At 12:38 +0100 on 09 Aug (1281357513), George Dunlap wrote:
> VMs that don't use their full timeslice are guaranteed to flip back
> and forth between "active" and "inactive". If we set credit to 0
> when setting "inactive", then when the VM comes back to "active"
> again, it will effectively be behind most other vcpus in credit.
> This causes the credit1 to effectively discriminate *against*
> VMs which use less than their full timeslice.
> Instead of setting credit to 0, divide it in half (shifting
> by one bit for more efficiency).
"credit" here is signed, and the compiler is allowed to use either a
signed or an unsigned shift for >>. Better to code it as /=2 and trust
the compiler to DTRT with a constant division.
> This gets rid of some of the
> system credit while allowing non-cpu-bound VMs to keep some priority
> Signed-off-by: George Dunlap <george.dunlap@xxxxxxxxxxxxx>
> diff -r 6f07d9ac1e7c -r 1a29b22ef2e9 xen/common/sched_credit.c
> --- a/xen/common/sched_credit.c Thu Aug 05 14:41:14 2010 +0100
> +++ b/xen/common/sched_credit.c Mon Aug 09 12:38:12 2010 +0100
> @@ -1069,7 +1069,9 @@
> if ( credit > CSCHED_CREDITS_PER_TSLICE )
> __csched_vcpu_acct_stop_locked(prv, svc);
> - credit = 0;
> + /* Divide credits in half, so that when it starts
> + * accounting again, it starts a little bit "ahead" */
> + credit >>= 1;
> atomic_set(&svc->credit, credit);
> Xen-devel mailing list
Tim Deegan <Tim.Deegan@xxxxxxxxxx>
Principal Software Engineer, XenServer Engineering
Citrix Systems UK Ltd. (Company #02937203, SL9 0BG)
Xen-devel mailing list