Re: 2.5.60 "Badness in kobject_register at lib/kobject.c:152"

[Guennadi Liakhovetski]

> Known problem. Its probably fixed in the 2.4 changes I made to the
> probe and flash bits yesterday. Its two bugs together. The vanishing
> disk is definitely fixed, the oops from drive->id = NULL should be
> sorted too (and the general noprobe, cdrom cases)

Yep! Works! Thanks! There are still errors coming:
hda: task_no_data_intr: status=0x51 { DriveReady SeekComplete Error }
hda: task_no_data_intr: error=0x04 { DriveStatusError }
hda: 31488 sectors (16 MB) w/1KiB Cache, CHS=246/4/32, BUG <=============
Then same for hdb, then
Partition chack:
 hda:hda: dma_intr: status=0x51 { DriveReady SeekComplete Error }
hda: dma_intr: error=0x04 { DriveStatusError }
repeated 4 times, then
hda: DMA disabled
hdb: DMA disabled
ide0: reset: success
 hda1
 hdb: hdb1

then, on mounting root again
hda: task_no_data_intr: status=0x51 { DriveReady SeekComplete Error }
hda: task_no_data_intr: error=0x04 { DriveStatusError }
hda: Write Cache FAILED Flushing!
/dev/hda1: clean...
hda: task_no_data_intr: status=0x51 { DriveReady SeekComplete Error }
hda: task_no_data_intr: error=0x04 { DriveStatusError }
hda: Write Cache FAILED Flushing!
VFS: busy inodes on changed media.
hda: task_no_data_intr: status=0x51 { DriveReady SeekComplete Error }
hda: task_no_data_intr: error=0x04 { DriveStatusError }
hda: Write Cache FAILED Flushing!

And then these errors appear again on some disk-opoerations, e.g. when
running lilo, doing dd if=/dev/hda, and some others (raw access?). Can
this errors be disk-specific? (it's a SiliconTech disk, reported as
Hitachi) I can try some others, e.g. SunDisk.

Thanks
Guennadi
---------------------------------
Guennadi Liakhovetski, Ph.D.
DSA Daten- und Systemtechnik GmbH
Pascalstr. 28
D-52076 Aachen
Germany