On 20.09.22 15:23, Marek Marczykowski-Górecki wrote:
On Tue, Sep 20, 2022 at 11:06:57AM +0200, Jan Beulich wrote:On 19.09.2022 17:09, Marek Marczykowski-Górecki wrote:--- a/xen/common/sched/credit2.c +++ b/xen/common/sched/credit2.c @@ -996,9 +996,13 @@ cpu_add_to_runqueue(const struct scheduler *ops, unsigned int cpu) * * Otherwise, let's try to make sure that siblings stay in the * same runqueue, pretty much under any cinrcumnstances. + * + * Furthermore, try to respect credit2_runqueue=all, as long as + * max_cpus_runq isn't violated.This last part is questionable, partly because the command line doc is ambiguous as to which of the two options is intended to "win". I guess one needs to know the original intentions to resolve this.Right, I've chosen this approach, because you can still emulate the other by setting sufficiently large max_cpus_runq. I can add doc clarification in v2.
I think this is the better approach, as it allows more flexibility. Updating the doc would be mandatory, though. With that added you can have my: Reviewed-by: Juergen Gross <jgr...@suse.com> Juergen
OpenPGP_0xB0DE9DD628BF132F.asc
Description: OpenPGP public key
OpenPGP_signature
Description: OpenPGP digital signature
