On Sun, 2010-05-30 at 19:35 +0200, Moritz Muehlenhoff wrote: > On Mon, Mar 22, 2010 at 02:32:44AM +0000, Ben Hutchings wrote: > > We don't want memory cgroup support to be enabled by default, due to the > > run-time overhead. Below is a patch that allows the code to be included > > but disabled by default. However, memory cgroup support still has some > > run-time cost when included but disabled, which we need to evaluate > > before making this change. > > Did you submit this upstream? No, and I'm not sure it is suitable for upstream as enabling memory cgroups is said to have less of a performance cost. Note that we don't apply this yet, anyway. I'm going to run some more tests to find the performance cost of including them disabled-by-default. Ben. -- Ben Hutchings Once a job is fouled up, anything done to improve it makes it worse.
Attachment:
signature.asc
Description: This is a digitally signed message part