Re: 4.0.0-rc4: panic in free_block

[David Ahern <david.ahern@oracle.com>]

On 3/20/15 10:48 AM, Linus Torvalds wrote:
> [ Added Davem and the sparc mailing list, since it happens on sparc
> and that just makes me suspicious ]
>
> On Fri, Mar 20, 2015 at 8:07 AM, David Ahern <david.ahern@oracle.com> wrote:
>> I can easily reproduce the panic below doing a kernel build with make -j N,
>> N\x128, 256, etc. This is a 1024 cpu system running 4.0.0-rc4.
>
> 3.19 is fine? Because I dont' think I've seen any reports like this
> for others, and what stands out is sparc (and to a lesser degree "1024
> cpus", which obviously gets a lot less testing)

I haven't tried 3.19 yet. Just backed up to 3.18 and it shows the same 
problem. And I can reproduce the 4.0 crash in a 128 cpu ldom (VM).

>
>> The top 3 frames are consistently:
>>      free_block+0x60
>>      cache_flusharray+0xac
>>      kmem_cache_free+0xfc
>>
>> After that one path has been from __mmdrop and the others are like below,
>> from remove_vma.
>>
>> Unable to handle kernel paging request at virtual address 0006100000000000
>
> One thing you *might* check is if the problem goes away if you select
> CONFIG_SLUB instead of CONFIG_SLAB. I'd really like to just get rid of
> SLAB. The whole "we have multiple different allocators" is a mess and
> causes test coverage issues.
>
> Apart from testing with CONFIG_SLUB, if 3.19 is ok and you seem to be
> able to "easily reproduce" this, the obvious thing to do is to try to
> bisect it.

I'll try SLUB. The ldom reboots 1000 times faster then resetting the h/w 
so a better chance of bisecting - if I can find a known good release.

David