From: Gavin Shan <gshan@redhat.com>
To: Will Deacon <will@kernel.org>, kvmarm@lists.cs.columbia.edu
Cc: Marc Zyngier <maz@kernel.org>,
Catalin Marinas <catalin.marinas@arm.com>,
kernel-team@android.com, linux-arm-kernel@lists.infradead.org
Subject: Re: [PATCH v3 05/21] KVM: arm64: Add support for creating kernel-agnostic stage-2 page tables
Date: Wed, 2 Sep 2020 16:40:03 +1000 [thread overview]
Message-ID: <e0b07673-bc32-95e4-be7a-382c80240000@redhat.com> (raw)
In-Reply-To: <20200825093953.26493-6-will@kernel.org>
Hi Will,
On 8/25/20 7:39 PM, Will Deacon wrote:
> Introduce alloc() and free() functions to the generic page-table code
> for guest stage-2 page-tables and plumb these into the existing KVM
> page-table allocator. Subsequent patches will convert other operations
> within the KVM allocator over to the generic code.
>
> Cc: Marc Zyngier <maz@kernel.org>
> Cc: Quentin Perret <qperret@google.com>
> Signed-off-by: Will Deacon <will@kernel.org>
> ---
> arch/arm64/include/asm/kvm_host.h | 1 +
> arch/arm64/include/asm/kvm_pgtable.h | 18 +++++++++
> arch/arm64/kvm/hyp/pgtable.c | 51 ++++++++++++++++++++++++++
> arch/arm64/kvm/mmu.c | 55 +++++++++++++++-------------
> 4 files changed, 99 insertions(+), 26 deletions(-)
>
With the following one question resolved:
Reviewed-by: Gavin Shan <gshan@redhat.com>
> diff --git a/arch/arm64/include/asm/kvm_host.h b/arch/arm64/include/asm/kvm_host.h
> index e52c927aade5..0b7c702b2151 100644
> --- a/arch/arm64/include/asm/kvm_host.h
> +++ b/arch/arm64/include/asm/kvm_host.h
> @@ -81,6 +81,7 @@ struct kvm_s2_mmu {
> */
> pgd_t *pgd;
> phys_addr_t pgd_phys;
> + struct kvm_pgtable *pgt;
>
> /* The last vcpu id that ran on each physical CPU */
> int __percpu *last_vcpu_ran;
> diff --git a/arch/arm64/include/asm/kvm_pgtable.h b/arch/arm64/include/asm/kvm_pgtable.h
> index 2af84ab78cb8..3389f978d573 100644
> --- a/arch/arm64/include/asm/kvm_pgtable.h
> +++ b/arch/arm64/include/asm/kvm_pgtable.h
> @@ -116,6 +116,24 @@ void kvm_pgtable_hyp_destroy(struct kvm_pgtable *pgt);
> int kvm_pgtable_hyp_map(struct kvm_pgtable *pgt, u64 addr, u64 size, u64 phys,
> enum kvm_pgtable_prot prot);
>
> +/**
> + * kvm_pgtable_stage2_init() - Initialise a guest stage-2 page-table.
> + * @pgt: Uninitialised page-table structure to initialise.
> + * @kvm: KVM structure representing the guest virtual machine.
> + *
> + * Return: 0 on success, negative error code on failure.
> + */
> +int kvm_pgtable_stage2_init(struct kvm_pgtable *pgt, struct kvm *kvm);
> +
> +/**
> + * kvm_pgtable_stage2_destroy() - Destroy an unused guest stage-2 page-table.
> + * @pgt: Page-table structure initialised by kvm_pgtable_stage2_init().
> + *
> + * The page-table is assumed to be unreachable by any hardware walkers prior
> + * to freeing and therefore no TLB invalidation is performed.
> + */
> +void kvm_pgtable_stage2_destroy(struct kvm_pgtable *pgt);
> +
> /**
> * kvm_pgtable_walk() - Walk a page-table.
> * @pgt: Page-table structure initialised by kvm_pgtable_*_init().
> diff --git a/arch/arm64/kvm/hyp/pgtable.c b/arch/arm64/kvm/hyp/pgtable.c
> index d75166823ad9..b8550ccaef4d 100644
> --- a/arch/arm64/kvm/hyp/pgtable.c
> +++ b/arch/arm64/kvm/hyp/pgtable.c
> @@ -419,3 +419,54 @@ void kvm_pgtable_hyp_destroy(struct kvm_pgtable *pgt)
> free_page((unsigned long)pgt->pgd);
> pgt->pgd = NULL;
> }
> +
> +int kvm_pgtable_stage2_init(struct kvm_pgtable *pgt, struct kvm *kvm)
> +{
> + size_t pgd_sz;
> + u64 vtcr = kvm->arch.vtcr;
> + u32 ia_bits = VTCR_EL2_IPA(vtcr);
> + u32 sl0 = FIELD_GET(VTCR_EL2_SL0_MASK, vtcr);
> + u32 start_level = VTCR_EL2_TGRAN_SL0_BASE - sl0;
> +
> + pgd_sz = kvm_pgd_pages(ia_bits, start_level) * PAGE_SIZE;
> + pgt->pgd = alloc_pages_exact(pgd_sz, GFP_KERNEL | __GFP_ZERO);
> + if (!pgt->pgd)
> + return -ENOMEM;
> +
> + pgt->ia_bits = ia_bits;
> + pgt->start_level = start_level;
> + pgt->mmu = &kvm->arch.mmu;
> + return 0;
> +}
> +
> +static int stage2_free_walker(u64 addr, u64 end, u32 level, kvm_pte_t *ptep,
> + enum kvm_pgtable_walk_flags flag,
> + void * const arg)
> +{
> + kvm_pte_t pte = *ptep;
> +
> + if (!kvm_pte_valid(pte))
> + return 0;
> +
> + put_page(virt_to_page(ptep));
> +
> + if (kvm_pte_table(pte, level))
> + free_page((unsigned long)kvm_pte_follow(pte));
> +
> + return 0;
> +}
> +
> +void kvm_pgtable_stage2_destroy(struct kvm_pgtable *pgt)
> +{
> + size_t pgd_sz;
> + struct kvm_pgtable_walker walker = {
> + .cb = stage2_free_walker,
> + .flags = KVM_PGTABLE_WALK_LEAF |
> + KVM_PGTABLE_WALK_TABLE_POST,
> + };
> +
> + WARN_ON(kvm_pgtable_walk(pgt, 0, BIT(pgt->ia_bits), &walker));
> + pgd_sz = kvm_pgd_pages(pgt->ia_bits, pgt->start_level) * PAGE_SIZE;
> + free_pages_exact(pgt->pgd, pgd_sz);
> + pgt->pgd = NULL;
> +}
> diff --git a/arch/arm64/kvm/mmu.c b/arch/arm64/kvm/mmu.c
> index fabd72b0c8a4..4607e9ca60a2 100644
> --- a/arch/arm64/kvm/mmu.c
> +++ b/arch/arm64/kvm/mmu.c
> @@ -668,47 +668,49 @@ int create_hyp_exec_mappings(phys_addr_t phys_addr, size_t size,
> * @kvm: The pointer to the KVM structure
> * @mmu: The pointer to the s2 MMU structure
> *
> - * Allocates only the stage-2 HW PGD level table(s) of size defined by
> - * stage2_pgd_size(mmu->kvm).
> - *
> + * Allocates only the stage-2 HW PGD level table(s).
> * Note we don't need locking here as this is only called when the VM is
> * created, which can only be done once.
> */
> int kvm_init_stage2_mmu(struct kvm *kvm, struct kvm_s2_mmu *mmu)
> {
> - phys_addr_t pgd_phys;
> - pgd_t *pgd;
> - int cpu;
> + int cpu, err;
> + struct kvm_pgtable *pgt;
>
> - if (mmu->pgd != NULL) {
> + if (mmu->pgt != NULL) {
> kvm_err("kvm_arch already initialized?\n");
> return -EINVAL;
> }
>
> - /* Allocate the HW PGD, making sure that each page gets its own refcount */
> - pgd = alloc_pages_exact(stage2_pgd_size(kvm), GFP_KERNEL | __GFP_ZERO);
> - if (!pgd)
> + pgt = kzalloc(sizeof(*pgt), GFP_KERNEL);
> + if (!pgt)
> return -ENOMEM;
>
> - pgd_phys = virt_to_phys(pgd);
> - if (WARN_ON(pgd_phys & ~kvm_vttbr_baddr_mask(kvm)))
> - return -EINVAL;
> + err = kvm_pgtable_stage2_init(pgt, kvm);
> + if (err)
> + goto out_free_pgtable;
>
> mmu->last_vcpu_ran = alloc_percpu(typeof(*mmu->last_vcpu_ran));
> if (!mmu->last_vcpu_ran) {
> - free_pages_exact(pgd, stage2_pgd_size(kvm));
> - return -ENOMEM;
> + err = -ENOMEM;
> + goto out_destroy_pgtable;
> }
>
> for_each_possible_cpu(cpu)
> *per_cpu_ptr(mmu->last_vcpu_ran, cpu) = -1;
>
> mmu->kvm = kvm;
> - mmu->pgd = pgd;
> - mmu->pgd_phys = pgd_phys;
> + mmu->pgt = pgt;
> + mmu->pgd_phys = __pa(pgt->pgd);
> + mmu->pgd = (void *)pgt->pgd;
> mmu->vmid.vmid_gen = 0;
> -
> return 0;
> +
> +out_destroy_pgtable:
> + kvm_pgtable_stage2_destroy(pgt);
> +out_free_pgtable:
> + kfree(pgt);
> + return err;
> }
>
kvm_pgtable_stage2_destroy() might not needed here because
the stage2 page pgtable is empty so far. However, it should
be rare to hit the case. If I'm correct, what we need to do
is just freeing the PGDs.
> static void stage2_unmap_memslot(struct kvm *kvm,
> @@ -781,20 +783,21 @@ void stage2_unmap_vm(struct kvm *kvm)
> void kvm_free_stage2_pgd(struct kvm_s2_mmu *mmu)
> {
> struct kvm *kvm = mmu->kvm;
> - void *pgd = NULL;
> + struct kvm_pgtable *pgt = NULL;
>
> spin_lock(&kvm->mmu_lock);
> - if (mmu->pgd) {
> - unmap_stage2_range(mmu, 0, kvm_phys_size(kvm));
> - pgd = READ_ONCE(mmu->pgd);
> + pgt = mmu->pgt;
> + if (pgt) {
> mmu->pgd = NULL;
> + mmu->pgd_phys = 0;
> + mmu->pgt = NULL;
> + free_percpu(mmu->last_vcpu_ran);
> }
> spin_unlock(&kvm->mmu_lock);
>
> - /* Free the HW pgd, one page at a time */
> - if (pgd) {
> - free_pages_exact(pgd, stage2_pgd_size(kvm));
> - free_percpu(mmu->last_vcpu_ran);
> + if (pgt) {
> + kvm_pgtable_stage2_destroy(pgt);
> + kfree(pgt);
> }
> }
>
Thanks,
Gavin
_______________________________________________
kvmarm mailing list
kvmarm@lists.cs.columbia.edu
https://lists.cs.columbia.edu/mailman/listinfo/kvmarm
next prev parent reply other threads:[~2020-09-02 6:40 UTC|newest]
Thread overview: 86+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-08-25 9:39 [PATCH v3 00/21] KVM: arm64: Rewrite page-table code and fault handling Will Deacon
2020-08-25 9:39 ` [PATCH v3 01/21] KVM: arm64: Remove kvm_mmu_free_memory_caches() Will Deacon
2020-08-25 9:39 ` [PATCH v3 02/21] KVM: arm64: Add stand-alone page-table walker infrastructure Will Deacon
2020-08-27 16:27 ` Alexandru Elisei
2020-08-28 15:43 ` Alexandru Elisei
2020-09-02 10:36 ` Will Deacon
2020-08-28 15:51 ` Alexandru Elisei
2020-09-02 10:49 ` Will Deacon
2020-09-02 6:31 ` Gavin Shan
2020-09-02 11:02 ` Will Deacon
2020-09-03 1:11 ` Gavin Shan
2020-08-25 9:39 ` [PATCH v3 03/21] KVM: arm64: Add support for creating kernel-agnostic stage-1 page tables Will Deacon
2020-08-28 15:35 ` Alexandru Elisei
2020-09-02 10:06 ` Will Deacon
2020-08-25 9:39 ` [PATCH v3 04/21] KVM: arm64: Use generic allocator for hyp stage-1 page-tables Will Deacon
2020-08-28 16:32 ` Alexandru Elisei
2020-09-02 11:35 ` Will Deacon
2020-09-02 14:48 ` Alexandru Elisei
2020-08-25 9:39 ` [PATCH v3 05/21] KVM: arm64: Add support for creating kernel-agnostic stage-2 page tables Will Deacon
2020-09-02 6:40 ` Gavin Shan [this message]
2020-09-02 11:30 ` Will Deacon
2020-08-25 9:39 ` [PATCH v3 06/21] KVM: arm64: Add support for stage-2 map()/unmap() in generic page-table Will Deacon
2020-09-01 16:24 ` Alexandru Elisei
2020-09-02 11:46 ` Will Deacon
2020-09-03 2:57 ` Gavin Shan
2020-09-03 5:27 ` Gavin Shan
2020-09-03 11:18 ` Gavin Shan
2020-09-03 12:30 ` Will Deacon
2020-09-03 16:15 ` Will Deacon
2020-09-04 0:47 ` Gavin Shan
2020-08-25 9:39 ` [PATCH v3 07/21] KVM: arm64: Convert kvm_phys_addr_ioremap() to generic page-table API Will Deacon
2020-09-01 17:08 ` Alexandru Elisei
2020-09-02 11:48 ` Will Deacon
2020-09-03 3:57 ` Gavin Shan
2020-08-25 9:39 ` [PATCH v3 08/21] KVM: arm64: Convert kvm_set_spte_hva() " Will Deacon
2020-09-02 15:37 ` Alexandru Elisei
2020-09-03 16:37 ` Will Deacon
2020-09-03 4:13 ` Gavin Shan
2020-08-25 9:39 ` [PATCH v3 09/21] KVM: arm64: Convert unmap_stage2_range() " Will Deacon
2020-09-02 16:23 ` Alexandru Elisei
2020-09-02 18:44 ` Alexandru Elisei
2020-09-03 17:57 ` Will Deacon
2020-09-08 13:07 ` Alexandru Elisei
2020-09-09 10:57 ` Alexandru Elisei
2020-09-03 4:19 ` Gavin Shan
2020-08-25 9:39 ` [PATCH v3 10/21] KVM: arm64: Add support for stage-2 page-aging in generic page-table Will Deacon
2020-09-03 4:33 ` Gavin Shan
2020-09-03 16:48 ` Will Deacon
2020-09-04 1:01 ` Gavin Shan
2020-08-25 9:39 ` [PATCH v3 11/21] KVM: arm64: Convert page-aging and access faults to generic page-table API Will Deacon
2020-09-03 4:37 ` Gavin Shan
2020-08-25 9:39 ` [PATCH v3 12/21] KVM: arm64: Add support for stage-2 write-protect in generic page-table Will Deacon
2020-09-03 4:47 ` Gavin Shan
2020-08-25 9:39 ` [PATCH v3 13/21] KVM: arm64: Convert write-protect operation to generic page-table API Will Deacon
2020-09-03 4:48 ` Gavin Shan
2020-08-25 9:39 ` [PATCH v3 14/21] KVM: arm64: Add support for stage-2 cache flushing in generic page-table Will Deacon
2020-09-03 4:51 ` Gavin Shan
2020-08-25 9:39 ` [PATCH v3 15/21] KVM: arm64: Convert memslot cache-flushing code to generic page-table API Will Deacon
2020-09-03 4:52 ` Gavin Shan
2020-08-25 9:39 ` [PATCH v3 16/21] KVM: arm64: Add support for relaxing stage-2 perms in generic page-table code Will Deacon
2020-09-03 4:55 ` Gavin Shan
2020-08-25 9:39 ` [PATCH v3 17/21] KVM: arm64: Convert user_mem_abort() to generic page-table API Will Deacon
2020-09-03 6:05 ` Gavin Shan
2020-08-25 9:39 ` [PATCH v3 18/21] KVM: arm64: Check the pgt instead of the pgd when modifying page-table Will Deacon
2020-09-03 5:00 ` Gavin Shan
2020-08-25 9:39 ` [PATCH v3 19/21] KVM: arm64: Remove unused page-table code Will Deacon
2020-09-03 6:02 ` Gavin Shan
2020-08-25 9:39 ` [PATCH v3 20/21] KVM: arm64: Remove unused 'pgd' field from 'struct kvm_s2_mmu' Will Deacon
2020-09-03 5:07 ` Gavin Shan
2020-09-03 16:50 ` Will Deacon
2020-09-04 0:59 ` Gavin Shan
2020-09-04 10:02 ` Marc Zyngier
2020-08-25 9:39 ` [PATCH v3 21/21] KVM: arm64: Don't constrain maximum IPA size based on host configuration Will Deacon
2020-09-03 5:09 ` Gavin Shan
2020-08-27 16:26 ` [PATCH v3 00/21] KVM: arm64: Rewrite page-table code and fault handling Alexandru Elisei
2020-09-01 16:15 ` Will Deacon
2020-09-03 7:34 ` Gavin Shan
2020-09-03 11:13 ` Gavin Shan
2020-09-03 11:48 ` Gavin Shan
2020-09-03 12:16 ` Will Deacon
2020-09-04 0:51 ` Gavin Shan
2020-09-04 10:07 ` Marc Zyngier
2020-09-05 3:56 ` Gavin Shan
2020-09-05 9:33 ` Marc Zyngier
2020-09-07 9:27 ` Will Deacon
2020-09-03 18:52 ` Will Deacon
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=e0b07673-bc32-95e4-be7a-382c80240000@redhat.com \
--to=gshan@redhat.com \
--cc=catalin.marinas@arm.com \
--cc=kernel-team@android.com \
--cc=kvmarm@lists.cs.columbia.edu \
--cc=linux-arm-kernel@lists.infradead.org \
--cc=maz@kernel.org \
--cc=will@kernel.org \
/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