Messages in this thread | | | Date | Tue, 17 Apr 2012 12:36:51 +0530 | From | Raghavendra K T <> | Subject | Re: [PATCH RFC V5 2/6] kvm hypervisor : Add a hypercall to KVM hypervisor to support pv-ticketlocks |
| |
On 04/12/2012 05:59 AM, Marcelo Tosatti wrote: > On Wed, Apr 11, 2012 at 09:06:29PM -0300, Marcelo Tosatti wrote: >> On Fri, Mar 23, 2012 at 01:37:04PM +0530, Raghavendra K T wrote: >>> From: Srivatsa Vaddagiri<vatsa@linux.vnet.ibm.com> >>> [...]
barrier(); >> >> Is it always OK to erase the notification, even in case an unrelated >> event such as interrupt was the source of wakeup? > > Note i am only asking whether it is OK to lose a notification, not > requesting a change to atomic test-and-clear.
Yes.. got your point. IMO, this is the only (safe) place where it can clear kicked(pv_unhalted) flag. Since it is going to be runnable.
and you are also right in having concern on unwanted clear of flag since that would result in vcpu /vm hangs eventually.
Hope I did not miss anything.
> > It would be nice to have a comment explaining it. >
Sure will do that
>>
| |