DragonFlyBSD threading model

Matthew Dillon dillon at apollo.backplane.com
Tue Aug 2 09:31:08 PDT 2005


:
:Matthew Dillon wrote:
:>     This separation also allows us to theoretically implement multiple
:>     userland schedulers, either exclusively or in tandem, or even locked
:>     to a subset of available cpus.  These are features I want to add.
:How do you plan to add granularity for this? Per jail, per application, etc?
:
:Toma¾

   Don't know yet.  The current scheduler structure pointer is inherited by
   children of a fork, it's mostly a matter of building the infrastructure
   to manage multiple schedulers and select which cpu(s) they run on.

					-Matt
					Matthew Dillon 
					<dillon at xxxxxxxxxxxxx>





More information about the Kernel mailing list