> -Original Message-
> From: Jan Beulich [mailto:jbeul...@suse.com]
> Sent: Friday, September 23, 2016 10:19 PM
> To: Wu, Feng
> Cc: andrew.coop...@citrix.com; dario.faggi...@citrix.com;
> george.dun...@eu.citrix.com; Tian, Kevin ; xen-
>
>>> On 02.09.16 at 15:15, wrote:
> And that is another reason I use pause/unpause here, it can address
> all the races. And this is an one-time action (Only occurs at the first
> device gets assigned), do you think it is that unacceptable?
Btw. please see George's very nice
> -Original Message-
> From: Dario Faggioli [mailto:dario.faggi...@citrix.com]
> Sent: Tuesday, September 20, 2016 4:14 PM
> To: Jan Beulich
> Cc: andrew.coop...@citrix.com; george.dun...@eu.citrix.com; Wu, Feng
> ; Tian, Kevin
On Tue, 2016-09-20 at 01:31 -0600, Jan Beulich wrote:
> > > > On 20.09.16 at 01:12, wrote:
> > How's possible for the solution here to be "either remove the
> > ASSERT()
> > _OR_ change the code"? That really makes few sense to me... :-O
>
> I disagree: Whether
> -Original Message-
> From: Jan Beulich [mailto:jbeul...@suse.com]
> Sent: Tuesday, September 20, 2016 3:32 PM
> To: Dario Faggioli
> Cc: andrew.coop...@citrix.com; george.dun...@eu.citrix.com; Wu, Feng
> ; Tian, Kevin
>>> On 20.09.16 at 01:12, wrote:
> On Sun, 2016-09-18 at 08:37 +, Wu, Feng wrote:
>> > From: Dario Faggioli [mailto:dario.faggi...@citrix.com]
>> However, I found this is not true in some scenarios, such as,
>> vmx_pi_switch_to() hasn't been installed or executed
> -Original Message-
> From: Dario Faggioli [mailto:dario.faggi...@citrix.com]
> Sent: Tuesday, September 20, 2016 7:12 AM
> To: Wu, Feng ; Jan Beulich
> Cc: andrew.coop...@citrix.com; george.dun...@eu.citrix.com; Tian, Kevin
>
On Sun, 2016-09-18 at 08:37 +, Wu, Feng wrote:
> > From: Dario Faggioli [mailto:dario.faggi...@citrix.com]
> > So why this is all of the sudden becoming one? Am I completely off
> > with
> > my recollection (or in general :-P)? Or what am I missing about the
> > issue we are trying to address
Hi Dario,
> -Original Message-
> From: Dario Faggioli [mailto:dario.faggi...@citrix.com]
> Sent: Wednesday, September 14, 2016 10:52 PM
> To: Wu, Feng ; Jan Beulich
> Cc: andrew.coop...@citrix.com; george.dun...@eu.citrix.com; Tian, Kevin
>
On Wed, 2016-09-14 at 02:23 +, Wu, Feng wrote:
> Then I tried to implement the function like the following:
>
> /* This function is called when pcidevs_lock is held */
> void vmx_pi_hooks_assign(struct domain *d)
> {
> if ( !iommu_intpost || !has_hvm_container_domain(d) )
>
>>> On 14.09.16 at 04:23, wrote:
>> From: Wu, Feng
>> Sent: Monday, September 5, 2016 11:12 AM
>> I still have trouble to fully understand this, please see the following
>> scenario:
>>
>> 1) Set 'NDST' to the pCPU0 (which vCPU is currently running on, but
>> it may changes
> -Original Message-
> From: Wu, Feng
> Sent: Monday, September 5, 2016 11:12 AM
> To: Jan Beulich
> Cc: andrew.coop...@citrix.com; dario.faggi...@citrix.com;
> george.dun...@eu.citrix.com; Tian, Kevin ; xen-
> de...@lists.xen.org; Wu, Feng
>>> On 05.09.16 at 05:11, wrote:
>
>> -Original Message-
>> From: Jan Beulich [mailto:jbeul...@suse.com]
>> Sent: Friday, September 2, 2016 9:55 PM
>> To: Wu, Feng
>> Cc: andrew.coop...@citrix.com; dario.faggi...@citrix.com;
>>
> -Original Message-
> From: Jan Beulich [mailto:jbeul...@suse.com]
> Sent: Friday, September 2, 2016 9:55 PM
> To: Wu, Feng
> Cc: andrew.coop...@citrix.com; dario.faggi...@citrix.com;
> george.dun...@eu.citrix.com; Tian, Kevin ; xen-
>
>>> On 02.09.16 at 15:15, wrote:
>
>> -Original Message-
>> From: Jan Beulich [mailto:jbeul...@suse.com]
>> Sent: Friday, September 2, 2016 6:46 PM
>> To: Wu, Feng
>> Cc: andrew.coop...@citrix.com; dario.faggi...@citrix.com;
>>
> -Original Message-
> From: Jan Beulich [mailto:jbeul...@suse.com]
> Sent: Friday, September 2, 2016 6:46 PM
> To: Wu, Feng
> Cc: andrew.coop...@citrix.com; dario.faggi...@citrix.com;
> george.dun...@eu.citrix.com; Tian, Kevin ; xen-
>
>>> On 02.09.16 at 12:30, wrote:
>
>> -Original Message-
>> From: Jan Beulich [mailto:jbeul...@suse.com]
>> Sent: Friday, September 2, 2016 5:26 PM
>> To: Wu, Feng
>> Cc: andrew.coop...@citrix.com; dario.faggi...@citrix.com;
>>
> -Original Message-
> From: Jan Beulich [mailto:jbeul...@suse.com]
> Sent: Friday, September 2, 2016 5:26 PM
> To: Wu, Feng
> Cc: andrew.coop...@citrix.com; dario.faggi...@citrix.com;
> george.dun...@eu.citrix.com; Tian, Kevin ; xen-
>
>>> On 02.09.16 at 10:40, wrote:
>
>> -Original Message-
>> From: Jan Beulich [mailto:jbeul...@suse.com]
>> Sent: Friday, September 2, 2016 4:16 PM
>> To: Wu, Feng
>> Cc: andrew.coop...@citrix.com; dario.faggi...@citrix.com;
>>
> -Original Message-
> From: Jan Beulich [mailto:jbeul...@suse.com]
> Sent: Friday, September 2, 2016 4:16 PM
> To: Wu, Feng
> Cc: andrew.coop...@citrix.com; dario.faggi...@citrix.com;
> george.dun...@eu.citrix.com; Tian, Kevin ; xen-
>
>>> On 02.09.16 at 09:31, wrote:
>
>> -Original Message-
>> From: Jan Beulich [mailto:jbeul...@suse.com]
>> Sent: Friday, September 2, 2016 3:04 PM
>> To: Wu, Feng
>> Cc: andrew.coop...@citrix.com; dario.faggi...@citrix.com;
>>
> -Original Message-
> From: Jan Beulich [mailto:jbeul...@suse.com]
> Sent: Friday, September 2, 2016 3:04 PM
> To: Wu, Feng
> Cc: andrew.coop...@citrix.com; dario.faggi...@citrix.com;
> george.dun...@eu.citrix.com; Tian, Kevin ; xen-
>
>>> On 02.09.16 at 03:46, wrote:
>
>> -Original Message-
>> From: Jan Beulich [mailto:jbeul...@suse.com]
>> Sent: Thursday, September 1, 2016 4:30 PM
>> To: Wu, Feng
>> Cc: andrew.coop...@citrix.com; dario.faggi...@citrix.com;
>>
> -Original Message-
> From: Jan Beulich [mailto:jbeul...@suse.com]
> Sent: Thursday, September 1, 2016 4:30 PM
> To: Wu, Feng
> Cc: andrew.coop...@citrix.com; dario.faggi...@citrix.com;
> george.dun...@eu.citrix.com; Tian, Kevin ; xen-
>
>>> On 31.08.16 at 05:56, wrote:
> --- a/xen/arch/x86/hvm/vmx/vmx.c
> +++ b/xen/arch/x86/hvm/vmx/vmx.c
> @@ -219,8 +219,19 @@ void vmx_pi_hooks_assign(struct domain *d)
>
> ASSERT(!d->arch.hvm_domain.vmx.vcpu_block);
>
> +/*
> + * Pausing the domain can make
Pausing the domain can make sure the vCPU is not running and
hence calling the hooks simultaneously when deassigning the
PI hooks. This makes sure that all the appropriate state of
PI descriptor is actually set up for all vCPus before leaving
this function.
Signed-off-by: Feng Wu
26 matches
Mail list logo