxxx ide driver

[Pascal Schmidt]

Hello all!

I am seeing system crashes while using the via82cxxx driver on a VIA MVP3
motherboard. I can reproduce the crashes with

	e2fsck -f -p /dev/hda3 -C 0

where hda3 is a 3 GB ext2 filesystem. e2fsck is 1.23, but it also happens
with any other version between 1.19 and 1.23, so it's unlikely to be an
e2fsck problem.

I have tried it about 10 times now, and the machine always crashes at the
moment when e2fsck reaches 75.6% percent. The machine just powers off, no
chance to read syslog or to see whether there is a kernel oops or panic.

How do I know that this is because of the via82cxxx driver? Well, it
always happens with 2.4.x and this driver enabled. It never happens to me
with 2.2.19, which has an older driver for this kind of ide controller.
But when I apply the ide backport patch to 2.2.19, giving me the via82cxxx
driver on 2.2.19, the machine once again crashes at the same 75.6% of the
e2fsck run. When I disable the driver in the kernel config, e2fsck does
not crash the machine. So it's not the newer IDE stuff in general, but
specifically this driver.

My guess is that the newer via82cxxx driver in 2.4.x stresses the chipset
in some way that it doesn't really like.

The problem also strikes during other heavy disk activity, such as
installing a multi-megabyte RPM file. The Red Hat 7.1 installation in fact
crashed twice with the same symptoms before it got through on the third
try.

I can work around the problem by not using the via8cxxx driver, but I'd
rather like to have UDMA33 for my hard disk because of the CRC error
detection.

Let me know if I should supply more information.

lspci -v output for the motherboard chipset:

00:00.0 Host bridge: VIA Technologies, Inc. VT82C598 [Apollo MVP3] (rev
04)
        Flags: bus master, medium devsel, latency 16
        Memory at e0000000 (32-bit, prefetchable)
        Capabilities: <available only to root>

00:01.0 PCI bridge: VIA Technologies, Inc. VT82C598/694x [Apollo
MVP3/Pro133x AGP] (prog-if 00 [Normal decode])
        Flags: bus master, 66Mhz, medium devsel, latency 0
        Bus: primary=00, secondary=01, subordinate=01, sec-latency=0
        I/O behind bridge: 0000b000-0000bfff
        Memory behind bridge: e4000000-e5ffffff
        Prefetchable memory behind bridge: e6000000-e6ffffff

00:07.0 ISA bridge: VIA Technologies, Inc. VT82C586/A/B PCI-to-ISA [Apollo
VP] (rev 47)
        Subsystem: VIA Technologies, Inc. MVP3 ISA Bridge
        Flags: bus master, stepping, medium devsel, latency 0

00:07.1 IDE interface: VIA Technologies, Inc. Bus Master IDE (rev 06)
(prog-if 8a [Master SecP PriP])
        Flags: bus master, medium devsel, latency 64
        I/O ports at c000

00:07.3 Host bridge: VIA Technologies, Inc. VT82C586B ACPI (rev 10)
        Flags: medium devsel

-- 
Ciao, Pascal

-<[ pharao90@tzi.de, netmail 2:241/215.72, home http://cobol.cjb.net/) ]>-

Re: System crashes with via82cxxx ide driver

[Tim Moore]

> How do I know that this is because of the via82cxxx driver? Well, it
> always happens with 2.4.x and this driver enabled. It never happens to me
> with 2.2.19, which has an older driver for this kind of ide controller.
> But when I apply the ide backport patch to 2.2.19, giving me the via82cxxx
> driver on 2.2.19, the machine once again crashes at the same 75.6% of the
> e2fsck run. When I disable the driver in the kernel config, e2fsck does
> not crash the machine. So it's not the newer IDE stuff in general, but
> specifically this driver.

FWIW no problems with this setup:

2.2.20p9, ide.2.2.19.05042001.patch.bz2, e2fsck v1.19.

[tim@abit 2.2]# e2fsck -f -p /dev/hdc4 -C 0
/dev/hdc4: 30/1411488 files (16.7% non-contiguous), 1754402/2819407
blocks     

00:00.0 Host bridge: VIA Technologies, Inc.: Unknown device 0391 (rev
02)
00:01.0 PCI bridge: VIA Technologies, Inc.: Unknown device 8391
00:07.0 ISA bridge: VIA Technologies, Inc. VT82C686 [Apollo Super] (rev
22)
00:07.1 IDE interface: VIA Technologies, Inc. VT82C586 IDE [Apollo] (rev
10)
00:07.4 Host bridge: VIA Technologies, Inc. VT82C686 [Apollo Super ACPI]
(rev 30)
00:08.0 FireWire (IEEE 1394): Texas Instruments TSB12LV23 OHCI Compliant
IEEE-1394 Controller
00:09.0 Multimedia audio controller: Cirrus Logic CS 4614/22/24
[CrystalClear SoundFusion Audio Accelerator] (rev 01)
00:0f.0 Ethernet controller: Lite-On Communications Inc LNE100TX (rev
20)
01:00.0 VGA compatible controller: nVidia Corporation Riva TNT2 Model 64
(rev 11)

rgds,
tim


--

Re: System crashes with via82cxxx ide driver

[Pascal Schmidt]

On Thu, 30 Aug 2001, Tim Moore wrote:

> 2.2.20p9, ide.2.2.19.05042001.patch.bz2, e2fsck v1.19.
How did you get both the 2.2.20 pre-patch and the ide patch to apply? I
get a couple of rejects.

> 00:07.1 IDE interface: VIA Technologies, Inc. VT82C586 IDE [Apollo] (rev
> 10)
The IDE part of my chipset is only revision 06, so perhaps this is a bug
fixed in your revision.

-- 
Ciao, Pascal

-<[ pharao90@tzi.de, netmail 2:241/215.72, home http://cobol.cjb.net/) ]>-

Re: System crashes with via82cxxx ide driver

[Tim Moore]

> > 2.2.20p9, ide.2.2.19.05042001.patch.bz2, e2fsck v1.19.
> How did you get both the 2.2.20 pre-patch and the ide patch to apply? I
> get a couple of rejects.

1. Do the pre patch first.
2. The rejects didn't make any difference.  init/main.c.rej was SCSI
hd[m-t] addresses which were easy to drop in by hand,
drivers/block/ide.c.rej was comments & spacing, and the others were in
Configure.help or Makefile EXTRAVERSION conflicts.  These don't matter
in most cases where the ide patch makes sense.

> > 00:07.1 IDE interface: VIA Technologies, Inc. VT82C586 IDE [Apollo] (rev
> > 10)
> The IDE part of my chipset is only revision 06, so perhaps this is a bug
> fixed in your revision.

I've another machine that uses the same kernel+ide:

00:00.0 Host bridge: VIA Technologies, Inc. VT82C585VP [Apollo VP1/VPX]
(rev 10)
00:07.0 ISA bridge: VIA Technologies, Inc. VT82C586/A/B PCI-to-ISA
[Apollo VP] (rev 02)
00:07.1 IDE interface: VIA Technologies, Inc. VT82C586 IDE [Apollo] (rev
02)
00:08.0 VGA compatible unclassified device: 3DLabs Permedia II 2D+3D
(rev 01)
00:09.0 Ethernet controller: Lite-On Communications Inc LNE100TX (rev
20)
00:0a.0 Ethernet controller: Lite-On Communications Inc LNE100TX (rev
20)

rgds,
tim.
--

Re: System crashes with via82cxxx ide driver

[Pascal Schmidt]

On Thu, 30 Aug 2001, Tim Moore wrote:

> 2. The rejects didn't make any difference.  init/main.c.rej was SCSI
> hd[m-t] addresses which were easy to drop in by hand,
> drivers/block/ide.c.rej was comments & spacing, and the others were in
> Configure.help or Makefile EXTRAVERSION conflicts.  These don't matter
> in most cases where the ide patch makes sense.
Did that and tested. 2.2.20pre9 with ide-05042001 patch still crashes at
the same point with the same symptoms. I even have to pull out the power
chord for a few seconds before the system can be started again.

> I've another machine that uses the same kernel+ide:
> 00:07.1 IDE interface: VIA Technologies, Inc. VT82C586 IDE [Apollo] (rev
> 02)
So it works for you with revision 02 and 10, but doesn't work for me with
revision 06. Strange.

-- 
Ciao, Pascal

-<[ pharao90@tzi.de, netmail 2:241/215.72, home http://cobol.cjb.net/) ]>-