Have you searched Deja.com for this error message? I had a similar problem with
"Drive Status Error Bad CRC" IDE warnings and then getting kicked out of DMA. A
search on Deja.com revealed that this was probably caused by cables which were
too long for the UDMA standard. And indeed, once I changed to shorter cables,
the problem went away.
A quick search on Deja.com for "status=0x58" revealed quite a few people having
the same error messages as you, but only one answer. It's from Andre Hedrick,
the Linux IDE guy, so it should be pretty authoritative, but it's all Dutch to
me, so I'll just quote it verbatim and see if you can make more sense of it than
me.
Andre's message:
---------------
2.0.36 does not know about DMAing with ALI.
Second you have to dork with ISA-bridge at init-time to fix PnP
wonder-blunders.
kernel-mirrors/pub/linux/kernel/people/hedrick/
Andre Hedrick
The Linux IDE guy
On Thu, 19 Aug 1999, Csabai Csaba wrote:
>
> A few days ago I got a new main board, I put it a linux server with kernel
> version 2.2.11, and the kernel said: Unknown IDE interface on PCI, then
> I examined it under kernel version 2.0.36 too, and with that version it
> worked well.
>
>
> -------------------------------------------------------------------------
> On 2.0.36 boot time:
>
> pcibios_init : BIOS 32 Service Directory structure at 0x000fdae0
> pcibios_init : BIOS 32 Service Directory entry at 0xfdaf0
> pcibios_init : PCI BIOS revision 2,10 entry at 0xfdb11
> Probing PCI hardware.
>
> And after boot the PCI bios is work correctly, no problems
>
> -------------------------------------------------------------------------
> On 2.2.11 boot time: (and other 2.2.x)
>
> PCI_IDE: unknown IDE controller on PCI bus 00 device 58, VID=10b9, DID=5229
> PCI_IDE: not 100% native mode: will probe irqs later
> PCI_IDE: simplex device: DMA disable
> ide0: PCI_IDE Bus-Master DMA disable (BIOS)
> hda: ST31720A , ATA DISK drive
>
> and after some time:
>
> hda: lost interrupt
> ide0:unexpected interrupt, status=0x58, count=1
> hda: lost interrupt
> ide0:unexpected interrupt, status=0x58, count=2
> hda: lost interrupt
> ide0:unexpected interrupt, status=0x58, count=3
> [...]
> hda: status error: status=0x58 {DriveReady SeekComplete DataRequest }
> hda: drive not ready for command
>
> and this messages is looping, and system is "busy". :)
>
>
> ------------------------------------------------------------------------
>
> Thiz bug fixing is very important, because my server is need 2.2.x kernel.
>
> Any idea ?
I have no idea how you "dork with ISA-bridge at init-time", but if you can
figure it out, maybe that's your answer. I believe Andre occasionally lurks in
this list, so if he is, maybe he can explain what needs doing. Otherwise, you
could try forwarding your message to him at [EMAIL PROTECTED]
> I think my next move will be to give up raid 0.90 and try the stock 0.36
> unless someone else can suggest something else to try.....
I wouldn't have thought this would help. This error message looks like an IDE
problem, not a raid problem. But you never know....
Just one other thought: You have set the jumpers on the disk for DMA, haven't
you?
Cheers,
Bruno Prior [EMAIL PROTECTED]
> -----Original Message-----
> From: [EMAIL PROTECTED]
> [mailto:[EMAIL PROTECTED]]On Behalf Of Glenn McGrath
> Sent: 21 August 1999 07:15
> To: Linux Raid
> Subject: dma,irq timouts when using raid partitions /
> Quantum(KA18GB&&hpt366) hangs
>
>
> Hi, i keep getting the following message when i try and do anything usefull
> on my raid0 array.
>
> hde: timeout waiting for DMA
> hde: irq timeout: status=0x58 { DriveReady SeekComplete DataRequest }
> hde: DMA disabled
>
> I never get this problem if i only work on a "normal" non-raid partition on
> this drive.
>
> Ive been using bonnie and hdparm to test the drives.
>
> Ive tried a few tweaks using hdparm but havent done anything that warns of
> great danger.
>
> hde is the slowest of my 4 ide drives, its a quantum ex, as opposed to the
> others which a quantum cr and a ka (more on the ka later), doing a hdparm -t
> on hde gives 13MB/s, the fastest is 14.5 MB/s or so, so this doesnt seem to
> me like this drives performance would be a major bottleneck (correct me if
> im wrong).
>
> I have a 4-way ide raid, each drive is on its own channel and is master. Im
> using an external UDMA/66 controller card that uses the hpt366 chipset.
> To get the drivers of this card requires the uniform ide patch which im
> using (v6.20).
>
> Im tried many kernels and patch combinations, starting from 2.2.10 to 2.2.12
> all using the latest raid0.90 patch and tools. and uniform ide driver patch.
> I havent tried any develpment kernels as i dont know about getting raid0.90
> for 2.3.x.
>
> I think my next move will be to give up raid 0.90 and try the stock 0.36
> unless someone else can suggest something else to try.....
>
> Oh and i did select different options in the kernel, like the md by default
> option, didnt seem to make any difference.
>
> This next one is probably off-topic for this list, maybe i should get it to
> the main linux kernel mail list somehow, but because im totally rambling on,
> you may as well hear of ALL my problem :)
> With my 18GB Quantum KA drive and the Uniform IDE driver v6.2 (which
> supports my hpt622 chipset ide controller) the kernel just hangs halfway
> through the boot (when trying to detect drive paramaters)!! I saw in the
> code that the 13GB Quantum KA is on the bad list, maybe the 18GB should go
> there too !
>
> So, is anyone else having/had problems like this, or am i totally alone ?
>
> Thanks
> Glenn McGrath
>
>
