On Wed, Oct 16, 2013 at 12:12:11PM +0300, Gleb Natapov wrote:
> On Tue, Oct 15, 2013 at 07:21:19PM -0300, Marcelo Tosatti wrote:
> > On Tue, Oct 15, 2013 at 06:57:05AM +0300, Gleb Natapov wrote:
> > > >
> > > > Why is it safe to allow access, by the lockless page write protect
> > > > side, to
On Tue, Oct 15, 2013 at 07:21:19PM -0300, Marcelo Tosatti wrote:
> On Tue, Oct 15, 2013 at 06:57:05AM +0300, Gleb Natapov wrote:
> > >
> > > Why is it safe to allow access, by the lockless page write protect
> > > side, to spt pointer for shadow page A that can change to a shadow page
> > >
On Tue, Oct 15, 2013 at 07:21:19PM -0300, Marcelo Tosatti wrote:
On Tue, Oct 15, 2013 at 06:57:05AM +0300, Gleb Natapov wrote:
Why is it safe to allow access, by the lockless page write protect
side, to spt pointer for shadow page A that can change to a shadow page
pointer of shadow
On Wed, Oct 16, 2013 at 12:12:11PM +0300, Gleb Natapov wrote:
On Tue, Oct 15, 2013 at 07:21:19PM -0300, Marcelo Tosatti wrote:
On Tue, Oct 15, 2013 at 06:57:05AM +0300, Gleb Natapov wrote:
Why is it safe to allow access, by the lockless page write protect
side, to spt pointer for
On Oct 16, 2013, at 6:21 AM, Marcelo Tosatti wrote:
> On Tue, Oct 15, 2013 at 06:57:05AM +0300, Gleb Natapov wrote:
>>>
>>> Why is it safe to allow access, by the lockless page write protect
>>> side, to spt pointer for shadow page A that can change to a shadow page
>>> pointer of shadow page
On Tue, Oct 15, 2013 at 06:57:05AM +0300, Gleb Natapov wrote:
> >
> > Why is it safe to allow access, by the lockless page write protect
> > side, to spt pointer for shadow page A that can change to a shadow page
> > pointer of shadow page B?
> >
> > Write protect spte of any page at will? Or
On Oct 16, 2013, at 6:21 AM, Marcelo Tosatti mtosa...@redhat.com wrote:
On Tue, Oct 15, 2013 at 06:57:05AM +0300, Gleb Natapov wrote:
Why is it safe to allow access, by the lockless page write protect
side, to spt pointer for shadow page A that can change to a shadow page
pointer of
On Tue, Oct 15, 2013 at 06:57:05AM +0300, Gleb Natapov wrote:
Why is it safe to allow access, by the lockless page write protect
side, to spt pointer for shadow page A that can change to a shadow page
pointer of shadow page B?
Write protect spte of any page at will? Or verify that
On Mon, Oct 14, 2013 at 04:29:45PM -0300, Marcelo Tosatti wrote:
> On Sat, Oct 12, 2013 at 08:53:56AM +0300, Gleb Natapov wrote:
> > On Fri, Oct 11, 2013 at 05:30:17PM -0300, Marcelo Tosatti wrote:
> > > On Fri, Oct 11, 2013 at 08:38:31AM +0300, Gleb Natapov wrote:
> > > > > n_max_mmu_pages is not
On Sat, Oct 12, 2013 at 08:53:56AM +0300, Gleb Natapov wrote:
> On Fri, Oct 11, 2013 at 05:30:17PM -0300, Marcelo Tosatti wrote:
> > On Fri, Oct 11, 2013 at 08:38:31AM +0300, Gleb Natapov wrote:
> > > > n_max_mmu_pages is not a suitable limit to throttle freeing of pages via
> > > > RCU (its too
On Sat, Oct 12, 2013 at 08:53:56AM +0300, Gleb Natapov wrote:
On Fri, Oct 11, 2013 at 05:30:17PM -0300, Marcelo Tosatti wrote:
On Fri, Oct 11, 2013 at 08:38:31AM +0300, Gleb Natapov wrote:
n_max_mmu_pages is not a suitable limit to throttle freeing of pages via
RCU (its too large). If
On Mon, Oct 14, 2013 at 04:29:45PM -0300, Marcelo Tosatti wrote:
On Sat, Oct 12, 2013 at 08:53:56AM +0300, Gleb Natapov wrote:
On Fri, Oct 11, 2013 at 05:30:17PM -0300, Marcelo Tosatti wrote:
On Fri, Oct 11, 2013 at 08:38:31AM +0300, Gleb Natapov wrote:
n_max_mmu_pages is not a suitable
On Fri, Oct 11, 2013 at 05:30:17PM -0300, Marcelo Tosatti wrote:
> On Fri, Oct 11, 2013 at 08:38:31AM +0300, Gleb Natapov wrote:
> > > n_max_mmu_pages is not a suitable limit to throttle freeing of pages via
> > > RCU (its too large). If the free memory watermarks are smaller than
> > >
On Fri, Oct 11, 2013 at 08:38:31AM +0300, Gleb Natapov wrote:
> > n_max_mmu_pages is not a suitable limit to throttle freeing of pages via
> > RCU (its too large). If the free memory watermarks are smaller than
> > n_max_mmu_pages for all guests, OOM is possible.
> >
> Ah, yes. I am not saying
On Fri, Oct 11, 2013 at 08:38:31AM +0300, Gleb Natapov wrote:
n_max_mmu_pages is not a suitable limit to throttle freeing of pages via
RCU (its too large). If the free memory watermarks are smaller than
n_max_mmu_pages for all guests, OOM is possible.
Ah, yes. I am not saying
On Fri, Oct 11, 2013 at 05:30:17PM -0300, Marcelo Tosatti wrote:
On Fri, Oct 11, 2013 at 08:38:31AM +0300, Gleb Natapov wrote:
n_max_mmu_pages is not a suitable limit to throttle freeing of pages via
RCU (its too large). If the free memory watermarks are smaller than
n_max_mmu_pages for
On Thu, Oct 10, 2013 at 06:03:01PM -0300, Marcelo Tosatti wrote:
> On Thu, Oct 10, 2013 at 10:16:46PM +0300, Gleb Natapov wrote:
> > On Thu, Oct 10, 2013 at 01:42:22PM -0300, Marcelo Tosatti wrote:
> > > On Thu, Oct 10, 2013 at 03:08:45PM +0300, Gleb Natapov wrote:
> > > > On Wed, Oct 09, 2013 at
On Thu, Oct 10, 2013 at 10:16:46PM +0300, Gleb Natapov wrote:
> On Thu, Oct 10, 2013 at 01:42:22PM -0300, Marcelo Tosatti wrote:
> > On Thu, Oct 10, 2013 at 03:08:45PM +0300, Gleb Natapov wrote:
> > > On Wed, Oct 09, 2013 at 10:47:10PM -0300, Marcelo Tosatti wrote:
> > > > > >> Gleb has a idea
On Thu, Oct 10, 2013 at 01:42:22PM -0300, Marcelo Tosatti wrote:
> On Thu, Oct 10, 2013 at 03:08:45PM +0300, Gleb Natapov wrote:
> > On Wed, Oct 09, 2013 at 10:47:10PM -0300, Marcelo Tosatti wrote:
> > > > >> Gleb has a idea that uses RCU_DESTORY to protect the shadow page
> > > > >> table
> > >
On Thu, Oct 10, 2013 at 03:08:45PM +0300, Gleb Natapov wrote:
> On Wed, Oct 09, 2013 at 10:47:10PM -0300, Marcelo Tosatti wrote:
> > > >> Gleb has a idea that uses RCU_DESTORY to protect the shadow page table
> > > >> and encodes the page-level into the spte (since we need to check if
> > > >>
On Wed, Oct 09, 2013 at 10:47:10PM -0300, Marcelo Tosatti wrote:
> > >> Gleb has a idea that uses RCU_DESTORY to protect the shadow page table
> > >> and encodes the page-level into the spte (since we need to check if the
> > >> spte
> > >> is the last-spte. ). How about this?
> > >
> > >
On Wed, Oct 09, 2013 at 10:47:10PM -0300, Marcelo Tosatti wrote:
Gleb has a idea that uses RCU_DESTORY to protect the shadow page table
and encodes the page-level into the spte (since we need to check if the
spte
is the last-spte. ). How about this?
Pointer please? Why is
On Thu, Oct 10, 2013 at 03:08:45PM +0300, Gleb Natapov wrote:
On Wed, Oct 09, 2013 at 10:47:10PM -0300, Marcelo Tosatti wrote:
Gleb has a idea that uses RCU_DESTORY to protect the shadow page table
and encodes the page-level into the spte (since we need to check if
the spte
is
On Thu, Oct 10, 2013 at 01:42:22PM -0300, Marcelo Tosatti wrote:
On Thu, Oct 10, 2013 at 03:08:45PM +0300, Gleb Natapov wrote:
On Wed, Oct 09, 2013 at 10:47:10PM -0300, Marcelo Tosatti wrote:
Gleb has a idea that uses RCU_DESTORY to protect the shadow page
table
and encodes the
On Thu, Oct 10, 2013 at 10:16:46PM +0300, Gleb Natapov wrote:
On Thu, Oct 10, 2013 at 01:42:22PM -0300, Marcelo Tosatti wrote:
On Thu, Oct 10, 2013 at 03:08:45PM +0300, Gleb Natapov wrote:
On Wed, Oct 09, 2013 at 10:47:10PM -0300, Marcelo Tosatti wrote:
Gleb has a idea that uses
On Thu, Oct 10, 2013 at 06:03:01PM -0300, Marcelo Tosatti wrote:
On Thu, Oct 10, 2013 at 10:16:46PM +0300, Gleb Natapov wrote:
On Thu, Oct 10, 2013 at 01:42:22PM -0300, Marcelo Tosatti wrote:
On Thu, Oct 10, 2013 at 03:08:45PM +0300, Gleb Natapov wrote:
On Wed, Oct 09, 2013 at 10:47:10PM
On Wed, Oct 09, 2013 at 06:45:47PM +0800, Xiao Guangrong wrote:
> On 10/09/2013 09:56 AM, Marcelo Tosatti wrote:
> > On Tue, Oct 08, 2013 at 12:02:32PM +0800, Xiao Guangrong wrote:
> >>
> >> Hi Marcelo,
> >>
> >> On Oct 8, 2013, at 9:23 AM, Marcelo Tosatti wrote:
> >>
>
> +if
On 10/09/2013 09:56 AM, Marcelo Tosatti wrote:
> On Tue, Oct 08, 2013 at 12:02:32PM +0800, Xiao Guangrong wrote:
>>
>> Hi Marcelo,
>>
>> On Oct 8, 2013, at 9:23 AM, Marcelo Tosatti wrote:
>>
+ if (kvm->arch.rcu_free_shadow_page) {
+ kvm_mmu_isolate_pages(invalid_list);
On 10/09/2013 09:56 AM, Marcelo Tosatti wrote:
On Tue, Oct 08, 2013 at 12:02:32PM +0800, Xiao Guangrong wrote:
Hi Marcelo,
On Oct 8, 2013, at 9:23 AM, Marcelo Tosatti mtosa...@redhat.com wrote:
+ if (kvm-arch.rcu_free_shadow_page) {
+ kvm_mmu_isolate_pages(invalid_list);
+
On Wed, Oct 09, 2013 at 06:45:47PM +0800, Xiao Guangrong wrote:
On 10/09/2013 09:56 AM, Marcelo Tosatti wrote:
On Tue, Oct 08, 2013 at 12:02:32PM +0800, Xiao Guangrong wrote:
Hi Marcelo,
On Oct 8, 2013, at 9:23 AM, Marcelo Tosatti mtosa...@redhat.com wrote:
+if
On Tue, Oct 08, 2013 at 12:02:32PM +0800, Xiao Guangrong wrote:
>
> Hi Marcelo,
>
> On Oct 8, 2013, at 9:23 AM, Marcelo Tosatti wrote:
>
> >>
> >> + if (kvm->arch.rcu_free_shadow_page) {
> >> + kvm_mmu_isolate_pages(invalid_list);
> >> + sp = list_first_entry(invalid_list,
On Tue, Oct 08, 2013 at 12:02:32PM +0800, Xiao Guangrong wrote:
Hi Marcelo,
On Oct 8, 2013, at 9:23 AM, Marcelo Tosatti mtosa...@redhat.com wrote:
+ if (kvm-arch.rcu_free_shadow_page) {
+ kvm_mmu_isolate_pages(invalid_list);
+ sp = list_first_entry(invalid_list,
Hi Marcelo,
On Oct 8, 2013, at 9:23 AM, Marcelo Tosatti wrote:
>>
>> +if (kvm->arch.rcu_free_shadow_page) {
>> +kvm_mmu_isolate_pages(invalid_list);
>> +sp = list_first_entry(invalid_list, struct kvm_mmu_page, link);
>> +list_del_init(invalid_list);
>>
On Thu, Sep 05, 2013 at 06:29:15PM +0800, Xiao Guangrong wrote:
> It is easy if the handler is in the vcpu context, in that case we can use
> walk_shadow_page_lockless_begin() and walk_shadow_page_lockless_end() that
> disable interrupt to stop shadow page being freed. But we are on the ioctl
>
On Thu, Sep 05, 2013 at 06:29:15PM +0800, Xiao Guangrong wrote:
It is easy if the handler is in the vcpu context, in that case we can use
walk_shadow_page_lockless_begin() and walk_shadow_page_lockless_end() that
disable interrupt to stop shadow page being freed. But we are on the ioctl
Hi Marcelo,
On Oct 8, 2013, at 9:23 AM, Marcelo Tosatti mtosa...@redhat.com wrote:
+if (kvm-arch.rcu_free_shadow_page) {
+kvm_mmu_isolate_pages(invalid_list);
+sp = list_first_entry(invalid_list, struct kvm_mmu_page, link);
+
It is easy if the handler is in the vcpu context, in that case we can use
walk_shadow_page_lockless_begin() and walk_shadow_page_lockless_end() that
disable interrupt to stop shadow page being freed. But we are on the ioctl
context and the paths we are optimizing for have heavy workload, disabling
It is easy if the handler is in the vcpu context, in that case we can use
walk_shadow_page_lockless_begin() and walk_shadow_page_lockless_end() that
disable interrupt to stop shadow page being freed. But we are on the ioctl
context and the paths we are optimizing for have heavy workload, disabling
38 matches
Mail list logo
