Messages in this thread | | | Date | Wed, 29 Jun 2011 19:50:41 +0800 | From | Xiao Guangrong <> | Subject | Re: [PATCH v2 19/22] KVM: MMU: lockless walking shadow page table |
| |
On 06/29/2011 07:18 PM, Avi Kivity wrote: > On 06/29/2011 02:16 PM, Xiao Guangrong wrote: >> >> @@ -1767,6 +1874,14 @@ static void kvm_mmu_commit_zap_page(struct kvm *kvm, >> >> >> >> kvm_flush_remote_tlbs(kvm); >> >> >> >> + if (atomic_read(&kvm->arch.reader_counter)) { >> >> + kvm_mmu_isolate_pages(invalid_list); >> >> + sp = list_first_entry(invalid_list, struct kvm_mmu_page, link); >> >> + list_del_init(invalid_list); >> >> + call_rcu(&sp->rcu, free_pages_rcu); >> >> + return; >> >> + } >> >> + >> > >> > I think we should do this unconditionally. The cost of ping-ponging the shared cache line containing reader_counter will increase with large smp counts. On the other hand, zap_page is very rare, so it can be a little slower. Also, less code paths = easier to understand. >> > >> >> On soft mmu, zap_page is very frequently, it can cause performance regression in my test. > > Any idea what the cause of the regression is? It seems to me that simply deferring freeing shouldn't have a large impact. >
I guess it is because the page is freed too frequently, i have done the test, it shows about 3219 pages is freed per second
Kernbench performance comparing:
the origin way: 3m27.723 free all shadow page in rcu context: 3m30.519
| |