All of lore.kernel.org
 help / color / mirror / Atom feed
From: David Matlack <dmatlack@google.com>
To: Anish Moorthy <amoorthy@google.com>
Cc: Sean Christopherson <seanjc@google.com>,
	oliver.upton@linux.dev, kvm@vger.kernel.org,
	 kvmarm@lists.linux.dev, pbonzini@redhat.com, maz@kernel.org,
	 robert.hoo.linux@gmail.com, jthoughton@google.com,
	axelrasmussen@google.com,  peterx@redhat.com,
	nadav.amit@gmail.com, isaku.yamahata@gmail.com,
	 kconsul@linux.vnet.ibm.com
Subject: Re: [PATCH v5 10/17] KVM: Implement KVM_CAP_USERFAULT_ON_MISSING by atomizing __gfn_to_pfn_memslot() calls
Date: Wed, 1 Nov 2023 15:39:58 -0700	[thread overview]
Message-ID: <CALzav=ccmnY1kb3KatbSuYtXzGjsh7xFSKzmfv10pAOGJV3g7Q@mail.gmail.com> (raw)
In-Reply-To: <CAF7b7mqEU0rT9dqq5SXvE+XU0TdCbXWk0OW2ayrW5nBg3M_BFg@mail.gmail.com>

On Wed, Nov 1, 2023 at 3:26 PM Anish Moorthy <amoorthy@google.com> wrote:
>
> On Wed, Nov 1, 2023 at 3:03 PM Sean Christopherson <seanjc@google.com> wrote:
> >
> > On Wed, Nov 01, 2023, Anish Moorthy wrote:
> > > On Wed, Oct 4, 2023 at 6:44 PM Sean Christopherson <seanjc@google.com> wrote:
> > > >
> > > > Eh, the shortlog basically says "do work" with a lot of fancy words.  It really
> > > > just boils down to:
> > > >
> > > >   KVM: Let callers of __gfn_to_pfn_memslot() opt-out of USERFAULT_ON_MISSING
> > >
> > > Proposed commit message for v6:
> > >
> > > > KVM: Implement KVM_CAP EXIT_ON_MISSING by checking memslot flag in __gfn_to_pfn_memslot()
> > > >
> > > > When the slot flag is enabled, forbid __gfn_to_pfn_memslot() from
> > > > faulting in pages for which mappings are absent. However, some callers of
> > > > __gfn_to_pfn_memslot() (such as kvm_vcpu_map()) must be able to opt out
> > > > of this behavior: allow doing so via the new can_exit_on_missing
> > > > parameter.
> > >
> > > Although separately, I don't think the parameter should be named
> > > can_exit_on_missing (or, as you suggested, can_do_userfault)-
> > > __gfn_to_pfn_memslot() shouldn't know or care how its callers are
> > > setting up KVM exits, after all.
> >
> > Why not?  __gfn_to_pfn_memslot() gets passed all kinds of constraints, I don't
> > see how "I can't handle exits to userspace" is any different.
>
> Well the thing is that __gfn_to_pfn_memslot() is orthogonal to KVM
> exits. Its callers are just using it to try resolving a pfn, and what
> they do with the results is up to them.
>
> Put more concretely, __gfn_to_pfn_memslot() has many callers of which
> only two (the stage-2 fault handlers) actually use it to set up a KVM
> exit- how does a parameter named "can_exit_on_missing" make sense to
> its callers in general?

"exit" is shorthand for "return_to_userspace" and all KVM code runs in
the context of ioctls (which eventually return to userspace). So I
think it makes sense in general.

And FWIW I found the MEMSLOT_ACCESS_ enum hard to read and agree a
boolean would be simpler.

  reply	other threads:[~2023-11-01 22:40 UTC|newest]

Thread overview: 53+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-09-08 22:28 [PATCH v5 00/17] Improve KVM + userfaultfd live migration via annotated memory faults Anish Moorthy
2023-09-08 22:28 ` [PATCH v5 01/17] KVM: Clarify documentation of hva_to_pfn()'s 'atomic' parameter Anish Moorthy
2023-09-08 22:28 ` [PATCH v5 02/17] KVM: Add docstrings to __kvm_read/write_guest_page() Anish Moorthy
2023-10-05  1:18   ` Sean Christopherson
2023-09-08 22:28 ` [PATCH v5 03/17] KVM: Simplify error handling in __gfn_to_pfn_memslot() Anish Moorthy
2023-09-08 22:28 ` [PATCH v5 04/17] KVM: Add KVM_CAP_MEMORY_FAULT_INFO Anish Moorthy
2023-10-05  1:14   ` Sean Christopherson
2023-10-05 18:45     ` Anish Moorthy
2023-10-05 22:13       ` Sean Christopherson
2023-10-10 22:58   ` David Matlack
2023-10-10 23:40     ` Sean Christopherson
2023-10-16 17:07       ` David Matlack
2023-10-16 19:14         ` Sean Christopherson
2023-09-08 22:28 ` [PATCH v5 05/17] KVM: Annotate -EFAULTs from kvm_vcpu_read/write_guest_page() Anish Moorthy
2023-09-14  8:04   ` kernel test robot
2023-10-05  1:53   ` Sean Christopherson
2023-10-05 23:03   ` Anish Moorthy
2023-09-08 22:28 ` [PATCH v5 06/17] KVM: x86: Annotate -EFAULTs from kvm_handle_error_pfn() Anish Moorthy
2023-10-05  1:26   ` Sean Christopherson
2023-10-05 23:57     ` Anish Moorthy
2023-10-06  0:36       ` Sean Christopherson
2023-09-08 22:28 ` [PATCH v5 07/17] KVM: arm64: Annotate -EFAULT from user_mem_abort() Anish Moorthy
2023-09-28 21:42   ` Anish Moorthy
2023-10-05  1:26   ` Sean Christopherson
2023-10-10 23:01   ` David Matlack
2023-09-08 22:28 ` [PATCH v5 08/17] KVM: Allow hva_pfn_fast() to resolve read faults Anish Moorthy
2023-09-08 22:28 ` [PATCH v5 09/17] KVM: Introduce KVM_CAP_USERFAULT_ON_MISSING without implementation Anish Moorthy
2023-10-10 23:16   ` David Matlack
2023-10-11 17:54     ` Anish Moorthy
2023-10-16 19:38       ` Sean Christopherson
2023-09-08 22:28 ` [PATCH v5 10/17] KVM: Implement KVM_CAP_USERFAULT_ON_MISSING by atomizing __gfn_to_pfn_memslot() calls Anish Moorthy
2023-10-05  1:44   ` Sean Christopherson
2023-10-05 18:58     ` Anish Moorthy
2023-10-06  0:17       ` Sean Christopherson
2023-10-11 22:04         ` Anish Moorthy
2023-11-01 21:53     ` Anish Moorthy
2023-11-01 22:03       ` Sean Christopherson
2023-11-01 22:25         ` Anish Moorthy
2023-11-01 22:39           ` David Matlack [this message]
2023-11-01 22:42           ` Sean Christopherson
2023-11-02 19:14       ` Anish Moorthy
2023-11-02 20:25         ` Anish Moorthy
2023-11-03 20:05         ` Sean Christopherson
2023-09-08 22:28 ` [PATCH v5 11/17] KVM: x86: Enable KVM_CAP_USERFAULT_ON_MISSING Anish Moorthy
2023-10-05  1:52   ` Sean Christopherson
2023-11-01 22:55     ` Anish Moorthy
2023-11-02 14:31       ` Sean Christopherson
2023-09-08 22:28 ` [PATCH v5 12/17] KVM: arm64: " Anish Moorthy
2023-09-08 22:29 ` [PATCH v5 13/17] KVM: selftests: Report per-vcpu demand paging rate from demand paging test Anish Moorthy
2023-09-08 22:29 ` [PATCH v5 14/17] KVM: selftests: Allow many vCPUs and reader threads per UFFD in " Anish Moorthy
2023-09-08 22:29 ` [PATCH v5 15/17] KVM: selftests: Use EPOLL in userfaultfd_util reader threads and signal errors via TEST_ASSERT Anish Moorthy
2023-09-08 22:29 ` [PATCH v5 16/17] KVM: selftests: Add memslot_flags parameter to memstress_create_vm() Anish Moorthy
2023-09-08 22:29 ` [PATCH v5 17/17] KVM: selftests: Handle memory fault exits in demand_paging_test Anish Moorthy

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='CALzav=ccmnY1kb3KatbSuYtXzGjsh7xFSKzmfv10pAOGJV3g7Q@mail.gmail.com' \
    --to=dmatlack@google.com \
    --cc=amoorthy@google.com \
    --cc=axelrasmussen@google.com \
    --cc=isaku.yamahata@gmail.com \
    --cc=jthoughton@google.com \
    --cc=kconsul@linux.vnet.ibm.com \
    --cc=kvm@vger.kernel.org \
    --cc=kvmarm@lists.linux.dev \
    --cc=maz@kernel.org \
    --cc=nadav.amit@gmail.com \
    --cc=oliver.upton@linux.dev \
    --cc=pbonzini@redhat.com \
    --cc=peterx@redhat.com \
    --cc=robert.hoo.linux@gmail.com \
    --cc=seanjc@google.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.