| Message ID | 20230729004144.1054885-1-seanjc@google.com (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
| Series | KVM: Wrap kvm_{gfn,hva}_range.pte in a per-action union | expand |
On Fri, Jul 28, 2023 at 05:41:44PM -0700, Sean Christopherson wrote: > Wrap kvm_{gfn,hva}_range.pte in a union so that future notifier events can > pass event specific information up and down the stack without needing to > constantly expand and churn the APIs. Lockless aging of SPTEs will pass > around a bitmap, and support for memory attributes will pass around the > new attributes for the range. > > Add a "KVM_NO_ARG" placeholder to simplify handling events without an > argument (creating a dummy union variable is midly annoying). > > Opportunstically drop explicit zero-initialization of the "pte" field, as > omitting the field (now a union) has the same effect. > > Cc: Yu Zhao <yuzhao@google.com> > Link: https://lore.kernel.org/all/CAOUHufagkd2Jk3_HrVoFFptRXM=hX2CV8f+M-dka-hJU4bP8kw@mail.gmail.com > Signed-off-by: Sean Christopherson <seanjc@google.com> Looks good, and I don't think it'll conflict with anything on the arm64 side. Reviewed-by: Oliver Upton <oliver.upton@linux.dev> > --- > > If this looks good, my thought is to squeeze it into 6.6 so that the MGLRU > and guest_memfd() series can build on it. Or those series could just > include it? Eh, I'm not a huge fan of having two series independently reposting a common base. It can be a bit annoying when the two authors have slightly different interpretations on how to improve it... > diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c > index dfbaafbe3a00..f84ef9399aee 100644 > --- a/virt/kvm/kvm_main.c > +++ b/virt/kvm/kvm_main.c > @@ -526,7 +526,7 @@ typedef void (*on_unlock_fn_t)(struct kvm *kvm); > struct kvm_hva_range { > unsigned long start; > unsigned long end; > - pte_t pte; > + union kvm_mmu_notifier_arg arg; > hva_handler_t handler; > on_lock_fn_t on_lock; > on_unlock_fn_t on_unlock; > @@ -547,6 +547,8 @@ static void kvm_null_fn(void) > } > #define IS_KVM_NULL_FN(fn) ((fn) == (void *)kvm_null_fn) > > +static const union kvm_mmu_notifier_arg KVM_NO_ARG; > + I'm guessing you were trying to keep this short, but it might be nice to use MMU_NOTIFIER_ (or similar) as the prefix to make the scope immediately obvious.
On Mon, Jul 31, 2023, Oliver Upton wrote: > On Fri, Jul 28, 2023 at 05:41:44PM -0700, Sean Christopherson wrote: > > If this looks good, my thought is to squeeze it into 6.6 so that the MGLRU > > and guest_memfd() series can build on it. Or those series could just > > include it? > > Eh, I'm not a huge fan of having two series independently reposting a > common base. It can be a bit annoying when the two authors have slightly > different interpretations on how to improve it... That suggests that there's something to improve upon ;-) > > diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c > > index dfbaafbe3a00..f84ef9399aee 100644 > > --- a/virt/kvm/kvm_main.c > > +++ b/virt/kvm/kvm_main.c > > @@ -526,7 +526,7 @@ typedef void (*on_unlock_fn_t)(struct kvm *kvm); > > struct kvm_hva_range { > > unsigned long start; > > unsigned long end; > > - pte_t pte; > > + union kvm_mmu_notifier_arg arg; > > hva_handler_t handler; > > on_lock_fn_t on_lock; > > on_unlock_fn_t on_unlock; > > @@ -547,6 +547,8 @@ static void kvm_null_fn(void) > > } > > #define IS_KVM_NULL_FN(fn) ((fn) == (void *)kvm_null_fn) > > > > +static const union kvm_mmu_notifier_arg KVM_NO_ARG; > > + > > I'm guessing you were trying to keep this short, but it might be nice to > use MMU_NOTIFIER_ (or similar) as the prefix to make the scope > immediately obvious. Yeah, agreed, it's worth the extra line in kvm_mmu_notifier_clear_flush_young().
On Fri, Jul 28, 2023 at 6:41 PM Sean Christopherson <seanjc@google.com> wrote: > > Wrap kvm_{gfn,hva}_range.pte in a union so that future notifier events can > pass event specific information up and down the stack without needing to > constantly expand and churn the APIs. Lockless aging of SPTEs will pass > around a bitmap, and support for memory attributes will pass around the > new attributes for the range. > > Add a "KVM_NO_ARG" placeholder to simplify handling events without an > argument (creating a dummy union variable is midly annoying). > > Opportunstically drop explicit zero-initialization of the "pte" field, as > omitting the field (now a union) has the same effect. > > Cc: Yu Zhao <yuzhao@google.com> > Link: https://lore.kernel.org/all/CAOUHufagkd2Jk3_HrVoFFptRXM=hX2CV8f+M-dka-hJU4bP8kw@mail.gmail.com > Signed-off-by: Sean Christopherson <seanjc@google.com> Acked-by: Yu Zhao <yuzhao@google.com>
On Fri, 28 Jul 2023 17:41:44 -0700, Sean Christopherson wrote: > Wrap kvm_{gfn,hva}_range.pte in a union so that future notifier events can > pass event specific information up and down the stack without needing to > constantly expand and churn the APIs. Lockless aging of SPTEs will pass > around a bitmap, and support for memory attributes will pass around the > new attributes for the range. > > Add a "KVM_NO_ARG" placeholder to simplify handling events without an > argument (creating a dummy union variable is midly annoying). > > [...] Applied to kvm-x86 generic, with KVM_MMU_NOTIFIER_NO_ARG as suggested by Oliver. [1/1] KVM: Wrap kvm_{gfn,hva}_range.pte in a per-action union https://github.com/kvm-x86/linux/commit/3e1efe2b67d3 -- https://github.com/kvm-x86/linux/tree/next https://github.com/kvm-x86/linux/tree/fixes
diff --git a/arch/arm64/kvm/mmu.c b/arch/arm64/kvm/mmu.c index 6db9ef288ec3..55f03a68f1cd 100644 --- a/arch/arm64/kvm/mmu.c +++ b/arch/arm64/kvm/mmu.c @@ -1721,7 +1721,7 @@ bool kvm_unmap_gfn_range(struct kvm *kvm, struct kvm_gfn_range *range) bool kvm_set_spte_gfn(struct kvm *kvm, struct kvm_gfn_range *range) { - kvm_pfn_t pfn = pte_pfn(range->pte); + kvm_pfn_t pfn = pte_pfn(range->arg.pte); if (!kvm->arch.mmu.pgt) return false; diff --git a/arch/mips/kvm/mmu.c b/arch/mips/kvm/mmu.c index e8c08988ed37..7b2ac1319d70 100644 --- a/arch/mips/kvm/mmu.c +++ b/arch/mips/kvm/mmu.c @@ -447,7 +447,7 @@ bool kvm_unmap_gfn_range(struct kvm *kvm, struct kvm_gfn_range *range) bool kvm_set_spte_gfn(struct kvm *kvm, struct kvm_gfn_range *range) { gpa_t gpa = range->start << PAGE_SHIFT; - pte_t hva_pte = range->pte; + pte_t hva_pte = range->arg.pte; pte_t *gpa_pte = kvm_mips_pte_for_gpa(kvm, NULL, gpa); pte_t old_pte; diff --git a/arch/riscv/kvm/mmu.c b/arch/riscv/kvm/mmu.c index f2eb47925806..857f4312b0f8 100644 --- a/arch/riscv/kvm/mmu.c +++ b/arch/riscv/kvm/mmu.c @@ -559,7 +559,7 @@ bool kvm_unmap_gfn_range(struct kvm *kvm, struct kvm_gfn_range *range) bool kvm_set_spte_gfn(struct kvm *kvm, struct kvm_gfn_range *range) { int ret; - kvm_pfn_t pfn = pte_pfn(range->pte); + kvm_pfn_t pfn = pte_pfn(range->arg.pte); if (!kvm->arch.pgd) return false; diff --git a/arch/x86/kvm/mmu/mmu.c b/arch/x86/kvm/mmu/mmu.c index ec169f5c7dce..d72f2b20f430 100644 --- a/arch/x86/kvm/mmu/mmu.c +++ b/arch/x86/kvm/mmu/mmu.c @@ -1588,7 +1588,7 @@ static __always_inline bool kvm_handle_gfn_range(struct kvm *kvm, for_each_slot_rmap_range(range->slot, PG_LEVEL_4K, KVM_MAX_HUGEPAGE_LEVEL, range->start, range->end - 1, &iterator) ret |= handler(kvm, iterator.rmap, range->slot, iterator.gfn, - iterator.level, range->pte); + iterator.level, range->arg.pte); return ret; } diff --git a/arch/x86/kvm/mmu/tdp_mmu.c b/arch/x86/kvm/mmu/tdp_mmu.c index 512163d52194..6250bd3d20c1 100644 --- a/arch/x86/kvm/mmu/tdp_mmu.c +++ b/arch/x86/kvm/mmu/tdp_mmu.c @@ -1241,7 +1241,7 @@ static bool set_spte_gfn(struct kvm *kvm, struct tdp_iter *iter, u64 new_spte; /* Huge pages aren't expected to be modified without first being zapped. */ - WARN_ON(pte_huge(range->pte) || range->start + 1 != range->end); + WARN_ON(pte_huge(range->arg.pte) || range->start + 1 != range->end); if (iter->level != PG_LEVEL_4K || !is_shadow_present_pte(iter->old_spte)) @@ -1255,9 +1255,9 @@ static bool set_spte_gfn(struct kvm *kvm, struct tdp_iter *iter, */ tdp_mmu_iter_set_spte(kvm, iter, 0); - if (!pte_write(range->pte)) { + if (!pte_write(range->arg.pte)) { new_spte = kvm_mmu_changed_pte_notifier_make_spte(iter->old_spte, - pte_pfn(range->pte)); + pte_pfn(range->arg.pte)); tdp_mmu_iter_set_spte(kvm, iter, new_spte); } diff --git a/include/linux/kvm_host.h b/include/linux/kvm_host.h index 9d3ac7720da9..9125d0ab642d 100644 --- a/include/linux/kvm_host.h +++ b/include/linux/kvm_host.h @@ -256,11 +256,15 @@ int kvm_async_pf_wakeup_all(struct kvm_vcpu *vcpu); #endif #ifdef KVM_ARCH_WANT_MMU_NOTIFIER +union kvm_mmu_notifier_arg { + pte_t pte; +}; + struct kvm_gfn_range { struct kvm_memory_slot *slot; gfn_t start; gfn_t end; - pte_t pte; + union kvm_mmu_notifier_arg arg; bool may_block; }; bool kvm_unmap_gfn_range(struct kvm *kvm, struct kvm_gfn_range *range); diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c index dfbaafbe3a00..f84ef9399aee 100644 --- a/virt/kvm/kvm_main.c +++ b/virt/kvm/kvm_main.c @@ -526,7 +526,7 @@ typedef void (*on_unlock_fn_t)(struct kvm *kvm); struct kvm_hva_range { unsigned long start; unsigned long end; - pte_t pte; + union kvm_mmu_notifier_arg arg; hva_handler_t handler; on_lock_fn_t on_lock; on_unlock_fn_t on_unlock; @@ -547,6 +547,8 @@ static void kvm_null_fn(void) } #define IS_KVM_NULL_FN(fn) ((fn) == (void *)kvm_null_fn) +static const union kvm_mmu_notifier_arg KVM_NO_ARG; + /* Iterate over each memslot intersecting [start, last] (inclusive) range */ #define kvm_for_each_memslot_in_hva_range(node, slots, start, last) \ for (node = interval_tree_iter_first(&slots->hva_tree, start, last); \ @@ -591,7 +593,7 @@ static __always_inline int __kvm_handle_hva_range(struct kvm *kvm, * bother making these conditional (to avoid writes on * the second or later invocation of the handler). */ - gfn_range.pte = range->pte; + gfn_range.arg = range->arg; gfn_range.may_block = range->may_block; /* @@ -632,14 +634,14 @@ static __always_inline int __kvm_handle_hva_range(struct kvm *kvm, static __always_inline int kvm_handle_hva_range(struct mmu_notifier *mn, unsigned long start, unsigned long end, - pte_t pte, + union kvm_mmu_notifier_arg arg, hva_handler_t handler) { struct kvm *kvm = mmu_notifier_to_kvm(mn); const struct kvm_hva_range range = { .start = start, .end = end, - .pte = pte, + .arg = arg, .handler = handler, .on_lock = (void *)kvm_null_fn, .on_unlock = (void *)kvm_null_fn, @@ -659,7 +661,6 @@ static __always_inline int kvm_handle_hva_range_no_flush(struct mmu_notifier *mn const struct kvm_hva_range range = { .start = start, .end = end, - .pte = __pte(0), .handler = handler, .on_lock = (void *)kvm_null_fn, .on_unlock = (void *)kvm_null_fn, @@ -693,6 +694,7 @@ static void kvm_mmu_notifier_change_pte(struct mmu_notifier *mn, pte_t pte) { struct kvm *kvm = mmu_notifier_to_kvm(mn); + const union kvm_mmu_notifier_arg arg = { .pte = pte }; trace_kvm_set_spte_hva(address); @@ -708,7 +710,7 @@ static void kvm_mmu_notifier_change_pte(struct mmu_notifier *mn, if (!READ_ONCE(kvm->mmu_invalidate_in_progress)) return; - kvm_handle_hva_range(mn, address, address + 1, pte, kvm_change_spte_gfn); + kvm_handle_hva_range(mn, address, address + 1, arg, kvm_change_spte_gfn); } void kvm_mmu_invalidate_begin(struct kvm *kvm, unsigned long start, @@ -747,7 +749,6 @@ static int kvm_mmu_notifier_invalidate_range_start(struct mmu_notifier *mn, const struct kvm_hva_range hva_range = { .start = range->start, .end = range->end, - .pte = __pte(0), .handler = kvm_unmap_gfn_range, .on_lock = kvm_mmu_invalidate_begin, .on_unlock = kvm_arch_guest_memory_reclaimed, @@ -812,7 +813,6 @@ static void kvm_mmu_notifier_invalidate_range_end(struct mmu_notifier *mn, const struct kvm_hva_range hva_range = { .start = range->start, .end = range->end, - .pte = __pte(0), .handler = (void *)kvm_null_fn, .on_lock = kvm_mmu_invalidate_end, .on_unlock = (void *)kvm_null_fn, @@ -845,7 +845,7 @@ static int kvm_mmu_notifier_clear_flush_young(struct mmu_notifier *mn, { trace_kvm_age_hva(start, end); - return kvm_handle_hva_range(mn, start, end, __pte(0), kvm_age_gfn); + return kvm_handle_hva_range(mn, start, end, KVM_NO_ARG, kvm_age_gfn); } static int kvm_mmu_notifier_clear_young(struct mmu_notifier *mn,
Wrap kvm_{gfn,hva}_range.pte in a union so that future notifier events can pass event specific information up and down the stack without needing to constantly expand and churn the APIs. Lockless aging of SPTEs will pass around a bitmap, and support for memory attributes will pass around the new attributes for the range. Add a "KVM_NO_ARG" placeholder to simplify handling events without an argument (creating a dummy union variable is midly annoying). Opportunstically drop explicit zero-initialization of the "pte" field, as omitting the field (now a union) has the same effect. Cc: Yu Zhao <yuzhao@google.com> Link: https://lore.kernel.org/all/CAOUHufagkd2Jk3_HrVoFFptRXM=hX2CV8f+M-dka-hJU4bP8kw@mail.gmail.com Signed-off-by: Sean Christopherson <seanjc@google.com> --- If this looks good, my thought is to squeeze it into 6.6 so that the MGLRU and guest_memfd() series can build on it. Or those series could just include it? arch/arm64/kvm/mmu.c | 2 +- arch/mips/kvm/mmu.c | 2 +- arch/riscv/kvm/mmu.c | 2 +- arch/x86/kvm/mmu/mmu.c | 2 +- arch/x86/kvm/mmu/tdp_mmu.c | 6 +++--- include/linux/kvm_host.h | 6 +++++- virt/kvm/kvm_main.c | 18 +++++++++--------- 7 files changed, 21 insertions(+), 17 deletions(-) base-commit: fdf0eaf11452d72945af31804e2a1048ee1b574c