Re: [PATCH v2 7/9] KVM: X86: MMU: Tune PTE_LIST_EXT to be bigger

[Sean Christopherson <seanjc@google.com>]

On Fri, Jun 25, 2021, Peter Xu wrote:
> Currently rmap array element only contains 3 entries.  However for EPT=N there
> could have a lot of guest pages that got tens of even hundreds of rmap entry.
> 
> A normal distribution of a 6G guest (even if idle) shows this with rmap count
> statistics:
> 
> Rmap_Count:     0       1       2-3     4-7     8-15    16-31   32-63   64-127  128-255 256-511 512-1023
> Level=4K:       3089171 49005   14016   1363    235     212     15      7       0       0       0
> Level=2M:       5951    227     0       0       0       0       0       0       0       0       0
> Level=1G:       32      0       0       0       0       0       0       0       0       0       0
> 
> If we do some more fork some pages will grow even larger rmap counts.
> 
> This patch makes PTE_LIST_EXT bigger so it'll be more efficient for the general
> use case of EPT=N as we do list reference less and the loops over PTE_LIST_EXT
> will be slightly more efficient; but still not too large so less waste when
> array not full.
> 
> It should not affecting EPT=Y since EPT normally only has zero or one rmap
> entry for each page, so no array is even allocated.
> 
> With a test case to fork 500 child and recycle them ("./rmap_fork 500" [1]),
> this patch speeds up fork time of about 22%.
> 
>     Before: 367.20 (+-4.58%)
>     After:  302.00 (+-5.30%)
> 
> [1] https://github.com/xzpeter/clibs/commit/825436f825453de2ea5aaee4bdb1c92281efe5b3
> 
> Signed-off-by: Peter Xu <peterx@redhat.com>
> ---
>  arch/x86/kvm/mmu/mmu.c | 4 ++--
>  1 file changed, 2 insertions(+), 2 deletions(-)
> 
> diff --git a/arch/x86/kvm/mmu/mmu.c b/arch/x86/kvm/mmu/mmu.c
> index b3f738a7c05e..9b093985a2ef 100644
> --- a/arch/x86/kvm/mmu/mmu.c
> +++ b/arch/x86/kvm/mmu/mmu.c
> @@ -137,8 +137,8 @@ module_param(dbg, bool, 0644);
>  
>  #include <trace/events/kvm.h>
>  
> -/* make pte_list_desc fit well in cache line */
> -#define PTE_LIST_EXT 3
> +/* make pte_list_desc fit well in cache lines */
> +#define PTE_LIST_EXT 15

Ha, I was going to say that this should be '14' to fit pte_list_desc within two
cache lines, but looks like Paolo fixed it up on commit.

Also, if the whole cache line thing actually matters, sptes[] and spte_count
should be swapped since spte_count is always read, whereas spte_count[7:14] will
be read iff there are 8+ SPTEs.

>  struct pte_list_desc {
>  	u64 *sptes[PTE_LIST_EXT];
> -- 
> 2.31.1
>