Re: KVM: MMU: add KVM_ZAP_GFN ioctl

[Avi Kivity <avi@qumranet.com>]

Marcelo Tosatti wrote:
> Add an ioctl to zap all mappings to a given gfn. This allows userspace
> remove the QEMU process mappings and the page without causing
> inconsistency.
>
>   

I'm thinking of comitting rmap_nuke() to kvm.git, and the rest to the 
external module, since this is only needed on kernels without mmu notifiers.

Andrea, is rmap_nuke() suitable for the mmu notifiers pte clear callback?


Oh, and a single gfn may have multiple hvas, so we need to iterate over 
something here.

> Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
>
>
> diff --git a/arch/x86/kvm/mmu.c b/arch/x86/kvm/mmu.c
> index f0cdfba..c41464f 100644
> --- a/arch/x86/kvm/mmu.c
> +++ b/arch/x86/kvm/mmu.c
> @@ -642,6 +642,67 @@ static void rmap_write_protect(struct kvm *kvm, u64 gfn)
>  	account_shadowed(kvm, gfn);
>  }
>  
> +static void rmap_nuke(struct kvm *kvm, u64 gfn)
> +{
> +	unsigned long *rmapp;
> +	u64 *spte;
> +	int nuked = 0;
> +
> +	gfn = unalias_gfn(kvm, gfn);
> +	rmapp = gfn_to_rmap(kvm, gfn, 0);
> +
> +	spte = rmap_next(kvm, rmapp, NULL);
> +	while (spte) {
> +		BUG_ON(!spte);
> +		BUG_ON(!(*spte & PT_PRESENT_MASK));
> +		rmap_printk("rmap_nuke: spte %p %llx\n", spte, *spte);
> +		rmap_remove(kvm, spte);
> +		set_shadow_pte(spte, shadow_trap_nonpresent_pte);
> +                nuked = 1;
> +		spte = rmap_next(kvm, rmapp, spte);
> +	}
> +	/* check for huge page mappings */
> +	rmapp = gfn_to_rmap(kvm, gfn, 1);
> +	spte = rmap_next(kvm, rmapp, NULL);
> +	while (spte) {
> +		BUG_ON(!spte);
> +		BUG_ON(!(*spte & PT_PRESENT_MASK));
> +		BUG_ON((*spte & (PT_PAGE_SIZE_MASK|PT_PRESENT_MASK)) != (PT_PAGE_SIZE_MASK|PT_PRESENT_MASK));
> +		pgprintk("rmap_nuke(large): spte %p %llx %lld\n", spte, *spte, gfn);
> +		rmap_remove(kvm, spte);
> +		--kvm->stat.lpages;
> +		set_shadow_pte(spte, shadow_trap_nonpresent_pte);
> +		nuked = 1;
> +		spte = rmap_next(kvm, rmapp, spte);
> +	}
> +
> +	if (nuked)
> +		kvm_flush_remote_tlbs(kvm);
> +}
> +
> +int kvm_zap_single_gfn(struct kvm *kvm, gfn_t gfn)
> +{
> +	unsigned long addr;
> +	int have_mmu_notifiers = 0;
> +
> +	down_read(&kvm->slots_lock);
> +	addr = gfn_to_hva(kvm, gfn);
> +
> +	if (kvm_is_error_hva(addr)) {
> +		up_read(&kvm->slots_lock);
> +		return -EINVAL;
> +	}
> +
> +	if (!have_mmu_notifiers) {
> +		spin_lock(&kvm->mmu_lock);
> +		rmap_nuke(kvm, gfn);
> +		spin_unlock(&kvm->mmu_lock);
> +	}
> +	up_read(&kvm->slots_lock);
> +
> +	return 0;
> +}
> +
>  #ifdef MMU_DEBUG
>  static int is_empty_shadow_page(u64 *spt)
>  {
> diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
> index e65a9d6..d982ca1 100644
> --- a/arch/x86/kvm/x86.c
> +++ b/arch/x86/kvm/x86.c
> @@ -816,6 +816,9 @@ int kvm_dev_ioctl_check_extension(long ext)
>  	case KVM_CAP_NR_MEMSLOTS:
>  		r = KVM_MEMORY_SLOTS;
>  		break;
> +	case KVM_CAP_ZAP_GFN:
> +		r = 1;
> +		break;
>  	default:
>  		r = 0;
>  		break;
> @@ -1636,6 +1639,15 @@ long kvm_arch_vm_ioctl(struct file *filp,
>  		r = 0;
>  		break;
>  	}
> +	case KVM_ZAP_GFN: {
> +		gfn_t gfn;
> +
> +		r = -EFAULT;
> +		if (copy_from_user(&gfn, argp, sizeof gfn))
> +			goto out;
> +		r = kvm_zap_single_gfn(kvm, gfn);
> +		break;
> +        } 
>  	default:
>  		;
>  	}
> diff --git a/include/asm-x86/kvm_host.h b/include/asm-x86/kvm_host.h
> index 024b57c..4e45bd2 100644
> --- a/include/asm-x86/kvm_host.h
> +++ b/include/asm-x86/kvm_host.h
> @@ -425,6 +425,7 @@ void kvm_mmu_set_nonpresent_ptes(u64 trap_pte, u64 notrap_pte);
>  int kvm_mmu_reset_context(struct kvm_vcpu *vcpu);
>  void kvm_mmu_slot_remove_write_access(struct kvm *kvm, int slot);
>  void kvm_mmu_zap_all(struct kvm *kvm);
> +int  kvm_zap_single_gfn(struct kvm *kvm, gfn_t gfn);
>  unsigned int kvm_mmu_calculate_mmu_pages(struct kvm *kvm);
>  void kvm_mmu_change_mmu_pages(struct kvm *kvm, unsigned int kvm_nr_mmu_pages);
>  
> diff --git a/include/linux/kvm.h b/include/linux/kvm.h
> index e92e703..9ea714f 100644
> --- a/include/linux/kvm.h
> +++ b/include/linux/kvm.h
> @@ -236,6 +236,7 @@ struct kvm_vapic_addr {
>  #define KVM_CAP_CLOCKSOURCE 8
>  #define KVM_CAP_NR_VCPUS 9       /* returns max vcpus per vm */
>  #define KVM_CAP_NR_MEMSLOTS 10   /* returns max memory slots per vm */
> +#define KVM_CAP_ZAP_GFN	11
>  
>  /*
>   * ioctls for VM fds
> @@ -258,6 +259,7 @@ struct kvm_vapic_addr {
>  #define KVM_IRQ_LINE		  _IOW(KVMIO, 0x61, struct kvm_irq_level)
>  #define KVM_GET_IRQCHIP		  _IOWR(KVMIO, 0x62, struct kvm_irqchip)
>  #define KVM_SET_IRQCHIP		  _IOR(KVMIO,  0x63, struct kvm_irqchip)
> +#define KVM_ZAP_GFN	   	  _IOR(KVMIO,  0x64, unsigned long)
>  
>  /*
>   * ioctls for vcpu fds
>
> -------------------------------------------------------------------------
> This SF.net email is sponsored by: Microsoft
> Defy all challenges. Microsoft(R) Visual Studio 2008.
> http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
> _______________________________________________
> kvm-devel mailing list
> kvm-devel@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/kvm-devel
>   


-- 
error compiling committee.c: too many arguments to function


-------------------------------------------------------------------------
This SF.net email is sponsored by: Microsoft
Defy all challenges. Microsoft(R) Visual Studio 2008.
http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/