2020-04-28 00:56:17

by Sean Christopherson

Subject: [PATCH 1/3] KVM: x86/mmu: Tweak PSE hugepage handling to avoid 2M vs 4M conundrum

Change the PSE hugepage handling in walk_addr_generic() to fire on any
page level greater than PT_PAGE_TABLE_LEVEL, a.k.a. PG_LEVEL_4K. PSE
paging only has two levels, so "== 2" and "> 1" are functionally the
seam, i.e. this is a nop.

A future patch will drop KVM's PT_*_LEVEL enums in favor of the kernel's
PG_LEVEL_* enums, at which point "walker->level == PG_LEVEL_2M" is
semantically incorrect (though still functionally ok).

No functional change intended.

Suggested-by: Paolo Bonzini <[email protected]>
Signed-off-by: Sean Christopherson <[email protected]>
---
arch/x86/kvm/mmu/paging_tmpl.h | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/arch/x86/kvm/mmu/paging_tmpl.h b/arch/x86/kvm/mmu/paging_tmpl.h
index efec7d27b8c5..ca39bd315f70 100644
--- a/arch/x86/kvm/mmu/paging_tmpl.h
+++ b/arch/x86/kvm/mmu/paging_tmpl.h
@@ -436,7 +436,7 @@ static int FNAME(walk_addr_generic)(struct guest_walker *walker,
gfn = gpte_to_gfn_lvl(pte, walker->level);
gfn += (addr & PT_LVL_OFFSET_MASK(walker->level)) >> PAGE_SHIFT;

- if (PTTYPE == 32 && walker->level == PT_DIRECTORY_LEVEL && is_cpuid_PSE36())
+ if (PTTYPE == 32 && walker->level > PT_PAGE_TABLE_LEVEL && is_cpuid_PSE36())
gfn += pse36_gfn_delta(pte);

real_gpa = mmu->translate_gpa(vcpu, gfn_to_gpa(gfn), access, &walker->fault);
--
2.26.0

2020-04-28 14:10:58

by Barret Rhoden

Subject: Re: [PATCH 1/3] KVM: x86/mmu: Tweak PSE hugepage handling to avoid 2M vs 4M conundrum

On 4/27/20 8:54 PM, Sean Christopherson wrote:
> Change the PSE hugepage handling in walk_addr_generic() to fire on any
> page level greater than PT_PAGE_TABLE_LEVEL, a.k.a. PG_LEVEL_4K. PSE
> paging only has two levels, so "== 2" and "> 1" are functionally the
> seam, i.e. this is a nop.
^ s/seam/same/

Barret

>
> A future patch will drop KVM's PT_*_LEVEL enums in favor of the kernel's
> PG_LEVEL_* enums, at which point "walker->level == PG_LEVEL_2M" is
> semantically incorrect (though still functionally ok).
>
> No functional change intended.
>
> Suggested-by: Paolo Bonzini <[email protected]>
> Signed-off-by: Sean Christopherson <[email protected]>
> ---
> arch/x86/kvm/mmu/paging_tmpl.h | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/arch/x86/kvm/mmu/paging_tmpl.h b/arch/x86/kvm/mmu/paging_tmpl.h
> index efec7d27b8c5..ca39bd315f70 100644
> --- a/arch/x86/kvm/mmu/paging_tmpl.h
> +++ b/arch/x86/kvm/mmu/paging_tmpl.h
> @@ -436,7 +436,7 @@ static int FNAME(walk_addr_generic)(struct guest_walker *walker,
> gfn = gpte_to_gfn_lvl(pte, walker->level);
> gfn += (addr & PT_LVL_OFFSET_MASK(walker->level)) >> PAGE_SHIFT;
>
> - if (PTTYPE == 32 && walker->level == PT_DIRECTORY_LEVEL && is_cpuid_PSE36())
> + if (PTTYPE == 32 && walker->level > PT_PAGE_TABLE_LEVEL && is_cpuid_PSE36())
> gfn += pse36_gfn_delta(pte);
>
> real_gpa = mmu->translate_gpa(vcpu, gfn_to_gpa(gfn), access, &walker->fault);
>