2.3.41-4 / hda: lost interrupt

[Zdenek Kabelac]

Andre Hedrick wrote:
> 
> What chipset and drive vendor?
> 
Hi

I'm having same kind of problems for two months (BP6 2x492MHz (400))

Just now I've made some new tests - for them I use this setup:

build SMP kernel (with AGP patch)

boot SMP kernel with nosmp option into single mode.

turn DMA off  (# hdparm -d 0 /dev/hda)
test the speed (# while : ; do hdparm -t /dev/hda ; done)

Two months ago I've got "hda: lost interrupt message" even
on my home computer (BP6 2x556 (400)) computer.
Now I don't have such problems with this test at home 
- so I would say there were some IDE improvements which have
fixed this.

However this problem remains on my computer at University.

And it looks like it's network related - as when I do not insert
eth0 module (eth0: 3Com 3c905C Tornado at 0xd000 100Mbps)
(or I remove network card completely) I couldn't get this messages
with this test. But as soon is eth0 is used the problem is back.
I've also tried other network card: tulip 10Mbps board and with the
same results (so I think it is not 3com related).

As this computer is being used in network environment
it is quite problematic to use it reliable - also
I don't have any problems with 2.2.14pre12 on a single CPU
system.

And one more comment - problems are visible much sooner
when I'm using RTLinux extension - especially when nfs server
gets busy (like package upgrade (running Debian)) this client
machine locks quite fast. Sometimes I could see the message:
"nfs server ... busy" in the xconsole window (and it isn't, 
because it's serving other clients without any problems) 
 - but thats all I could see and do with this computer as it is no
longer
usable - as even disks can't be remounted read-only (even with magic
key).

Is there any tool which could help to finaly catch this bug??

bye

-- 
         binary drivers cause sclerosis of the kernel (mark hahn)
  Zdenek Kabelac  http://i.am/kabi/ kabi en i.am {debian.org; fi.muni.cz}


-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo en vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/