From patchwork Sun Jan 19 04:04:58 2020 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: "Yang, Weijiang" X-Patchwork-Id: 11340517 Return-Path: Received: from mail.kernel.org (pdx-korg-mail-1.web.codeaurora.org [172.30.200.123]) by pdx-korg-patchwork-2.web.codeaurora.org (Postfix) with ESMTP id 31E6514B4 for ; Sun, 19 Jan 2020 04:00:39 +0000 (UTC) Received: from vger.kernel.org (vger.kernel.org [209.132.180.67]) by mail.kernel.org (Postfix) with ESMTP id EF9682081E for ; Sun, 19 Jan 2020 04:00:38 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1726605AbgASEAZ (ORCPT ); Sat, 18 Jan 2020 23:00:25 -0500 Received: from mga14.intel.com ([192.55.52.115]:62796 "EHLO mga14.intel.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1725980AbgASEAZ (ORCPT ); Sat, 18 Jan 2020 23:00:25 -0500 X-Amp-Result: SKIPPED(no attachment in message) X-Amp-File-Uploaded: False Received: from orsmga007.jf.intel.com ([10.7.209.58]) by fmsmga103.fm.intel.com with ESMTP/TLS/DHE-RSA-AES256-GCM-SHA384; 18 Jan 2020 20:00:24 -0800 X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="5.70,336,1574150400"; d="scan'208";a="214910461" Received: from local-michael-cet-test.sh.intel.com ([10.239.159.128]) by orsmga007.jf.intel.com with ESMTP; 18 Jan 2020 20:00:22 -0800 From: Yang Weijiang To: kvm@vger.kernel.org, linux-kernel@vger.kernel.org, pbonzini@redhat.com, jmattson@google.com, sean.j.christopherson@intel.com Cc: yu.c.zhang@linux.intel.com, alazar@bitdefender.com, edwin.zhai@intel.com, Yang Weijiang Subject: [PATCH v11 01/10] Documentation: Add EPT based Subpage Protection and related APIs Date: Sun, 19 Jan 2020 12:04:58 +0800 Message-Id: <20200119040507.23113-2-weijiang.yang@intel.com> X-Mailer: git-send-email 2.17.2 In-Reply-To: <20200119040507.23113-1-weijiang.yang@intel.com> References: <20200119040507.23113-1-weijiang.yang@intel.com> Sender: kvm-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: kvm@vger.kernel.org Co-developed-by: yi.z.zhang@linux.intel.com Signed-off-by: yi.z.zhang@linux.intel.com Signed-off-by: Yang Weijiang --- Documentation/virt/kvm/api.txt | 39 ++++++ Documentation/virtual/kvm/spp_kvm.txt | 179 ++++++++++++++++++++++++++ 2 files changed, 218 insertions(+) create mode 100644 Documentation/virtual/kvm/spp_kvm.txt diff --git a/Documentation/virt/kvm/api.txt b/Documentation/virt/kvm/api.txt index ebb37b34dcfc..a08c944d8eb6 100644 --- a/Documentation/virt/kvm/api.txt +++ b/Documentation/virt/kvm/api.txt @@ -4168,6 +4168,45 @@ This ioctl issues an ultravisor call to terminate the secure guest, unpins the VPA pages and releases all the device pages that are used to track the secure pages by hypervisor. +4.122 KVM_SUBPAGES_GET_ACCESS + +Architectures: x86 +Type: vm ioctl +Parameters: struct kvm_subpage_info (in/out) +Returns: 0 on success, < 0 on error + +#define KVM_SUBPAGE_MAX_PAGES 512 +struct kvm_subpage { + __u64 gfn_base; /* the first page gfn of the contiguous pages */ + __u32 npages; /* number of 4K pages */ + __u32 flags; /* reserved to 0 now */ + __u32 access_map[0]; /* start place of bitmap array */ +}; + +This ioctl fetches subpage permission from contiguous pages starting with +gfn. npages is the number of contiguous pages to fetch. access_map contains permission +vectors fetched for all the pages. + +4.123 KVM_SUBPAGES_SET_ACCESS + +Architectures: x86 +Type: vm ioctl +Parameters: struct kvm_subpage_info (in/out) +Returns: 0 on success, < 0 on error + +#define KVM_SUBPAGE_MAX_PAGES 512 +struct kvm_subpage { + __u64 gfn_base; /* the first page gfn of the contiguous pages */ + __u32 npages; /* number of 4K pages */ + __u32 flags; /* reserved to 0 now */ + __u32 access_map[0]; /* start place of bitmap array */ +}; + +This ioctl sets subpage permission for contiguous pages starting with gfn. npages is +the number of contiguous pages to set. access_map contains permission vectors for all the +pages. Since during execution of the ioctl, it holds mmu_lock, so limits the MAX pages +to 512 to reduce the impact to EPT. + 5. The kvm_run structure ------------------------ diff --git a/Documentation/virtual/kvm/spp_kvm.txt b/Documentation/virtual/kvm/spp_kvm.txt new file mode 100644 index 000000000000..1b41125e0cb1 --- /dev/null +++ b/Documentation/virtual/kvm/spp_kvm.txt @@ -0,0 +1,179 @@ +EPT-Based Sub-Page Protection (SPP) for KVM +==================================================== + +1.Overview + EPT-based Sub-Page Protection(SPP) allows VMM to specify + fine-grained(128byte per sub-page) write-protection for guest physical + memory. When it's enabled, the CPU enforces write-access permission + for the sub-pages within a 4KB page, if corresponding bit is set in + permission vector, write to sub-page region is allowed, otherwise, + it's prevented with a EPT violation. + + *Note*: In current implementation, SPP is exclusive with nested flag, + if it's on, SPP feature won't work. + +2.SPP Operation + Sub-Page Protection Table (SPPT) is introduced to manage sub-page + write-access permission. + + It is active when: + a) nested flag is turned off. + b) "sub-page write protection" VM-execution control is 1. + c) SPP is initialized with KVM_ENABLE_CAP ioctl and sub-class KVM_CAP_X86_SPP. + d) Sub-page permissions are set with KVM_SUBPAGES_SET_ACCESS ioctl. + see below sections for details. + + __________________________________________________________________________ + + How SPP hardware works: + __________________________________________________________________________ + + Guest write access --> GPA --> Walk EPT --> EPT leaf entry -----| + |---------------------------------------------------------------| + |-> if VMexec_control.spp && ept_leaf_entry.spp_bit (bit 61) + | + |-> --> EPT legacy behavior + | + | + |-> --> if ept_leaf_entry.writable + | + |-> --> Ignore SPP + | + |-> --> GPA --> Walk SPP 4-level table--| + | + |------------<----------get-the-SPPT-point-from-VMCS-field-----<------| + | + Walk SPP L4E table + | + |---> if-entry-misconfiguration ------------>-------|-------<---------| + | | | + else | | + | | | + | |------------------SPP VMexit<-----------------| | + | | | + | |-> exit_qualification & sppt_misconfig --> sppt misconfig | + | | | + | |-> exit_qualification & sppt_miss --> sppt miss | + |---| | + | | + walk SPPT L3E--|--> if-entry-misconfiguration------------>------------| + | | + else | + | | + | | + walk SPPT L2E --|--> if-entry-misconfiguration-------->-------| + | | + else | + | | + | | + walk SPPT L1E --|-> if-entry-misconfiguration--->----| + | + else + | + |-> if sub-page writable + |-> allow, write access + |-> disallow, EPT violation + ______________________________________________________________________________ + +3.IOCTL Interfaces + + KVM_ENABLE_CAP(capability: KVM_CAP_X86_SPP): + Allocate storage for sub-page permission vectors and SPPT root page. + + KVM_SUBPAGES_GET_ACCESS: + Get sub-page write permission vectors for given contiguous guest pages. + + KVM_SUBPAGES_SET_ACCESS + Set SPP bit in EPT leaf entries for given contiguous guest pages. The + actual SPPT setup is triggered when SPP miss vm-exit is handled. + + struct kvm_subpage{ + __u64 gfn_base; /* the first page gfn of the contiguous pages */ + __u32 npages; /* number of 4K pages */ + __u32 flags; /* reserved to 0 now */ + __u32 access_map[0]; /* start place of bitmap array */ + }; + + #define KVM_SUBPAGES_GET_ACCESS _IOR(KVMIO, 0x49, __u64) + #define KVM_SUBPAGES_SET_ACCESS _IOW(KVMIO, 0x4a, __u64) + +4.Set Sub-Page Permission + + * To enable SPP protection, KVM user-space application sets sub-page permission + via KVM_SUBPAGES_SET_ACCESS ioctl: + (1) It first stores the access permissions in bitmap array. + + (2) Then, if the target 4KB pages are mapped as PT_PAGE_TABLE_LEVEL entry in EPT, + it sets SPP bit of the corresponding entry to mark sub-page protection. + If the 4KB pages are mapped within PT_DIRECTORY_LEVEL or PT_PDPE_LEVEL entry, + it first zaps the hugepage entries so as to let following memory access to trigger + EPT violation, there the gfn is check against SPP permission bitmap and + proper level is selected to set up EPT entry. + + + The SPPT paging structure format is as below: + + Format of the SPPT L4E, L3E, L2E: + | Bit | Contents | + | :----- | :------------------------------------------------------------------------| + | 0 | Valid entry when set; indicates whether the entry is present | + | 11:1 | Reserved (0) | + | N-1:12 | Physical address of 4KB aligned SPPT LX-1 Table referenced by this entry | + | 51:N | Reserved (0) | + | 63:52 | Reserved (0) | + Note: N is the physical address width supported by the processor. X is the page level + + Format of the SPPT L1E: + | Bit | Contents | + | :---- | :---------------------------------------------------------------- | + | 0+2i | Write permission for i-th 128 byte sub-page region. | + | 1+2i | Reserved (0). | + Note: 0<=i<=31 + +5.SPPT-induced VM exit + + * SPPT miss and misconfiguration induced VM exit + + A SPPT missing VM exit occurs when walk the SPPT, there is no SPPT + misconfiguration but a paging-structure entry is not + present in any of L4E/L3E/L2E entries. + + A SPPT misconfiguration VM exit occurs when reserved bits or unsupported values + are set in SPPT entry. + + *NOTE* SPPT miss and SPPT misconfigurations can occur only due to + "eligible" memory write, this excludes, e.g., guest paging structure, + please refer to SDM 28.2 for details of "non-eligible" cases. + + * SPP permission induced VM exit + SPP sub-page permission induced violation is reported as EPT violation + therefore causes VM exit. + +6.SPPT-induced VM exit handling + + #define EXIT_REASON_SPP 66 + + static int (*const kvm_vmx_exit_handlers[])(struct kvm_vcpu *vcpu) = { + ... + [EXIT_REASON_SPP] = handle_spp, + ... + }; + + New exit qualification for SPPT-induced vmexits. + + | Bit | Contents | + | :---- | :---------------------------------------------------------------- | + | 10:0 | Reserved (0). | + | 11 | SPPT VM exit type. Set for SPPT Miss, cleared for SPPT Misconfig. | + | 12 | NMI unblocking due to IRET | + | 63:13 | Reserved (0) | + + * SPPT miss induced VM exit + Set up SPPT entries correctly. + + * SPPT misconfiguration induced VM exit + This is left to user-space application to handle. + + * SPP permission induced VM exit + This is left to user-space application to handle, e.g., + retry the fault instruction or skip it. From patchwork Sun Jan 19 04:04:59 2020 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: "Yang, Weijiang" X-Patchwork-Id: 11340519 Return-Path: Received: from mail.kernel.org (pdx-korg-mail-1.web.codeaurora.org [172.30.200.123]) by pdx-korg-patchwork-2.web.codeaurora.org (Postfix) with ESMTP id 46A5692A for ; Sun, 19 Jan 2020 04:00:39 +0000 (UTC) Received: from vger.kernel.org (vger.kernel.org [209.132.180.67]) by mail.kernel.org (Postfix) with ESMTP id 2FB12206D5 for ; Sun, 19 Jan 2020 04:00:39 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1726761AbgASEA2 (ORCPT ); Sat, 18 Jan 2020 23:00:28 -0500 Received: from mga14.intel.com ([192.55.52.115]:62796 "EHLO mga14.intel.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1725980AbgASEA1 (ORCPT ); Sat, 18 Jan 2020 23:00:27 -0500 X-Amp-Result: SKIPPED(no attachment in message) X-Amp-File-Uploaded: False Received: from orsmga007.jf.intel.com ([10.7.209.58]) by fmsmga103.fm.intel.com with ESMTP/TLS/DHE-RSA-AES256-GCM-SHA384; 18 Jan 2020 20:00:26 -0800 X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="5.70,336,1574150400"; d="scan'208";a="214910467" Received: from local-michael-cet-test.sh.intel.com ([10.239.159.128]) by orsmga007.jf.intel.com with ESMTP; 18 Jan 2020 20:00:24 -0800 From: Yang Weijiang To: kvm@vger.kernel.org, linux-kernel@vger.kernel.org, pbonzini@redhat.com, jmattson@google.com, sean.j.christopherson@intel.com Cc: yu.c.zhang@linux.intel.com, alazar@bitdefender.com, edwin.zhai@intel.com, Yang Weijiang Subject: [PATCH v11 02/10] mmu: spp: Implement SPPT setup functions Date: Sun, 19 Jan 2020 12:04:59 +0800 Message-Id: <20200119040507.23113-3-weijiang.yang@intel.com> X-Mailer: git-send-email 2.17.2 In-Reply-To: <20200119040507.23113-1-weijiang.yang@intel.com> References: <20200119040507.23113-1-weijiang.yang@intel.com> Sender: kvm-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: kvm@vger.kernel.org SPPT(Sub-Page Permission Table) is a 4-level structure similar to EPT. If SPP is enabled in secondary execution control, WP and SPP bit(61) are set in a 4KB EPT leaf entry, then SPPT is traversed with the gfn, the leaf entry of SPPT contains the permission vector, every subpage within the 4KB page owns one bit. SPPT setup is similar to that of EPT therefore a lot of EPT helpers are re-used in spp.c. To make least change to mmu.c and keep the patch clean meanwhile, spp.c is embedded at the end of mmu.c. Specific to SPPT: 1)The leaf entry contains a 64-bit permission vector. Subpage is 128B each so 4KB/128B = 32bits are required for write permission. The even bits(2*i) corrrespond to write-permission to subpage(i), if it's 1, subpage(i) is writable, otherwise, it's write-protected. The odd bits are reserved and must be 0. 2)When permission vectors are updated, it first flushes the corresponding entry at SPPT L2E by making the entry invalid so that following SPPT walk can trigger SPP-miss handling, there the permission vectors are rebuilt. Co-developed-by: He Chen Signed-off-by: He Chen Co-developed-by: Zhang Yi Signed-off-by: Zhang Yi Signed-off-by: Yang Weijiang --- arch/x86/include/asm/kvm_host.h | 5 +- arch/x86/kvm/mmu/mmu.c | 7 ++ arch/x86/kvm/mmu/spp.c | 126 ++++++++++++++++++++++++++++++++ arch/x86/kvm/mmu/spp.h | 5 ++ 4 files changed, 142 insertions(+), 1 deletion(-) create mode 100644 arch/x86/kvm/mmu/spp.c create mode 100644 arch/x86/kvm/mmu/spp.h diff --git a/arch/x86/include/asm/kvm_host.h b/arch/x86/include/asm/kvm_host.h index b79cd6aa4075..9506c9d40895 100644 --- a/arch/x86/include/asm/kvm_host.h +++ b/arch/x86/include/asm/kvm_host.h @@ -261,7 +261,8 @@ union kvm_mmu_page_role { unsigned smap_andnot_wp:1; unsigned ad_disabled:1; unsigned guest_mode:1; - unsigned :6; + unsigned spp:1; + unsigned reserved:5; /* * This is left at the top of the word so that @@ -956,6 +957,8 @@ struct kvm_arch { struct kvm_pmu_event_filter *pmu_event_filter; struct task_struct *nx_lpage_recovery_thread; + + hpa_t sppt_root; }; struct kvm_vm_stat { diff --git a/arch/x86/kvm/mmu/mmu.c b/arch/x86/kvm/mmu/mmu.c index 6f92b40d798c..dff52763e05c 100644 --- a/arch/x86/kvm/mmu/mmu.c +++ b/arch/x86/kvm/mmu/mmu.c @@ -6500,3 +6500,10 @@ void kvm_mmu_pre_destroy_vm(struct kvm *kvm) if (kvm->arch.nx_lpage_recovery_thread) kthread_stop(kvm->arch.nx_lpage_recovery_thread); } + +/* + * SPP Table structure is similar to EPT, so a lot of MMU functions + * defined in mmu.c are re-used in spp.c. To make least + * change to mmu.c, spp.c is embedded in mmu.c here. + */ +#include "spp.c" diff --git a/arch/x86/kvm/mmu/spp.c b/arch/x86/kvm/mmu/spp.c new file mode 100644 index 000000000000..4247d6b1c6f7 --- /dev/null +++ b/arch/x86/kvm/mmu/spp.c @@ -0,0 +1,126 @@ +// SPDX-License-Identifier: GPL-2.0 + +#include "spp.h" + +#define for_each_shadow_spp_entry(_vcpu, _addr, _walker) \ + for (shadow_spp_walk_init(&(_walker), _vcpu, _addr); \ + shadow_walk_okay(&(_walker)); \ + shadow_walk_next(&(_walker))) + +static void shadow_spp_walk_init(struct kvm_shadow_walk_iterator *iterator, + struct kvm_vcpu *vcpu, u64 addr) +{ + iterator->addr = addr; + iterator->shadow_addr = vcpu->kvm->arch.sppt_root; + + /* SPP Table is a 4-level paging structure */ + iterator->level = PT64_ROOT_4LEVEL; +} + +struct kvm_mmu_page *kvm_spp_get_page(struct kvm_vcpu *vcpu, + gfn_t gfn, + unsigned int level) +{ + struct kvm_mmu_page *sp; + union kvm_mmu_page_role role; + + role = vcpu->arch.mmu->mmu_role.base; + role.level = level; + role.direct = true; + role.spp = true; + + for_each_valid_sp(vcpu->kvm, sp, gfn) { + if (sp->gfn != gfn) + continue; + if (sp->role.word != role.word) + continue; + if (sp->role.spp && sp->role.level == level) + goto out; + } + + sp = kvm_mmu_alloc_page(vcpu, true); + sp->gfn = gfn; + sp->role = role; + hlist_add_head(&sp->hash_link, + &vcpu->kvm->arch.mmu_page_hash + [kvm_page_table_hashfn(gfn)]); + clear_page(sp->spt); +out: + return sp; +} + +static void link_spp_shadow_page(struct kvm_vcpu *vcpu, u64 *sptep, + struct kvm_mmu_page *sp) +{ + u64 spte; + + spte = __pa(sp->spt) | PT_PRESENT_MASK; + + mmu_spte_set(sptep, spte); + + mmu_page_add_parent_pte(vcpu, sp, sptep); +} + +static u64 format_spp_spte(u32 spp_wp_bitmap) +{ + u64 new_spte = 0; + int i = 0; + + /* + * One 4K-page contains 32 sub-pages, they're flagged in even bits in + * SPPT L4E, the odd bits are reserved now, so convert 4-byte write + * permission bitmap to 8-byte SPP L4E format. + */ + for (i = 0; i < 32; i++) + new_spte |= (spp_wp_bitmap & BIT_ULL(i)) << i; + + return new_spte; +} + +static void spp_spte_set(u64 *sptep, u64 new_spte) +{ + __set_spte(sptep, new_spte); +} + +int kvm_spp_setup_structure(struct kvm_vcpu *vcpu, + u32 access_map, gfn_t gfn) +{ + struct kvm_shadow_walk_iterator iter; + struct kvm_mmu_page *sp; + gfn_t pseudo_gfn; + u64 old_spte, spp_spte; + int ret = -EFAULT; + + if (!VALID_PAGE(vcpu->kvm->arch.sppt_root)) + return -EFAULT; + + for_each_shadow_spp_entry(vcpu, (u64)gfn << PAGE_SHIFT, iter) { + if (iter.level == PT_PAGE_TABLE_LEVEL) { + spp_spte = format_spp_spte(access_map); + old_spte = mmu_spte_get_lockless(iter.sptep); + if (old_spte != spp_spte) + spp_spte_set(iter.sptep, spp_spte); + ret = 0; + break; + } + + if (!is_shadow_present_pte(*iter.sptep)) { + u64 base_addr = iter.addr; + + base_addr &= PT64_LVL_ADDR_MASK(iter.level); + pseudo_gfn = base_addr >> PAGE_SHIFT; + sp = kvm_spp_get_page(vcpu, pseudo_gfn, + iter.level - 1); + link_spp_shadow_page(vcpu, iter.sptep, sp); + } else if (iter.level == PT_DIRECTORY_LEVEL && + !(spp_spte & PT_PRESENT_MASK) && + (spp_spte & PT64_BASE_ADDR_MASK)) { + spp_spte = mmu_spte_get_lockless(iter.sptep); + spp_spte |= PT_PRESENT_MASK; + spp_spte_set(iter.sptep, spp_spte); + } + } + + kvm_flush_remote_tlbs(vcpu->kvm); + return ret; +} diff --git a/arch/x86/kvm/mmu/spp.h b/arch/x86/kvm/mmu/spp.h new file mode 100644 index 000000000000..03e4dfad595a --- /dev/null +++ b/arch/x86/kvm/mmu/spp.h @@ -0,0 +1,5 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +#ifndef __KVM_X86_VMX_SPP_H +#define __KVM_X86_VMX_SPP_H + +#endif /* __KVM_X86_VMX_SPP_H */ From patchwork Sun Jan 19 04:05:00 2020 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: "Yang, Weijiang" X-Patchwork-Id: 11340521 Return-Path: Received: from mail.kernel.org (pdx-korg-mail-1.web.codeaurora.org [172.30.200.123]) by pdx-korg-patchwork-2.web.codeaurora.org (Postfix) with ESMTP id 67A79188B for ; Sun, 19 Jan 2020 04:00:39 +0000 (UTC) Received: from vger.kernel.org (vger.kernel.org [209.132.180.67]) by mail.kernel.org (Postfix) with ESMTP id 505C6206D5 for ; Sun, 19 Jan 2020 04:00:39 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1726857AbgASEAa (ORCPT ); Sat, 18 Jan 2020 23:00:30 -0500 Received: from mga14.intel.com ([192.55.52.115]:62796 "EHLO mga14.intel.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1726811AbgASEAa (ORCPT ); Sat, 18 Jan 2020 23:00:30 -0500 X-Amp-Result: SKIPPED(no attachment in message) X-Amp-File-Uploaded: False Received: from orsmga007.jf.intel.com ([10.7.209.58]) by fmsmga103.fm.intel.com with ESMTP/TLS/DHE-RSA-AES256-GCM-SHA384; 18 Jan 2020 20:00:29 -0800 X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="5.70,336,1574150400"; d="scan'208";a="214910478" Received: from local-michael-cet-test.sh.intel.com ([10.239.159.128]) by orsmga007.jf.intel.com with ESMTP; 18 Jan 2020 20:00:26 -0800 From: Yang Weijiang To: kvm@vger.kernel.org, linux-kernel@vger.kernel.org, pbonzini@redhat.com, jmattson@google.com, sean.j.christopherson@intel.com Cc: yu.c.zhang@linux.intel.com, alazar@bitdefender.com, edwin.zhai@intel.com, Yang Weijiang Subject: [PATCH v11 03/10] mmu: spp: Implement functions to {get|set}_subpage permission Date: Sun, 19 Jan 2020 12:05:00 +0800 Message-Id: <20200119040507.23113-4-weijiang.yang@intel.com> X-Mailer: git-send-email 2.17.2 In-Reply-To: <20200119040507.23113-1-weijiang.yang@intel.com> References: <20200119040507.23113-1-weijiang.yang@intel.com> Sender: kvm-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: kvm@vger.kernel.org Userspace application can {set|get} subpage permission via IOCTLs if SPP has been initialized. Steps for set_permission: 1)Store the permission vectors to SPP bitmap buffer. 2)Flush existing hugepage mapping in rmap so as to avoid stale mapping. 3)Walk EPT to check if gfn->pfn 4KB mapping is there, mark the existing entry as WP and SPP protected. 4)Zap the entry if gfn->pfn is hugepage mapping so that following memory access can trigger EPT page_fault() to set up SPP protection. Co-developed-by: He Chen Signed-off-by: He Chen Co-developed-by: Zhang Yi Signed-off-by: Zhang Yi Signed-off-by: Yang Weijiang --- arch/x86/include/asm/kvm_host.h | 3 + arch/x86/kvm/mmu.h | 2 + arch/x86/kvm/mmu/spp.c | 242 ++++++++++++++++++++++++++++++++ arch/x86/kvm/mmu/spp.h | 12 ++ include/uapi/linux/kvm.h | 8 ++ 5 files changed, 267 insertions(+) diff --git a/arch/x86/include/asm/kvm_host.h b/arch/x86/include/asm/kvm_host.h index 9506c9d40895..4fb8816a328a 100644 --- a/arch/x86/include/asm/kvm_host.h +++ b/arch/x86/include/asm/kvm_host.h @@ -712,6 +712,7 @@ struct kvm_vcpu_arch { unsigned nmi_pending; /* NMI queued after currently running handler */ bool nmi_injected; /* Trying to inject an NMI this entry */ bool smi_pending; /* SMI queued after currently running handler */ + bool spp_pending; /* SPP has been requested, need to update VMCS */ struct kvm_mtrr mtrr_state; u64 pat; @@ -812,6 +813,7 @@ struct kvm_lpage_info { struct kvm_arch_memory_slot { struct kvm_rmap_head *rmap[KVM_NR_PAGE_SIZES]; + u32 *subpage_wp_info; struct kvm_lpage_info *lpage_info[KVM_NR_PAGE_SIZES - 1]; unsigned short *gfn_track[KVM_PAGE_TRACK_MAX]; }; @@ -959,6 +961,7 @@ struct kvm_arch { struct task_struct *nx_lpage_recovery_thread; hpa_t sppt_root; + bool spp_active; }; struct kvm_vm_stat { diff --git a/arch/x86/kvm/mmu.h b/arch/x86/kvm/mmu.h index d55674f44a18..a09ea3896efc 100644 --- a/arch/x86/kvm/mmu.h +++ b/arch/x86/kvm/mmu.h @@ -28,6 +28,8 @@ #define PT_GLOBAL_MASK (1ULL << 8) #define PT64_NX_SHIFT 63 #define PT64_NX_MASK (1ULL << PT64_NX_SHIFT) +#define PT_SPP_SHIFT 61 +#define PT_SPP_MASK (1ULL << PT_SPP_SHIFT) #define PT_PAT_SHIFT 7 #define PT_DIR_PAT_SHIFT 12 diff --git a/arch/x86/kvm/mmu/spp.c b/arch/x86/kvm/mmu/spp.c index 4247d6b1c6f7..55899eee4398 100644 --- a/arch/x86/kvm/mmu/spp.c +++ b/arch/x86/kvm/mmu/spp.c @@ -17,6 +17,68 @@ static void shadow_spp_walk_init(struct kvm_shadow_walk_iterator *iterator, iterator->level = PT64_ROOT_4LEVEL; } +u32 *gfn_to_subpage_wp_info(struct kvm_memory_slot *slot, gfn_t gfn) +{ + unsigned long idx; + + if (!slot->arch.subpage_wp_info) + return NULL; + + idx = gfn_to_index(gfn, slot->base_gfn, PT_PAGE_TABLE_LEVEL); + if (idx > slot->npages - 1) + return NULL; + + return &slot->arch.subpage_wp_info[idx]; +} + +static bool __rmap_update_subpage_bit(struct kvm *kvm, + struct kvm_rmap_head *rmap_head, + bool setbit) +{ + struct rmap_iterator iter; + bool flush = false; + u64 *sptep; + u64 spte; + + for_each_rmap_spte(rmap_head, &iter, sptep) { + /* + * SPP works only when the page is write-protected + * and SPP bit is set in EPT leaf entry. + */ + flush |= spte_write_protect(sptep, false); + spte = setbit ? (*sptep | PT_SPP_MASK) : + (*sptep & ~PT_SPP_MASK); + flush |= mmu_spte_update(sptep, spte); + } + + return flush; +} + +static int kvm_spp_update_write_protect(struct kvm *kvm, + struct kvm_memory_slot *slot, + gfn_t gfn, + bool enable) +{ + struct kvm_rmap_head *rmap_head; + bool flush = false; + + /* + * SPP is only supported with 4KB level1 memory page, check + * if the page is mapped in EPT leaf entry. + */ + rmap_head = __gfn_to_rmap(gfn, PT_PAGE_TABLE_LEVEL, slot); + + if (!rmap_head->val) + return -EFAULT; + + flush |= __rmap_update_subpage_bit(kvm, rmap_head, enable); + + if (flush) + kvm_flush_remote_tlbs(kvm); + + return 0; +} + struct kvm_mmu_page *kvm_spp_get_page(struct kvm_vcpu *vcpu, gfn_t gfn, unsigned int level) @@ -82,6 +144,20 @@ static void spp_spte_set(u64 *sptep, u64 new_spte) __set_spte(sptep, new_spte); } +static int kvm_spp_level_pages(gfn_t gfn_lower, gfn_t gfn_upper, int level) +{ + int page_num = KVM_PAGES_PER_HPAGE(level); + gfn_t gfn_max = (gfn_lower & ~(page_num - 1)) + page_num - 1; + int ret; + + if (gfn_upper <= gfn_max) + ret = gfn_upper - gfn_lower + 1; + else + ret = gfn_max - gfn_lower + 1; + + return ret; +} + int kvm_spp_setup_structure(struct kvm_vcpu *vcpu, u32 access_map, gfn_t gfn) { @@ -124,3 +200,169 @@ int kvm_spp_setup_structure(struct kvm_vcpu *vcpu, kvm_flush_remote_tlbs(vcpu->kvm); return ret; } + +int kvm_spp_get_permission(struct kvm *kvm, u64 gfn, u32 npages, + u32 *access_map) +{ + u32 *access; + struct kvm_memory_slot *slot; + int i; + + if (!kvm->arch.spp_active) + return -ENODEV; + + for (i = 0; i < npages; i++, gfn++) { + slot = gfn_to_memslot(kvm, gfn); + if (!slot) + return -EFAULT; + access = gfn_to_subpage_wp_info(slot, gfn); + if (!access) + return -EFAULT; + access_map[i] = *access; + } + + return i; +} + +static void kvm_spp_zap_pte(struct kvm *kvm, u64 *spte, int level) +{ + u64 pte; + + pte = *spte; + if (is_shadow_present_pte(pte) && is_last_spte(pte, level)) { + drop_spte(kvm, spte); + if (is_large_pte(pte)) + --kvm->stat.lpages; + } +} + +static bool kvm_spp_flush_rmap(struct kvm *kvm, u64 gfn_min, u64 gfn_max) +{ + u64 *sptep; + struct rmap_iterator iter; + struct kvm_rmap_head *rmap_head; + int level; + struct kvm_memory_slot *slot; + bool flush = false; + + slot = gfn_to_memslot(kvm, gfn_min); + if (!slot) + return false; + + for (; gfn_min <= gfn_max; gfn_min++) { + for (level = PT_PAGE_TABLE_LEVEL; + level <= PT_DIRECTORY_LEVEL; level++) { + rmap_head = __gfn_to_rmap(gfn_min, level, slot); + for_each_rmap_spte(rmap_head, &iter, sptep) { + pte_list_remove(rmap_head, sptep); + flush = true; + } + } + } + + return flush; +} + +int kvm_spp_set_permission(struct kvm *kvm, u64 gfn, u32 npages, + u32 *access_map) +{ + gfn_t old_gfn = gfn; + u32 *access; + struct kvm_memory_slot *slot; + struct kvm_shadow_walk_iterator iterator; + struct kvm_vcpu *vcpu; + gfn_t gfn_end; + int i, count, level; + bool flush = false; + + if (!kvm->arch.spp_active) + return -ENODEV; + + vcpu = kvm_get_vcpu(kvm, 0); + if (!VALID_PAGE(vcpu->kvm->arch.sppt_root)) + return -EFAULT; + + for (i = 0; i < npages; i++, gfn++) { + slot = gfn_to_memslot(kvm, gfn); + if (!slot) + return -EFAULT; + + access = gfn_to_subpage_wp_info(slot, gfn); + if (!access) + return -EFAULT; + *access = access_map[i]; + } + + gfn = old_gfn; + gfn_end = gfn + npages - 1; + vcpu = kvm_get_vcpu(kvm, 0); + + if (!vcpu || (vcpu && !VALID_PAGE(vcpu->arch.mmu->root_hpa))) + goto out; + + /* Flush any existing stale mappings in EPT before set up SPP */ + flush = kvm_spp_flush_rmap(kvm, gfn, gfn_end); + + for (i = 0; gfn <= gfn_end; i++, gfn++) { + for_each_shadow_entry(vcpu, (u64)gfn << PAGE_SHIFT, iterator) { + if (!is_shadow_present_pte(*iterator.sptep)) + break; + + if (iterator.level == PT_PAGE_TABLE_LEVEL) { + if (kvm_spp_mark_protection(kvm, + gfn, + access_map[i]) < 0) + return -EFAULT; + break; + } else if (is_large_pte(*iterator.sptep)) { + level = iterator.level; + if (access_map[i] == FULL_SPP_ACCESS) + break; + count = kvm_spp_level_pages(gfn, + gfn_end, + level); + /* + * Zap existing hugepage entry so that eligible + * 4KB mappings can be rebuilt in page_fault. + */ + kvm_spp_zap_pte(kvm, iterator.sptep, level); + flush = true; + if (count >= npages) + goto out; + gfn += count - 1; + } + } + } +out: + if (flush) + kvm_flush_remote_tlbs(kvm); + return npages; +} + +int kvm_spp_mark_protection(struct kvm *kvm, u64 gfn, u32 access) +{ + struct kvm_memory_slot *slot; + struct kvm_rmap_head *rmap_head; + int ret = 0; + bool enable; + + if (!kvm->arch.spp_active) + return -ENODEV; + + slot = gfn_to_memslot(kvm, gfn); + if (!slot) + return -EFAULT; + + /* + * check whether the target 4KB page exists in EPT leaf + * entry.If it's there, just flag SPP bit of the entry, + * defer the setup to SPPT miss induced vm-exit handler. + */ + rmap_head = __gfn_to_rmap(gfn, PT_PAGE_TABLE_LEVEL, slot); + + if (rmap_head->val) { + enable = access != FULL_SPP_ACCESS; + ret = kvm_spp_update_write_protect(kvm, slot, gfn, enable); + } + return ret; +} diff --git a/arch/x86/kvm/mmu/spp.h b/arch/x86/kvm/mmu/spp.h index 03e4dfad595a..9171e682be1f 100644 --- a/arch/x86/kvm/mmu/spp.h +++ b/arch/x86/kvm/mmu/spp.h @@ -2,4 +2,16 @@ #ifndef __KVM_X86_VMX_SPP_H #define __KVM_X86_VMX_SPP_H +#define FULL_SPP_ACCESS (u32)(BIT_ULL(32) - 1) + +int kvm_spp_get_permission(struct kvm *kvm, u64 gfn, u32 npages, + u32 *access_map); +int kvm_spp_set_permission(struct kvm *kvm, u64 gfn, u32 npages, + u32 *access_map); +int kvm_spp_mark_protection(struct kvm *kvm, u64 gfn, u32 access); + +int kvm_spp_setup_structure(struct kvm_vcpu *vcpu, + u32 access_map, gfn_t gfn); +u32 *gfn_to_subpage_wp_info(struct kvm_memory_slot *slot, gfn_t gfn); + #endif /* __KVM_X86_VMX_SPP_H */ diff --git a/include/uapi/linux/kvm.h b/include/uapi/linux/kvm.h index f0a16b4adbbd..eabd55ec5af7 100644 --- a/include/uapi/linux/kvm.h +++ b/include/uapi/linux/kvm.h @@ -102,6 +102,14 @@ struct kvm_userspace_memory_region { __u64 userspace_addr; /* start of the userspace allocated memory */ }; +/* for KVM_SUBPAGES_GET_ACCESS and KVM_SUBPAGES_SET_ACCESS */ +struct kvm_subpage { + __u64 gfn_base; /* the first page gfn of the contiguous pages */ + __u32 npages; /* number of 4K pages */ + __u32 flags; /* reserved to 0 now */ + __u32 access_map[0]; /* start place of bitmap array */ +}; + /* * The bit 0 ~ bit 15 of kvm_memory_region::flags are visible for userspace, * other bits are reserved for kvm internal use which are defined in From patchwork Sun Jan 19 04:05:01 2020 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: "Yang, Weijiang" X-Patchwork-Id: 11340523 Return-Path: Received: from mail.kernel.org (pdx-korg-mail-1.web.codeaurora.org [172.30.200.123]) by pdx-korg-patchwork-2.web.codeaurora.org (Postfix) with ESMTP id 93272184C for ; Sun, 19 Jan 2020 04:00:39 +0000 (UTC) Received: from vger.kernel.org (vger.kernel.org [209.132.180.67]) by mail.kernel.org (Postfix) with ESMTP id 71757206D5 for ; Sun, 19 Jan 2020 04:00:39 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1726915AbgASEAc (ORCPT ); Sat, 18 Jan 2020 23:00:32 -0500 Received: from mga14.intel.com ([192.55.52.115]:62796 "EHLO mga14.intel.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1726811AbgASEAb (ORCPT ); Sat, 18 Jan 2020 23:00:31 -0500 X-Amp-Result: SKIPPED(no attachment in message) X-Amp-File-Uploaded: False Received: from orsmga007.jf.intel.com ([10.7.209.58]) by fmsmga103.fm.intel.com with ESMTP/TLS/DHE-RSA-AES256-GCM-SHA384; 18 Jan 2020 20:00:31 -0800 X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="5.70,336,1574150400"; d="scan'208";a="214910486" Received: from local-michael-cet-test.sh.intel.com ([10.239.159.128]) by orsmga007.jf.intel.com with ESMTP; 18 Jan 2020 20:00:29 -0800 From: Yang Weijiang To: kvm@vger.kernel.org, linux-kernel@vger.kernel.org, pbonzini@redhat.com, jmattson@google.com, sean.j.christopherson@intel.com Cc: yu.c.zhang@linux.intel.com, alazar@bitdefender.com, edwin.zhai@intel.com, Yang Weijiang Subject: [PATCH v11 04/10] x86: spp: Introduce user-space SPP IOCTLs Date: Sun, 19 Jan 2020 12:05:01 +0800 Message-Id: <20200119040507.23113-5-weijiang.yang@intel.com> X-Mailer: git-send-email 2.17.2 In-Reply-To: <20200119040507.23113-1-weijiang.yang@intel.com> References: <20200119040507.23113-1-weijiang.yang@intel.com> Sender: kvm-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: kvm@vger.kernel.org Userspace application, e.g., QEMU or VMI, must initialize SPP before {gets|sets} SPP subpages. Co-developed-by: He Chen Signed-off-by: He Chen Co-developed-by: Zhang Yi Signed-off-by: Zhang Yi Signed-off-by: Yang Weijiang --- arch/x86/kvm/mmu/spp.c | 79 ++++++++++++++++++++++++++++++++++++ arch/x86/kvm/mmu/spp.h | 11 ++++- arch/x86/kvm/x86.c | 88 ++++++++++++++++++++++++++++++++++++++++ include/uapi/linux/kvm.h | 2 + 4 files changed, 179 insertions(+), 1 deletion(-) diff --git a/arch/x86/kvm/mmu/spp.c b/arch/x86/kvm/mmu/spp.c index 55899eee4398..c9f5180c403b 100644 --- a/arch/x86/kvm/mmu/spp.c +++ b/arch/x86/kvm/mmu/spp.c @@ -201,6 +201,46 @@ int kvm_spp_setup_structure(struct kvm_vcpu *vcpu, return ret; } +int spp_flush_sppt(struct kvm *kvm, u64 gfn_base, u32 npages) +{ + struct kvm_shadow_walk_iterator iter; + struct kvm_vcpu *vcpu; + gfn_t gfn = gfn_base; + gfn_t gfn_end = gfn_base + npages - 1; + u64 spde; + int count; + bool flush = false; + + vcpu = kvm_get_vcpu(kvm, 0); + if (!VALID_PAGE(vcpu->kvm->arch.sppt_root)) + return -EFAULT; + + for (; gfn <= gfn_end; gfn++) { + for_each_shadow_spp_entry(vcpu, (u64)gfn << PAGE_SHIFT, iter) { + if (!is_shadow_present_pte(*iter.sptep)) + break; + + if (iter.level != PT_DIRECTORY_LEVEL) + continue; + + spde = *iter.sptep; + spde &= ~PT_PRESENT_MASK; + spp_spte_set(iter.sptep, spde); + count = kvm_spp_level_pages(gfn, gfn_end, + PT_DIRECTORY_LEVEL); + flush = true; + if (count >= npages) + goto out; + gfn += count; + break; + } + } +out: + if (flush) + kvm_flush_remote_tlbs(kvm); + return 0; +} + int kvm_spp_get_permission(struct kvm *kvm, u64 gfn, u32 npages, u32 *access_map) { @@ -366,3 +406,42 @@ int kvm_spp_mark_protection(struct kvm *kvm, u64 gfn, u32 access) } return ret; } + +int kvm_vm_ioctl_get_subpages(struct kvm *kvm, + u64 gfn, + u32 npages, + u32 *access_map) +{ + int ret; + + mutex_lock(&kvm->slots_lock); + ret = kvm_spp_get_permission(kvm, gfn, npages, access_map); + mutex_unlock(&kvm->slots_lock); + + return ret; +} + +int kvm_vm_ioctl_set_subpages(struct kvm *kvm, + u64 gfn, + u32 npages, + u32 *access_map) +{ + int ret; + + spin_lock(&kvm->mmu_lock); + ret = spp_flush_sppt(kvm, gfn, npages); + spin_unlock(&kvm->mmu_lock); + + if (ret < 0) + return ret; + + mutex_lock(&kvm->slots_lock); + spin_lock(&kvm->mmu_lock); + + ret = kvm_spp_set_permission(kvm, gfn, npages, access_map); + + spin_unlock(&kvm->mmu_lock); + mutex_unlock(&kvm->slots_lock); + + return ret; +} diff --git a/arch/x86/kvm/mmu/spp.h b/arch/x86/kvm/mmu/spp.h index 9171e682be1f..75d4bfd64dbd 100644 --- a/arch/x86/kvm/mmu/spp.h +++ b/arch/x86/kvm/mmu/spp.h @@ -3,15 +3,24 @@ #define __KVM_X86_VMX_SPP_H #define FULL_SPP_ACCESS (u32)(BIT_ULL(32) - 1) +#define KVM_SUBPAGE_MAX_PAGES 512 int kvm_spp_get_permission(struct kvm *kvm, u64 gfn, u32 npages, u32 *access_map); int kvm_spp_set_permission(struct kvm *kvm, u64 gfn, u32 npages, u32 *access_map); int kvm_spp_mark_protection(struct kvm *kvm, u64 gfn, u32 access); - +int kvm_vm_ioctl_get_subpages(struct kvm *kvm, + u64 gfn, + u32 npages, + u32 *access_map); +int kvm_vm_ioctl_set_subpages(struct kvm *kvm, + u64 gfn, + u32 npages, + u32 *access_map); int kvm_spp_setup_structure(struct kvm_vcpu *vcpu, u32 access_map, gfn_t gfn); u32 *gfn_to_subpage_wp_info(struct kvm_memory_slot *slot, gfn_t gfn); +int spp_flush_sppt(struct kvm *kvm, u64 gfn_base, u32 npages); #endif /* __KVM_X86_VMX_SPP_H */ diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c index cf917139de6b..48871882a00c 100644 --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -26,6 +26,7 @@ #include "cpuid.h" #include "pmu.h" #include "hyperv.h" +#include "mmu/spp.h" #include #include @@ -5134,6 +5135,93 @@ long kvm_arch_vm_ioctl(struct file *filp, case KVM_SET_PMU_EVENT_FILTER: r = kvm_vm_ioctl_set_pmu_event_filter(kvm, argp); break; + case KVM_SUBPAGES_GET_ACCESS: { + struct kvm_subpage spp_info, *pinfo; + u32 total; + + r = -ENODEV; + if (!kvm->arch.spp_active) + goto out; + + r = -EFAULT; + if (copy_from_user(&spp_info, argp, sizeof(spp_info))) + goto out; + + r = -EINVAL; + if (spp_info.flags != 0 || + spp_info.npages > KVM_SUBPAGE_MAX_PAGES) + goto out; + r = 0; + if (!spp_info.npages) + goto out; + + total = sizeof(spp_info) + + sizeof(spp_info.access_map[0]) * spp_info.npages; + pinfo = kvzalloc(total, GFP_KERNEL_ACCOUNT); + + r = -ENOMEM; + if (!pinfo) + goto out; + + r = -EFAULT; + if (copy_from_user(pinfo, argp, total)) + goto out; + + r = kvm_vm_ioctl_get_subpages(kvm, + pinfo->gfn_base, + pinfo->npages, + pinfo->access_map); + if (r != pinfo->npages) + goto out; + + r = -EFAULT; + if (copy_to_user(argp, pinfo, total)) + goto out; + + r = pinfo->npages; + kfree(pinfo); + break; + } + case KVM_SUBPAGES_SET_ACCESS: { + struct kvm_subpage spp_info, *pinfo; + u32 total; + + r = -ENODEV; + if (!kvm->arch.spp_active) + goto out; + + r = -EFAULT; + if (copy_from_user(&spp_info, argp, sizeof(spp_info))) + goto out; + + r = -EINVAL; + if (spp_info.flags != 0 || + spp_info.npages > KVM_SUBPAGE_MAX_PAGES) + goto out; + + r = 0; + if (!spp_info.npages) + goto out; + + total = sizeof(spp_info) + + sizeof(spp_info.access_map[0]) * spp_info.npages; + pinfo = kvzalloc(total, GFP_KERNEL_ACCOUNT); + + r = -ENOMEM; + if (!pinfo) + goto out; + + r = -EFAULT; + if (copy_from_user(pinfo, argp, total)) + goto out; + + r = kvm_vm_ioctl_set_subpages(kvm, + pinfo->gfn_base, + pinfo->npages, + pinfo->access_map); + kfree(pinfo); + break; + } default: r = -ENOTTY; } diff --git a/include/uapi/linux/kvm.h b/include/uapi/linux/kvm.h index eabd55ec5af7..04af48ee6452 100644 --- a/include/uapi/linux/kvm.h +++ b/include/uapi/linux/kvm.h @@ -1272,6 +1272,8 @@ struct kvm_vfio_spapr_tce { struct kvm_userspace_memory_region) #define KVM_SET_TSS_ADDR _IO(KVMIO, 0x47) #define KVM_SET_IDENTITY_MAP_ADDR _IOW(KVMIO, 0x48, __u64) +#define KVM_SUBPAGES_GET_ACCESS _IOR(KVMIO, 0x49, __u64) +#define KVM_SUBPAGES_SET_ACCESS _IOW(KVMIO, 0x4a, __u64) /* enable ucontrol for s390 */ struct kvm_s390_ucas_mapping { From patchwork Sun Jan 19 04:05:02 2020 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: "Yang, Weijiang" X-Patchwork-Id: 11340535 Return-Path: Received: from mail.kernel.org (pdx-korg-mail-1.web.codeaurora.org [172.30.200.123]) by pdx-korg-patchwork-2.web.codeaurora.org (Postfix) with ESMTP id B073992A for ; Sun, 19 Jan 2020 04:01:05 +0000 (UTC) Received: from vger.kernel.org (vger.kernel.org [209.132.180.67]) by mail.kernel.org (Postfix) with ESMTP id 84646208E4 for ; Sun, 19 Jan 2020 04:01:05 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1726999AbgASEAf (ORCPT ); Sat, 18 Jan 2020 23:00:35 -0500 Received: from mga14.intel.com ([192.55.52.115]:62796 "EHLO mga14.intel.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1726982AbgASEAe (ORCPT ); Sat, 18 Jan 2020 23:00:34 -0500 X-Amp-Result: SKIPPED(no attachment in message) X-Amp-File-Uploaded: False Received: from orsmga007.jf.intel.com ([10.7.209.58]) by fmsmga103.fm.intel.com with ESMTP/TLS/DHE-RSA-AES256-GCM-SHA384; 18 Jan 2020 20:00:33 -0800 X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="5.70,336,1574150400"; d="scan'208";a="214910497" Received: from local-michael-cet-test.sh.intel.com ([10.239.159.128]) by orsmga007.jf.intel.com with ESMTP; 18 Jan 2020 20:00:31 -0800 From: Yang Weijiang To: kvm@vger.kernel.org, linux-kernel@vger.kernel.org, pbonzini@redhat.com, jmattson@google.com, sean.j.christopherson@intel.com Cc: yu.c.zhang@linux.intel.com, alazar@bitdefender.com, edwin.zhai@intel.com, Yang Weijiang Subject: [PATCH v11 05/10] vmx: spp: Handle SPP induced vmexit and EPT violation Date: Sun, 19 Jan 2020 12:05:02 +0800 Message-Id: <20200119040507.23113-6-weijiang.yang@intel.com> X-Mailer: git-send-email 2.17.2 In-Reply-To: <20200119040507.23113-1-weijiang.yang@intel.com> References: <20200119040507.23113-1-weijiang.yang@intel.com> Sender: kvm-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: kvm@vger.kernel.org If write to subpage is prohibited, EPT violation is generated and handled in fast_page_fault(). In current implementation, SPPT setup is handled in handle_spp() handler, it's triggered when SPP is enabled in EPT leaf entry while SPPT entry is invalid. There could be two kinds of SPP usages, one is for write-protection, the other is for access-tracking, the differece is the former keeps memory unchange while the latter just records the memory access and may let the write take effect. To fit these two cases, when SPP induced vmexit to userspace, the fault instruction length is returned, the application may take action according to the specific use-case, re-do write operation or discard it. To make SPP operatable with dirty-logging, introduce a free bit in EPT entry to store SPP bit, after dirty-logging happened, it restores SPP bit and make entry SPP protected again so that a retry write will trigger a normal SPP induced vmexit. Suggested-by: Paolo Bonzini Co-developed-by: He Chen Signed-off-by: He Chen Co-developed-by: Zhang Yi Signed-off-by: Zhang Yi Signed-off-by: Yang Weijiang --- arch/x86/include/asm/kvm_host.h | 2 + arch/x86/include/asm/vmx.h | 9 +++++ arch/x86/include/uapi/asm/vmx.h | 2 + arch/x86/kvm/mmu/mmu.c | 66 ++++++++++++++++++++++++++---- arch/x86/kvm/mmu/spp.c | 28 +++++++++++++ arch/x86/kvm/mmu/spp.h | 5 +++ arch/x86/kvm/trace.h | 66 ++++++++++++++++++++++++++++++ arch/x86/kvm/vmx/vmx.c | 72 +++++++++++++++++++++++++++++++++ arch/x86/kvm/x86.c | 3 ++ include/uapi/linux/kvm.h | 6 +++ 10 files changed, 252 insertions(+), 7 deletions(-) diff --git a/arch/x86/include/asm/kvm_host.h b/arch/x86/include/asm/kvm_host.h index 4fb8816a328a..0cf886e58004 100644 --- a/arch/x86/include/asm/kvm_host.h +++ b/arch/x86/include/asm/kvm_host.h @@ -1239,6 +1239,8 @@ struct kvm_x86_ops { bool (*apic_init_signal_blocked)(struct kvm_vcpu *vcpu); int (*enable_direct_tlbflush)(struct kvm_vcpu *vcpu); + + int (*get_insn_len)(struct kvm_vcpu *vcpu); }; struct kvm_arch_async_pf { diff --git a/arch/x86/include/asm/vmx.h b/arch/x86/include/asm/vmx.h index 1835767aa335..fc69ea8035fb 100644 --- a/arch/x86/include/asm/vmx.h +++ b/arch/x86/include/asm/vmx.h @@ -212,6 +212,8 @@ enum vmcs_field { XSS_EXIT_BITMAP_HIGH = 0x0000202D, ENCLS_EXITING_BITMAP = 0x0000202E, ENCLS_EXITING_BITMAP_HIGH = 0x0000202F, + SPPT_POINTER = 0x00002030, + SPPT_POINTER_HIGH = 0x00002031, TSC_MULTIPLIER = 0x00002032, TSC_MULTIPLIER_HIGH = 0x00002033, GUEST_PHYSICAL_ADDRESS = 0x00002400, @@ -533,6 +535,13 @@ struct vmx_msr_entry { #define EPT_VIOLATION_EXECUTABLE (1 << EPT_VIOLATION_EXECUTABLE_BIT) #define EPT_VIOLATION_GVA_TRANSLATED (1 << EPT_VIOLATION_GVA_TRANSLATED_BIT) +/* + * Exit Qualifications for SPPT-Induced vmexits + */ +#define SPPT_INDUCED_EXIT_TYPE_BIT 11 +#define SPPT_INDUCED_EXIT_TYPE (1 << SPPT_INDUCED_EXIT_TYPE_BIT) +#define SPPT_INTR_INFO_UNBLOCK_NMI INTR_INFO_UNBLOCK_NMI + /* * VM-instruction error numbers */ diff --git a/arch/x86/include/uapi/asm/vmx.h b/arch/x86/include/uapi/asm/vmx.h index 3eb8411ab60e..4833a39a799b 100644 --- a/arch/x86/include/uapi/asm/vmx.h +++ b/arch/x86/include/uapi/asm/vmx.h @@ -86,6 +86,7 @@ #define EXIT_REASON_PML_FULL 62 #define EXIT_REASON_XSAVES 63 #define EXIT_REASON_XRSTORS 64 +#define EXIT_REASON_SPP 66 #define EXIT_REASON_UMWAIT 67 #define EXIT_REASON_TPAUSE 68 @@ -145,6 +146,7 @@ { EXIT_REASON_ENCLS, "ENCLS" }, \ { EXIT_REASON_RDSEED, "RDSEED" }, \ { EXIT_REASON_PML_FULL, "PML_FULL" }, \ + { EXIT_REASON_SPP, "SPP" }, \ { EXIT_REASON_XSAVES, "XSAVES" }, \ { EXIT_REASON_XRSTORS, "XRSTORS" }, \ { EXIT_REASON_UMWAIT, "UMWAIT" }, \ diff --git a/arch/x86/kvm/mmu/mmu.c b/arch/x86/kvm/mmu/mmu.c index dff52763e05c..665923deb4a9 100644 --- a/arch/x86/kvm/mmu/mmu.c +++ b/arch/x86/kvm/mmu/mmu.c @@ -20,6 +20,7 @@ #include "x86.h" #include "kvm_cache_regs.h" #include "cpuid.h" +#include "spp.h" #include #include @@ -177,6 +178,7 @@ module_param(dbg, bool, 0644); /* The mask for the R/X bits in EPT PTEs */ #define PT64_EPT_READABLE_MASK 0x1ull #define PT64_EPT_EXECUTABLE_MASK 0x4ull +#define PT64_SPP_SAVED_BIT (1ULL << (PT64_SECOND_AVAIL_BITS_SHIFT + 1)) #include @@ -200,6 +202,7 @@ enum { RET_PF_RETRY = 0, RET_PF_EMULATE = 1, RET_PF_INVALID = 2, + RET_PF_USERSPACE = 3, }; struct pte_list_desc { @@ -979,6 +982,9 @@ static u64 mark_spte_for_access_track(u64 spte) shadow_acc_track_saved_bits_shift; spte &= ~shadow_acc_track_mask; + if (spte & PT_SPP_MASK) + save_spp_bit(&spte); + return spte; } @@ -1634,9 +1640,16 @@ static void drop_large_spte(struct kvm_vcpu *vcpu, u64 *sptep) static bool spte_write_protect(u64 *sptep, bool pt_protect) { u64 spte = *sptep; + bool spp_protected = false; + + if (spte & PT_SPP_MASK) { + save_spp_bit(&spte); + spp_protected = true; + } if (!is_writable_pte(spte) && - !(pt_protect && spte_can_locklessly_be_made_writable(spte))) + !(pt_protect && spte_can_locklessly_be_made_writable(spte)) && + !spp_protected) return false; rmap_printk("rmap_write_protect: spte %p %llx\n", sptep, *sptep); @@ -1677,9 +1690,15 @@ static bool spte_wrprot_for_clear_dirty(u64 *sptep) { bool was_writable = test_and_clear_bit(PT_WRITABLE_SHIFT, (unsigned long *)sptep); + bool was_spp_armed = test_and_clear_bit(PT_SPP_SHIFT, + (unsigned long *)sptep); + if (was_writable && !spte_ad_enabled(*sptep)) kvm_set_pfn_dirty(spte_to_pfn(*sptep)); + if (was_spp_armed) + *sptep |= PT64_SPP_SAVED_BIT; + return was_writable; } @@ -3478,7 +3497,8 @@ static bool page_fault_can_be_fast(u32 error_code) */ static bool fast_pf_fix_direct_spte(struct kvm_vcpu *vcpu, struct kvm_mmu_page *sp, - u64 *sptep, u64 old_spte, u64 new_spte) + u64 *sptep, u64 old_spte, u64 new_spte, + bool spp_protected) { gfn_t gfn; @@ -3499,7 +3519,8 @@ fast_pf_fix_direct_spte(struct kvm_vcpu *vcpu, struct kvm_mmu_page *sp, if (cmpxchg64(sptep, old_spte, new_spte) != old_spte) return false; - if (is_writable_pte(new_spte) && !is_writable_pte(old_spte)) { + if ((is_writable_pte(new_spte) && !is_writable_pte(old_spte)) || + spp_protected) { /* * The gfn of direct spte is stable since it is * calculated by sp->gfn. @@ -3534,6 +3555,7 @@ static bool fast_page_fault(struct kvm_vcpu *vcpu, gva_t gva, int level, struct kvm_shadow_walk_iterator iterator; struct kvm_mmu_page *sp; bool fault_handled = false; + bool spp_protected = false; u64 spte = 0ull; uint retry_count = 0; @@ -3585,7 +3607,30 @@ static bool fast_page_fault(struct kvm_vcpu *vcpu, gva_t gva, int level, if ((error_code & PFERR_WRITE_MASK) && spte_can_locklessly_be_made_writable(spte)) { - new_spte |= PT_WRITABLE_MASK; + /* + * Record write protect fault caused by + * Sub-page Protection, let VMI decide + * the next step. + */ + if (spte & PT_SPP_MASK) { + int len = kvm_x86_ops->get_insn_len(vcpu); + + fault_handled = true; + vcpu->run->exit_reason = KVM_EXIT_SPP; + vcpu->run->spp.addr = gva; + vcpu->run->spp.insn_len = len; + trace_kvm_spp_induced_page_fault(vcpu, + gva, + len); + break; + } + + if (was_spp_armed(new_spte)) { + restore_spp_bit(&new_spte); + spp_protected = true; + } else { + new_spte |= PT_WRITABLE_MASK; + } /* * Do not fix write-permission on the large spte. Since @@ -3604,7 +3649,8 @@ static bool fast_page_fault(struct kvm_vcpu *vcpu, gva_t gva, int level, /* Verify that the fault can be handled in the fast path */ if (new_spte == spte || - !is_access_allowed(error_code, new_spte)) + (!is_access_allowed(error_code, new_spte) && + !spp_protected)) break; /* @@ -3614,7 +3660,8 @@ static bool fast_page_fault(struct kvm_vcpu *vcpu, gva_t gva, int level, */ fault_handled = fast_pf_fix_direct_spte(vcpu, sp, iterator.sptep, spte, - new_spte); + new_spte, + spp_protected); if (fault_handled) break; @@ -5223,7 +5270,6 @@ void kvm_init_mmu(struct kvm_vcpu *vcpu, bool reset_roots) uint i; vcpu->arch.mmu->root_hpa = INVALID_PAGE; - for (i = 0; i < KVM_MMU_NUM_PREV_ROOTS; i++) vcpu->arch.mmu->prev_roots[i] = KVM_MMU_ROOT_INFO_INVALID; } @@ -5539,6 +5585,10 @@ int kvm_mmu_page_fault(struct kvm_vcpu *vcpu, gva_t cr2, u64 error_code, r = vcpu->arch.mmu->page_fault(vcpu, cr2, lower_32_bits(error_code), false); + + if (vcpu->run->exit_reason == KVM_EXIT_SPP) + r = RET_PF_USERSPACE; + WARN_ON(r == RET_PF_INVALID); } @@ -5547,6 +5597,8 @@ int kvm_mmu_page_fault(struct kvm_vcpu *vcpu, gva_t cr2, u64 error_code, if (r < 0) return r; + if (r == RET_PF_USERSPACE) + return 0; /* * Before emulating the instruction, check if the error code * was due to a RO violation while translating the guest page. diff --git a/arch/x86/kvm/mmu/spp.c b/arch/x86/kvm/mmu/spp.c index c9f5180c403b..2f2558c0041d 100644 --- a/arch/x86/kvm/mmu/spp.c +++ b/arch/x86/kvm/mmu/spp.c @@ -17,6 +17,25 @@ static void shadow_spp_walk_init(struct kvm_shadow_walk_iterator *iterator, iterator->level = PT64_ROOT_4LEVEL; } +/* Save reserved bit for SPP armed PTE */ +void save_spp_bit(u64 *spte) +{ + *spte |= PT64_SPP_SAVED_BIT; + *spte &= ~PT_SPP_MASK; +} + +/* Restore reserved bit for SPP armed PTE */ +void restore_spp_bit(u64 *spte) +{ + *spte &= ~PT64_SPP_SAVED_BIT; + *spte |= PT_SPP_MASK; +} + +bool was_spp_armed(u64 spte) +{ + return !!(spte & PT64_SPP_SAVED_BIT); +} + u32 *gfn_to_subpage_wp_info(struct kvm_memory_slot *slot, gfn_t gfn) { unsigned long idx; @@ -30,6 +49,7 @@ u32 *gfn_to_subpage_wp_info(struct kvm_memory_slot *slot, gfn_t gfn) return &slot->arch.subpage_wp_info[idx]; } +EXPORT_SYMBOL_GPL(gfn_to_subpage_wp_info); static bool __rmap_update_subpage_bit(struct kvm *kvm, struct kvm_rmap_head *rmap_head, @@ -200,6 +220,7 @@ int kvm_spp_setup_structure(struct kvm_vcpu *vcpu, kvm_flush_remote_tlbs(vcpu->kvm); return ret; } +EXPORT_SYMBOL_GPL(kvm_spp_setup_structure); int spp_flush_sppt(struct kvm *kvm, u64 gfn_base, u32 npages) { @@ -331,6 +352,7 @@ int kvm_spp_set_permission(struct kvm *kvm, u64 gfn, u32 npages, if (!access) return -EFAULT; *access = access_map[i]; + trace_kvm_spp_set_subpages(vcpu, gfn, *access); } gfn = old_gfn; @@ -445,3 +467,9 @@ int kvm_vm_ioctl_set_subpages(struct kvm *kvm, return ret; } + +inline u64 construct_spptp(unsigned long root_hpa) +{ + return root_hpa & PAGE_MASK; +} +EXPORT_SYMBOL_GPL(construct_spptp); diff --git a/arch/x86/kvm/mmu/spp.h b/arch/x86/kvm/mmu/spp.h index 75d4bfd64dbd..c3588c20be52 100644 --- a/arch/x86/kvm/mmu/spp.h +++ b/arch/x86/kvm/mmu/spp.h @@ -4,6 +4,7 @@ #define FULL_SPP_ACCESS (u32)(BIT_ULL(32) - 1) #define KVM_SUBPAGE_MAX_PAGES 512 +#define MAX_ENTRIES_PER_MMUPAGE BIT(9) int kvm_spp_get_permission(struct kvm *kvm, u64 gfn, u32 npages, u32 *access_map); @@ -22,5 +23,9 @@ int kvm_spp_setup_structure(struct kvm_vcpu *vcpu, u32 access_map, gfn_t gfn); u32 *gfn_to_subpage_wp_info(struct kvm_memory_slot *slot, gfn_t gfn); int spp_flush_sppt(struct kvm *kvm, u64 gfn_base, u32 npages); +void save_spp_bit(u64 *spte); +void restore_spp_bit(u64 *spte); +bool was_spp_armed(u64 spte); +u64 construct_spptp(unsigned long root_hpa); #endif /* __KVM_X86_VMX_SPP_H */ diff --git a/arch/x86/kvm/trace.h b/arch/x86/kvm/trace.h index 7c741a0c5f80..afd2ddcc8ebd 100644 --- a/arch/x86/kvm/trace.h +++ b/arch/x86/kvm/trace.h @@ -1496,6 +1496,72 @@ TRACE_EVENT(kvm_nested_vmenter_failed, __print_symbolic(__entry->err, VMX_VMENTER_INSTRUCTION_ERRORS)) ); +TRACE_EVENT(kvm_spp_set_subpages, + TP_PROTO(struct kvm_vcpu *vcpu, gfn_t gfn, u32 access), + TP_ARGS(vcpu, gfn, access), + + TP_STRUCT__entry( + __field(int, vcpu_id) + __field(gfn_t, gfn) + __field(u32, access) + ), + + TP_fast_assign( + __entry->vcpu_id = vcpu->vcpu_id; + __entry->gfn = gfn; + __entry->access = access; + ), + + TP_printk("vcpu %d gfn %llx access %x", + __entry->vcpu_id, + __entry->gfn, + __entry->access) +); + +TRACE_EVENT(kvm_spp_induced_exit, + TP_PROTO(struct kvm_vcpu *vcpu, gpa_t gpa, u32 qualification), + TP_ARGS(vcpu, gpa, qualification), + + TP_STRUCT__entry( + __field(int, vcpu_id) + __field(gpa_t, gpa) + __field(u32, qualification) + ), + + TP_fast_assign( + __entry->vcpu_id = vcpu->vcpu_id; + __entry->gpa = gpa; + __entry->qualification = qualification; + ), + + TP_printk("vcpu %d gpa %llx qualificaiton %x", + __entry->vcpu_id, + __entry->gpa, + __entry->qualification) +); + +TRACE_EVENT(kvm_spp_induced_page_fault, + TP_PROTO(struct kvm_vcpu *vcpu, gpa_t gpa, int insn_len), + TP_ARGS(vcpu, gpa, insn_len), + + TP_STRUCT__entry( + __field(int, vcpu_id) + __field(gpa_t, gpa) + __field(int, insn_len) + ), + + TP_fast_assign( + __entry->vcpu_id = vcpu->vcpu_id; + __entry->gpa = gpa; + __entry->insn_len = insn_len; + ), + + TP_printk("vcpu %d gpa %llx insn_len %d", + __entry->vcpu_id, + __entry->gpa, + __entry->insn_len) +); + #endif /* _TRACE_KVM_H */ #undef TRACE_INCLUDE_PATH diff --git a/arch/x86/kvm/vmx/vmx.c b/arch/x86/kvm/vmx/vmx.c index e3394c839dea..4032e615ca85 100644 --- a/arch/x86/kvm/vmx/vmx.c +++ b/arch/x86/kvm/vmx/vmx.c @@ -60,6 +60,7 @@ #include "vmcs12.h" #include "vmx.h" #include "x86.h" +#include "mmu/spp.h" MODULE_AUTHOR("Qumranet"); MODULE_LICENSE("GPL"); @@ -1422,6 +1423,11 @@ static bool emulation_required(struct kvm_vcpu *vcpu) return emulate_invalid_guest_state && !guest_state_valid(vcpu); } +static int vmx_get_insn_len(struct kvm_vcpu *vcpu) +{ + return vmcs_read32(VM_EXIT_INSTRUCTION_LEN); +} + static void vmx_decache_cr0_guest_bits(struct kvm_vcpu *vcpu); unsigned long vmx_get_rflags(struct kvm_vcpu *vcpu) @@ -5350,6 +5356,69 @@ static int handle_monitor_trap(struct kvm_vcpu *vcpu) return 1; } +static int handle_spp(struct kvm_vcpu *vcpu) +{ + unsigned long exit_qualification; + struct kvm_memory_slot *slot; + gfn_t gfn, gfn_end; + u32 *access; + gpa_t gpa; + + exit_qualification = vmcs_readl(EXIT_QUALIFICATION); + + /* + * SPP VM exit happened while executing iret from NMI, + * "blocked by NMI" bit has to be set before next VM entry. + * There are errata that may cause this bit to not be set: + * AAK134, BY25. + */ + if (!(to_vmx(vcpu)->idt_vectoring_info & VECTORING_INFO_VALID_MASK) && + (exit_qualification & SPPT_INTR_INFO_UNBLOCK_NMI)) + vmcs_set_bits(GUEST_INTERRUPTIBILITY_INFO, + GUEST_INTR_STATE_NMI); + + vcpu->arch.exit_qualification = exit_qualification; + if (WARN_ON(!(exit_qualification & SPPT_INDUCED_EXIT_TYPE))) + goto out_err; + /* + * SPPT missing + * We don't set SPP write access for the corresponding + * GPA, if we haven't setup, we need to construct + * SPP table here. + */ + gpa = vmcs_read64(GUEST_PHYSICAL_ADDRESS); + gfn = gpa_to_gfn(gpa); + trace_kvm_spp_induced_exit(vcpu, gpa, exit_qualification); + /* + * In level 1 of SPPT, there's no PRESENT bit, all data is + * regarded as permission vector, so need to check from + * level 2 to set up the vector if target page is protected. + */ + spin_lock(&vcpu->kvm->mmu_lock); + gfn &= ~(MAX_ENTRIES_PER_MMUPAGE - 1); + gfn_end = gfn + MAX_ENTRIES_PER_MMUPAGE; + for (; gfn < gfn_end; gfn++) { + slot = gfn_to_memslot(vcpu->kvm, gfn); + if (!slot) + continue; + access = gfn_to_subpage_wp_info(slot, gfn); + if (access && *access != FULL_SPP_ACCESS) + kvm_spp_setup_structure(vcpu, *access, gfn); + } + spin_unlock(&vcpu->kvm->mmu_lock); + return 1; +out_err: + /* + * SPPT Misconfig + * This is probably caused by some mis-configuration in SPPT + * entries, cannot handle it here, escalate the fault to + * emulator. + */ + vcpu->run->exit_reason = KVM_EXIT_UNKNOWN; + vcpu->run->hw.hardware_exit_reason = EXIT_REASON_SPP; + return 0; +} + static int handle_monitor(struct kvm_vcpu *vcpu) { printk_once(KERN_WARNING "kvm: MONITOR instruction emulated as NOP!\n"); @@ -5564,6 +5633,7 @@ static int (*kvm_vmx_exit_handlers[])(struct kvm_vcpu *vcpu) = { [EXIT_REASON_INVVPID] = handle_vmx_instruction, [EXIT_REASON_RDRAND] = handle_invalid_op, [EXIT_REASON_RDSEED] = handle_invalid_op, + [EXIT_REASON_SPP] = handle_spp, [EXIT_REASON_PML_FULL] = handle_pml_full, [EXIT_REASON_INVPCID] = handle_invpcid, [EXIT_REASON_VMFUNC] = handle_vmx_instruction, @@ -7908,6 +7978,8 @@ static struct kvm_x86_ops vmx_x86_ops __ro_after_init = { .nested_get_evmcs_version = NULL, .need_emulation_on_page_fault = vmx_need_emulation_on_page_fault, .apic_init_signal_blocked = vmx_apic_init_signal_blocked, + + .get_insn_len = vmx_get_insn_len, }; static void vmx_cleanup_l1d_flush(void) diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c index 48871882a00c..e93641d631cb 100644 --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -10401,3 +10401,6 @@ EXPORT_TRACEPOINT_SYMBOL_GPL(kvm_pml_full); EXPORT_TRACEPOINT_SYMBOL_GPL(kvm_pi_irte_update); EXPORT_TRACEPOINT_SYMBOL_GPL(kvm_avic_unaccelerated_access); EXPORT_TRACEPOINT_SYMBOL_GPL(kvm_avic_incomplete_ipi); +EXPORT_TRACEPOINT_SYMBOL_GPL(kvm_spp_set_subpages); +EXPORT_TRACEPOINT_SYMBOL_GPL(kvm_spp_induced_exit); +EXPORT_TRACEPOINT_SYMBOL_GPL(kvm_spp_induced_page_fault); diff --git a/include/uapi/linux/kvm.h b/include/uapi/linux/kvm.h index 04af48ee6452..280a6d52e5ff 100644 --- a/include/uapi/linux/kvm.h +++ b/include/uapi/linux/kvm.h @@ -244,6 +244,7 @@ struct kvm_hyperv_exit { #define KVM_EXIT_IOAPIC_EOI 26 #define KVM_EXIT_HYPERV 27 #define KVM_EXIT_ARM_NISV 28 +#define KVM_EXIT_SPP 29 /* For KVM_EXIT_INTERNAL_ERROR */ /* Emulate instruction failed. */ @@ -401,6 +402,11 @@ struct kvm_run { struct { __u8 vector; } eoi; + /* KVM_EXIT_SPP */ + struct { + __u64 addr; + __u8 insn_len; + } spp; /* KVM_EXIT_HYPERV */ struct kvm_hyperv_exit hyperv; /* KVM_EXIT_ARM_NISV */ From patchwork Sun Jan 19 04:05:03 2020 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: "Yang, Weijiang" X-Patchwork-Id: 11340525 Return-Path: Received: from mail.kernel.org (pdx-korg-mail-1.web.codeaurora.org [172.30.200.123]) by pdx-korg-patchwork-2.web.codeaurora.org (Postfix) with ESMTP id BECE71921 for ; Sun, 19 Jan 2020 04:00:39 +0000 (UTC) Received: from vger.kernel.org (vger.kernel.org [209.132.180.67]) by mail.kernel.org (Postfix) with ESMTP id 9C256206D5 for ; Sun, 19 Jan 2020 04:00:39 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1727048AbgASEAg (ORCPT ); Sat, 18 Jan 2020 23:00:36 -0500 Received: from mga14.intel.com ([192.55.52.115]:62796 "EHLO mga14.intel.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1727031AbgASEAg (ORCPT ); Sat, 18 Jan 2020 23:00:36 -0500 X-Amp-Result: SKIPPED(no attachment in message) X-Amp-File-Uploaded: False Received: from orsmga007.jf.intel.com ([10.7.209.58]) by fmsmga103.fm.intel.com with ESMTP/TLS/DHE-RSA-AES256-GCM-SHA384; 18 Jan 2020 20:00:35 -0800 X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="5.70,336,1574150400"; d="scan'208";a="214910507" Received: from local-michael-cet-test.sh.intel.com ([10.239.159.128]) by orsmga007.jf.intel.com with ESMTP; 18 Jan 2020 20:00:33 -0800 From: Yang Weijiang To: kvm@vger.kernel.org, linux-kernel@vger.kernel.org, pbonzini@redhat.com, jmattson@google.com, sean.j.christopherson@intel.com Cc: yu.c.zhang@linux.intel.com, alazar@bitdefender.com, edwin.zhai@intel.com, Yang Weijiang Subject: [PATCH v11 06/10] mmu: spp: Enable Lazy mode SPP protection Date: Sun, 19 Jan 2020 12:05:03 +0800 Message-Id: <20200119040507.23113-7-weijiang.yang@intel.com> X-Mailer: git-send-email 2.17.2 In-Reply-To: <20200119040507.23113-1-weijiang.yang@intel.com> References: <20200119040507.23113-1-weijiang.yang@intel.com> Sender: kvm-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: kvm@vger.kernel.org When SPP protection is set but the gfn->pfn mapping isn't there, we need to check and mark SPP protection in EPT while gfn->pfn mapping is being built, but the setup for SPPT is deferred to handle_spp() handler. From HW's capability, SPP only works for 4KB mappings, to apply SPP protection for hugepage(2MB,1GB) cases, hugepage entries need to be zapped before SPP set up. In tdp_page_fault(), there's check for SPP protection before sets 4KB, 2MB or 1GB mapping, the target is introducing least impact to hugepage setup, i.e., falls back to most possible hugepage mapping. Suggested-by: Paolo Bonzini Signed-off-by: Yang Weijiang --- arch/x86/kvm/mmu/mmu.c | 20 ++++++++++++++++++++ arch/x86/kvm/mmu/spp.c | 43 ++++++++++++++++++++++++++++++++++++++++++ arch/x86/kvm/mmu/spp.h | 2 ++ 3 files changed, 65 insertions(+) diff --git a/arch/x86/kvm/mmu/mmu.c b/arch/x86/kvm/mmu/mmu.c index 665923deb4a9..fe14f60928a2 100644 --- a/arch/x86/kvm/mmu/mmu.c +++ b/arch/x86/kvm/mmu/mmu.c @@ -3251,6 +3251,7 @@ static int direct_pte_prefetch_many(struct kvm_vcpu *vcpu, unsigned access = sp->role.access; int i, ret; gfn_t gfn; + u32 *wp_bitmap; gfn = kvm_mmu_page_get_gfn(sp, start - sp->spt); slot = gfn_to_memslot_dirty_bitmap(vcpu, gfn, access & ACC_WRITE_MASK); @@ -3264,6 +3265,13 @@ static int direct_pte_prefetch_many(struct kvm_vcpu *vcpu, for (i = 0; i < ret; i++, gfn++, start++) { mmu_set_spte(vcpu, start, access, 0, sp->role.level, gfn, page_to_pfn(pages[i]), true, true); + if (vcpu->kvm->arch.spp_active) { + wp_bitmap = gfn_to_subpage_wp_info(slot, gfn); + if (wp_bitmap && *wp_bitmap != FULL_SPP_ACCESS) + kvm_spp_mark_protection(vcpu->kvm, + gfn, + *wp_bitmap); + } put_page(pages[i]); } @@ -3377,6 +3385,15 @@ static int __direct_map(struct kvm_vcpu *vcpu, gpa_t gpa, int write, map_writable); direct_pte_prefetch(vcpu, it.sptep); ++vcpu->stat.pf_fixed; + if (level == PT_PAGE_TABLE_LEVEL) { + int ret; + u32 access; + + ret = kvm_spp_get_permission(vcpu->kvm, gfn, 1, &access); + if (ret == 1 && access != FULL_SPP_ACCESS) + kvm_spp_mark_protection(vcpu->kvm, gfn, access); + } + return ret; } @@ -4337,6 +4354,9 @@ static int tdp_page_fault(struct kvm_vcpu *vcpu, gva_t gpa, u32 error_code, if (level > PT_DIRECTORY_LEVEL && !check_hugepage_cache_consistency(vcpu, gfn, level)) level = PT_DIRECTORY_LEVEL; + + check_spp_protection(vcpu, gfn, &force_pt_level, &level); + gfn &= ~(KVM_PAGES_PER_HPAGE(level) - 1); } diff --git a/arch/x86/kvm/mmu/spp.c b/arch/x86/kvm/mmu/spp.c index 2f2558c0041d..eb540e1b5133 100644 --- a/arch/x86/kvm/mmu/spp.c +++ b/arch/x86/kvm/mmu/spp.c @@ -429,6 +429,49 @@ int kvm_spp_mark_protection(struct kvm *kvm, u64 gfn, u32 access) return ret; } +static bool is_spp_protected(struct kvm_memory_slot *slot, gfn_t gfn, int level) +{ + int page_num = KVM_PAGES_PER_HPAGE(level); + u32 *access; + gfn_t gfn_max; + + gfn &= ~(page_num - 1); + gfn_max = gfn + page_num - 1; + for (; gfn <= gfn_max; gfn++) { + access = gfn_to_subpage_wp_info(slot, gfn); + if (access && *access != FULL_SPP_ACCESS) + return true; + } + return false; +} + +bool check_spp_protection(struct kvm_vcpu *vcpu, gfn_t gfn, + bool *force_pt_level, int *level) +{ + struct kvm *kvm = vcpu->kvm; + struct kvm_memory_slot *slot; + bool protected; + int old_level = *level; + + if (!kvm->arch.spp_active) + return false; + + slot = gfn_to_memslot(kvm, gfn); + + if (!slot) + return false; + protected = is_spp_protected(slot, gfn, PT_DIRECTORY_LEVEL); + + if (protected) { + *level = PT_PAGE_TABLE_LEVEL; + *force_pt_level = true; + } else if (*level == PT_PDPE_LEVEL && + is_spp_protected(slot, gfn, PT_PDPE_LEVEL)) + *level = PT_DIRECTORY_LEVEL; + + return (old_level != *level); +} + int kvm_vm_ioctl_get_subpages(struct kvm *kvm, u64 gfn, u32 npages, diff --git a/arch/x86/kvm/mmu/spp.h b/arch/x86/kvm/mmu/spp.h index c3588c20be52..51a209a04863 100644 --- a/arch/x86/kvm/mmu/spp.h +++ b/arch/x86/kvm/mmu/spp.h @@ -11,6 +11,8 @@ int kvm_spp_get_permission(struct kvm *kvm, u64 gfn, u32 npages, int kvm_spp_set_permission(struct kvm *kvm, u64 gfn, u32 npages, u32 *access_map); int kvm_spp_mark_protection(struct kvm *kvm, u64 gfn, u32 access); +bool check_spp_protection(struct kvm_vcpu *vcpu, gfn_t gfn, + bool *force_pt_level, int *level); int kvm_vm_ioctl_get_subpages(struct kvm *kvm, u64 gfn, u32 npages, From patchwork Sun Jan 19 04:05:04 2020 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: "Yang, Weijiang" X-Patchwork-Id: 11340533 Return-Path: Received: from mail.kernel.org (pdx-korg-mail-1.web.codeaurora.org [172.30.200.123]) by pdx-korg-patchwork-2.web.codeaurora.org (Postfix) with ESMTP id BEA0292A for ; Sun, 19 Jan 2020 04:01:00 +0000 (UTC) Received: from vger.kernel.org (vger.kernel.org [209.132.180.67]) by mail.kernel.org (Postfix) with ESMTP id 9CEBC2064C for ; Sun, 19 Jan 2020 04:01:00 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1727031AbgASEAk (ORCPT ); Sat, 18 Jan 2020 23:00:40 -0500 Received: from mga14.intel.com ([192.55.52.115]:62796 "EHLO mga14.intel.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1727066AbgASEAi (ORCPT ); Sat, 18 Jan 2020 23:00:38 -0500 X-Amp-Result: SKIPPED(no attachment in message) X-Amp-File-Uploaded: False Received: from orsmga007.jf.intel.com ([10.7.209.58]) by fmsmga103.fm.intel.com with ESMTP/TLS/DHE-RSA-AES256-GCM-SHA384; 18 Jan 2020 20:00:37 -0800 X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="5.70,336,1574150400"; d="scan'208";a="214910512" Received: from local-michael-cet-test.sh.intel.com ([10.239.159.128]) by orsmga007.jf.intel.com with ESMTP; 18 Jan 2020 20:00:35 -0800 From: Yang Weijiang To: kvm@vger.kernel.org, linux-kernel@vger.kernel.org, pbonzini@redhat.com, jmattson@google.com, sean.j.christopherson@intel.com Cc: yu.c.zhang@linux.intel.com, alazar@bitdefender.com, edwin.zhai@intel.com, Yang Weijiang Subject: [PATCH v11 07/10] mmu: spp: Re-enable SPP protection when EPT mapping changes Date: Sun, 19 Jan 2020 12:05:04 +0800 Message-Id: <20200119040507.23113-8-weijiang.yang@intel.com> X-Mailer: git-send-email 2.17.2 In-Reply-To: <20200119040507.23113-1-weijiang.yang@intel.com> References: <20200119040507.23113-1-weijiang.yang@intel.com> Sender: kvm-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: kvm@vger.kernel.org Host page swapping/migration may change the translation in EPT leaf entry, if the target page is SPP protected, re-enable SPP protection. When SPPT mmu-page is reclaimed, no need to clear rmap as no memory-mapping is in SPPT L4E. Signed-off-by: Yang Weijiang --- arch/x86/kvm/mmu/mmu.c | 17 +++++++++++++++++ 1 file changed, 17 insertions(+) diff --git a/arch/x86/kvm/mmu/mmu.c b/arch/x86/kvm/mmu/mmu.c index fe14f60928a2..099f92f0c42a 100644 --- a/arch/x86/kvm/mmu/mmu.c +++ b/arch/x86/kvm/mmu/mmu.c @@ -1918,6 +1918,19 @@ static int kvm_set_pte_rmapp(struct kvm *kvm, struct kvm_rmap_head *rmap_head, new_spte &= ~PT_WRITABLE_MASK; new_spte &= ~SPTE_HOST_WRITEABLE; + /* + * if it's EPT leaf entry and the physical page is + * SPP protected, then re-enable SPP protection for + * the page. + */ + if (kvm->arch.spp_active && + level == PT_PAGE_TABLE_LEVEL) { + u32 *access = gfn_to_subpage_wp_info(slot, gfn); + + if (access && *access != FULL_SPP_ACCESS) + new_spte |= PT_SPP_MASK; + } + new_spte = mark_spte_for_access_track(new_spte); mmu_spte_clear_track_bits(sptep); @@ -2768,6 +2781,10 @@ static bool mmu_page_zap_pte(struct kvm *kvm, struct kvm_mmu_page *sp, pte = *spte; if (is_shadow_present_pte(pte)) { if (is_last_spte(pte, sp->role.level)) { + /* SPPT leaf entries don't have rmaps*/ + if (sp->role.spp && sp->role.level == + PT_PAGE_TABLE_LEVEL) + return true; drop_spte(kvm, spte); if (is_large_pte(pte)) --kvm->stat.lpages; From patchwork Sun Jan 19 04:05:05 2020 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: "Yang, Weijiang" X-Patchwork-Id: 11340529 Return-Path: Received: from mail.kernel.org (pdx-korg-mail-1.web.codeaurora.org [172.30.200.123]) by pdx-korg-patchwork-2.web.codeaurora.org (Postfix) with ESMTP id 85F0A92A for ; Sun, 19 Jan 2020 04:00:54 +0000 (UTC) Received: from vger.kernel.org (vger.kernel.org [209.132.180.67]) by mail.kernel.org (Postfix) with ESMTP id 6D8B120730 for ; Sun, 19 Jan 2020 04:00:54 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1727403AbgASEAl (ORCPT ); Sat, 18 Jan 2020 23:00:41 -0500 Received: from mga14.intel.com ([192.55.52.115]:62803 "EHLO mga14.intel.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1727117AbgASEAj (ORCPT ); Sat, 18 Jan 2020 23:00:39 -0500 X-Amp-Result: SKIPPED(no attachment in message) X-Amp-File-Uploaded: False Received: from orsmga007.jf.intel.com ([10.7.209.58]) by fmsmga103.fm.intel.com with ESMTP/TLS/DHE-RSA-AES256-GCM-SHA384; 18 Jan 2020 20:00:39 -0800 X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="5.70,336,1574150400"; d="scan'208";a="214910515" Received: from local-michael-cet-test.sh.intel.com ([10.239.159.128]) by orsmga007.jf.intel.com with ESMTP; 18 Jan 2020 20:00:37 -0800 From: Yang Weijiang To: kvm@vger.kernel.org, linux-kernel@vger.kernel.org, pbonzini@redhat.com, jmattson@google.com, sean.j.christopherson@intel.com Cc: yu.c.zhang@linux.intel.com, alazar@bitdefender.com, edwin.zhai@intel.com, Yang Weijiang Subject: [PATCH v11 08/10] x86: spp: Add SPP protection check in instruction emulation Date: Sun, 19 Jan 2020 12:05:05 +0800 Message-Id: <20200119040507.23113-9-weijiang.yang@intel.com> X-Mailer: git-send-email 2.17.2 In-Reply-To: <20200119040507.23113-1-weijiang.yang@intel.com> References: <20200119040507.23113-1-weijiang.yang@intel.com> Sender: kvm-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: kvm@vger.kernel.org In instruction/mmio emulation cases, if the target write memroy is SPP protected, "fake" an vmexit to userspace to let application handle it. Signed-off-by: Yang Weijiang --- arch/x86/kvm/x86.c | 37 +++++++++++++++++++++++++++++++++++++ 1 file changed, 37 insertions(+) diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c index e93641d631cb..102a3ff8f690 100644 --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -5726,6 +5726,37 @@ static const struct read_write_emulator_ops write_emultor = { .write = true, }; +static bool is_emulator_spp_protected(struct kvm_vcpu *vcpu, + gpa_t gpa, + unsigned int bytes) +{ + gfn_t gfn, gfn_start, gfn_end; + struct kvm *kvm = vcpu->kvm; + struct kvm_memory_slot *slot; + u32 *access; + + if (!kvm->arch.spp_active) + return false; + + gfn_start = gpa_to_gfn(gpa); + gfn_end = gpa_to_gfn(gpa + bytes); + for (gfn = gfn_start; gfn <= gfn_end; gfn++) { + slot = gfn_to_memslot(kvm, gfn); + if (slot) { + access = gfn_to_subpage_wp_info(slot, gfn); + if (access && *access != FULL_SPP_ACCESS) { + vcpu->run->exit_reason = KVM_EXIT_SPP; + vcpu->run->spp.addr = gfn; + vcpu->run->spp.insn_len = + kvm_x86_ops->get_insn_len(vcpu); + return true; + } + } + } + + return false; +} + static int emulator_read_write_onepage(unsigned long addr, void *val, unsigned int bytes, struct x86_exception *exception, @@ -5756,6 +5787,9 @@ static int emulator_read_write_onepage(unsigned long addr, void *val, return X86EMUL_PROPAGATE_FAULT; } + if (write && is_emulator_spp_protected(vcpu, gpa, bytes)) + return X86EMUL_UNHANDLEABLE; + if (!ret && ops->read_write_emulate(vcpu, gpa, val, bytes)) return X86EMUL_CONTINUE; @@ -6832,6 +6866,9 @@ int x86_emulate_instruction(struct kvm_vcpu *vcpu, return 1; if (r == EMULATION_FAILED) { + if (vcpu->run->exit_reason == KVM_EXIT_SPP) + return 0; + if (reexecute_instruction(vcpu, cr2, write_fault_to_spt, emulation_type)) return 1; From patchwork Sun Jan 19 04:05:06 2020 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: "Yang, Weijiang" X-Patchwork-Id: 11340531 Return-Path: Received: from mail.kernel.org (pdx-korg-mail-1.web.codeaurora.org [172.30.200.123]) by pdx-korg-patchwork-2.web.codeaurora.org (Postfix) with ESMTP id 40E9F92A for ; Sun, 19 Jan 2020 04:00:58 +0000 (UTC) Received: from vger.kernel.org (vger.kernel.org [209.132.180.67]) by mail.kernel.org (Postfix) with ESMTP id 163D220CC7 for ; Sun, 19 Jan 2020 04:00:58 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1728775AbgASEAy (ORCPT ); Sat, 18 Jan 2020 23:00:54 -0500 Received: from mga14.intel.com ([192.55.52.115]:62807 "EHLO mga14.intel.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1727138AbgASEAm (ORCPT ); Sat, 18 Jan 2020 23:00:42 -0500 X-Amp-Result: SKIPPED(no attachment in message) X-Amp-File-Uploaded: False Received: from orsmga007.jf.intel.com ([10.7.209.58]) by fmsmga103.fm.intel.com with ESMTP/TLS/DHE-RSA-AES256-GCM-SHA384; 18 Jan 2020 20:00:41 -0800 X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="5.70,336,1574150400"; d="scan'208";a="214910523" Received: from local-michael-cet-test.sh.intel.com ([10.239.159.128]) by orsmga007.jf.intel.com with ESMTP; 18 Jan 2020 20:00:39 -0800 From: Yang Weijiang To: kvm@vger.kernel.org, linux-kernel@vger.kernel.org, pbonzini@redhat.com, jmattson@google.com, sean.j.christopherson@intel.com Cc: yu.c.zhang@linux.intel.com, alazar@bitdefender.com, edwin.zhai@intel.com, Yang Weijiang Subject: [PATCH v11 09/10] vmx: spp: Initialize SPP bitmap and SPP protection Date: Sun, 19 Jan 2020 12:05:06 +0800 Message-Id: <20200119040507.23113-10-weijiang.yang@intel.com> X-Mailer: git-send-email 2.17.2 In-Reply-To: <20200119040507.23113-1-weijiang.yang@intel.com> References: <20200119040507.23113-1-weijiang.yang@intel.com> Sender: kvm-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: kvm@vger.kernel.org For each memoryslot, there's a SPP bitmap buffer allocated, every 4Byte corresponds to subpages within a 4KB page. The original default value for each 4byte is all 1s, meaning the whole 4KB page is not SPP protected, this eases following SPP protection check. To support SPP enablement on-demand, SPP initialization can be done via KVM_ENABLE_CAP with subcode KVM_CAP_X86_SPP. KVM_SUBPAGE_MAX_PAGES is set to 512 to reduce the impact to EPT page_fault() handling because when SPP protection is configured, mmu-lock is held. All vcpus share the same SPPT, a KVM_REQ_LOAD_CR3 request is issued to each vcpu after SPP is initialized, in handling of the request, SPPTP and VMX SPP execution control bit are configured in VMCS. Signed-off-by: Yang Weijiang --- arch/x86/include/asm/kvm_host.h | 1 + arch/x86/include/asm/vmx.h | 1 + arch/x86/kvm/mmu/mmu.c | 9 ++++ arch/x86/kvm/mmu/spp.c | 96 +++++++++++++++++++++++++++++++++ arch/x86/kvm/mmu/spp.h | 5 ++ arch/x86/kvm/vmx/capabilities.h | 5 ++ arch/x86/kvm/vmx/vmx.c | 36 +++++++++++++ arch/x86/kvm/x86.c | 7 +++ include/uapi/linux/kvm.h | 1 + 9 files changed, 161 insertions(+) diff --git a/arch/x86/include/asm/kvm_host.h b/arch/x86/include/asm/kvm_host.h index 0cf886e58004..977bfedf3a1a 100644 --- a/arch/x86/include/asm/kvm_host.h +++ b/arch/x86/include/asm/kvm_host.h @@ -1241,6 +1241,7 @@ struct kvm_x86_ops { int (*enable_direct_tlbflush)(struct kvm_vcpu *vcpu); int (*get_insn_len)(struct kvm_vcpu *vcpu); + u32 (*get_spp_status)(struct kvm_vcpu *vcpu); }; struct kvm_arch_async_pf { diff --git a/arch/x86/include/asm/vmx.h b/arch/x86/include/asm/vmx.h index fc69ea8035fb..51eef174b8c1 100644 --- a/arch/x86/include/asm/vmx.h +++ b/arch/x86/include/asm/vmx.h @@ -68,6 +68,7 @@ #define SECONDARY_EXEC_XSAVES 0x00100000 #define SECONDARY_EXEC_PT_USE_GPA 0x01000000 #define SECONDARY_EXEC_MODE_BASED_EPT_EXEC 0x00400000 +#define SECONDARY_EXEC_SPP 0x00800000 #define SECONDARY_EXEC_TSC_SCALING 0x02000000 #define SECONDARY_EXEC_ENABLE_USR_WAIT_PAUSE 0x04000000 diff --git a/arch/x86/kvm/mmu/mmu.c b/arch/x86/kvm/mmu/mmu.c index 099f92f0c42a..a8b0c3849c4a 100644 --- a/arch/x86/kvm/mmu/mmu.c +++ b/arch/x86/kvm/mmu/mmu.c @@ -3821,6 +3821,12 @@ void kvm_mmu_free_roots(struct kvm_vcpu *vcpu, struct kvm_mmu *mmu, (mmu->root_level >= PT64_ROOT_4LEVEL || mmu->direct_map)) { mmu_free_root_page(vcpu->kvm, &mmu->root_hpa, &invalid_list); + if (vcpu->kvm->arch.spp_active) { + vcpu->kvm->arch.spp_active = false; + mmu_free_root_page(vcpu->kvm, + &vcpu->kvm->arch.sppt_root, + &invalid_list); + } } else { for (i = 0; i < 4; ++i) if (mmu->pae_root[i] != 0) @@ -5307,6 +5313,9 @@ void kvm_init_mmu(struct kvm_vcpu *vcpu, bool reset_roots) uint i; vcpu->arch.mmu->root_hpa = INVALID_PAGE; + if (!vcpu->kvm->arch.spp_active) + vcpu->kvm->arch.sppt_root = INVALID_PAGE; + for (i = 0; i < KVM_MMU_NUM_PREV_ROOTS; i++) vcpu->arch.mmu->prev_roots[i] = KVM_MMU_ROOT_INFO_INVALID; } diff --git a/arch/x86/kvm/mmu/spp.c b/arch/x86/kvm/mmu/spp.c index eb540e1b5133..55a39c787827 100644 --- a/arch/x86/kvm/mmu/spp.c +++ b/arch/x86/kvm/mmu/spp.c @@ -262,6 +262,102 @@ int spp_flush_sppt(struct kvm *kvm, u64 gfn_base, u32 npages) return 0; } +static int kvm_spp_create_bitmaps(struct kvm *kvm) +{ + struct kvm_memslots *slots; + struct kvm_memory_slot *memslot; + int i, j, ret; + u32 *buff; + + for (i = 0; i < KVM_ADDRESS_SPACE_NUM; i++) { + slots = __kvm_memslots(kvm, i); + kvm_for_each_memslot(memslot, slots) { + buff = kvzalloc(memslot->npages * + sizeof(*memslot->arch.subpage_wp_info), + GFP_KERNEL); + + if (!buff) { + ret = -ENOMEM; + goto out_free; + } + memslot->arch.subpage_wp_info = buff; + + for (j = 0; j < memslot->npages; j++) + buff[j] = FULL_SPP_ACCESS; + } + } + + return 0; +out_free: + for (i = 0; i < KVM_ADDRESS_SPACE_NUM; i++) { + slots = __kvm_memslots(kvm, i); + kvm_for_each_memslot(memslot, slots) { + if (memslot->arch.subpage_wp_info) { + kvfree(memslot->arch.subpage_wp_info); + memslot->arch.subpage_wp_info = NULL; + } + } + } + + return ret; +} + +void kvm_spp_free_memslot(struct kvm_memory_slot *free, + struct kvm_memory_slot *dont) +{ + if (!dont || free->arch.subpage_wp_info != + dont->arch.subpage_wp_info) { + kvfree(free->arch.subpage_wp_info); + free->arch.subpage_wp_info = NULL; + } +} + +int spp_init(struct kvm *kvm) +{ + bool first_root = true; + int i, ret; + int root_level; + u32 status; + struct kvm_vcpu *vcpu; + struct kvm_mmu_page *ssp_sp; + + /* SPP feature is exclusive with nested VM.*/ + if (kvm_x86_ops->get_nested_state) + return -EPERM; + + vcpu = kvm_get_vcpu(kvm, 0); + status = kvm_x86_ops->get_spp_status(vcpu); + + if ((status & (SPP_STATUS_VMX_SUPPORT | SPP_STATUS_EPT_SUPPORT)) != + (SPP_STATUS_VMX_SUPPORT | SPP_STATUS_EPT_SUPPORT)) + return -ENODEV; + + if (kvm->arch.spp_active) + return 0; + + ret = kvm_spp_create_bitmaps(kvm); + + if (ret) + return ret; + + kvm_for_each_vcpu(i, vcpu, kvm) { + if (first_root) { + /* prepare caches for SPP setup.*/ + mmu_topup_memory_caches(vcpu); + root_level = vcpu->arch.mmu->shadow_root_level; + ssp_sp = kvm_spp_get_page(vcpu, 0, root_level); + first_root = false; + vcpu->kvm->arch.sppt_root = __pa(ssp_sp->spt); + } + vcpu->arch.spp_pending = true; + ++ssp_sp->root_count; + kvm_make_request(KVM_REQ_LOAD_CR3, vcpu); + } + + kvm->arch.spp_active = true; + return 0; +} + int kvm_spp_get_permission(struct kvm *kvm, u64 gfn, u32 npages, u32 *access_map) { diff --git a/arch/x86/kvm/mmu/spp.h b/arch/x86/kvm/mmu/spp.h index 51a209a04863..ad37221bb306 100644 --- a/arch/x86/kvm/mmu/spp.h +++ b/arch/x86/kvm/mmu/spp.h @@ -5,7 +5,12 @@ #define FULL_SPP_ACCESS (u32)(BIT_ULL(32) - 1) #define KVM_SUBPAGE_MAX_PAGES 512 #define MAX_ENTRIES_PER_MMUPAGE BIT(9) +#define SPP_STATUS_VMX_SUPPORT 0x1 +#define SPP_STATUS_EPT_SUPPORT 0x2 +int spp_init(struct kvm *kvm); +void kvm_spp_free_memslot(struct kvm_memory_slot *free, + struct kvm_memory_slot *dont); int kvm_spp_get_permission(struct kvm *kvm, u64 gfn, u32 npages, u32 *access_map); int kvm_spp_set_permission(struct kvm *kvm, u64 gfn, u32 npages, diff --git a/arch/x86/kvm/vmx/capabilities.h b/arch/x86/kvm/vmx/capabilities.h index 7aa69716d516..78be4390180a 100644 --- a/arch/x86/kvm/vmx/capabilities.h +++ b/arch/x86/kvm/vmx/capabilities.h @@ -247,6 +247,11 @@ static inline bool vmx_xsaves_supported(void) SECONDARY_EXEC_XSAVES; } +static inline bool cpu_has_vmx_ept_spp(void) +{ + return vmcs_config.cpu_based_2nd_exec_ctrl & SECONDARY_EXEC_SPP; +} + static inline bool vmx_waitpkg_supported(void) { return vmcs_config.cpu_based_2nd_exec_ctrl & diff --git a/arch/x86/kvm/vmx/vmx.c b/arch/x86/kvm/vmx/vmx.c index 4032e615ca85..e54bafa8b887 100644 --- a/arch/x86/kvm/vmx/vmx.c +++ b/arch/x86/kvm/vmx/vmx.c @@ -113,6 +113,9 @@ module_param_named(pml, enable_pml, bool, S_IRUGO); static bool __read_mostly dump_invalid_vmcs = 0; module_param(dump_invalid_vmcs, bool, 0644); +/* SPP is disabled by default unless it's enabled via KVM_ENABLE_CAP. */ +static bool __read_mostly enable_spp = 0; + #define MSR_BITMAP_MODE_X2APIC 1 #define MSR_BITMAP_MODE_X2APIC_APICV 2 @@ -1428,6 +1431,17 @@ static int vmx_get_insn_len(struct kvm_vcpu *vcpu) return vmcs_read32(VM_EXIT_INSTRUCTION_LEN); } +static u32 vmx_get_spp_status(struct kvm_vcpu *vcpu) +{ + u32 status = 0; + + if (cpu_has_vmx_ept_spp()) + status |= SPP_STATUS_VMX_SUPPORT; + if (enable_ept) + status |= SPP_STATUS_EPT_SUPPORT; + return status; +} + static void vmx_decache_cr0_guest_bits(struct kvm_vcpu *vcpu); unsigned long vmx_get_rflags(struct kvm_vcpu *vcpu) @@ -2397,6 +2411,7 @@ static __init int setup_vmcs_config(struct vmcs_config *vmcs_conf, SECONDARY_EXEC_RDSEED_EXITING | SECONDARY_EXEC_RDRAND_EXITING | SECONDARY_EXEC_ENABLE_PML | + SECONDARY_EXEC_SPP | SECONDARY_EXEC_TSC_SCALING | SECONDARY_EXEC_ENABLE_USR_WAIT_PAUSE | SECONDARY_EXEC_PT_USE_GPA | @@ -2997,6 +3012,7 @@ void vmx_set_cr3(struct kvm_vcpu *vcpu, unsigned long cr3) bool update_guest_cr3 = true; unsigned long guest_cr3; u64 eptp; + u64 spptp; guest_cr3 = cr3; if (enable_ept) { @@ -3025,6 +3041,20 @@ void vmx_set_cr3(struct kvm_vcpu *vcpu, unsigned long cr3) if (update_guest_cr3) vmcs_writel(GUEST_CR3, guest_cr3); + + if (kvm->arch.spp_active && VALID_PAGE(vcpu->kvm->arch.sppt_root)) { + spptp = construct_spptp(vcpu->kvm->arch.sppt_root); + vmcs_write64(SPPT_POINTER, spptp); + + if (vcpu->arch.spp_pending && cpu_has_secondary_exec_ctrls()) { + struct vcpu_vmx *vmx = to_vmx(vcpu); + + secondary_exec_controls_setbit(vmx, + SECONDARY_EXEC_SPP); + enable_spp = 1; + vcpu->arch.spp_pending = false; + } + } } int vmx_set_cr4(struct kvm_vcpu *vcpu, unsigned long cr4) @@ -4045,6 +4075,9 @@ static void vmx_compute_secondary_exec_control(struct vcpu_vmx *vmx) if (!enable_pml) exec_control &= ~SECONDARY_EXEC_ENABLE_PML; + if (!enable_spp) + exec_control &= ~SECONDARY_EXEC_SPP; + if (vmx_xsaves_supported()) { /* Exposing XSAVES only when XSAVE is exposed */ bool xsaves_enabled = @@ -5866,6 +5899,8 @@ void dump_vmcs(void) pr_err("PostedIntrVec = 0x%02x\n", vmcs_read16(POSTED_INTR_NV)); if ((secondary_exec_control & SECONDARY_EXEC_ENABLE_EPT)) pr_err("EPT pointer = 0x%016llx\n", vmcs_read64(EPT_POINTER)); + if ((secondary_exec_control & SECONDARY_EXEC_SPP)) + pr_err("SPPT pointer = 0x%016llx\n", vmcs_read64(SPPT_POINTER)); n = vmcs_read32(CR3_TARGET_COUNT); for (i = 0; i + 1 < n; i += 4) pr_err("CR3 target%u=%016lx target%u=%016lx\n", @@ -7980,6 +8015,7 @@ static struct kvm_x86_ops vmx_x86_ops __ro_after_init = { .apic_init_signal_blocked = vmx_apic_init_signal_blocked, .get_insn_len = vmx_get_insn_len, + .get_spp_status = vmx_get_spp_status, }; static void vmx_cleanup_l1d_flush(void) diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c index 102a3ff8f690..8ed67f91947b 100644 --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -3336,6 +3336,9 @@ int kvm_vm_ioctl_check_extension(struct kvm *kvm, long ext) case KVM_CAP_HYPERV_ENLIGHTENED_VMCS: r = kvm_x86_ops->nested_enable_evmcs != NULL; break; + case KVM_CAP_X86_SPP: + r = KVM_SUBPAGE_MAX_PAGES; + break; default: break; } @@ -4831,6 +4834,9 @@ int kvm_vm_ioctl_enable_cap(struct kvm *kvm, kvm->arch.exception_payload_enabled = cap->args[0]; r = 0; break; + case KVM_CAP_X86_SPP: + r = spp_init(kvm); + break; default: r = -EINVAL; break; @@ -9814,6 +9820,7 @@ void kvm_arch_free_memslot(struct kvm *kvm, struct kvm_memory_slot *free, } kvm_page_track_free_memslot(free, dont); + kvm_spp_free_memslot(free, dont); } int kvm_arch_create_memslot(struct kvm *kvm, struct kvm_memory_slot *slot, diff --git a/include/uapi/linux/kvm.h b/include/uapi/linux/kvm.h index 280a6d52e5ff..e595a233a9fa 100644 --- a/include/uapi/linux/kvm.h +++ b/include/uapi/linux/kvm.h @@ -1023,6 +1023,7 @@ struct kvm_ppc_resize_hpt { #define KVM_CAP_PPC_GUEST_DEBUG_SSTEP 176 #define KVM_CAP_ARM_NISV_TO_USER 177 #define KVM_CAP_ARM_INJECT_EXT_DABT 178 +#define KVM_CAP_X86_SPP 179 #ifdef KVM_CAP_IRQ_ROUTING From patchwork Sun Jan 19 04:05:07 2020 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: "Yang, Weijiang" X-Patchwork-Id: 11340527 Return-Path: Received: from mail.kernel.org (pdx-korg-mail-1.web.codeaurora.org [172.30.200.123]) by pdx-korg-patchwork-2.web.codeaurora.org (Postfix) with ESMTP id BCD3614B4 for ; Sun, 19 Jan 2020 04:00:49 +0000 (UTC) Received: from vger.kernel.org (vger.kernel.org [209.132.180.67]) by mail.kernel.org (Postfix) with ESMTP id 9C2002064C for ; Sun, 19 Jan 2020 04:00:49 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1728721AbgASEAs (ORCPT ); Sat, 18 Jan 2020 23:00:48 -0500 Received: from mga14.intel.com ([192.55.52.115]:62807 "EHLO mga14.intel.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1728512AbgASEAn (ORCPT ); Sat, 18 Jan 2020 23:00:43 -0500 X-Amp-Result: SKIPPED(no attachment in message) X-Amp-File-Uploaded: False Received: from orsmga007.jf.intel.com ([10.7.209.58]) by fmsmga103.fm.intel.com with ESMTP/TLS/DHE-RSA-AES256-GCM-SHA384; 18 Jan 2020 20:00:43 -0800 X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="5.70,336,1574150400"; d="scan'208";a="214910530" Received: from local-michael-cet-test.sh.intel.com ([10.239.159.128]) by orsmga007.jf.intel.com with ESMTP; 18 Jan 2020 20:00:41 -0800 From: Yang Weijiang To: kvm@vger.kernel.org, linux-kernel@vger.kernel.org, pbonzini@redhat.com, jmattson@google.com, sean.j.christopherson@intel.com Cc: yu.c.zhang@linux.intel.com, alazar@bitdefender.com, edwin.zhai@intel.com, Yang Weijiang Subject: [PATCH v11 10/10] kvm: selftests: selftest for Sub-Page protection Date: Sun, 19 Jan 2020 12:05:07 +0800 Message-Id: <20200119040507.23113-11-weijiang.yang@intel.com> X-Mailer: git-send-email 2.17.2 In-Reply-To: <20200119040507.23113-1-weijiang.yang@intel.com> References: <20200119040507.23113-1-weijiang.yang@intel.com> Sender: kvm-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: kvm@vger.kernel.org Sub-Page Permission(SPP) is to protect finer granularity subpages (128Byte each) within a 4KB page. It's not enabled in KVM by default, the test first initializes the SPP runtime environment with KVM_ENABLE_CAP ioctl, then sets protection with KVM_SUBPAGES_SET_ACCESS for the target guest page, check permissions with KVM_SUBPAGES_GET_ACCESS to make sure they are set as expected. Two steps in guest code to very whether SPP is working: 1) protect all 128byte subpages, write data to each subpage to see if SPP induced EPT violation happening. 2)unprotect all subpages, again write data to each subpage to see if SPP still works or not. Signed-off-by: Yang Weijiang --- tools/testing/selftests/kvm/Makefile | 1 + tools/testing/selftests/kvm/lib/kvm_util.c | 1 + tools/testing/selftests/kvm/x86_64/spp_test.c | 235 ++++++++++++++++++ 3 files changed, 237 insertions(+) create mode 100644 tools/testing/selftests/kvm/x86_64/spp_test.c diff --git a/tools/testing/selftests/kvm/Makefile b/tools/testing/selftests/kvm/Makefile index 3138a916574a..c9ad445cd8a9 100644 --- a/tools/testing/selftests/kvm/Makefile +++ b/tools/testing/selftests/kvm/Makefile @@ -30,6 +30,7 @@ TEST_GEN_PROGS_x86_64 += clear_dirty_log_test TEST_GEN_PROGS_x86_64 += dirty_log_test TEST_GEN_PROGS_x86_64 += kvm_create_max_vcpus +TEST_GEN_PROGS_x86_64 += x86_64/spp_test TEST_GEN_PROGS_aarch64 += clear_dirty_log_test TEST_GEN_PROGS_aarch64 += dirty_log_test TEST_GEN_PROGS_aarch64 += kvm_create_max_vcpus diff --git a/tools/testing/selftests/kvm/lib/kvm_util.c b/tools/testing/selftests/kvm/lib/kvm_util.c index 41cf45416060..bc0a25f4276b 100644 --- a/tools/testing/selftests/kvm/lib/kvm_util.c +++ b/tools/testing/selftests/kvm/lib/kvm_util.c @@ -1486,6 +1486,7 @@ static struct exit_reason { {KVM_EXIT_UNKNOWN, "UNKNOWN"}, {KVM_EXIT_EXCEPTION, "EXCEPTION"}, {KVM_EXIT_IO, "IO"}, + {KVM_EXIT_SPP, "SPP"}, {KVM_EXIT_HYPERCALL, "HYPERCALL"}, {KVM_EXIT_DEBUG, "DEBUG"}, {KVM_EXIT_HLT, "HLT"}, diff --git a/tools/testing/selftests/kvm/x86_64/spp_test.c b/tools/testing/selftests/kvm/x86_64/spp_test.c new file mode 100644 index 000000000000..c0e29c101cbf --- /dev/null +++ b/tools/testing/selftests/kvm/x86_64/spp_test.c @@ -0,0 +1,235 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * Sub-Page Permission test + * + * Copyright (C) 2019, Intel Corp. + * + */ + +#include +#include +#include +#include +#include + +#include "test_util.h" +#include "kvm_util.h" +#include "processor.h" +#include "../../lib/kvm_util_internal.h" +#include "linux/kvm.h" + +#define VCPU_ID 1 +#define PAGE_SIZE (4096) +#define SPP_GUARD_SIZE (16 * PAGE_SIZE) +#define SPP_GUARD_MEMSLOT (1) +#define SPP_GUARD_PAGES (SPP_GUARD_SIZE / PAGE_SIZE) +#define SPP_GUARD_GPA 0x10000000 + +#define SUBPAGE_ACCESS_DEFAULT (0x0) +#define SUBPAGE_ACCESS_FULL (0xFFFFFFFF) +#define START_SPP_VM_ADDR (0x700000) +#define SUBPAGE_SIZE (128) + +vm_vaddr_t vspp_start; +vm_paddr_t pspp_start; + +void guest_code(void) +{ + uint8_t *iterator = (uint8_t *)vspp_start; + int count; + + GUEST_SYNC(1); + /* + * expect EPT violation induced by SPP in each interation since + * the full page is protected by SPP. + */ + for (count = 0; count < PAGE_SIZE / SUBPAGE_SIZE; count++) { + *(uint32_t *)(iterator) = 0x99; + iterator += SUBPAGE_SIZE; + } + GUEST_SYNC(2); + iterator = (uint8_t *)vspp_start; + + /* + * don't expect EPT violation happen since SPP is disabled + * for the page + */ + for (count = 0; count < PAGE_SIZE / SUBPAGE_SIZE; count++) { + *(uint32_t *)(iterator) = 0x99; + iterator += SUBPAGE_SIZE; + } +} + +void prepare_test(struct kvm_vm **g_vm, struct kvm_run **g_run) +{ + void *spp_hva; + struct kvm_vm *vm; + struct kvm_run *run; + /* Create VM, SPP is only valid for 4KB page mode */ + *g_vm = vm_create_default(VCPU_ID, 0, guest_code); + vm = *g_vm; + + *g_run = vcpu_state(vm, VCPU_ID); + run = *g_run; + + vm_userspace_mem_region_add(vm, VM_MEM_SRC_ANONYMOUS, SPP_GUARD_GPA, + SPP_GUARD_MEMSLOT, SPP_GUARD_PAGES, 0); + + pspp_start = vm_phy_pages_alloc(vm, 1, SPP_GUARD_GPA, + SPP_GUARD_MEMSLOT); + + memset(addr_gpa2hva(vm, SPP_GUARD_GPA), 0x0, PAGE_SIZE); + + virt_map(vm, START_SPP_VM_ADDR, SPP_GUARD_GPA, PAGE_SIZE, 0); + + vspp_start = vm_vaddr_alloc(vm, PAGE_SIZE, START_SPP_VM_ADDR, + SPP_GUARD_MEMSLOT, 0); + + spp_hva = addr_gva2hva(vm, vspp_start); + + pspp_start = addr_hva2gpa(vm, spp_hva); + + printf("SPP protected zone: size = %d, gva = 0x%lx, gpa = 0x%lx, " + "hva = 0x%p\n", PAGE_SIZE, vspp_start, pspp_start, spp_hva); + + /* make sure the virtual address is visible to VM. */ + sync_global_to_guest(vm, vspp_start); + + vcpu_run(vm, VCPU_ID); + + TEST_ASSERT(run->exit_reason == KVM_EXIT_IO, + "exit reason: %u (%s),\n", run->exit_reason, + exit_reason_str(run->exit_reason)); +} + +void setup_spp(struct kvm_vm *vm) +{ + struct kvm_enable_cap cap; + int ret = 0; + struct kvm_subpage *sp; + int len; + + memset(&cap, 0, sizeof(cap)); + cap.cap = KVM_CAP_X86_SPP; + cap.flags = 0; + + /* initialize the SPP runtime environment.*/ + ret = ioctl(vm->fd, KVM_ENABLE_CAP, &cap); + TEST_ASSERT(ret == 0, "KVM_CAP_X86_SPP failed."); + len = sizeof(*sp) + sizeof(__u32); + printf("SPP initialized successfully.\n"); + + sp = malloc(len); + TEST_ASSERT(sp > 0, "Low memory 1!"); + memset(sp, 0, len); + /* set up SPP protection for the page. */ + sp->npages = 1; + sp->gfn_base = pspp_start >> 12; + sp->access_map[0] = SUBPAGE_ACCESS_DEFAULT; + ret = ioctl(vm->fd, KVM_SUBPAGES_SET_ACCESS, sp); + + TEST_ASSERT(ret == 1, "KVM_SUBPAGES_SET_ACCESS failed. ret = 0x%x, " + "gfn_base = 0x%llx\n", ret, sp->gfn_base); + printf("set spp protection info: gfn = 0x%llx, access = 0x%x, " + "npages = %d\n", sp->gfn_base, sp->access_map[0], + sp->npages); + + memset(sp, 0, len); + /* make sure the SPP permission bits are actully set as expected. */ + sp->npages = 1; + sp->gfn_base = pspp_start >> 12; + + ret = ioctl(vm->fd, KVM_SUBPAGES_GET_ACCESS, sp); + + TEST_ASSERT(ret == 1, "KVM_SUBPAGES_GET_ACCESS failed."); + + TEST_ASSERT(sp->access_map[0] == SUBPAGE_ACCESS_DEFAULT, + "subpage access didn't match."); + printf("get spp protection info: gfn = 0x%llx, access = 0x%x, " + "npages = %d\n", sp->gfn_base, + sp->access_map[0], sp->npages); + + free(sp); + printf("got matched subpage permission vector.\n"); + printf("expect VM exits caused by SPP below.\n"); +} + +void unset_spp(struct kvm_vm *vm) +{ + struct kvm_subpage *sp; + int len; + + len = sizeof(*sp) + sizeof(__u32); + sp = malloc(len); + TEST_ASSERT(sp > 0, "Low memory 2!"); + memset(sp, 0, len); + + /* now unprotect the SPP to the page.*/ + sp->npages = 1; + sp->gfn_base = pspp_start >> 12; + sp->access_map[0] = SUBPAGE_ACCESS_FULL; + ioctl(vm->fd, KVM_SUBPAGES_SET_ACCESS, sp); + + printf("unset SPP protection at gfn: 0x%llx\n", sp->gfn_base); + printf("expect NO VM exits caused by SPP below.\n"); + free(sp); +} + +#define TEST_SYNC_FIELDS KVM_SYNC_X86_REGS + +void run_test(struct kvm_vm *vm, struct kvm_run *run) +{ + int loop; + int ept_fault = 0; + struct kvm_regs regs; + + run->kvm_valid_regs = TEST_SYNC_FIELDS; + vcpu_run(vm, VCPU_ID); + + for (loop = 0; loop < PAGE_SIZE / SUBPAGE_SIZE; loop++) { + /* + * if everything goes correctly, should get VM exit + * with KVM_EXIT_SPP. + */ + TEST_ASSERT(run->exit_reason == KVM_EXIT_SPP, + "exit reason: %u (%s),\n", run->exit_reason, + exit_reason_str(run->exit_reason)); + printf("%d - exit reason: %s\n", loop + 1, + exit_reason_str(run->exit_reason)); + ept_fault++; + + vcpu_regs_get(vm, VCPU_ID, ®s); + + run->s.regs.regs.rip += run->spp.insn_len; + + run->kvm_valid_regs = TEST_SYNC_FIELDS; + run->kvm_dirty_regs = KVM_SYNC_X86_REGS; + + vcpu_run(vm, VCPU_ID); + } + + printf("total EPT violation count: %d\n", ept_fault); +} + +int main(int argc, char *argv[]) +{ + struct kvm_vm *vm; + struct kvm_run *run; + + prepare_test(&vm, &run); + + setup_spp(vm); + + run_test(vm, run); + + unset_spp(vm); + + vcpu_run(vm, VCPU_ID); + + printf("completed SPP test successfully!\n"); + + kvm_vm_free(vm); + + return 0; +} +