Accessing IDE disk with bad sectors freezes the box

[Daniel Eriksson]

I wrote:

> I have a 6-CURRENT installation on a spare disk that I will 
> hook up to the
> machine later and see how it handles the bad sectors using the same
> controller.

Using a week-old CURRENT to read the disk with the bad sectors when it's
hooked up to a SiL 3112 controller yields the following result:

...
ad4: FAILURE - READ_DMA status=51<READY,DSC,ERROR> error=40<UNCORRECTABLE>
LBA=478486335
g_vfs_done():ad4s1d[READ(offset=244984971264, length=16384)]error = 5
ad4: FAILURE - READ_DMA status=51<READY,DSC,ERROR> error=40<UNCORRECTABLE>
LBA=478507071
g_vfs_done():ad4s1d[READ(offset=244995522560, length=131072)]error = 5
ata2: SATA connect ready time=1480ms
ata2: sata_connect devices=0x1<ATA_MASTER>
ata2: SATA connect ready time=10000ms
ata2: sata_connect devices=0x0
ad4: FAILURE - device detached
subdisk4: detached
ad4: detached
unknown: timeout waiting to issue command
unknown: error issueing SETFEATURES SET TRANSFER MODE command
unknown: req=0xc16e4000 SETFEATURES SET TRANSFER MODE semaphore timeout !!
DANGER Will Robinson !!
unknown: req=0xc16e4000 SETFEATURES SET TRANSFER MODE semaphore timeout !!
DANGER Will Robinson !!

The process reading the disk gets stuck in "getblk D+" status according to
'top' and cannot be killed. Compared to 5.4-RC4 this is a regression imho.

/Daniel Eriksson