Re: [PATCH v3] KVM: x86/mmu: Always drop mmu_lock to allocate TDP MMU SPs for eager splitting

[Sean Christopherson <seanjc@google.com>]

On Tue, Jun 11, 2024, David Matlack wrote:
> On 2024-06-03 11:13 AM, Sean Christopherson wrote:
> > On Thu, May 09, 2024, David Matlack wrote:
> > > diff --git a/arch/x86/kvm/mmu/tdp_mmu.c b/arch/x86/kvm/mmu/tdp_mmu.c
> > > index aaa2369a9479..2089d696e3c6 100644
> > > --- a/arch/x86/kvm/mmu/tdp_mmu.c
> > > +++ b/arch/x86/kvm/mmu/tdp_mmu.c
> > > @@ -1385,11 +1385,11 @@ bool kvm_tdp_mmu_wrprot_slot(struct kvm *kvm,
> > >  	return spte_set;
> > >  }
> > >  
> > > -static struct kvm_mmu_page *__tdp_mmu_alloc_sp_for_split(gfp_t gfp)
> > > +static struct kvm_mmu_page *__tdp_mmu_alloc_sp_for_split(void)
> > >  {
> > > +	gfp_t gfp = GFP_KERNEL_ACCOUNT | __GFP_ZERO;
> > >  	struct kvm_mmu_page *sp;
> > >  
> > > -	gfp |= __GFP_ZERO;
> > >  
> > >  	sp = kmem_cache_alloc(mmu_page_header_cache, gfp);
> > 
> > This can more simply and cleary be:
> > 
> > 	sp = kmem_cache_zalloc(mmu_page_header_cache, GFP_KERNEL_ACCOUNT);
> 
> Will do. And I assume you'd prefer get_zeroed_page(GFP_KERNEL_ACCOUNT)
> as well below?

Ah, yeah, good catch!