From: David Hildenbrand <david@redhat.com>
To: Quentin Perret <qperret@google.com>
Cc: Fuad Tabba <tabba@google.com>,
Matthew Wilcox <willy@infradead.org>,
kvm@vger.kernel.org, kvmarm@lists.linux.dev, pbonzini@redhat.com,
chenhuacai@kernel.org, mpe@ellerman.id.au, anup@brainfault.org,
paul.walmsley@sifive.com, palmer@dabbelt.com,
aou@eecs.berkeley.edu, seanjc@google.com, brauner@kernel.org,
akpm@linux-foundation.org, xiaoyao.li@intel.com,
yilun.xu@intel.com, chao.p.peng@linux.intel.com,
jarkko@kernel.org, amoorthy@google.com, dmatlack@google.com,
yu.c.zhang@linux.intel.com, isaku.yamahata@intel.com,
mic@digikod.net, vbabka@suse.cz, vannapurve@google.com,
ackerleytng@google.com, mail@maciej.szmigiero.name,
michael.roth@amd.com, wei.w.wang@intel.com,
liam.merwick@oracle.com, isaku.yamahata@gmail.com,
kirill.shutemov@linux.intel.com, suzuki.poulose@arm.com,
steven.price@arm.com, quic_mnalajal@quicinc.com,
quic_tsoni@quicinc.com, quic_svaddagi@quicinc.com,
quic_cvanscha@quicinc.com, quic_pderrin@quicinc.com,
quic_pheragu@quicinc.com, catalin.marinas@arm.com,
james.morse@arm.com, yuzenghui@huawei.com,
oliver.upton@linux.dev, maz@kernel.org, will@kernel.org,
keirf@google.com, linux-mm@kvack.org
Subject: Re: folio_mmapped
Date: Mon, 4 Mar 2024 21:22:44 +0100 [thread overview]
Message-ID: <2773d2e1-9d73-4d38-97f1-4d90c17b2e26@redhat.com> (raw)
In-Reply-To: <ZeXEUMPn27J5je8T@google.com>
On 04.03.24 13:53, Quentin Perret wrote:
> On Friday 01 Mar 2024 at 12:16:54 (+0100), David Hildenbrand wrote:
>>>> I don't think that we can assume that only a single VMA covers a page.
>>>>
>>>>> But of course, no rmap walk is always better.
>>>>
>>>> We've been thinking some more about how to handle the case where the
>>>> host userspace has a mapping of a page that later becomes private.
>>>>
>>>> One idea is to refuse to run the guest (i.e., exit vcpu_run() to back
>>>> to the host with a meaningful exit reason) until the host unmaps that
>>>> page, and check for the refcount to the page as you mentioned earlier.
>>>> This is essentially what the RFC I sent does (minus the bugs :) ) .
>>>>
>>>> The other idea is to use the rmap walk as you suggested to zap that
>>>> page. If the host tries to access that page again, it would get a
>>>> SIGBUS on the fault. This has the advantage that, as you'd mentioned,
>>>> the host doesn't need to constantly mmap() and munmap() pages. It
>>>> could potentially be optimised further as suggested if we have a
>>>> cooperating VMM that would issue a MADV_DONTNEED or something like
>>>> that, but that's just an optimisation and we would still need to have
>>>> the option of the rmap walk. However, I was wondering how practical
>>>> this idea would be if more than a single VMA covers a page?
>>>>
>>>
>>> Agree with all your points here. I changed Gunyah's implementation to do
>>> the unmap instead of erroring out. I didn't observe a significant
>>> performance difference. However, doing unmap might be a little faster
>>> because we can check folio_mapped() before doing the rmap walk. When
>>> erroring out at mmap() level, we always have to do the walk.
>>
>> Right. On the mmap() level you won't really have to walk page tables, as the
>> the munmap() already zapped the page and removed the "problematic" VMA.
>>
>> Likely, you really want to avoid repeatedly calling mmap()+munmap() just to
>> access shared memory; but that's just my best guess about your user space
>> app :)
>
> Ack, and expecting userspace to munmap the pages whenever we hit a valid
> mapping in userspace page-tables in the KVM faults path makes for a
> somewhat unusual interface IMO. Userspace can munmap, mmap again, and if
> it doesn't touch the pages, it can proceed to run the guest just fine,
> is that the expectation? If so, it feels like we're 'leaking' internal
It would be weird, and I would not suggest that. It's either
(1) you can leave it mmap'ed, but any access to private memory will
SIGBUS. The kernel will try zapping pages inside a VMA itself.
(2) you cannot leave it mmap'ed. In order to convert shared -> private,
you have to munmap. mmap will fail if it would cover a currently-private
page.
So for (1) you could mmap once in user space and be done with it. For
(2) you would have to mmap+munmap when accessing shared memory.
> kernel state somehow. The kernel is normally well within its rights to
> zap userspace mappings if it wants to e.g. swap. (Obviously mlock is a
> weird case, but even in that case, IIRC the kernel still has a certain
> amount of flexibility and can use compaction and friends). Similarly,
> it should be well within its right to proactively create them. How
> would this scheme work if, 10 years from now, something like
> Speculative Page Faults makes it into the kernel in a different form?
At least with (2), speculative page faults would never generate a
SIGBUS. Just like HW must not generate a fault on speculative access.
>
> Not requiring to userspace to unmap makes the userspace interface a lot
> simpler I think -- once a protected guest starts, you better not touch
> its memory if it's not been shared back or you'll get slapped on the
> wrist. Whether or not those pages have been accessed beforehand for
> example is irrelevant.
Yes. So the theory :)
--
Cheers,
David / dhildenb
next prev parent reply other threads:[~2024-03-04 20:22 UTC|newest]
Thread overview: 96+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-02-22 16:10 [RFC PATCH v1 00/26] KVM: Restricted mapping of guest_memfd at the host and pKVM/arm64 support Fuad Tabba
2024-02-22 16:10 ` [RFC PATCH v1 01/26] KVM: Split KVM memory attributes into user and kernel attributes Fuad Tabba
2024-02-22 16:10 ` [RFC PATCH v1 02/26] KVM: Introduce kvm_gmem_get_pfn_locked(), which retains the folio lock Fuad Tabba
2024-02-22 16:10 ` [RFC PATCH v1 03/26] KVM: Add restricted support for mapping guestmem by the host Fuad Tabba
2024-02-22 16:28 ` David Hildenbrand
2024-02-26 8:58 ` Fuad Tabba
2024-02-26 9:57 ` David Hildenbrand
2024-02-26 17:30 ` Fuad Tabba
2024-02-27 7:40 ` David Hildenbrand
2024-02-22 16:10 ` [RFC PATCH v1 04/26] KVM: Don't allow private attribute to be set if mapped by host Fuad Tabba
2024-04-17 23:27 ` Sean Christopherson
2024-04-18 10:54 ` David Hildenbrand
2024-02-22 16:10 ` [RFC PATCH v1 05/26] KVM: Don't allow private attribute to be removed for unmappable memory Fuad Tabba
2024-02-22 16:10 ` [RFC PATCH v1 06/26] KVM: Implement kvm_(read|/write)_guest_page for private memory slots Fuad Tabba
2024-02-22 16:10 ` [RFC PATCH v1 07/26] KVM: arm64: Turn llist of pinned pages into an rb-tree Fuad Tabba
2024-02-22 16:10 ` [RFC PATCH v1 08/26] KVM: arm64: Implement MEM_RELINQUISH SMCCC hypercall Fuad Tabba
2024-02-22 16:10 ` [RFC PATCH v1 09/26] KVM: arm64: Strictly check page type in MEM_RELINQUISH hypercall Fuad Tabba
2024-02-22 16:10 ` [RFC PATCH v1 10/26] KVM: arm64: Avoid unnecessary unmap walk " Fuad Tabba
2024-02-22 16:10 ` [RFC PATCH v1 11/26] KVM: arm64: Add initial support for KVM_CAP_EXIT_HYPERCALL Fuad Tabba
2024-02-22 16:10 ` [RFC PATCH v1 12/26] KVM: arm64: Allow userspace to receive SHARE and UNSHARE notifications Fuad Tabba
2024-02-22 16:10 ` [RFC PATCH v1 13/26] KVM: arm64: Create hypercall return handler Fuad Tabba
2024-02-22 16:10 ` [RFC PATCH v1 14/26] KVM: arm64: Refactor code around handling return from host to guest Fuad Tabba
2024-02-22 16:10 ` [RFC PATCH v1 15/26] KVM: arm64: Rename kvm_pinned_page to kvm_guest_page Fuad Tabba
2024-02-22 16:10 ` [RFC PATCH v1 16/26] KVM: arm64: Add a field to indicate whether the guest page was pinned Fuad Tabba
2024-02-22 16:10 ` [RFC PATCH v1 17/26] KVM: arm64: Do not allow changes to private memory slots Fuad Tabba
2024-02-22 16:10 ` [RFC PATCH v1 18/26] KVM: arm64: Skip VMA checks for slots without userspace address Fuad Tabba
2024-02-22 16:10 ` [RFC PATCH v1 19/26] KVM: arm64: Handle guest_memfd()-backed guest page faults Fuad Tabba
2024-02-22 16:10 ` [RFC PATCH v1 20/26] KVM: arm64: Track sharing of memory from protected guest to host Fuad Tabba
2024-02-22 16:10 ` [RFC PATCH v1 21/26] KVM: arm64: Mark a protected VM's memory as unmappable at initialization Fuad Tabba
2024-02-22 16:10 ` [RFC PATCH v1 22/26] KVM: arm64: Handle unshare on way back to guest entry rather than exit Fuad Tabba
2024-02-22 16:10 ` [RFC PATCH v1 23/26] KVM: arm64: Check that host unmaps memory unshared by guest Fuad Tabba
2024-02-22 16:10 ` [RFC PATCH v1 24/26] KVM: arm64: Add handlers for kvm_arch_*_set_memory_attributes() Fuad Tabba
2024-02-22 16:10 ` [RFC PATCH v1 25/26] KVM: arm64: Enable private memory support when pKVM is enabled Fuad Tabba
2024-02-22 16:10 ` [RFC PATCH v1 26/26] KVM: arm64: Enable private memory kconfig for arm64 Fuad Tabba
2024-02-22 23:43 ` [RFC PATCH v1 00/26] KVM: Restricted mapping of guest_memfd at the host and pKVM/arm64 support Elliot Berman
2024-02-23 0:35 ` folio_mmapped Matthew Wilcox
2024-02-26 9:28 ` folio_mmapped David Hildenbrand
2024-02-26 21:14 ` folio_mmapped Elliot Berman
2024-02-27 14:59 ` folio_mmapped David Hildenbrand
2024-02-28 10:48 ` folio_mmapped Quentin Perret
2024-02-28 11:11 ` folio_mmapped David Hildenbrand
2024-02-28 12:44 ` folio_mmapped Quentin Perret
2024-02-28 13:00 ` folio_mmapped David Hildenbrand
2024-02-28 13:34 ` folio_mmapped Quentin Perret
2024-02-28 18:43 ` folio_mmapped Elliot Berman
2024-02-28 18:51 ` Quentin Perret
2024-02-29 10:04 ` folio_mmapped David Hildenbrand
2024-02-29 19:01 ` folio_mmapped Fuad Tabba
2024-03-01 0:40 ` folio_mmapped Elliot Berman
2024-03-01 11:16 ` folio_mmapped David Hildenbrand
2024-03-04 12:53 ` folio_mmapped Quentin Perret
2024-03-04 20:22 ` David Hildenbrand [this message]
2024-03-01 11:06 ` folio_mmapped David Hildenbrand
2024-03-04 12:36 ` folio_mmapped Quentin Perret
2024-03-04 19:04 ` folio_mmapped Sean Christopherson
2024-03-04 20:17 ` folio_mmapped David Hildenbrand
2024-03-04 21:43 ` folio_mmapped Elliot Berman
2024-03-04 21:58 ` folio_mmapped David Hildenbrand
2024-03-19 9:47 ` folio_mmapped Quentin Perret
2024-03-19 9:54 ` folio_mmapped David Hildenbrand
2024-03-18 17:06 ` folio_mmapped Vishal Annapurve
2024-03-18 22:02 ` folio_mmapped David Hildenbrand
2024-03-18 23:07 ` folio_mmapped Vishal Annapurve
2024-03-19 0:10 ` folio_mmapped Sean Christopherson
2024-03-19 10:26 ` folio_mmapped David Hildenbrand
2024-03-19 13:19 ` folio_mmapped David Hildenbrand
2024-03-19 14:31 ` folio_mmapped Will Deacon
2024-03-19 23:54 ` folio_mmapped Elliot Berman
2024-03-22 16:36 ` Will Deacon
2024-03-22 18:46 ` Elliot Berman
2024-03-27 19:31 ` Will Deacon
2024-03-22 17:52 ` folio_mmapped David Hildenbrand
2024-03-22 21:21 ` folio_mmapped David Hildenbrand
2024-03-26 22:04 ` folio_mmapped Elliot Berman
2024-03-27 17:50 ` folio_mmapped David Hildenbrand
2024-03-27 19:34 ` folio_mmapped Will Deacon
2024-03-28 9:06 ` folio_mmapped David Hildenbrand
2024-03-28 10:10 ` folio_mmapped Quentin Perret
2024-03-28 10:32 ` folio_mmapped David Hildenbrand
2024-03-28 10:58 ` folio_mmapped Quentin Perret
2024-03-28 11:41 ` folio_mmapped David Hildenbrand
2024-03-29 18:38 ` folio_mmapped Vishal Annapurve
2024-04-04 0:15 ` folio_mmapped Sean Christopherson
2024-03-19 15:04 ` folio_mmapped Sean Christopherson
2024-03-22 17:16 ` folio_mmapped David Hildenbrand
2024-02-26 9:03 ` [RFC PATCH v1 00/26] KVM: Restricted mapping of guest_memfd at the host and pKVM/arm64 support Fuad Tabba
2024-02-23 12:00 ` Alexandru Elisei
2024-02-26 9:05 ` Fuad Tabba
2024-02-26 9:47 ` David Hildenbrand
2024-02-27 9:37 ` Fuad Tabba
2024-02-27 14:41 ` David Hildenbrand
2024-02-27 14:49 ` David Hildenbrand
2024-02-28 9:57 ` Fuad Tabba
2024-02-28 10:12 ` David Hildenbrand
2024-02-28 14:01 ` Quentin Perret
2024-02-29 9:51 ` David Hildenbrand
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=2773d2e1-9d73-4d38-97f1-4d90c17b2e26@redhat.com \
--to=david@redhat.com \
--cc=ackerleytng@google.com \
--cc=akpm@linux-foundation.org \
--cc=amoorthy@google.com \
--cc=anup@brainfault.org \
--cc=aou@eecs.berkeley.edu \
--cc=brauner@kernel.org \
--cc=catalin.marinas@arm.com \
--cc=chao.p.peng@linux.intel.com \
--cc=chenhuacai@kernel.org \
--cc=dmatlack@google.com \
--cc=isaku.yamahata@gmail.com \
--cc=isaku.yamahata@intel.com \
--cc=james.morse@arm.com \
--cc=jarkko@kernel.org \
--cc=keirf@google.com \
--cc=kirill.shutemov@linux.intel.com \
--cc=kvm@vger.kernel.org \
--cc=kvmarm@lists.linux.dev \
--cc=liam.merwick@oracle.com \
--cc=linux-mm@kvack.org \
--cc=mail@maciej.szmigiero.name \
--cc=maz@kernel.org \
--cc=mic@digikod.net \
--cc=michael.roth@amd.com \
--cc=mpe@ellerman.id.au \
--cc=oliver.upton@linux.dev \
--cc=palmer@dabbelt.com \
--cc=paul.walmsley@sifive.com \
--cc=pbonzini@redhat.com \
--cc=qperret@google.com \
--cc=quic_cvanscha@quicinc.com \
--cc=quic_mnalajal@quicinc.com \
--cc=quic_pderrin@quicinc.com \
--cc=quic_pheragu@quicinc.com \
--cc=quic_svaddagi@quicinc.com \
--cc=quic_tsoni@quicinc.com \
--cc=seanjc@google.com \
--cc=steven.price@arm.com \
--cc=suzuki.poulose@arm.com \
--cc=tabba@google.com \
--cc=vannapurve@google.com \
--cc=vbabka@suse.cz \
--cc=wei.w.wang@intel.com \
--cc=will@kernel.org \
--cc=willy@infradead.org \
--cc=xiaoyao.li@intel.com \
--cc=yilun.xu@intel.com \
--cc=yu.c.zhang@linux.intel.com \
--cc=yuzenghui@huawei.com \
/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 read-only IMAP folder(s) and NNTP newsgroup(s).