exception Emask 0x0 SAct 0x0 SErr 0x0 action 0x2 frozen

[Richard Heck]

Nestor Waldyd Alvarez Villa wrote:
> Hello...
>
> I am having a problem with my TSSTcorp CD-RW/DVD-ROM TSL462D. The error
> below is causing a parallel application to fail. Here is the relevant
> output of my system log (dmesg command)
>
>
> ata5.00: ATAPI: TSSTcorpCD-RW/DVD-ROM TSL462D, DE01, max UDMA/33
>
> ...
>
> ata5.00: exception Emask 0x0 SAct 0x0 SErr 0x0 action 0x2 frozen
> ata5.00: cmd a0/00:00:00:00:20/00:00:00:00:00/a0 tag 0 cdb 0x0 data 0
>             res 51/20:03:00:00:00/00:00:00:00:00/a0 Emask 0x3 (HSM
> violation)
> ata5: soft resetting port
> ata5.00: configured for UDMA/33
> ata5: EH pending after completion, repeating EH (cnt=4)
> ata5: EH complete
> ata5.00: exception Emask 0x0 SAct 0x0 SErr 0x0 action 0x2 frozen
> ata5.00: cmd a0/00:00:00:00:20/00:00:00:00:00/a0 tag 0 cdb 0x0 data 0
>             res 51/20:03:00:00:00/00:00:00:00:00/a0 Emask 0x3 (HSM
> violation)
> ata5: soft resetting port
> ata5.00: configured for UDMA/33
> ata5: EH pending after completion, repeating EH (cnt=4)
> ata5: EH complete
>
> I have noticed that when i run a parallel MPI application with more than
> one thread per machine, the problem is triggered... I also have 
> noticed that
> the prompt above comes from the HAL daemon because when i disable the hal
> daemon, this is not prompted anymore but the problem persists. Using the
> acpi=off kernel option, the system tends to be more stable but the 
> problem
> does not solve...
>
> Here is my system info (uname -a):
>
> [waldyd at nodo5 ~]$ uname -a
> Linux nodo5.gialea 2.6.23.14 #1 SMP Mon Feb 4 16:12:24 COT 2008 x86_64
> x86_64 x86_64 GNU/Linux
>
> Any ideas about why is this happening? How can i solve this?
>
Try upgrading to the 2.6.24 kernel. That fixed this sort of problem for me.

rh