Re: [PATCH] i386/kvm: Prefault memory on page state change

2025-06-03 Thread Xiaoyao Li
On 6/3/2025 11:00 PM, Paolo Bonzini wrote: I'm applying Tom's patch to get it out of his queue, but will delay sending a pull request until the Linux-side fix is accepted. BTW, for the patch itself. Reviewed-by: Xiaoyao Li Tested-by: Xiaoyao Li

Re: [PATCH] i386/kvm: Prefault memory on page state change

2025-06-03 Thread Xiaoyao Li
On 6/3/2025 11:00 PM, Paolo Bonzini wrote: On 6/3/25 13:47, Xiaoyao Li wrote: On 6/3/2025 3:41 PM, Xiaoyao Li wrote: On 3/29/2025 4:30 AM, Tom Lendacky wrote: A page state change is typically followed by an access of the page(s) and results in another VMEXIT in order to map the page into the

Re: [PATCH] i386/kvm: Prefault memory on page state change

2025-06-03 Thread Paolo Bonzini
On 6/3/25 13:47, Xiaoyao Li wrote: On 6/3/2025 3:41 PM, Xiaoyao Li wrote: On 3/29/2025 4:30 AM, Tom Lendacky wrote: A page state change is typically followed by an access of the page(s) and results in another VMEXIT in order to map the page into the nested page table. Depending on the size of

Re: [PATCH] i386/kvm: Prefault memory on page state change

2025-06-03 Thread Tom Lendacky
On 6/3/25 06:47, Xiaoyao Li wrote: > On 6/3/2025 3:41 PM, Xiaoyao Li wrote: >> On 3/29/2025 4:30 AM, Tom Lendacky wrote: >>> A page state change is typically followed by an access of the page(s) and >>> results in another VMEXIT in order to map the page into the nested page >>> table. Depending on

Re: [PATCH] i386/kvm: Prefault memory on page state change

2025-06-03 Thread Xiaoyao Li
On 6/3/2025 3:41 PM, Xiaoyao Li wrote: On 3/29/2025 4:30 AM, Tom Lendacky wrote: A page state change is typically followed by an access of the page(s) and results in another VMEXIT in order to map the page into the nested page table. Depending on the size of page state change request, this can g

Re: [PATCH] i386/kvm: Prefault memory on page state change

2025-06-03 Thread Xiaoyao Li
On 3/29/2025 4:30 AM, Tom Lendacky wrote: A page state change is typically followed by an access of the page(s) and results in another VMEXIT in order to map the page into the nested page table. Depending on the size of page state change request, this can generate a number of additional VMEXITs.

Re: [PATCH] i386/kvm: Prefault memory on page state change

2025-06-02 Thread Tom Lendacky
On 3/28/25 15:30, Lendacky, Thomas wrote: > A page state change is typically followed by an access of the page(s) and > results in another VMEXIT in order to map the page into the nested page > table. Depending on the size of page state change request, this can > generate a number of additional VME

[PATCH] i386/kvm: Prefault memory on page state change

2025-03-28 Thread Tom Lendacky
A page state change is typically followed by an access of the page(s) and results in another VMEXIT in order to map the page into the nested page table. Depending on the size of page state change request, this can generate a number of additional VMEXITs. For example, under SNP, when Linux is utiliz