[Crash-utility] problems running crash on recent rawhide live kernels

[Dave Anderson]

Jeff Layton wrote:
> I was trying to run crash against a live rawhide kernel and am getting
> some errors when I try to start it:
> 
>     crash: invalid task address in pid_hash: ccccccccccccca84
>     crash: invalid task address in pid_hash: ccccccccccccca84
>     ...
>     crash: invalid task address in pid_hash: ccccccccccccca84
> 
>     crash: cannot gather a stable task list via pid_hash (500 retries)
> 
> ...and then crash exits. There are a few messages in dmesg:
> 
>     crash memory driver: version 1.0
>     crash memory driver: !page_is_ram(pfn: 61000000c)
> 
> Relevant packages:
> 
>     kernel-2.6.24-0.62.rc3.git5.fc9.x86_64
>     kernel-debuginfo-2.6.24-0.62.rc3.git5.fc9.x86_64
>     crash-4.0-4.10.x86_64
> 
> ... the host is a FV xen guest (but that shouldn't matter, should it?).
> 
> I'm hoping I'm just doing something wrong, but I don't see anything
> right offhand.
> 

Hard to do anything wrong, i.e., you just have to enter "crash".

With a kernel that new, it certainly could be shifting sands syndrome.

It did get quite a long way before failing.  I wonder if something's
changed upstream associated with the pid_hash code?  Can you verify
that the tt->refresh_hash_table gets set to refresh_hlist_task_table_v2()?

If the system is accessible to me (reply offline), I'll be happy to
take a look.

Dave