From: Fuad Tabba <tabba@google.com>
To: Quentin Perret <qperret@google.com>
Cc: maz@kernel.org, james.morse@arm.com, alexandru.elisei@arm.com,
suzuki.poulose@arm.com, catalin.marinas@arm.com, will@kernel.org,
linux-arm-kernel@lists.infradead.org,
kvmarm@lists.cs.columbia.edu, linux-kernel@vger.kernel.org,
ardb@kernel.org, qwandor@google.com, dbrazdil@google.com,
kernel-team@android.com
Subject: Re: [PATCH v3 06/21] KVM: arm64: Optimize host memory aborts
Date: Mon, 2 Aug 2021 11:37:01 +0200 [thread overview]
Message-ID: <CA+EHjTz20cf78XtAEgTZ4sqaL=N3fY0hd=_=rZge53nUv=vsjQ@mail.gmail.com> (raw)
In-Reply-To: <20210729132818.4091769-7-qperret@google.com>
Hi Quentin,
On Thu, Jul 29, 2021 at 3:28 PM Quentin Perret <qperret@google.com> wrote:
>
> The kvm_pgtable_stage2_find_range() function is used in the host memory
> abort path to try and look for the largest block mapping that can be
> used to map the faulting address. In order to do so, the function
> currently walks the stage-2 page-table and looks for existing
> incompatible mappings within the range of the largest possible block.
> If incompatible mappings are found, it tries the same procedure again,
> but using a smaller block range, and repeats until a matching range is
> found (potentially up to page granularity). While this approach has
> benefits (mostly in the fact that it proactively coalesces host stage-2
> mappings), it can be slow if the ranges are fragmented, and it isn't
> optimized to deal with CPUs faulting on the same IPA as all of them will
> do all the work every time.
>
> To avoid these issues, remove kvm_pgtable_stage2_find_range(), and walk
> the page-table only once in the host_mem_abort() path to find the
> closest leaf to the input address. With this, use the corresponding
> range if it is invalid and not owned by another entity. If a valid leaf
> is found, return -EAGAIN similar to what is done in the
> kvm_pgtable_stage2_map() path to optimize concurrent faults.
>
> Signed-off-by: Quentin Perret <qperret@google.com>
Reviewing the code it seems to work as described, with the lock
assertion ensuring that the caller knows which lock to hold.
Reviewed-by: Fuad Tabba <tabba@google.com>
Thanks,
/fuad
> ---
> arch/arm64/include/asm/kvm_pgtable.h | 30 -----------
> arch/arm64/kvm/hyp/nvhe/mem_protect.c | 45 +++++++++++++++-
> arch/arm64/kvm/hyp/pgtable.c | 74 ---------------------------
> 3 files changed, 44 insertions(+), 105 deletions(-)
>
> diff --git a/arch/arm64/include/asm/kvm_pgtable.h b/arch/arm64/include/asm/kvm_pgtable.h
> index 6938eac72c1f..83c5c97d9eac 100644
> --- a/arch/arm64/include/asm/kvm_pgtable.h
> +++ b/arch/arm64/include/asm/kvm_pgtable.h
> @@ -154,16 +154,6 @@ enum kvm_pgtable_prot {
> #define PAGE_HYP_RO (KVM_PGTABLE_PROT_R)
> #define PAGE_HYP_DEVICE (PAGE_HYP | KVM_PGTABLE_PROT_DEVICE)
>
> -/**
> - * struct kvm_mem_range - Range of Intermediate Physical Addresses
> - * @start: Start of the range.
> - * @end: End of the range.
> - */
> -struct kvm_mem_range {
> - u64 start;
> - u64 end;
> -};
> -
> /**
> * enum kvm_pgtable_walk_flags - Flags to control a depth-first page-table walk.
> * @KVM_PGTABLE_WALK_LEAF: Visit leaf entries, including invalid
> @@ -490,24 +480,4 @@ int kvm_pgtable_walk(struct kvm_pgtable *pgt, u64 addr, u64 size,
> */
> int kvm_pgtable_get_leaf(struct kvm_pgtable *pgt, u64 addr,
> kvm_pte_t *ptep, u32 *level);
> -
> -/**
> - * kvm_pgtable_stage2_find_range() - Find a range of Intermediate Physical
> - * Addresses with compatible permission
> - * attributes.
> - * @pgt: Page-table structure initialised by kvm_pgtable_stage2_init*().
> - * @addr: Address that must be covered by the range.
> - * @prot: Protection attributes that the range must be compatible with.
> - * @range: Range structure used to limit the search space at call time and
> - * that will hold the result.
> - *
> - * The offset of @addr within a page is ignored. An IPA is compatible with @prot
> - * iff its corresponding stage-2 page-table entry has default ownership and, if
> - * valid, is mapped with protection attributes identical to @prot.
> - *
> - * Return: 0 on success, negative error code on failure.
> - */
> -int kvm_pgtable_stage2_find_range(struct kvm_pgtable *pgt, u64 addr,
> - enum kvm_pgtable_prot prot,
> - struct kvm_mem_range *range);
> #endif /* __ARM64_KVM_PGTABLE_H__ */
> diff --git a/arch/arm64/kvm/hyp/nvhe/mem_protect.c b/arch/arm64/kvm/hyp/nvhe/mem_protect.c
> index 74280a753efb..2148d3968aa5 100644
> --- a/arch/arm64/kvm/hyp/nvhe/mem_protect.c
> +++ b/arch/arm64/kvm/hyp/nvhe/mem_protect.c
> @@ -159,6 +159,11 @@ static int host_stage2_unmap_dev_all(void)
> return kvm_pgtable_stage2_unmap(pgt, addr, BIT(pgt->ia_bits) - addr);
> }
>
> +struct kvm_mem_range {
> + u64 start;
> + u64 end;
> +};
> +
> static bool find_mem_range(phys_addr_t addr, struct kvm_mem_range *range)
> {
> int cur, left = 0, right = hyp_memblock_nr;
> @@ -227,6 +232,44 @@ static inline int __host_stage2_idmap(u64 start, u64 end,
> __ret; \
> })
>
> +static inline bool range_included(struct kvm_mem_range *child,
> + struct kvm_mem_range *parent)
> +{
> + return parent->start <= child->start && child->end <= parent->end;
> +}
> +
> +static int host_stage2_adjust_range(u64 addr, struct kvm_mem_range *range)
> +{
> + struct kvm_mem_range cur;
> + kvm_pte_t pte;
> + u32 level;
> + int ret;
> +
> + hyp_assert_lock_held(&host_kvm.lock);
> + ret = kvm_pgtable_get_leaf(&host_kvm.pgt, addr, &pte, &level);
> + if (ret)
> + return ret;
> +
> + if (kvm_pte_valid(pte))
> + return -EAGAIN;
> +
> + if (pte)
> + return -EPERM;
> +
> + do {
> + u64 granule = kvm_granule_size(level);
> + cur.start = ALIGN_DOWN(addr, granule);
> + cur.end = cur.start + granule;
> + level++;
> + } while ((level < KVM_PGTABLE_MAX_LEVELS) &&
> + !(kvm_level_supports_block_mapping(level) &&
> + range_included(&cur, range)));
> +
> + *range = cur;
> +
> + return 0;
> +}
> +
> static int host_stage2_idmap(u64 addr)
> {
> enum kvm_pgtable_prot prot = KVM_PGTABLE_PROT_R | KVM_PGTABLE_PROT_W;
> @@ -238,7 +281,7 @@ static int host_stage2_idmap(u64 addr)
> prot |= KVM_PGTABLE_PROT_X;
>
> hyp_spin_lock(&host_kvm.lock);
> - ret = kvm_pgtable_stage2_find_range(&host_kvm.pgt, addr, prot, &range);
> + ret = host_stage2_adjust_range(addr, &range);
> if (ret)
> goto unlock;
>
> diff --git a/arch/arm64/kvm/hyp/pgtable.c b/arch/arm64/kvm/hyp/pgtable.c
> index 49d768b92997..4dff2ad39ee4 100644
> --- a/arch/arm64/kvm/hyp/pgtable.c
> +++ b/arch/arm64/kvm/hyp/pgtable.c
> @@ -1102,77 +1102,3 @@ void kvm_pgtable_stage2_destroy(struct kvm_pgtable *pgt)
> pgt->mm_ops->free_pages_exact(pgt->pgd, pgd_sz);
> pgt->pgd = NULL;
> }
> -
> -#define KVM_PTE_LEAF_S2_COMPAT_MASK (KVM_PTE_LEAF_ATTR_S2_PERMS | \
> - KVM_PTE_LEAF_ATTR_LO_S2_MEMATTR | \
> - KVM_PTE_LEAF_ATTR_S2_IGNORED)
> -
> -static int stage2_check_permission_walker(u64 addr, u64 end, u32 level,
> - kvm_pte_t *ptep,
> - enum kvm_pgtable_walk_flags flag,
> - void * const arg)
> -{
> - kvm_pte_t old_attr, pte = *ptep, *new_attr = arg;
> -
> - /*
> - * Compatible mappings are either invalid and owned by the page-table
> - * owner (whose id is 0), or valid with matching permission attributes.
> - */
> - if (kvm_pte_valid(pte)) {
> - old_attr = pte & KVM_PTE_LEAF_S2_COMPAT_MASK;
> - if (old_attr != *new_attr)
> - return -EEXIST;
> - } else if (pte) {
> - return -EEXIST;
> - }
> -
> - return 0;
> -}
> -
> -int kvm_pgtable_stage2_find_range(struct kvm_pgtable *pgt, u64 addr,
> - enum kvm_pgtable_prot prot,
> - struct kvm_mem_range *range)
> -{
> - kvm_pte_t attr;
> - struct kvm_pgtable_walker check_perm_walker = {
> - .cb = stage2_check_permission_walker,
> - .flags = KVM_PGTABLE_WALK_LEAF,
> - .arg = &attr,
> - };
> - u64 granule, start, end;
> - u32 level;
> - int ret;
> -
> - ret = stage2_set_prot_attr(pgt, prot, &attr);
> - if (ret)
> - return ret;
> - attr &= KVM_PTE_LEAF_S2_COMPAT_MASK;
> -
> - for (level = pgt->start_level; level < KVM_PGTABLE_MAX_LEVELS; level++) {
> - granule = kvm_granule_size(level);
> - start = ALIGN_DOWN(addr, granule);
> - end = start + granule;
> -
> - if (!kvm_level_supports_block_mapping(level))
> - continue;
> -
> - if (start < range->start || range->end < end)
> - continue;
> -
> - /*
> - * Check the presence of existing mappings with incompatible
> - * permissions within the current block range, and try one level
> - * deeper if one is found.
> - */
> - ret = kvm_pgtable_walk(pgt, start, granule, &check_perm_walker);
> - if (ret != -EEXIST)
> - break;
> - }
> -
> - if (!ret) {
> - range->start = start;
> - range->end = end;
> - }
> -
> - return ret;
> -}
> --
> 2.32.0.432.gabb21c7263-goog
>
next prev parent reply other threads:[~2021-08-02 9:37 UTC|newest]
Thread overview: 45+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-07-29 13:27 [PATCH v3 00/21] Track shared pages at EL2 in protected mode Quentin Perret
2021-07-29 13:27 ` [PATCH v3 01/21] KVM: arm64: Add hyp_spin_is_locked() for basic locking assertions at EL2 Quentin Perret
2021-07-29 13:27 ` [PATCH v3 02/21] KVM: arm64: Introduce hyp_assert_lock_held() Quentin Perret
2021-07-29 13:28 ` [PATCH v3 03/21] KVM: arm64: Provide the host_stage2_try() helper macro Quentin Perret
2021-08-02 9:36 ` Fuad Tabba
2021-07-29 13:28 ` [PATCH v3 04/21] KVM: arm64: Introduce helper to retrieve a PTE and its level Quentin Perret
2021-07-29 13:28 ` [PATCH v3 05/21] KVM: arm64: Expose page-table helpers Quentin Perret
2021-07-29 13:28 ` [PATCH v3 06/21] KVM: arm64: Optimize host memory aborts Quentin Perret
2021-08-02 9:37 ` Fuad Tabba [this message]
2021-07-29 13:28 ` [PATCH v3 07/21] KVM: arm64: Rename KVM_PTE_LEAF_ATTR_S2_IGNORED Quentin Perret
2021-08-02 9:37 ` Fuad Tabba
2021-07-29 13:28 ` [PATCH v3 08/21] KVM: arm64: Don't overwrite software bits with owner id Quentin Perret
2021-08-02 9:38 ` Fuad Tabba
2021-07-29 13:28 ` [PATCH v3 09/21] KVM: arm64: Tolerate re-creating hyp mappings to set software bits Quentin Perret
2021-08-02 9:50 ` Fuad Tabba
2021-07-29 13:28 ` [PATCH v3 10/21] KVM: arm64: Enable forcing page-level stage-2 mappings Quentin Perret
2021-08-02 9:49 ` Fuad Tabba
2021-08-03 10:13 ` Quentin Perret
2021-08-03 10:43 ` Fuad Tabba
2021-07-29 13:28 ` [PATCH v3 11/21] KVM: arm64: Allow populating software bits Quentin Perret
2021-07-29 13:28 ` [PATCH v3 12/21] KVM: arm64: Add helpers to tag shared pages in SW bits Quentin Perret
2021-08-02 10:30 ` Fuad Tabba
2021-07-29 13:28 ` [PATCH v3 13/21] KVM: arm64: Expose host stage-2 manipulation helpers Quentin Perret
2021-08-02 11:13 ` Fuad Tabba
2021-08-03 10:20 ` Quentin Perret
2021-07-29 13:28 ` [PATCH v3 14/21] KVM: arm64: Expose pkvm_hyp_id Quentin Perret
2021-07-29 13:28 ` [PATCH v3 15/21] KVM: arm64: Introduce addr_is_memory() Quentin Perret
2021-08-02 14:52 ` Fuad Tabba
2021-08-03 10:23 ` Quentin Perret
2021-07-29 13:28 ` [PATCH v3 16/21] KVM: arm64: Enable retrieving protections attributes of PTEs Quentin Perret
2021-08-02 14:52 ` Fuad Tabba
2021-08-03 10:24 ` Quentin Perret
2021-07-29 13:28 ` [PATCH v3 17/21] KVM: arm64: Mark host bss and rodata section as shared Quentin Perret
2021-08-03 5:02 ` Fuad Tabba
2021-08-03 10:34 ` Quentin Perret
2021-08-03 10:54 ` Fuad Tabba
2021-07-29 13:28 ` [PATCH v3 18/21] KVM: arm64: Remove __pkvm_mark_hyp Quentin Perret
2021-07-29 13:28 ` [PATCH v3 19/21] KVM: arm64: Refactor protected nVHE stage-1 locking Quentin Perret
2021-08-03 5:31 ` Fuad Tabba
2021-08-03 10:37 ` Quentin Perret
2021-08-03 10:51 ` Fuad Tabba
2021-07-29 13:28 ` [PATCH v3 20/21] KVM: arm64: Restrict EL2 stage-1 changes in protected mode Quentin Perret
2021-08-03 8:22 ` Fuad Tabba
2021-08-03 10:43 ` Quentin Perret
2021-07-29 13:28 ` [PATCH v3 21/21] KVM: arm64: Make __pkvm_create_mappings static Quentin Perret
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='CA+EHjTz20cf78XtAEgTZ4sqaL=N3fY0hd=_=rZge53nUv=vsjQ@mail.gmail.com' \
--to=tabba@google.com \
--cc=alexandru.elisei@arm.com \
--cc=ardb@kernel.org \
--cc=catalin.marinas@arm.com \
--cc=dbrazdil@google.com \
--cc=james.morse@arm.com \
--cc=kernel-team@android.com \
--cc=kvmarm@lists.cs.columbia.edu \
--cc=linux-arm-kernel@lists.infradead.org \
--cc=linux-kernel@vger.kernel.org \
--cc=maz@kernel.org \
--cc=qperret@google.com \
--cc=qwandor@google.com \
--cc=suzuki.poulose@arm.com \
--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;
as well as URLs for NNTP newsgroup(s).