[PATCH 0/4] zsmalloc improvements

Seth Jennings sjenning at linux.vnet.ibm.com
Wed Jul 11 14:00:30 UTC 2012


On 07/11/2012 02:03 AM, Minchan Kim wrote:
> On 07/03/2012 06:15 AM, Seth Jennings wrote:
>> zsmapbench measures the copy-based mapping at ~560 cycles for a
>> map/unmap operation on spanned object for both KVM guest and bare-metal,
>> while the page table mapping was ~1500 cycles on a VM and ~760 cycles
>> bare-metal.  The cycles for the copy method will vary with
>> allocation size, however, it is still faster even for the largest
>> allocation that zsmalloc supports.
>>
>> The result is convenient though, as mempcy is very portable :)
> 
> Today, I tested zsmapbench in my embedded board(ARM).
> tlb-flush is 30% faster than copy-based so it's always not win.
> I think it depends on CPU speed/cache size.
> 
> zram is already very popular on embedded systems so I want to use
> it continuously without 30% big demage so I want to keep our old approach
> which supporting local tlb flush. 
> 
> Of course, in case of KVM guest, copy-based would be always bin win.
> So shouldn't we support both approach? It could make code very ugly
> but I think it has enough value.
> 
> Any thought?

Thanks for testing on ARM.

I can add the pgtable assisted method back in, no problem.
The question is by which criteria are we going to choose
which method to use? By arch (i.e. ARM -> pgtable assist,
x86 -> copy, other archs -> ?)?

Also, what changes did you make to zsmapbench to measure
elapsed time/cycles on ARM?  Afaik, rdtscll() is not
supported on ARM.

Thanks,
Seth




More information about the devel mailing list