[2.4.21-pre7-ac1] IDE Warning when booting

[2.4.21-pre7-ac1] IDE Warning when booting

[Henning P. Schmiedehausen]

Hi,

I get these since many 2.4.21-pre<x>-ac<y> versions.

ide: Assuming 33MHz system bus speed for PIO modes; override with idebus=xx
PIIX3: IDE controller at PCI slot 00:07.1
PIIX3: chipset revision 0
PIIX3: not 100% native mode: will probe irqs later
    ide0: BM-DMA at 0xf000-0xf007, BIOS settings: hda:pio, hdb:pio
    ide1: BM-DMA at 0xf008-0xf00f, BIOS settings: hdc:pio, hdd:pio
hda: QUANTUM BIGFOOT_CY6480A, ATA DISK drive
blk: queue c02d2bc0, I/O limit 4095Mb (mask 0xffffffff)
hdc: IBM-DTLA-307060, ATA DISK drive
blk: queue c02d3018, I/O limit 4095Mb (mask 0xffffffff)
ide0 at 0x1f0-0x1f7,0x3f6 on irq 14
ide1 at 0x170-0x177,0x376 on irq 15
hda: attached ide-disk driver.

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
vvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvv
hda: task_no_data_intr: status=0x51 { DriveReady SeekComplete Error }
hda: task_no_data_intr: error=0x04 { DriveStatusError }
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!

hda: 12706470 sectors (6506 MB) w/67KiB Cache, CHS=790/255/63, DMA
hdc: attached ide-disk driver.
hdc: host protected area => 1
hdc: 120103200 sectors (61493 MB) w/1916KiB Cache, CHS=119150/16/63, (U)DMA
ide-floppy driver 0.99.newide
Partition check:
 hda: hda1 hda2 < hda5 hda6 hda7 hda8 hda9 >
 hdc: [PTBL] [7943/240/63] hdc1 < hdc5 hdc6 >
ide-floppy driver 0.99.newide

The kernels itself are not too stable on my trustworthy old PPro 200
box (which runs rock solid with the RedHat 7.3 release kernels). Not
too stable means "crashes about every other night".

The is a PPro 200 with 256 MBytes RAM, Tyan Mainboard. Not
overclocked, well cooled, lightly loaded. It normally crashes when the
only really disk pushing operation (Backup via Amanda) is started at
night.

I sent some crash reports to Alan, I will post any with pre7-ac1
here. I got a serial console hooked up to that box and I intent to use
it. :-)

	Regards
		Henning


-- 
Dipl.-Inf. (Univ.) Henning P. Schmiedehausen          INTERMETA GmbH
hps@intermeta.de        +49 9131 50 654 0   http://www.intermeta.de/

Java, perl, Solaris, Linux, xSP Consulting, Web Services 
freelance consultant -- Jakarta Turbine Development  -- hero for hire

Re: [2.4.21-pre7-ac1] IDE Warning when booting

[Athanasius]

[-- Attachment #1: Type: text/plain, Size: 1116 bytes --]

On Wed, Apr 16, 2003 at 10:16:12AM +0000, Henning P. Schmiedehausen wrote:
> !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
> vvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvv
> hda: task_no_data_intr: status=0x51 { DriveReady SeekComplete Error }
> hda: task_no_data_intr: error=0x04 { DriveStatusError }
> ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
> !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!

   Not the exact, byte-for-byte, same, but, from Configure.help:

CONFIG_IDEDISK_MULTI_MODE
  If you get this error, try to say Y here:

  hda: set_multmode: status=0x51 { DriveReady SeekComplete Error }
  hda: set_multmode: error=0x04 { DriveStatusError }

  If in doubt, say N.

Do you have that option enabled or not?

-Ath
-- 
- Athanasius = Athanasius(at)miggy.org / http://www.miggy.org/
                  Finger athan(at)fysh.org for PGP key
	   "And it's me who is my enemy. Me who beats me up.
Me who makes the monsters. Me who strips my confidence." Paula Cole - ME

[-- Attachment #2: Type: application/pgp-signature, Size: 240 bytes --]

Re: [2.4.21-pre7-ac1] IDE Warning when booting

[Henning P. Schmiedehausen]

Athanasius <link@gurus.tf> writes:


>--ALfTUftag+2gvp1h
>Content-Type: text/plain; charset=us-ascii
>Content-Disposition: inline
>Content-Transfer-Encoding: quoted-printable

>On Wed, Apr 16, 2003 at 10:16:12AM +0000, Henning P. Schmiedehausen wrote:
>> !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
>> vvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvv
>> hda: task_no_data_intr: status=3D0x51 { DriveReady SeekComplete Error }
>> hda: task_no_data_intr: error=3D0x04 { DriveStatusError }
>> ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
>> !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!

>   Not the exact, byte-for-byte, same, but, from Configure.help:

Alan did post an explanation for these (which I haven't read before
posting this) that these are harmless. And yes, the task_no_data_intr
vs. set_multmode makes all the difference. :-) Getting these quieted
down would be nice, though.

	Regards
		Henning

-- 
Dipl.-Inf. (Univ.) Henning P. Schmiedehausen          INTERMETA GmbH
hps@intermeta.de        +49 9131 50 654 0   http://www.intermeta.de/

Java, perl, Solaris, Linux, xSP Consulting, Web Services 
freelance consultant -- Jakarta Turbine Development  -- hero for hire

RE: [2.4.21-pre7-ac1] IDE Warning when booting

[Mudama, Eric]

So the quantum bigfoot doesn't support setting the multiple block size?

Hrm, how old is that disk?  SET MULTIPLE MODE has been an ATA standard
command since rev 3 of the spec, circa 1997...

I guess that is what Alan meant when he said people throw everything at
linux hardware wise...

--eric


> -----Original Message-----
> From: Henning P. Schmiedehausen [mailto:hps@intermeta.de]
> Sent: Wednesday, April 16, 2003 5:17 AM
> To: linux-kernel@vger.kernel.org
> Subject: Re: [2.4.21-pre7-ac1] IDE Warning when booting
> 
> Alan did post an explanation for these (which I haven't read before
> posting this) that these are harmless. And yes, the task_no_data_intr
> vs. set_multmode makes all the difference. :-) Getting these quieted
> down would be nice, though.
> 
> 	Regards
> 		Henning

Re: [2.4.21-pre7-ac1] IDE Warning when booting

[Henning P. Schmiedehausen]

"Mudama, Eric" <eric_mudama@maxtor.com> writes:

>So the quantum bigfoot doesn't support setting the multiple block size?

>Hrm, how old is that disk?  SET MULTIPLE MODE has been an ATA standard
>command since rev 3 of the spec, circa 1997...

>I guess that is what Alan meant when he said people throw everything at
>linux hardware wise...

What makes you think that?

% hdparm -I /dev/hda
non-removable ATA device, with non-removable media
        Model Number:           QUANTUM BIGFOOT_CY6480A
        Serial Number:          166702123281        
        Firmware Revision:      A03.0500
[...]
        r/w multiple sector transfer: Max = 32  Current = 32
        DMA: sdma0 sdma1 sdma2 mdma0 mdma1 *mdma2 
             Cycle time: min=120ns recommended=120ns
        PIO: pio0 pio1 pio2 pio3 pio4 
             Cycle time: no flow control=333ns  IORDY flow control=120ns

To me it seems to do multi sector transfers just fine. It does it even
better than the 60 GB IBM drive in this box which is much newer but
can do only 16 sectors.

	Regards
		Henning

-- 
Dipl.-Inf. (Univ.) Henning P. Schmiedehausen          INTERMETA GmbH
hps@intermeta.de        +49 9131 50 654 0   http://www.intermeta.de/

Java, perl, Solaris, Linux, xSP Consulting, Web Services 
freelance consultant -- Jakarta Turbine Development  -- hero for hire

RE: [2.4.21-pre7-ac1] IDE Warning when booting

[Mudama, Eric]

Was thinking...

Do you know what attempted multi count was as an argument to SET MULTIPLE
MODE?

The specification says powers of 2 from 2 to 128 are all valid, however,
most drives I have looked at only support <= 16.  This has some sort of
historical oem reason for being the max and I haven't worked here long
enough to know the 'why'.

However, the ID block also reports the maximum multiple count in word 47
bits 7..0, so if that was non zero, the drive shouldn't abort it.

The engineer sitting next to me (former quantum employee) is reasonably
certain that the drive shouldn't be aborting that command.

It'd be interesting to look at the ID block and the task file for that
command that the drive aborted.

--eric

> > -----Original Message-----
> > From: Henning P. Schmiedehausen [mailto:hps@intermeta.de]
> > Sent: Wednesday, April 16, 2003 5:17 AM
> > To: linux-kernel@vger.kernel.org
> > Subject: Re: [2.4.21-pre7-ac1] IDE Warning when booting
> > 
> > Alan did post an explanation for these (which I haven't read before
> > posting this) that these are harmless. And yes, the 
> task_no_data_intr
> > vs. set_multmode makes all the difference. :-) Getting these quieted
> > down would be nice, though.
> > 
> > 	Regards
> > 		Henning

RE: [2.4.21-pre7-ac1] IDE Warning when booting

[Alan Cox]

On Mer, 2003-04-16 at 17:25, Mudama, Eric wrote:
> However, the ID block also reports the maximum multiple count in word 47
> bits 7..0, so if that was non zero, the drive shouldn't abort it.

I don't believe we check that word (I didn't know about it either). I'll
take a look at that right now

Thanks
Alan

RE: [2.4.21-pre7-ac1] IDE Warning when booting

[Pau Aliagas]

On Wed, 16 Apr 2003, Mudama, Eric wrote:

> Was thinking...
> 
> Do you know what attempted multi count was as an argument to SET MULTIPLE
> MODE?

This are the parametres rc.sysinit sets just before switching to the
selected run level:

# cat /etc/sysconfig/harddisks | grep -v "^#"
USE_DMA=0 MULTIPLE_IO=16
EIDE_32BIT=3
LOOKAHEAD=1
EXTRA_PARAMS="-u1"

I twicked them to improve performance, but the behaviour was the same 
(poor performance and ide=nodma to be able to boot) before these 
modifications. So, maybe they are wrong, but no the cause of the problem.

> The specification says powers of 2 from 2 to 128 are all valid, however,
> most drives I have looked at only support <= 16.  This has some sort of
> historical oem reason for being the max and I haven't worked here long
> enough to know the 'why'.

Sorry, I do not know either.

> However, the ID block also reports the maximum multiple count in word 47
> bits 7..0, so if that was non zero, the drive shouldn't abort it.
> 
> The engineer sitting next to me (former quantum employee) is reasonably
> certain that the drive shouldn't be aborting that command.
> 
> It'd be interesting to look at the ID block and the task file for that
> command that the drive aborted.

How can I get this information?

Pau

RE: [2.4.21-pre7-ac1] IDE Warning when booting

[Mudama, Eric]

> -----Original Message-----
> From: Pau Aliagas [mailto:linuxnow@newtral.org]
> Sent: Wednesday, April 16, 2003 10:40 AM
> To: lkml; Eric Mudama
> Cc: hps@intermeta.de
> Subject: RE: [2.4.21-pre7-ac1] IDE Warning when booting
> 
> # cat /etc/sysconfig/harddisks | grep -v "^#"
> USE_DMA=0 MULTIPLE_IO=16
> EIDE_32BIT=3
> LOOKAHEAD=1
> EXTRA_PARAMS="-u1"

For fun, try changing MULTIPLE_IO to 2 instead of 16, and see if the 0x5104
error still occurrs.

> How can I get this information?

At work I'd use a bus analyzer or something capable of issuing "raw" ATA
commands... however, from the /proc data that was posted, it appeared the
drive supported a maximum of 16 for its multi count, which should be fine.

That boggles me.

--eric