[Crash-utility] Abort in "kmem -i" command

Dave Anderson anderson at redhat.com
Thu Jun 21 20:17:42 UTC 2018



----- Original Message -----
> Hi Dave,
> 
> I'm faced with abort in "kmem -i" command with some vmcores
> and looking into it, but I haven't found the cause so far.
> Could you possibly take a look at this issue?
> 
> I attached the abort log. I can send you the core and vmcore.
> 
> Depending on execution, the occurrences vary.
> 
> * abort, <segmentation violation in gdb>, or no problem
> * If not reproduced with interactive "crash> kmem -i", you may
>   reproduce it with "echo 'kmem -i' | crash vmlinux vmcore".
> * observed in "kmem -V" and "kmem -z" as well
> * also observed on i686 with kernel 4.14 vmcore
> * latest crash (1926150e)
> 
> Thanks,
> Kazu

Hi Kazu,

I'm not clear on why it always generates the SIGSEGV when the kmem
command is piped directly into crash, whereas it rarely happens when
running the command interactively.

But anyway, as it turns out, there is a 1-byte buffer overflow in 
vm_stat_init() that has always been there.

Can you try the attached patch?

Thanks,
  Dave
-------------- next part --------------
A non-text attachment was scrubbed...
Name: vm_stat_init.patch
Type: text/x-patch
Size: 628 bytes
Desc: not available
URL: <http://listman.redhat.com/archives/crash-utility/attachments/20180621/efafa361/attachment.bin>


More information about the Crash-utility mailing list