From aa8d1f48d353b0469bff357183ee9df137d15ef0 Mon Sep 17 00:00:00 2001 From: Yan Zhao Date: Wed, 3 Jul 2024 10:10:43 +0800 Subject: KVM: x86/mmu: Introduce a quirk to control memslot zap behavior Introduce the quirk KVM_X86_QUIRK_SLOT_ZAP_ALL to allow users to select KVM's behavior when a memslot is moved or deleted for KVM_X86_DEFAULT_VM VMs. Make sure KVM behave as if the quirk is always disabled for non-KVM_X86_DEFAULT_VM VMs. The KVM_X86_QUIRK_SLOT_ZAP_ALL quirk offers two behavior options: - when enabled: Invalidate/zap all SPTEs ("zap-all"), - when disabled: Precisely zap only the leaf SPTEs within the range of the moving/deleting memory slot ("zap-slot-leafs-only"). "zap-all" is today's KVM behavior to work around a bug [1] where the changing the zapping behavior of memslot move/deletion would cause VM instability for VMs with an Nvidia GPU assigned; while "zap-slot-leafs-only" allows for more precise zapping of SPTEs within the memory slot range, improving performance in certain scenarios [2], and meeting the functional requirements for TDX. Previous attempts to select "zap-slot-leafs-only" include a per-VM capability approach [3] (which was not preferred because the root cause of the bug remained unidentified) and a per-memslot flag approach [4]. Sean and Paolo finally recommended the implementation of this quirk and explained that it's the least bad option [5]. By default, the quirk is enabled on KVM_X86_DEFAULT_VM VMs to use "zap-all". Users have the option to disable the quirk to select "zap-slot-leafs-only" for specific KVM_X86_DEFAULT_VM VMs that are unaffected by this bug. For non-KVM_X86_DEFAULT_VM VMs, the "zap-slot-leafs-only" behavior is always selected without user's opt-in, regardless of if the user opts for "zap-all". This is because it is assumed until proven otherwise that non- KVM_X86_DEFAULT_VM VMs will not be exposed to the bug [1], and most importantly, it's because TDX must have "zap-slot-leafs-only" always selected. In TDX's case a memslot's GPA range can be a mixture of "private" or "shared" memory. Shared is roughly analogous to how EPT is handled for normal VMs, but private GPAs need lots of special treatment: 1) "zap-all" would require to zap private root page or non-leaf entries or at least leaf-entries beyond the deleting memslot scope. However, TDX demands that the root page of the private page table remains unchanged, with leaf entries being zapped before non-leaf entries, and any dropped private guest pages must be re-accepted by the guest. 2) if "zap-all" zaps only shared page tables, it would result in private pages still being mapped when the memslot is gone. This may affect even other processes if later the gmem fd was whole punched, causing the pages being freed on the host while still mapped in the TD, because there's no pgoff to the gfn information to zap the private page table after memslot is gone. So, simply go "zap-slot-leafs-only" as if the quirk is always disabled for non-KVM_X86_DEFAULT_VM VMs to avoid manual opt-in for every VM type [6] or complicating quirk disabling interface (current quirk disabling interface is limited, no way to query quirks, or force them to be disabled). Add a new function kvm_mmu_zap_memslot_leafs() to implement "zap-slot-leafs-only". This function does not call kvm_unmap_gfn_range(), bypassing special handling to APIC_ACCESS_PAGE_PRIVATE_MEMSLOT, as 1) The APIC_ACCESS_PAGE_PRIVATE_MEMSLOT cannot be created by users, nor can it be moved. It is only deleted by KVM when APICv is permanently inhibited. 2) kvm_vcpu_reload_apic_access_page() effectively does nothing when APIC_ACCESS_PAGE_PRIVATE_MEMSLOT is deleted. 3) Avoid making all cpus request of KVM_REQ_APIC_PAGE_RELOAD can save on costly IPIs. Suggested-by: Kai Huang Suggested-by: Sean Christopherson Suggested-by: Paolo Bonzini Link: https://patchwork.kernel.org/project/kvm/patch/20190205210137.1377-11-sean.j.christopherson@intel.com [1] Link: https://patchwork.kernel.org/project/kvm/patch/20190205210137.1377-11-sean.j.christopherson@intel.com/#25054908 [2] Link: https://lore.kernel.org/kvm/20200713190649.GE29725@linux.intel.com/T/#mabc0119583dacf621025e9d873c85f4fbaa66d5c [3] Link: https://lore.kernel.org/all/20240515005952.3410568-3-rick.p.edgecombe@intel.com [4] Link: https://lore.kernel.org/all/7df9032d-83e4-46a1-ab29-6c7973a2ab0b@redhat.com [5] Link: https://lore.kernel.org/all/ZnGa550k46ow2N3L@google.com [6] Co-developed-by: Rick Edgecombe Signed-off-by: Rick Edgecombe Signed-off-by: Yan Zhao Message-ID: <20240703021043.13881-1-yan.y.zhao@intel.com> Signed-off-by: Paolo Bonzini --- Documentation/virt/kvm/api.rst | 8 ++++++++ arch/x86/include/asm/kvm_host.h | 3 ++- arch/x86/include/uapi/asm/kvm.h | 1 + arch/x86/kvm/mmu/mmu.c | 42 ++++++++++++++++++++++++++++++++++++++++- 4 files changed, 52 insertions(+), 2 deletions(-) diff --git a/Documentation/virt/kvm/api.rst b/Documentation/virt/kvm/api.rst index b3be87489108..b4d1cf2e4628 100644 --- a/Documentation/virt/kvm/api.rst +++ b/Documentation/virt/kvm/api.rst @@ -8082,6 +8082,14 @@ KVM_X86_QUIRK_MWAIT_NEVER_UD_FAULTS By default, KVM emulates MONITOR/MWAIT (if guest CPUID on writes to MISC_ENABLE if KVM_X86_QUIRK_MISC_ENABLE_NO_MWAIT is disabled. + +KVM_X86_QUIRK_SLOT_ZAP_ALL By default, KVM invalidates all SPTEs in + fast way for memslot deletion when VM type + is KVM_X86_DEFAULT_VM. + When this quirk is disabled or when VM type + is other than KVM_X86_DEFAULT_VM, KVM zaps + only leaf SPTEs that are within the range of + the memslot being deleted. =================================== ============================================ 7.32 KVM_CAP_MAX_VCPU_ID diff --git a/arch/x86/include/asm/kvm_host.h b/arch/x86/include/asm/kvm_host.h index 4a68cb3eba78..e4fc362ba3da 100644 --- a/arch/x86/include/asm/kvm_host.h +++ b/arch/x86/include/asm/kvm_host.h @@ -2345,7 +2345,8 @@ int memslot_rmap_alloc(struct kvm_memory_slot *slot, unsigned long npages); KVM_X86_QUIRK_OUT_7E_INC_RIP | \ KVM_X86_QUIRK_MISC_ENABLE_NO_MWAIT | \ KVM_X86_QUIRK_FIX_HYPERCALL_INSN | \ - KVM_X86_QUIRK_MWAIT_NEVER_UD_FAULTS) + KVM_X86_QUIRK_MWAIT_NEVER_UD_FAULTS | \ + KVM_X86_QUIRK_SLOT_ZAP_ALL) /* * KVM previously used a u32 field in kvm_run to indicate the hypercall was diff --git a/arch/x86/include/uapi/asm/kvm.h b/arch/x86/include/uapi/asm/kvm.h index bf57a824f722..a8debbf2f702 100644 --- a/arch/x86/include/uapi/asm/kvm.h +++ b/arch/x86/include/uapi/asm/kvm.h @@ -439,6 +439,7 @@ struct kvm_sync_regs { #define KVM_X86_QUIRK_MISC_ENABLE_NO_MWAIT (1 << 4) #define KVM_X86_QUIRK_FIX_HYPERCALL_INSN (1 << 5) #define KVM_X86_QUIRK_MWAIT_NEVER_UD_FAULTS (1 << 6) +#define KVM_X86_QUIRK_SLOT_ZAP_ALL (1 << 7) #define KVM_STATE_NESTED_FORMAT_VMX 0 #define KVM_STATE_NESTED_FORMAT_SVM 1 diff --git a/arch/x86/kvm/mmu/mmu.c b/arch/x86/kvm/mmu/mmu.c index 928cf84778b0..f107ec2557c1 100644 --- a/arch/x86/kvm/mmu/mmu.c +++ b/arch/x86/kvm/mmu/mmu.c @@ -6997,10 +6997,50 @@ void kvm_arch_flush_shadow_all(struct kvm *kvm) kvm_mmu_zap_all(kvm); } +/* + * Zapping leaf SPTEs with memslot range when a memslot is moved/deleted. + * + * Zapping non-leaf SPTEs, a.k.a. not-last SPTEs, isn't required, worst + * case scenario we'll have unused shadow pages lying around until they + * are recycled due to age or when the VM is destroyed. + */ +static void kvm_mmu_zap_memslot_leafs(struct kvm *kvm, struct kvm_memory_slot *slot) +{ + struct kvm_gfn_range range = { + .slot = slot, + .start = slot->base_gfn, + .end = slot->base_gfn + slot->npages, + .may_block = true, + }; + bool flush = false; + + write_lock(&kvm->mmu_lock); + + if (kvm_memslots_have_rmaps(kvm)) + flush = kvm_handle_gfn_range(kvm, &range, kvm_zap_rmap); + + if (tdp_mmu_enabled) + flush = kvm_tdp_mmu_unmap_gfn_range(kvm, &range, flush); + + if (flush) + kvm_flush_remote_tlbs_memslot(kvm, slot); + + write_unlock(&kvm->mmu_lock); +} + +static inline bool kvm_memslot_flush_zap_all(struct kvm *kvm) +{ + return kvm->arch.vm_type == KVM_X86_DEFAULT_VM && + kvm_check_has_quirk(kvm, KVM_X86_QUIRK_SLOT_ZAP_ALL); +} + void kvm_arch_flush_shadow_memslot(struct kvm *kvm, struct kvm_memory_slot *slot) { - kvm_mmu_zap_all_fast(kvm); + if (kvm_memslot_flush_zap_all(kvm)) + kvm_mmu_zap_all_fast(kvm); + else + kvm_mmu_zap_memslot_leafs(kvm, slot); } void kvm_mmu_invalidate_mmio_sptes(struct kvm *kvm, u64 gen) -- cgit v1.2.3 From b4ed2c67d275b85b2ab07d54f88bebd5998d61d8 Mon Sep 17 00:00:00 2001 From: Yan Zhao Date: Wed, 3 Jul 2024 10:11:19 +0800 Subject: KVM: selftests: Test slot move/delete with slot zap quirk enabled/disabled Update set_memory_region_test to make sure memslot move and deletion function correctly both when slot zap quirk KVM_X86_QUIRK_SLOT_ZAP_ALL is enabled and disabled. Signed-off-by: Yan Zhao Message-ID: <20240703021119.13904-1-yan.y.zhao@intel.com> Signed-off-by: Paolo Bonzini --- .../testing/selftests/kvm/set_memory_region_test.c | 29 ++++++++++++++++------ 1 file changed, 21 insertions(+), 8 deletions(-) diff --git a/tools/testing/selftests/kvm/set_memory_region_test.c b/tools/testing/selftests/kvm/set_memory_region_test.c index bb8002084f52..a8267628e9ed 100644 --- a/tools/testing/selftests/kvm/set_memory_region_test.c +++ b/tools/testing/selftests/kvm/set_memory_region_test.c @@ -175,7 +175,7 @@ static void guest_code_move_memory_region(void) GUEST_DONE(); } -static void test_move_memory_region(void) +static void test_move_memory_region(bool disable_slot_zap_quirk) { pthread_t vcpu_thread; struct kvm_vcpu *vcpu; @@ -184,6 +184,9 @@ static void test_move_memory_region(void) vm = spawn_vm(&vcpu, &vcpu_thread, guest_code_move_memory_region); + if (disable_slot_zap_quirk) + vm_enable_cap(vm, KVM_CAP_DISABLE_QUIRKS2, KVM_X86_QUIRK_SLOT_ZAP_ALL); + hva = addr_gpa2hva(vm, MEM_REGION_GPA); /* @@ -266,7 +269,7 @@ static void guest_code_delete_memory_region(void) GUEST_ASSERT(0); } -static void test_delete_memory_region(void) +static void test_delete_memory_region(bool disable_slot_zap_quirk) { pthread_t vcpu_thread; struct kvm_vcpu *vcpu; @@ -276,6 +279,9 @@ static void test_delete_memory_region(void) vm = spawn_vm(&vcpu, &vcpu_thread, guest_code_delete_memory_region); + if (disable_slot_zap_quirk) + vm_enable_cap(vm, KVM_CAP_DISABLE_QUIRKS2, KVM_X86_QUIRK_SLOT_ZAP_ALL); + /* Delete the memory region, the guest should not die. */ vm_mem_region_delete(vm, MEM_REGION_SLOT); wait_for_vcpu(); @@ -553,7 +559,10 @@ int main(int argc, char *argv[]) { #ifdef __x86_64__ int i, loops; + int j, disable_slot_zap_quirk = 0; + if (kvm_check_cap(KVM_CAP_DISABLE_QUIRKS2) & KVM_X86_QUIRK_SLOT_ZAP_ALL) + disable_slot_zap_quirk = 1; /* * FIXME: the zero-memslot test fails on aarch64 and s390x because * KVM_RUN fails with ENOEXEC or EFAULT. @@ -579,13 +588,17 @@ int main(int argc, char *argv[]) else loops = 10; - pr_info("Testing MOVE of in-use region, %d loops\n", loops); - for (i = 0; i < loops; i++) - test_move_memory_region(); + for (j = 0; j <= disable_slot_zap_quirk; j++) { + pr_info("Testing MOVE of in-use region, %d loops, slot zap quirk %s\n", + loops, j ? "disabled" : "enabled"); + for (i = 0; i < loops; i++) + test_move_memory_region(!!j); - pr_info("Testing DELETE of in-use region, %d loops\n", loops); - for (i = 0; i < loops; i++) - test_delete_memory_region(); + pr_info("Testing DELETE of in-use region, %d loops, slot zap quirk %s\n", + loops, j ? "disabled" : "enabled"); + for (i = 0; i < loops; i++) + test_delete_memory_region(!!j); + } #endif return 0; -- cgit v1.2.3 From 218f6415004a881d116e254eeb837358aced55ab Mon Sep 17 00:00:00 2001 From: Yan Zhao Date: Wed, 3 Jul 2024 10:12:06 +0800 Subject: KVM: selftests: Allow slot modification stress test with quirk disabled Add a new user option to memslot_modification_stress_test to allow testing with slot zap quirk KVM_X86_QUIRK_SLOT_ZAP_ALL disabled. Signed-off-by: Yan Zhao Message-ID: <20240703021206.13923-1-yan.y.zhao@intel.com> Signed-off-by: Paolo Bonzini --- .../selftests/kvm/memslot_modification_stress_test.c | 19 +++++++++++++++++-- 1 file changed, 17 insertions(+), 2 deletions(-) diff --git a/tools/testing/selftests/kvm/memslot_modification_stress_test.c b/tools/testing/selftests/kvm/memslot_modification_stress_test.c index 49f162573126..e3343f0df9e1 100644 --- a/tools/testing/selftests/kvm/memslot_modification_stress_test.c +++ b/tools/testing/selftests/kvm/memslot_modification_stress_test.c @@ -79,6 +79,7 @@ struct test_params { useconds_t delay; uint64_t nr_iterations; bool partition_vcpu_memory_access; + bool disable_slot_zap_quirk; }; static void run_test(enum vm_guest_mode mode, void *arg) @@ -89,6 +90,13 @@ static void run_test(enum vm_guest_mode mode, void *arg) vm = memstress_create_vm(mode, nr_vcpus, guest_percpu_mem_size, 1, VM_MEM_SRC_ANONYMOUS, p->partition_vcpu_memory_access); +#ifdef __x86_64__ + if (p->disable_slot_zap_quirk) + vm_enable_cap(vm, KVM_CAP_DISABLE_QUIRKS2, KVM_X86_QUIRK_SLOT_ZAP_ALL); + + pr_info("Memslot zap quirk %s\n", p->disable_slot_zap_quirk ? + "disabled" : "enabled"); +#endif pr_info("Finished creating vCPUs\n"); @@ -107,11 +115,12 @@ static void run_test(enum vm_guest_mode mode, void *arg) static void help(char *name) { puts(""); - printf("usage: %s [-h] [-m mode] [-d delay_usec]\n" + printf("usage: %s [-h] [-m mode] [-d delay_usec] [-q]\n" " [-b memory] [-v vcpus] [-o] [-i iterations]\n", name); guest_modes_help(); printf(" -d: add a delay between each iteration of adding and\n" " deleting a memslot in usec.\n"); + printf(" -q: Disable memslot zap quirk.\n"); printf(" -b: specify the size of the memory region which should be\n" " accessed by each vCPU. e.g. 10M or 3G.\n" " Default: 1G\n"); @@ -137,7 +146,7 @@ int main(int argc, char *argv[]) guest_modes_append_default(); - while ((opt = getopt(argc, argv, "hm:d:b:v:oi:")) != -1) { + while ((opt = getopt(argc, argv, "hm:d:qb:v:oi:")) != -1) { switch (opt) { case 'm': guest_modes_cmdline(optarg); @@ -160,6 +169,12 @@ int main(int argc, char *argv[]) case 'i': p.nr_iterations = atoi_positive("Number of iterations", optarg); break; + case 'q': + p.disable_slot_zap_quirk = true; + + TEST_REQUIRE(kvm_check_cap(KVM_CAP_DISABLE_QUIRKS2) & + KVM_X86_QUIRK_SLOT_ZAP_ALL); + break; case 'h': default: help(argv[0]); -- cgit v1.2.3 From 61de4c34b51c5b9c7ef8229eb246346254638446 Mon Sep 17 00:00:00 2001 From: Yan Zhao Date: Wed, 3 Jul 2024 10:12:19 +0800 Subject: KVM: selftests: Test memslot move in memslot_perf_test with quirk disabled Add a new user option to memslot_perf_test to allow testing memslot move with quirk KVM_X86_QUIRK_SLOT_ZAP_ALL disabled. Signed-off-by: Yan Zhao Message-ID: <20240703021219.13939-1-yan.y.zhao@intel.com> Signed-off-by: Paolo Bonzini --- tools/testing/selftests/kvm/memslot_perf_test.c | 12 +++++++++++- 1 file changed, 11 insertions(+), 1 deletion(-) diff --git a/tools/testing/selftests/kvm/memslot_perf_test.c b/tools/testing/selftests/kvm/memslot_perf_test.c index 579a64f97333..893366982f77 100644 --- a/tools/testing/selftests/kvm/memslot_perf_test.c +++ b/tools/testing/selftests/kvm/memslot_perf_test.c @@ -113,6 +113,7 @@ static_assert(ATOMIC_BOOL_LOCK_FREE == 2, "atomic bool is not lockless"); static sem_t vcpu_ready; static bool map_unmap_verify; +static bool disable_slot_zap_quirk; static bool verbose; #define pr_info_v(...) \ @@ -578,6 +579,9 @@ static bool test_memslot_move_prepare(struct vm_data *data, uint32_t guest_page_size = data->vm->page_size; uint64_t movesrcgpa, movetestgpa; + if (disable_slot_zap_quirk) + vm_enable_cap(data->vm, KVM_CAP_DISABLE_QUIRKS2, KVM_X86_QUIRK_SLOT_ZAP_ALL); + movesrcgpa = vm_slot2gpa(data, data->nslots - 1); if (isactive) { @@ -896,6 +900,7 @@ static void help(char *name, struct test_args *targs) pr_info(" -h: print this help screen.\n"); pr_info(" -v: enable verbose mode (not for benchmarking).\n"); pr_info(" -d: enable extra debug checks.\n"); + pr_info(" -q: Disable memslot zap quirk during memslot move.\n"); pr_info(" -s: specify memslot count cap (-1 means no cap; currently: %i)\n", targs->nslots); pr_info(" -f: specify the first test to run (currently: %i; max %zu)\n", @@ -954,7 +959,7 @@ static bool parse_args(int argc, char *argv[], uint32_t max_mem_slots; int opt; - while ((opt = getopt(argc, argv, "hvds:f:e:l:r:")) != -1) { + while ((opt = getopt(argc, argv, "hvdqs:f:e:l:r:")) != -1) { switch (opt) { case 'h': default: @@ -966,6 +971,11 @@ static bool parse_args(int argc, char *argv[], case 'd': map_unmap_verify = true; break; + case 'q': + disable_slot_zap_quirk = true; + TEST_REQUIRE(kvm_check_cap(KVM_CAP_DISABLE_QUIRKS2) & + KVM_X86_QUIRK_SLOT_ZAP_ALL); + break; case 's': targs->nslots = atoi_paranoid(optarg); if (targs->nslots <= 1 && targs->nslots != -1) { -- cgit v1.2.3