[devel] hard lockup on SW OSB4 IDE operations

[Michael Shigorin <mike@osdn.org.ua>]

[-- Attachment #1.1: Type: text/plain, Size: 2540 bytes --]

	Hello,
while copying CD-ROM contents from TEAC 32X drive on IDE channel
to SCSI HDD (no other IDE devices), the system froze up with the
following:

---

Oct  8 17:27:03 entry automount[992]: attempting to mount entry /mnt/auto/cdrom
Oct  8 17:27:03 entry kernel: hdb: ATAPI 32X CD-ROM drive, 128kB Cache, UDMA(33)
Oct  8 17:27:03 entry kernel: Uniform CD-ROM driver Revision: 3.12
Oct  8 17:27:11 entry automount[3814]: expired /mnt/auto/cdrom
Oct  8 17:27:12 entry automount[992]: attempting to mount entry /mnt/auto/cdrom
Oct  8 17:27:17 entry automount[3821]: expired /mnt/auto/cdrom
Oct  8 17:27:17 entry automount[992]: attempting to mount entry /mnt/auto/cdrom
Oct  8 17:30:53 entry automount[3934]: expired /mnt/auto/cdrom
Oct  8 17:34:40 entry automount[992]: attempting to mount entry /mnt/auto/cdrom
Oct  8 17:37:08 entry kernel: Serverworks OSB4 in impossible state.
Oct  8 17:37:08 entry kernel: Disable UDMA or if you are using Seagate then try switching disk types
Oct  8 17:37:08 entry kernel: on this controller. Please report this event to osb4-bug@ide.cabal.tm

---

Note that it was the second CD-ROM at a time, with ~456M copied
over before the crash.  First one was copied just before.

The device was configured as a slave somehow, and here's hdparm
info on it:

---

/dev/hdb:
 HDIO_GET_MULTCOUNT failed: Invalid argument
 I/O support  =  0 (default 16-bit)
 unmaskirq    =  0 (off)
 using_dma    =  1 (on)
 keepsettings =  0 (off)
 HDIO_GET_NOWERR failed: Invalid argument
 readonly     =  1 (on)
 readahead    =  8 (on)
 HDIO_GETGEO failed: Invalid argument
 busstate     =  1 (on)

 Model=CD-232E, FwRev=1.0A, SerialNo=
 Config={ Fixed Removeable DTR<=5Mbs DTR>10Mbs nonMagnetic }
 RawCHS=0/0/0, TrkSize=0, SectSize=0, ECCbytes=0
 BuffType=unknown, BuffSize=0kB, MaxMultSect=0
 (maybe): CurCHS=0/0/0, CurSects=0, LBA=yes, LBAsects=0
 IORDY=on/off, tPIO={min:120,w/IORDY:120}, tDMA={min:120,rec:120}
 PIO modes: pio0 pio1 pio2 pio3 pio4 
 DMA modes: sdma0 sdma1 sdma2 mdma0 mdma1 mdma2 udma0 udma1 *udma2 
 AdvancedPM=no

---

dmesg/lspci from the same kernel (patched one) on the same
hardware are attached; what else can be of any help?

For now I'll disable DMA on that IDE.

2 goldhead: we'd better do that as a workaround if the chipset is
found during installation... and/or warn if more than one device
is there (as people were complaining).

-- 
 ---- WBR, Michael Shigorin <mike@altlinux.ru>
  ------ Linux.Kiev http://www.linux.kiev.ua/

[-- Attachment #1.2: dmesg.gz --]
[-- Type: application/x-gunzip, Size: 4617 bytes --]

[-- Attachment #1.3: lspci-vv.gz --]
[-- Type: application/x-gunzip, Size: 1111 bytes --]

[-- Attachment #2: Type: application/pgp-signature, Size: 187 bytes --]