Re: 2.4.23 : timeouts and lost interrupts using IDE-SCSI

Re: 2.4.23 : timeouts and lost interrupts using IDE-SCSI

[Kristian Peters]

Jimmie Mayfield <mayfield+kernel@sackheads.org> schrieb:
> Since upgrading from 2.4.20 to 2.4.23, I can no longer mount my ATAPI CDROM and
> CDR/W drives using IDE-SCSI.  Attempts to mount those drives result in
> lost interrupts messages:

That is normal and intended. You should see some "dma disabled" messages in your kernel-log (unless if you haven't enabled "DMA only for disks"). ATAPI/cdrom does not work reliable with DMA on 2.4. I think Andre Hedrick has switched that off after 2.4.20.
If you really want ide-scsi with dma you should try 2.6.

This is how it should like:

SCSI subsystem driver Revision: 1.00
hdb: attached ide-scsi driver.
hdb: DMA disabled		<------
hdd: attached ide-scsi driver.
hdd: DMA disabled		<------
scsi0 : SCSI host adapter emulation for IDE ATAPI devices
  Vendor: COMPAQ    Model: CD-ROM LTN485     Rev: KQA4
  Type:   CD-ROM                             ANSI SCSI revision: 02
  Vendor: HP        Model: CD-Writer+ 9100b  Rev: 1.06
  Type:   CD-ROM                             ANSI SCSI revision: 02

So you should disable that option in your config or if you prefer using hdparm try it with "-d0 /dev/cdrom".

*Kristian

Re: 2.4.23 : timeouts and lost interrupts using IDE-SCSI

[Daniel B.]

Kristian Peters wrote:
> 
> Jimmie Mayfield <mayfield+kernel@sackheads.org> schrieb:
> > Since upgrading from 2.4.20 to 2.4.23, I can no longer mount my ATAPI CDROM and
> > CDR/W drives using IDE-SCSI.  Attempts to mount those drives result in
> > lost interrupts messages:
> 
> That is ... intended. 

Huh?  Losing interrupts is intended?


Daniel
-- 
Daniel Barclay
dsb@smart.net

2.4.23 : timeouts and lost interrupts using IDE-SCSI

[Jimmie Mayfield]

Since upgrading from 2.4.20 to 2.4.23, I can no longer mount my ATAPI CDROM and
CDR/W drives using IDE-SCSI.  Attempts to mount those drives result in
lost interrupts messages:

Dec  3 08:21:10 kaon kernel: scsi : aborting command due to timeout : pid 2114, scs
i2, channel 0, id 0, lun 0 Read (10) 00 00 00 00 10 00 00 01 00 
Dec  3 08:21:10 kaon kernel: hdc: DMA interrupt recovery
Dec  3 08:21:10 kaon kernel: hdc: lost interrupt


Kernel is able to successfully probe for and identify these devices, however:

Dec  3 08:20:40 kaon kernel: hdc: attached ide-scsi driver.
Dec  3 08:20:41 kaon kernel: hdd: attached ide-scsi driver.
Dec  3 08:20:41 kaon kernel: scsi2 : SCSI host adapter emulation for IDE ATAPI devi
ces
Dec  3 08:20:41 kaon kernel:   Vendor: HITACHI   Model: CDR-8430          Rev: 0024
Dec  3 08:20:41 kaon kernel:   Type:   CD-ROM                             ANSI SCSI
 revision: 02
Dec  3 08:20:41 kaon kernel:   Vendor:           Model: CD-R/RW RW7060A   Rev: 1.50
Dec  3 08:20:41 kaon kernel:   Type:   CD-ROM                             ANSI SCSI
 revision: 02
Dec  3 08:20:41 kaon kernel: Attached scsi CD-ROM sr0 at scsi2, channel 0, id 0, lu
n 0
Dec  3 08:20:41 kaon kernel: Attached scsi CD-ROM sr1 at scsi2, channel 0, id 1, lu
n 0
Dec  3 08:20:41 kaon kernel: sr0: scsi3-mmc drive: 14x/32x cd/rw xa/form2 cdda tray
Dec  3 08:20:41 kaon kernel: Uniform CD-ROM driver Revision: 3.12
Dec  3 08:20:41 kaon kernel: sr1: scsi3-mmc drive: 24x/24x writer cd/rw xa/form2 cd
da tray


I can revert back to 2.4.20 and the drives are mountable once again so I do not
believe it's a hardware failure.  Similarly, I can build 2.4.23 with IDE CDROM
support instead of relying on SCSI emulation and things work (though I suspect
not having SCSI emulation enabled will cause problems when burning CDs).


System specs:

IBM Intellistation MPro
SMP P2-400 (2 CPUs)
440BX chipset
ACPI automatically disabled "because your bios is from 98 and too old"
CDROM as hdc, CDR/W as hdd
   both these devices are attached to the onboard PIIX4 controller
kernels compiled with gcc 2.95.3


Jimmie

-- 
Jimmie Mayfield  
http://www.sackheads.org/mayfield       email: mayfield+kernel@sackheads.org
My mail provider does not welcome UCE -- http://www.sackheads.org/uce