On 02/22/2012 11:08 AM, Pekka Enberg wrote:

Hi Glauber,

On Tue, Feb 21, 2012 at 1:34 PM, Glauber Costa[off-list ref]  wrote:

quoted

This is a first structured approach to tracking general kernel
memory within the memory controller. Please tell me what you think.

I like it! I only skimmed through the SLUB changes but they seemed
reasonable enough. What kind of performance hit are we taking when
memcg configuration option is enabled but the feature is disabled?

                         Pekka

Thanks Pekka.

Well, I didn't took any numbers, because I don't consider the whole work 
any close to final form, but I wanted people to comment anyway.

In particular, I intend to use the same trick I used for tcp sock 
buffers here for this case - (static_branch()), so the performance hit 
should come from two pointers in the kmem_cache structure - and I 
believe it is possible to remove one of them.

I can definitely measure when I implement that, but I think it is 
reasonable to expect not that much of a hit.

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Fight unfair telecom internet charges in Canada: sign http://stopthemeter.ca/
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>