[dm-devel] [PATCH] dm-thin: optimize power of two block size

[Joe Thornber]

On Sun, Jun 24, 2012 at 09:53:22PM -0400, Mikulas Patocka wrote:
> So the conclusion is that is that that divide instruction degrades 
> transfer speed, especially on dm-stripe with 4k stripe size (on dm-thin it 
> is measurable only with raw read, the difference is smaller because it has 
> a minimum chunk size 64k).
> 
> 
> The question is why do you want to avoid such optimization?

You've conviced me.  I just wanted proof, which you've done very
nicely.  Thankyou.

> BTW. when unloading the dm-thin device with debugging enabled (the tests 
> were done with debugging disabled), I got this message:
> device-mapper: space map checker: free block counts differ, checker 
> 131060, sm-disk:130991
> --- so there is supposedly some bug? The kernel is 3.4.3.

That message is ok.  I'm going to remove the sm-checker in 3.6.  It's
not earning it's keep.

- Joe