From: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
To: KarimAllah Ahmed <karahmed@amazon.de>
Cc: linux-kernel@vger.kernel.org, kvm@vger.kernel.org,
pbonzini@redhat.com, rkrcmar@redhat.com,
Filippo Sironi <sironi@amazon.de>
Subject: Re: [PATCH v5 03/13] X86/KVM: Handle PFNs outside of kernel reach when touching GPTEs
Date: Wed, 23 Jan 2019 12:31:29 -0500 [thread overview]
Message-ID: <20190123173051.GH19289@Konrads-MacBook-Pro.local> (raw)
In-Reply-To: <1547026933-31226-4-git-send-email-karahmed@amazon.de>
On Wed, Jan 09, 2019 at 10:42:03AM +0100, KarimAllah Ahmed wrote:
> From: Filippo Sironi <sironi@amazon.de>
>
> cmpxchg_gpte() calls get_user_pages_fast() to retrieve the number of
> pages and the respective struct page to map in the kernel virtual
> address space.
> This doesn't work if get_user_pages_fast() is invoked with a userspace
> virtual address that's backed by PFNs outside of kernel reach (e.g., when
> limiting the kernel memory with mem= in the command line and using
> /dev/mem to map memory).
>
> If get_user_pages_fast() fails, look up the VMA that back the userspace
> virtual address, compute the PFN and the physical address, and map it in
> the kernel virtual address space with memremap().
>
> Signed-off-by: Filippo Sironi <sironi@amazon.de>
> Signed-off-by: KarimAllah Ahmed <karahmed@amazon.de>
I personally would have used some crafty goto statements to jump to
'err' label which would have
up_read(¤t->mm->mmap_sem);
return -EFAULT;
which would be after
154 return (ret != orig_pte);
But that is bike-shedding so feel free to ignore it.
Either way:
Reviewed-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
> ---
> arch/x86/kvm/paging_tmpl.h | 38 +++++++++++++++++++++++++++++---------
> 1 file changed, 29 insertions(+), 9 deletions(-)
>
> diff --git a/arch/x86/kvm/paging_tmpl.h b/arch/x86/kvm/paging_tmpl.h
> index 6bdca39..c40af67 100644
> --- a/arch/x86/kvm/paging_tmpl.h
> +++ b/arch/x86/kvm/paging_tmpl.h
> @@ -141,15 +141,35 @@ static int FNAME(cmpxchg_gpte)(struct kvm_vcpu *vcpu, struct kvm_mmu *mmu,
> struct page *page;
>
> npages = get_user_pages_fast((unsigned long)ptep_user, 1, 1, &page);
> - /* Check if the user is doing something meaningless. */
> - if (unlikely(npages != 1))
> - return -EFAULT;
> -
> - table = kmap_atomic(page);
> - ret = CMPXCHG(&table[index], orig_pte, new_pte);
> - kunmap_atomic(table);
> -
> - kvm_release_page_dirty(page);
> + if (likely(npages == 1)) {
> + table = kmap_atomic(page);
> + ret = CMPXCHG(&table[index], orig_pte, new_pte);
> + kunmap_atomic(table);
> +
> + kvm_release_page_dirty(page);
> + } else {
> + struct vm_area_struct *vma;
> + unsigned long vaddr = (unsigned long)ptep_user & PAGE_MASK;
> + unsigned long pfn;
> + unsigned long paddr;
> +
> + down_read(¤t->mm->mmap_sem);
> + vma = find_vma_intersection(current->mm, vaddr, vaddr + PAGE_SIZE);
> + if (!vma || !(vma->vm_flags & VM_PFNMAP)) {
> + up_read(¤t->mm->mmap_sem);
> + return -EFAULT;
> + }
> + pfn = ((vaddr - vma->vm_start) >> PAGE_SHIFT) + vma->vm_pgoff;
> + paddr = pfn << PAGE_SHIFT;
> + table = memremap(paddr, PAGE_SIZE, MEMREMAP_WB);
> + if (!table) {
> + up_read(¤t->mm->mmap_sem);
> + return -EFAULT;
> + }
> + ret = CMPXCHG(&table[index], orig_pte, new_pte);
> + memunmap(table);
> + up_read(¤t->mm->mmap_sem);
> + }
>
> return (ret != orig_pte);
> }
> --
> 2.7.4
>
next prev parent reply other threads:[~2019-01-23 17:32 UTC|newest]
Thread overview: 39+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-01-09 9:42 [PATCH v5 00/13] KVM/X86: Introduce a new guest mapping interface KarimAllah Ahmed
2019-01-09 9:42 ` [PATCH v5 01/13] X86/nVMX: handle_vmon: Read 4 bytes from guest memory KarimAllah Ahmed
2019-01-23 16:59 ` Konrad Rzeszutek Wilk
2019-01-09 9:42 ` [PATCH v5 02/13] X86/nVMX: Update the PML table without mapping and unmapping the page KarimAllah Ahmed
2019-01-23 17:17 ` Konrad Rzeszutek Wilk
2019-01-09 9:42 ` [PATCH v5 03/13] X86/KVM: Handle PFNs outside of kernel reach when touching GPTEs KarimAllah Ahmed
2019-01-23 17:31 ` Konrad Rzeszutek Wilk [this message]
2019-01-09 9:42 ` [PATCH v5 04/13] KVM: Introduce a new guest mapping API KarimAllah Ahmed
2019-01-10 13:07 ` David Hildenbrand
2019-01-25 18:01 ` Raslan, KarimAllah
2019-01-23 17:41 ` Konrad Rzeszutek Wilk
2019-01-23 17:50 ` Konrad Rzeszutek Wilk
2019-01-25 18:09 ` Raslan, KarimAllah
2019-01-30 17:08 ` Paolo Bonzini
2019-01-09 9:42 ` [PATCH v5 05/13] X86/nVMX: handle_vmptrld: Use kvm_vcpu_map when copying VMCS12 from guest memory KarimAllah Ahmed
2019-01-23 17:44 ` Konrad Rzeszutek Wilk
2019-01-09 9:42 ` [PATCH v5 06/13] KVM/nVMX: Use kvm_vcpu_map when mapping the L1 MSR bitmap KarimAllah Ahmed
2019-01-09 9:42 ` [PATCH v5 07/13] KVM/nVMX: Use kvm_vcpu_map when mapping the virtual APIC page KarimAllah Ahmed
2019-01-23 17:57 ` Konrad Rzeszutek Wilk
2019-01-25 18:14 ` Raslan, KarimAllah
2019-01-09 9:42 ` [PATCH v5 08/13] KVM/nVMX: Use kvm_vcpu_map when mapping the posted interrupt descriptor table KarimAllah Ahmed
2019-01-23 18:03 ` Konrad Rzeszutek Wilk
2019-01-25 18:15 ` Raslan, KarimAllah
2019-01-09 9:42 ` [PATCH v5 09/13] KVM/X86: Use kvm_vcpu_map in emulator_cmpxchg_emulated KarimAllah Ahmed
2019-01-23 18:04 ` Konrad Rzeszutek Wilk
2019-01-09 9:42 ` [PATCH v5 10/13] KVM/nSVM: Use the new mapping API for mapping guest memory KarimAllah Ahmed
2019-01-23 18:08 ` Konrad Rzeszutek Wilk
2019-01-09 9:42 ` [PATCH v5 11/13] KVM/nVMX: Use kvm_vcpu_map for accessing the shadow VMCS KarimAllah Ahmed
2019-01-23 18:10 ` Konrad Rzeszutek Wilk
2019-01-09 9:42 ` [PATCH v5 12/13] KVM/nVMX: Use kvm_vcpu_map for accessing the enlightened VMCS KarimAllah Ahmed
2019-01-23 18:11 ` Konrad Rzeszutek Wilk
2019-01-09 9:42 ` [PATCH v5 13/13] KVM/nVMX: Use page_address_valid in a few more locations KarimAllah Ahmed
2019-01-23 18:18 ` Konrad Rzeszutek Wilk
2019-01-25 18:17 ` Raslan, KarimAllah
2019-01-23 18:16 ` [PATCH v5 00/13] KVM/X86: Introduce a new guest mapping interface Konrad Rzeszutek Wilk
2019-01-25 18:28 ` Raslan, KarimAllah
2019-01-30 17:14 ` Paolo Bonzini
2019-01-31 13:03 ` Raslan, KarimAllah
2019-01-31 13:10 ` Paolo Bonzini
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20190123173051.GH19289@Konrads-MacBook-Pro.local \
--to=konrad.wilk@oracle.com \
--cc=karahmed@amazon.de \
--cc=kvm@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=pbonzini@redhat.com \
--cc=rkrcmar@redhat.com \
--cc=sironi@amazon.de \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).