tech-kern archive

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Old Index]

Re: System goes into complete hang



On Wednesday, 9 Apr 2008 12:24:20
"D'Arcy J.M. Cain" <darcy%NetBSD.org@localhost> wrote:

> I turned off all the debugging (it wasn't showing anything useful)
> andthe ACPI stuff.  The system still hung.

> Is it possible that I am looking at a hardware problem?

It's not impossible, in my experience overheating and bad memory
issues generally had the symptom of userland applications randomly
crashing every now and then with SIGBUS or SIGSEGV, but if hardware
has problems I guess that anything is possible...

The usual CPU temperature and memory check (using memtest86) is
probably always a good idea then.

However, since disabling ACPI seemed to help, it also could
potentially be a NetBSD ACPI framework issue and/or BIOS/firmware
issue, then making sure that BIOS version is the latest for that
motherboard is also a good idea.  With ACPI disabled, if your system
is SMP do all cores still activate?

Unfortunately I'm not versed enough into ACPI to help further debug
it, but I'm sure it would be rather straightforward to log all ACPI
events, with ACPI turned on to see which event occurs regularily (if
any) which triggers the rest of the code causing the crash...  This
could narrow down on what to look for, if the problem is software.
It's possible to query the firmware's ACPI bytecode as well if I
remember (acpidump, amldb?) but I couldn't help to diagnose it with my
limited ACPI knowledge.

It's quite sad to have to say it, but on many problem systems
disabling ACPI helped, which is why I thought it was among the
necessary things to try.

-- 
Matthew Mondor


Home | Main Index | Thread Index | Old Index