Re: [PATCH] xen/x86/pvh: Use HVM's flush_tlb_others op

2015-12-15 Thread Jan Beulich
>>> On 15.12.15 at 16:37, wrote: > On 12/15/2015 10:24 AM, Jan Beulich wrote: > On 15.12.15 at 16:14, wrote: >>> On 12/15/2015 10:03 AM, Jan Beulich wrote: >>> On 15.12.15 at 15:36, wrote: > On 12/14/2015 10:27 AM, Konrad Rzeszutek Wilk wrote: >> On Sat, Dec 12, 2015 at

Re: [PATCH] xen/x86/pvh: Use HVM's flush_tlb_others op

2015-12-15 Thread Boris Ostrovsky
On 12/15/2015 10:24 AM, Jan Beulich wrote: On 15.12.15 at 16:14, wrote: On 12/15/2015 10:03 AM, Jan Beulich wrote: On 15.12.15 at 15:36, wrote: On 12/14/2015 10:27 AM, Konrad Rzeszutek Wilk wrote: On Sat, Dec 12, 2015 at 07:25:55PM -0500, Boris Ostrovsky wrote: Using

Re: [PATCH] xen/x86/pvh: Use HVM's flush_tlb_others op

2015-12-15 Thread Jan Beulich
>>> On 15.12.15 at 16:14, wrote: > On 12/15/2015 10:03 AM, Jan Beulich wrote: > On 15.12.15 at 15:36, wrote: >>> On 12/14/2015 10:27 AM, Konrad Rzeszutek Wilk wrote: On Sat, Dec 12, 2015 at 07:25:55PM -0500, Boris Ostrovsky wrote: > Using MMUEXT_TLB_FLUSH_MULTI doesn't buy us much

Re: [PATCH] xen/x86/pvh: Use HVM's flush_tlb_others op

2015-12-15 Thread Boris Ostrovsky
On 12/15/2015 10:03 AM, Jan Beulich wrote: On 15.12.15 at 15:36, wrote: On 12/14/2015 10:27 AM, Konrad Rzeszutek Wilk wrote: On Sat, Dec 12, 2015 at 07:25:55PM -0500, Boris Ostrovsky wrote: Using MMUEXT_TLB_FLUSH_MULTI doesn't buy us much since the hypervisor will likely perform same IPIs as

Re: [PATCH] xen/x86/pvh: Use HVM's flush_tlb_others op

2015-12-15 Thread Jan Beulich
>>> On 15.12.15 at 15:36, wrote: > On 12/14/2015 10:27 AM, Konrad Rzeszutek Wilk wrote: >> On Sat, Dec 12, 2015 at 07:25:55PM -0500, Boris Ostrovsky wrote: >>> Using MMUEXT_TLB_FLUSH_MULTI doesn't buy us much since the hypervisor >>> will likely perform same IPIs as would have the guest. >>> >>

Re: [PATCH] xen/x86/pvh: Use HVM's flush_tlb_others op

2015-12-15 Thread Boris Ostrovsky
On 12/14/2015 10:27 AM, Konrad Rzeszutek Wilk wrote: On Sat, Dec 12, 2015 at 07:25:55PM -0500, Boris Ostrovsky wrote: Using MMUEXT_TLB_FLUSH_MULTI doesn't buy us much since the hypervisor will likely perform same IPIs as would have the guest. But if the VCPU is asleep, doing it via the

Re: [PATCH] xen/x86/pvh: Use HVM's flush_tlb_others op

2015-12-15 Thread Jan Beulich
>>> On 15.12.15 at 16:37, wrote: > On 12/15/2015 10:24 AM, Jan Beulich wrote: > On 15.12.15 at 16:14, wrote: >>> On 12/15/2015 10:03 AM, Jan Beulich wrote: >>> On 15.12.15 at 15:36, wrote: > On

Re: [PATCH] xen/x86/pvh: Use HVM's flush_tlb_others op

2015-12-15 Thread Boris Ostrovsky
On 12/14/2015 10:27 AM, Konrad Rzeszutek Wilk wrote: On Sat, Dec 12, 2015 at 07:25:55PM -0500, Boris Ostrovsky wrote: Using MMUEXT_TLB_FLUSH_MULTI doesn't buy us much since the hypervisor will likely perform same IPIs as would have the guest. But if the VCPU is asleep, doing it via the

Re: [PATCH] xen/x86/pvh: Use HVM's flush_tlb_others op

2015-12-15 Thread Jan Beulich
>>> On 15.12.15 at 15:36, wrote: > On 12/14/2015 10:27 AM, Konrad Rzeszutek Wilk wrote: >> On Sat, Dec 12, 2015 at 07:25:55PM -0500, Boris Ostrovsky wrote: >>> Using MMUEXT_TLB_FLUSH_MULTI doesn't buy us much since the hypervisor >>> will likely perform same IPIs as

Re: [PATCH] xen/x86/pvh: Use HVM's flush_tlb_others op

2015-12-15 Thread Boris Ostrovsky
On 12/15/2015 10:24 AM, Jan Beulich wrote: On 15.12.15 at 16:14, wrote: On 12/15/2015 10:03 AM, Jan Beulich wrote: On 15.12.15 at 15:36, wrote: On 12/14/2015 10:27 AM, Konrad Rzeszutek Wilk wrote: On Sat, Dec 12, 2015 at 07:25:55PM

Re: [PATCH] xen/x86/pvh: Use HVM's flush_tlb_others op

2015-12-15 Thread Boris Ostrovsky
On 12/15/2015 10:03 AM, Jan Beulich wrote: On 15.12.15 at 15:36, wrote: On 12/14/2015 10:27 AM, Konrad Rzeszutek Wilk wrote: On Sat, Dec 12, 2015 at 07:25:55PM -0500, Boris Ostrovsky wrote: Using MMUEXT_TLB_FLUSH_MULTI doesn't buy us much since the hypervisor will

Re: [PATCH] xen/x86/pvh: Use HVM's flush_tlb_others op

2015-12-15 Thread Jan Beulich
>>> On 15.12.15 at 16:14, wrote: > On 12/15/2015 10:03 AM, Jan Beulich wrote: > On 15.12.15 at 15:36, wrote: >>> On 12/14/2015 10:27 AM, Konrad Rzeszutek Wilk wrote: On Sat, Dec 12, 2015 at 07:25:55PM -0500, Boris Ostrovsky wrote:

Re: [Xen-devel] [PATCH] xen/x86/pvh: Use HVM's flush_tlb_others op

2015-12-14 Thread Boris Ostrovsky
On 12/14/2015 10:35 AM, Roger Pau Monné wrote: El 14/12/15 a les 16.27, Konrad Rzeszutek Wilk ha escrit: On Sat, Dec 12, 2015 at 07:25:55PM -0500, Boris Ostrovsky wrote: Using MMUEXT_TLB_FLUSH_MULTI doesn't buy us much since the hypervisor will likely perform same IPIs as would have the guest.

Re: [Xen-devel] [PATCH] xen/x86/pvh: Use HVM's flush_tlb_others op

2015-12-14 Thread Roger Pau Monné
El 14/12/15 a les 16.27, Konrad Rzeszutek Wilk ha escrit: > On Sat, Dec 12, 2015 at 07:25:55PM -0500, Boris Ostrovsky wrote: >> Using MMUEXT_TLB_FLUSH_MULTI doesn't buy us much since the hypervisor >> will likely perform same IPIs as would have the guest. >> > > But if the VCPU is asleep, doing

Re: [Xen-devel] [PATCH] xen/x86/pvh: Use HVM's flush_tlb_others op

2015-12-14 Thread Boris Ostrovsky
On 12/14/2015 08:58 AM, David Vrabel wrote: On 13/12/15 00:25, Boris Ostrovsky wrote: Using MMUEXT_TLB_FLUSH_MULTI doesn't buy us much since the hypervisor will likely perform same IPIs as would have the guest. More importantly, using MMUEXT_INVLPG_MULTI may not to invalidate the guest's

Re: [Xen-devel] [PATCH] xen/x86/pvh: Use HVM's flush_tlb_others op

2015-12-14 Thread David Vrabel
On 13/12/15 00:25, Boris Ostrovsky wrote: > Using MMUEXT_TLB_FLUSH_MULTI doesn't buy us much since the hypervisor > will likely perform same IPIs as would have the guest. > > More importantly, using MMUEXT_INVLPG_MULTI may not to invalidate the > guest's address on remote CPU (when, for example,

Re: [Xen-devel] [PATCH] xen/x86/pvh: Use HVM's flush_tlb_others op

2015-12-14 Thread Roger Pau Monné
El 14/12/15 a les 16.27, Konrad Rzeszutek Wilk ha escrit: > On Sat, Dec 12, 2015 at 07:25:55PM -0500, Boris Ostrovsky wrote: >> Using MMUEXT_TLB_FLUSH_MULTI doesn't buy us much since the hypervisor >> will likely perform same IPIs as would have the guest. >> > > But if the VCPU is asleep, doing

Re: [Xen-devel] [PATCH] xen/x86/pvh: Use HVM's flush_tlb_others op

2015-12-14 Thread Boris Ostrovsky
On 12/14/2015 10:35 AM, Roger Pau Monné wrote: El 14/12/15 a les 16.27, Konrad Rzeszutek Wilk ha escrit: On Sat, Dec 12, 2015 at 07:25:55PM -0500, Boris Ostrovsky wrote: Using MMUEXT_TLB_FLUSH_MULTI doesn't buy us much since the hypervisor will likely perform same IPIs as would have the guest.

Re: [Xen-devel] [PATCH] xen/x86/pvh: Use HVM's flush_tlb_others op

2015-12-14 Thread David Vrabel
On 13/12/15 00:25, Boris Ostrovsky wrote: > Using MMUEXT_TLB_FLUSH_MULTI doesn't buy us much since the hypervisor > will likely perform same IPIs as would have the guest. > > More importantly, using MMUEXT_INVLPG_MULTI may not to invalidate the > guest's address on remote CPU (when, for example,

Re: [Xen-devel] [PATCH] xen/x86/pvh: Use HVM's flush_tlb_others op

2015-12-14 Thread Boris Ostrovsky
On 12/14/2015 08:58 AM, David Vrabel wrote: On 13/12/15 00:25, Boris Ostrovsky wrote: Using MMUEXT_TLB_FLUSH_MULTI doesn't buy us much since the hypervisor will likely perform same IPIs as would have the guest. More importantly, using MMUEXT_INVLPG_MULTI may not to invalidate the guest's

[PATCH] xen/x86/pvh: Use HVM's flush_tlb_others op

2015-12-12 Thread Boris Ostrovsky
Using MMUEXT_TLB_FLUSH_MULTI doesn't buy us much since the hypervisor will likely perform same IPIs as would have the guest. More importantly, using MMUEXT_INVLPG_MULTI may not to invalidate the guest's address on remote CPU (when, for example, VCPU from another guest is running there).

[PATCH] xen/x86/pvh: Use HVM's flush_tlb_others op

2015-12-12 Thread Boris Ostrovsky
Using MMUEXT_TLB_FLUSH_MULTI doesn't buy us much since the hypervisor will likely perform same IPIs as would have the guest. More importantly, using MMUEXT_INVLPG_MULTI may not to invalidate the guest's address on remote CPU (when, for example, VCPU from another guest is running there).