From: Vincent Donnefort <vdonnefort@google.com>
To: Fuad Tabba <fuad.tabba@linux.dev>
Cc: maz@kernel.org, oliver.upton@linux.dev, joey.gouly@arm.com,
suzuki.poulose@arm.com, yuzenghui@huawei.com,
catalin.marinas@arm.com, will@kernel.org,
linux-arm-kernel@lists.infradead.org, kvmarm@lists.linux.dev,
kernel-team@android.com, qperret@google.com
Subject: Re: [PATCH v2 15/18] KVM: arm64: Alloc pkvm_hyp_vm using pKVM heap allocator
Date: Wed, 15 Jul 2026 13:55:23 +0100 [thread overview]
Message-ID: <aleDOyiP5E8TcoZg@google.com> (raw)
In-Reply-To: <CA+EHjTzNMCMAt=VDnCuR2+Pfz9Qh5HmbEm6tdjpdcPBO+UchEg@mail.gmail.com>
On Wed, Jul 15, 2026 at 01:33:00PM +0100, Fuad Tabba wrote:
> On Mon, 6 Jul 2026 at 18:54, 'Vincent Donnefort' via kernel-team
> <kernel-team@android.com> wrote:
>
> ...
>
> > --- a/arch/arm64/kvm/pkvm.c
> > +++ b/arch/arm64/kvm/pkvm.c
> > @@ -191,8 +191,8 @@ static int __pkvm_create_hyp_vcpu(struct kvm_vcpu *vcpu)
> > */
> > static int __pkvm_create_hyp_vm(struct kvm *kvm)
> > {
> > - size_t pgd_sz, hyp_vm_sz;
> > - void *pgd, *hyp_vm;
> > + size_t pgd_sz;
> > + void *pgd;
> > int ret;
> >
> > if (kvm->created_vcpus < 1)
> > @@ -209,28 +209,15 @@ static int __pkvm_create_hyp_vm(struct kvm *kvm)
> > if (!pgd)
> > return -ENOMEM;
> >
> > - /* Allocate memory to donate to hyp for vm and vcpu pointers. */
> > - hyp_vm_sz = PAGE_ALIGN(size_add(PKVM_HYP_VM_SIZE,
> > - size_mul(sizeof(void *),
> > - kvm->created_vcpus)));
> > - hyp_vm = alloc_pages_exact(hyp_vm_sz, GFP_KERNEL_ACCOUNT);
> > - if (!hyp_vm) {
> > - ret = -ENOMEM;
> > - goto free_pgd;
> > - }
> > -
> > - /* Donate the VM memory to hyp and let hyp initialize it. */
> > - ret = kvm_call_hyp_nvhe(__pkvm_init_vm, kvm, hyp_vm, pgd);
> > + ret = pkvm_call_hyp_req(__pkvm_init_vm, kvm, pgd);
>
> Sashiko might be right on the memcg point: hyp_vm used to be
> GFP_KERNEL_ACCOUNT and now comes from the heap unaccounted, while pgd
> next to it is still accounted, so it's inconsistent.
Yep that is correct but there is no way around that:
* Donated memory can now be shared be shared by several VM (or other parts of
the hypervisor)
* Donated memory is also not reclaimed until the shrinker fires.
So we don't really have a way to account for that meta-data anymore.
(I should at least mention this in the commit message)
>
> If we do want to fix it, would it go in pkvm_hyp_topup() to cover all
> consumers rather than here?
>
> Cheers,
> /fuad
next prev parent reply other threads:[~2026-07-15 12:55 UTC|newest]
Thread overview: 43+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-07-06 17:53 [PATCH v2 00/18] KVM: arm64: Introduce pKVM hypervisor heap allocator Vincent Donnefort
2026-07-06 17:53 ` [PATCH v2 01/18] KVM: arm64: Add __pkvm_private_range_pa Vincent Donnefort
2026-07-14 16:59 ` Fuad Tabba
2026-07-06 17:53 ` [PATCH v2 02/18] KVM: arm64: Add pkvm_remove_mappings Vincent Donnefort
2026-07-14 17:48 ` Fuad Tabba
2026-07-06 17:54 ` [PATCH v2 03/18] KVM: arm64: Add __hyp_allocator_map for the pKVM hyp Vincent Donnefort
2026-07-14 17:53 ` Fuad Tabba
2026-07-15 8:58 ` Vincent Donnefort
2026-07-15 9:24 ` Fuad Tabba
2026-07-06 17:54 ` [PATCH v2 04/18] KVM: arm64: Add a heap allocator " Vincent Donnefort
2026-07-14 18:35 ` Fuad Tabba
2026-07-15 10:59 ` Fuad Tabba
2026-07-06 17:54 ` [PATCH v2 05/18] KVM: arm64: Allow kvm_hyp_memcache usage outside of stage-2 Vincent Donnefort
2026-07-14 19:00 ` Fuad Tabba
2026-07-06 17:54 ` [PATCH v2 06/18] KVM: arm64: Add topup interface for the pKVM heap allocator Vincent Donnefort
2026-07-14 19:28 ` Fuad Tabba
2026-07-06 17:54 ` [PATCH v2 07/18] KVM: arm64: Add pkvm_hyp_req infrastructure Vincent Donnefort
2026-07-14 19:45 ` Fuad Tabba
2026-07-06 17:54 ` [PATCH v2 08/18] KVM: arm64: Handle PKVM_HYP_REQ_HYP_ALLOC request Vincent Donnefort
2026-07-15 6:28 ` Fuad Tabba
2026-07-06 17:54 ` [PATCH v2 09/18] KVM: arm64: Add reclaim interface for the pKVM heap alloc Vincent Donnefort
2026-07-14 20:10 ` Fuad Tabba
2026-07-06 17:54 ` [PATCH v2 10/18] KVM: arm64: Add selftests for the pKVM heap allocator Vincent Donnefort
2026-07-15 11:07 ` Fuad Tabba
2026-07-06 17:54 ` [PATCH v2 11/18] KVM: arm64: Add a shrinker for pKVM Vincent Donnefort
2026-07-15 11:20 ` Fuad Tabba
2026-07-15 11:35 ` Vincent Donnefort
2026-07-06 17:54 ` [PATCH v2 12/18] KVM: arm64: Filter out non-kernel addresses in kern_hyp_va Vincent Donnefort
2026-07-15 11:30 ` Fuad Tabba
2026-07-06 17:54 ` [PATCH v2 13/18] KVM: arm64: Move hyp_vm refcount into the structure Vincent Donnefort
2026-07-15 11:35 ` Fuad Tabba
2026-07-06 17:54 ` [PATCH v2 14/18] KVM: arm64: Use noclear for PGD in __pkvm_init_vm error path Vincent Donnefort
2026-07-15 12:09 ` Fuad Tabba
2026-07-06 17:54 ` [PATCH v2 15/18] KVM: arm64: Alloc pkvm_hyp_vm using pKVM heap allocator Vincent Donnefort
2026-07-15 12:33 ` Fuad Tabba
2026-07-15 12:55 ` Vincent Donnefort [this message]
2026-07-15 12:58 ` Fuad Tabba
2026-07-06 17:54 ` [PATCH v2 16/18] KVM: arm64: Alloc pkvm_hyp_vcpu " Vincent Donnefort
2026-07-15 13:12 ` Fuad Tabba
2026-07-06 17:54 ` [PATCH v2 17/18] KVM: arm64: Reject hyp trace descriptors with fewer CPUs than hyp_nr_cpus Vincent Donnefort
2026-07-15 13:54 ` Fuad Tabba
2026-07-06 17:54 ` [PATCH v2 18/18] KVM: arm64: Alloc simple_buffer_page using pKVM hyp allocator Vincent Donnefort
2026-07-15 15:07 ` Fuad Tabba
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=aleDOyiP5E8TcoZg@google.com \
--to=vdonnefort@google.com \
--cc=catalin.marinas@arm.com \
--cc=fuad.tabba@linux.dev \
--cc=joey.gouly@arm.com \
--cc=kernel-team@android.com \
--cc=kvmarm@lists.linux.dev \
--cc=linux-arm-kernel@lists.infradead.org \
--cc=maz@kernel.org \
--cc=oliver.upton@linux.dev \
--cc=qperret@google.com \
--cc=suzuki.poulose@arm.com \
--cc=will@kernel.org \
--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