- 01 Oct, 2021 3 commits
-
-
Maxim Levitsky authored
This was tested by booting a nested guest with TSC=1Ghz, observing the clocks, and doing about 100 cycles of migration. Note that qemu patch is needed to support migration because of a new MSR that needs to be placed in the migration state. The patch will be sent to the qemu mailing list soon. Signed-off-by:
Maxim Levitsky <mlevitsk@redhat.com> Message-Id: <20210914154825.104886-14-mlevitsk@redhat.com> Signed-off-by:
Paolo Bonzini <pbonzini@redhat.com>
-
Maxim Levitsky authored
This allows to easily simulate a CPU without this feature. Signed-off-by:
-
Paolo Bonzini authored
Commit adc2a237 ("KVM: nSVM: improve SYSENTER emulation on AMD"), made init_vmcb set vmload/vmsave intercepts unconditionally, and relied on svm_vcpu_after_set_cpuid to clear them when possible. However init_vmcb is also called when the vCPU is reset, and it is not followed by another call to svm_vcpu_after_set_cpuid because the CPUID is already set. This mistake makes the VMSAVE/VMLOAD intercept to be set when it is not needed, and harms performance of the nested guest. Extract the relevant parts of svm_vcpu_after_set_cpuid so that they can be called again on reset. Fixes: adc2a237 ("KVM: nSVM: improve SYSENTER emulation on AMD") Signed-off-by:
-
- 30 Sep, 2021 37 commits
-
-
Maxim Levitsky authored
This is useful for debug and also makes it consistent with the rest of the SVM optional features. Signed-off-by:
-
Maxim Levitsky authored
According to the SDM, the CPU never modifies these settings. It loads them on VM entry and updates an internal copy instead. Also don't load them from the vmcb12 as we don't expose these features to the nested guest yet. Signed-off-by:
-
Longpeng(Mike) authored
All of the irqfds would to be updated when update the irq routing, it's too expensive if there're too many irqfds. However we can reduce the cost by avoid some unnecessary updates. For irqs of MSI type on X86, the update can be saved if the msi values are not change. The vfio migration could receives benefit from this optimi- zaiton. The test VM has 128 vcpus and 8 VF (with 65 vectors enabled), so the VM has more than 520 irqfds. We mesure the cost of the vfio_msix_enable (in QEMU, it would set routing for each irqfd) for each VF, and we can see the total cost can be significantly reduced. Origin Apply this Patch 1st 8 4 2nd 15 5 3rd 22 6 4th 24 6 5th 36 7 6th 44 7 7th 51 8 8th 58 8 Total 258ms 51ms We're also tring to optimize the QEMU part [1], but it's still worth to optimize the KVM to gain more benefits. [1] https://lists.gnu.org/archive/html/qemu-devel/2021-08/msg04215.htmlSigned-off-by:Longpeng(Mike) <longpeng2@huawei.com> Message-Id: <20210827080003.2689-1-longpeng2@huawei.com> Signed-off-by:
-
Lai Jiangshan authored
If the original spte is writable, the target gfn should not be the gfn of synchronized shadowpage and can continue to be writable. When !can_unsync, speculative must be false. So when the check of "!can_unsync" is removed, we need to move the label of "out" up. Signed-off-by:
Lai Jiangshan <laijs@linux.alibaba.com> Signed-off-by:
-
Lai Jiangshan authored
We'd better only unsync the pagetable when there just was a really write fault on a level-1 pagetable. Signed-off-by:
-
Lai Jiangshan authored
Its solo caller is changed to use FNAME(prefetch_gpte) directly. Signed-off-by:
-
Lai Jiangshan authored
In mmu_sync_children(), it can zap the invalid list after remote tlb flushing. Emptifying the invalid list ASAP might help reduce a remote tlb flushing in some cases. Signed-off-by:
-
Lai Jiangshan authored
Currently kvm_sync_page() returns true when there is any present spte. But the return value is ignored in the callers. Changing kvm_sync_page() to return true when remote flush is needed and changing mmu->sync_page() not to directly flush can combine and reduce remote flush requests. Signed-off-by:
-
Lai Jiangshan authored
Because local_flush is useless, kvm_mmu_flush_or_zap() can be removed and kvm_mmu_remote_flush_or_zap is used instead. Signed-off-by:
-
Lai Jiangshan authored
After any shadow page modification, flushing tlb only on current VCPU is weird due to other VCPU's tlb might still be stale. In other words, if there is any mandatory tlb-flushing after shadow page modification, SET_SPTE_NEED_REMOTE_TLB_FLUSH or remote_flush should be set and the tlbs of all VCPUs should be flushed. There is not point to only flush current tlb except when the request is from vCPU's or pCPU's activities. If there was any bug that mandatory tlb-flushing is required and SET_SPTE_NEED_REMOTE_TLB_FLUSH/remote_flush is failed to set, this patch would expose the bug in a more destructive way. The related code paths are checked and no missing SET_SPTE_NEED_REMOTE_TLB_FLUSH is found yet. Currently, there is no optional tlb-flushing after sync page related code is changed to flush tlb timely. So we can just remove these local flushing code. Signed-off-by:
-
Sean Christopherson authored
Make a final call to direct_pte_prefetch_many() if there are "trailing" SPTEs to prefetch, i.e. SPTEs for GFNs following the faulting GFN. The call to direct_pte_prefetch_many() in the loop only handles the case where there are !PRESENT SPTEs preceding a PRESENT SPTE. E.g. if the faulting GFN is a multiple of 8 (the prefetch size) and all SPTEs for the following GFNs are !PRESENT, the loop will terminate with "start = sptep+1" and not prefetch any SPTEs. Prefetching trailing SPTEs as intended can drastically reduce the number of guest page faults, e.g. accessing the first byte of every 4kb page in a 6gb chunk of virtual memory, in a VM with 8gb of preallocated memory, the number of pf_fixed events observed in L0 drops from ~1.75M to <0.27M. Note, this only affects memory that is backed by 4kb pages as KVM doesn't prefetch when installing hugepages. Shadow paging prefetching is not affected as it does not batch the prefetches due to the need to process the corresponding guest PTE. The TDP MMU is not affected because it doesn't have prefetching, yet... Fixes: 957ed9ef ("KVM: MMU: prefetch ptes when intercepted guest #PF") Cc: Sergey Senozhatsky <senozhatsky@google.com> Cc: Ben Gardon <bgardon@google.com> Signed-off-by:
Sean Christopherson <seanjc@google.com> Message-Id: <20210818235615.2047588-1-seanjc@google.com> Signed-off-by:
-
Thomas Huth authored
The kvm_vm_free() statement here is currently dead code, since the loop in front of it can only be left with the "goto done" that jumps right after the kvm_vm_free(). Fix it by swapping the locations of the "done" label and the kvm_vm_free(). Signed-off-by:
Thomas Huth <thuth@redhat.com> Message-Id: <20210826074928.240942-1-thuth@redhat.com> Signed-off-by:
-
Colin Ian King authored
There is a spelling mistake in an error message. Fix it. Signed-off-by:
Colin Ian King <colin.king@canonical.com> Message-Id: <20210826120752.12633-1-colin.king@canonical.com> Signed-off-by:
-
Sean Christopherson authored
Manually look for a CPUID.0x1 entry instead of bouncing through kvm_cpuid() when retrieving the Family-Model-Stepping information for vCPU RESET/INIT. This fixes a potential undefined behavior bug due to kvm_cpuid() using the uninitialized "dummy" param as the ECX _input_, a.k.a. the index. A more minimal fix would be to simply zero "dummy", but the extra work in kvm_cpuid() is wasteful, and KVM should be treating the FMS retrieval as an out-of-band access, e.g. same as how KVM computes guest.MAXPHYADDR. Both Intel's SDM and AMD's APM describe the RDX value at RESET/INIT as holding the CPU's FMS information, not as holding CPUID.0x1.EAX. KVM's usage of CPUID entries to get FMS is simply a pragmatic approach to avoid having yet another way for userspace to provide inconsistent data. No functional change intended. Signed-off-by:
Jim Mattson <jmattson@google.com> Message-Id: <20210929222426.1855730-3-seanjc@google.com> Signed-off-by:
-
Sean Christopherson authored
WARN if CR0, CR3, or CR4 are non-zero at RESET, which given the current KVM implementation, really means WARN if they're not zeroed at vCPU creation. VMX in particular has several ->set_*() flows that read other registers to handle side effects, and because those flows are common to RESET and INIT, KVM subtly relies on emulated/virtualized registers to be zeroed at vCPU creation in order to do the right thing at RESET. Use CRs as a sentinel because they are most likely to be written as side effects, and because KVM specifically needs CR0.PG and CR0.PE to be '0' to correctly reflect the state of the vCPU's MMU. CRs are also loaded and stored from/to the VMCS, and so adds some level of coverage to verify that KVM doesn't conflate zero-allocating the VMCS with properly initializing the VMCS with VMWRITEs. Note, '0' is somewhat arbitrary, vCPU creation can technically stuff any value for a register so long as it's coherent with respect to the current vCPU state. In practice, '0' works for all registers and is convenient. Suggested-by:
Vitaly Kuznetsov <vkuznets@redhat.com> Signed-off-by:
-
Sean Christopherson authored
Move RESET emulation for SVM vCPUs to svm_vcpu_reset(), and drop an extra init_vmcb() from svm_create_vcpu() in the process. Hopefully KVM will someday expose a dedicated RESET ioctl(), and in the meantime separating "create" from "RESET" is a nice cleanup. Keep the call to svm_switch_vmcb() so that misuse of svm->vmcb at worst breaks the guest, e.g. premature accesses doesn't cause a NULL pointer dereference. Cc: Reiji Watanabe <reijiw@google.com> Signed-off-by:
-
Sean Christopherson authored
Move vCPU RESET emulation, including initializating of select VMCS state, to vmx_vcpu_reset(). Drop the open coded "vCPU load" sequence, as ->vcpu_reset() is invoked while the vCPU is properly loaded (which is kind of the point of ->vcpu_reset()...). Hopefully KVM will someday expose a dedicated RESET ioctl(), and in the meantime separating "create" from "RESET" is a nice cleanup. Deferring VMCS initialization is effectively a nop as it's impossible to safely access the VMCS between the current call site and its new home, as both the vCPU and the pCPU are put immediately after init_vmcs(), i.e. the VMCS isn't guaranteed to be loaded. Note, task preemption is not a problem as vmx_sched_in() _can't_ touch the VMCS as ->sched_in() is invoked before the vCPU, and thus VMCS, is reloaded. I.e. the preemption path also can't consume VMCS state. Cc: Reiji Watanabe <reijiw@google.com> Signed-off-by:
-
Sean Christopherson authored
Don't zero out user return and nested MSRs during vCPU creation, and instead rely on vcpu_vmx being zero-allocated. Explicitly zeroing MSRs is not wrong, and is in fact necessary if KVM ever emulates vCPU RESET outside of vCPU creation, but zeroing only a subset of MSRs is confusing. Poking directly into KVM's backing is also undesirable in that it doesn't scale and is error prone. Ideally KVM would have a common RESET path for all MSRs, e.g. by expanding kvm_set_msr(), which would obviate the need for this out-of-bad code (to support standalone RESET). No functional change intended. Signed-off-by:
-
Sean Christopherson authored
Move the few bits of relevant fx_init() code into kvm_arch_vcpu_create(), dropping the superfluous check on vcpu->arch.guest_fpu that was blindly and wrongly added by commit ed02b213 ("KVM: SVM: Guest FPU state save/restore not needed for SEV-ES guest"). Note, KVM currently allocates and then frees FPU state for SEV-ES guests, rather than avoid the allocation in the first place. While that approach is inarguably inefficient and unnecessary, it's a cleanup for the future. No functional change intended. Signed-off-by:
-
Sean Christopherson authored
Drop code to initialize XCR0 during fx_init(), a.k.a. vCPU creation, as XCR0 has been initialized during kvm_vcpu_reset() (for RESET) since commit a554d207 ("KVM: X86: Processor States following Reset or INIT"). Back when XCR0 support was added by commit 2acf923e ("KVM: VMX: Enable XSAVE/XRSTOR for guest"), KVM didn't differentiate between RESET and INIT. Ignoring the fact that calling fx_init() for INIT is obviously wrong, e.g. FPU state after INIT is not the same as after RESET, setting XCR0 in fx_init() was correct. Eventually fx_init() got moved to kvm_arch_vcpu_init(), a.k.a. vCPU creation (ignore the terrible name) by commit 0ee6a517 ("x86/fpu, kvm: Simplify fx_init()"). Finally, commit 95a0d01e ("KVM: x86: Move all vcpu init code into kvm_arch_vcpu_create()") killed off kvm_arch_vcpu_init(), leaving behind the oddity of redundant setting of guest state during vCPU creation. No functional change intended. Signed-off-by:
-
Sean Christopherson authored
Drop code to set CR0.ET for the guest during initialization of the guest FPU. The code was added as a misguided bug fix by commit 380102c8 ("KVM Set the ET flag in CR0 after initializing FX") to resolve an issue where vcpu->cr0 (now vcpu->arch.cr0) was not correctly initialized on SVM systems. While init_vmcb() did set CR0.ET, it only did so in the VMCB, and subtly did not update vcpu->cr0. Stuffing CR0.ET worked around the immediate problem, but did not fix the real bug of vcpu->cr0 and the VMCB being out of sync. That underlying bug was eventually remedied by commit 18fa000a ("KVM: SVM: Reset cr0 properly on vcpu reset"). No functional change intended. Signed-off-by:
-
Sean Christopherson authored
Do not blindly mark all registers as available+dirty at RESET/INIT, and instead rely on writes to registers to go through the proper mutators or to explicitly mark registers as dirty. INIT in particular does not blindly overwrite all registers, e.g. select bits in CR0 are preserved across INIT, thus marking registers available+dirty without first reading the register from hardware is incorrect. In practice this is a benign bug as KVM doesn't let the guest control CR0 bits that are preserved across INIT, and all other true registers are explicitly written during the RESET/INIT flows. The PDPTRs and EX_INFO "registers" are not explicitly written, but accessing those values during RESET/INIT is nonsensical and would be a KVM bug regardless of register caching. Fixes: 66f7b72e ("KVM: x86: Make register state after reset conform to specification") [sean: !!! NOT FOR STABLE !!!] Signed-off-by:
-
Sean Christopherson authored
Replace impressively complex "logic" for computing the page offset from CR3 when loading PDPTRs. Unlike other paging modes, the address held in CR3 for PAE paging is 32-byte aligned, i.e. occupies bits 31:5, thus bits 11:5 need to be used as the offset from the gfn when reading PDPTRs. The existing calculation originated in commit 1342d353 ("[PATCH] KVM: MMU: Load the pae pdptrs on cr3 change like the processor does"), which read the PDPTRs from guest memory as individual 8-byte loads. At the time, the so called "offset" was the base index of PDPTR0 as a _u64_, not a byte offset. Naming aside, the computation was useful and arguably simplified the overall flow. Unfortunately, when commit 195aefde ("KVM: Add general accessors to read and write guest memory") added accessors with offsets at byte granularity, the cleverness of the original code was lost and KVM was left with convoluted code for a simple operation. No functional change intended. Signed-off-by:
-
Sean Christopherson authored
Open code the call to mmu->translate_gpa() when loading nested PDPTRs and kill off the existing helper, kvm_read_guest_page_mmu(), to discourage incorrect use. Reading guest memory straight from an L2 GPA is extremely rare (as evidenced by the lack of users), as very few constructs in x86 specify physical addresses, even fewer are virtualized by KVM, and even fewer yet require emulation of L2 by L0 KVM. No functional change intended. Signed-off-by:
-
Juergen Gross authored
KVM_MAX_VCPU_ID is not specifying the highest allowed vcpu-id, but the number of allowed vcpu-ids. This has already led to confusion, so rename KVM_MAX_VCPU_ID to KVM_MAX_VCPU_IDS to make its semantics more clear Suggested-by:
Eduardo Habkost <ehabkost@redhat.com> Signed-off-by:
Juergen Gross <jgross@suse.com> Signed-off-by:
-
Juergen Gross authored
This reverts commit 76b4f357. The commit has the wrong reasoning, as KVM_MAX_VCPU_ID is not defining the maximum allowed vcpu-id as its name suggests, but the number of vcpu-ids. So revert this patch again. Suggested-by:
-
Vitaly Kuznetsov authored
kvm_make_vcpus_request_mask() already disables preemption so just like kvm_make_all_cpus_request_except() it can be switched to using pre-allocated per-cpu cpumasks. This allows for improvements for both users of the function: in Hyper-V emulation code 'tlb_flush' can now be dropped from 'struct kvm_vcpu_hv' and kvm_make_scan_ioapic_request_mask() gets rid of dynamic allocation. cpumask_available() checks in kvm_make_vcpu_request() and kvm_kick_many_cpus() can now be dropped as they checks for an impossible condition: kvm_init() makes sure per-cpu masks are allocated. Signed-off-by:
-
Vitaly Kuznetsov authored
Allocating cpumask dynamically in zalloc_cpumask_var() is not ideal. Allocation is somewhat slow and can (in theory and when CPUMASK_OFFSTACK) fail. kvm_make_all_cpus_request_except() already disables preemption so we can use pre-allocated per-cpu cpumasks instead. Signed-off-by:
-
Vitaly Kuznetsov authored
Both remaining callers of kvm_make_vcpus_request_mask() pass 'NULL' for 'except' parameter so it can just be dropped. No functional change intended
© . Suggested-by: -
Vitaly Kuznetsov authored
Iterating over set bits in 'vcpu_bitmap' should be faster than going through all vCPUs, especially when just a few bits are set. Drop kvm_make_vcpus_request_mask() call from kvm_make_all_cpus_request_except() to avoid handling the special case when 'vcpu_bitmap' is NULL, move the code to kvm_make_all_cpus_request_except() itself. Signed-off-by:
-
Vitaly Kuznetsov authored
In preparation to making kvm_make_vcpus_request_mask() use for_each_set_bit() switch kvm_hv_flush_tlb() to calling kvm_make_all_cpus_request() for 'all cpus' case. Note: kvm_make_all_cpus_request() (unlike kvm_make_vcpus_request_mask()) currently dynamically allocates cpumask on each call and this is suboptimal. Both kvm_make_all_cpus_request() and kvm_make_vcpus_request_mask() are going to be switched to using pre-allocated per-cpu masks. Reviewed-by:
-
Yang Li authored
Use vma_pages function on vma object instead of explicit computation. Fix the following coccicheck warning: ./virt/kvm/kvm_main.c:3526:29-35: WARNING: Consider using vma_pages helper on vma Reported-by:
Abaci Robot <abaci@linux.alibaba.com> Signed-off-by:
Yang Li <yang.lee@linux.alibaba.com> Message-Id: <1632900526-119643-1-git-send-email-yang.lee@linux.alibaba.com> Signed-off-by:
-
Vitaly Kuznetsov authored
Currently, 'vmx->nested.vmxon_ptr' is not reset upon VMXOFF emulation. This is not a problem per se as we never access it when !vmx->nested.vmxon. But this should be done to avoid any issue in the future. Also, initialize the vmxon_ptr when vcpu is created. Signed-off-by:
Yu Zhang <yu.c.zhang@linux.intel.com> Message-Id: <20210929175154.11396-3-yu.c.zhang@linux.intel.com> Signed-off-by:
-
Yu Zhang authored
Clean up nested.c and vmx.c by using INVALID_GPA instead of "-1ull", to denote an invalid address in nested VMX. Affected addresses are the ones of VMXON region, current VMCS, VMCS link pointer, virtual- APIC page, ENCLS-exiting bitmap, and IO bitmap etc. Suggested-by:
-
Sean Christopherson authored
Rework the CPU selection in the migration worker to ensure the specified number of migrations are performed when the test iteslf is affined to a subset of CPUs. The existing logic skips iterations if the target CPU is not in the original set of possible CPUs, which causes the test to fail if too many iterations are skipped. ==== Test Assertion Failure ==== rseq_test.c:228: i > (NR_TASK_MIGRATIONS / 2) pid=10127 tid=10127 errno=4 - Interrupted system call 1 0x00000000004018e5: main at rseq_test.c:227 2 0x00007fcc8fc66bf6: ?? ??:0 3 0x0000000000401959: _start at ??:? Only performed 4 KVM_RUNs, task stalled too much? Calculate the min/max possible CPUs as a cheap "best effort" to avoid high runtimes when the test is affined to a small percentage of CPUs. Alternatively, a list or xarray of the possible CPUs could be used, but even in a horrendously inefficient setup, such optimizations are not needed because the runtime is completely dominated by the cost of migrating the task, and the absolute runtime is well under a minute in even truly absurd setups, e.g. running on a subset of vCPUs in a VM that is heavily overcommited (16 vCPUs per pCPU). Fixes: 61e52f16 ("KVM: selftests: Add a test for KVM_RUN+rseq to detect task migration bugs") Reported-by:Dongli Zhang <dongli.zhang@oracle.com> Signed-off-by:
-
Sean Christopherson authored
Check whether a CPUID entry's index is significant before checking for a matching index to hack-a-fix an undefined behavior bug due to consuming uninitialized data. RESET/INIT emulation uses kvm_cpuid() to retrieve CPUID.0x1, which does _not_ have a significant index, and fails to initialize the dummy variable that doubles as EBX/ECX/EDX output _and_ ECX, a.k.a. index, input. Practically speaking, it's _extremely_ unlikely any compiler will yield code that causes problems, as the compiler would need to inline the kvm_cpuid() call to detect the uninitialized data, and intentionally hose the kernel, e.g. insert ud2, instead of simply ignoring the result of the index comparison. Although the sketchy "dummy" pattern was introduced in SVM by commit 66f7b72e ("KVM: x86: Make register state after reset conform to specification"), it wasn't actually broken until commit 7ff6c035 ("KVM: x86: Remove stateful CPUID handling") arbitrarily swapped the order of operations such that "index" was checked before the significant flag. Avoid consuming uninitialized data by reverting to checking the flag before the index purely so that the fix can be easily backported; the offending RESET/INIT code has been refactored, moved, and consolidated from vendor code to common x86 since the bug was introduced. A future patch will directly address the bad RESET/INIT behavior. The undefined behavior was detected by syzbot + KernelMemorySanitizer. BUG: KMSAN: uninit-value in cpuid_entry2_find arch/x86/kvm/cpuid.c:68 BUG: KMSAN: uninit-value in kvm_find_cpuid_entry arch/x86/kvm/cpuid.c:1103 BUG: KMSAN: uninit-value in kvm_cpuid+0x456/0x28f0 arch/x86/kvm/cpuid.c:1183 cpuid_entry2_find arch/x86/kvm/cpuid.c:68 [inline] kvm_find_cpuid_entry arch/x86/kvm/cpuid.c:1103 [inline] kvm_cpuid+0x456/0x28f0 arch/x86/kvm/cpuid.c:1183 kvm_vcpu_reset+0x13fb/0x1c20 arch/x86/kvm/x86.c:10885 kvm_apic_accept_events+0x58f/0x8c0 arch/x86/kvm/lapic.c:2923 vcpu_enter_guest+0xfd2/0x6d80 arch/x86/kvm/x86.c:9534 vcpu_run+0x7f5/0x18d0 arch/x86/kvm/x86.c:9788 kvm_arch_vcpu_ioctl_run+0x245b/0x2d10 arch/x86/kvm/x86.c:10020 Local variable ----dummy@kvm_vcpu_reset created at: kvm_vcpu_reset+0x1fb/0x1c20 arch/x86/kvm/x86.c:10812 kvm_apic_accept_events+0x58f/0x8c0 arch/x86/kvm/lapic.c:2923 Reported-by: syzbot+f3985126b746b3d59c9d@syzkaller.appspotmail.com Reported-by:
Alexander Potapenko <glider@google.com> Fixes: 2a24be79 ("KVM: VMX: Set EDX at INIT with CPUID.0x1, Family-Model-Stepping") Fixes: 7ff6c035 ("KVM: x86: Remove stateful CPUID handling") Cc: stable@vger.kernel.org Signed-off-by:
-
Zelin Deng authored
hv_clock is preallocated to have only HVC_BOOT_ARRAY_SIZE (64) elements; if the PTP_SYS_OFFSET_PRECISE ioctl is executed on vCPUs whose index is 64 of higher, retrieving the struct pvclock_vcpu_time_info pointer with "src = &hv_clock[cpu].pvti" will result in an out-of-bounds access and a wild pointer. Change it to "this_cpu_pvti()" which is guaranteed to be valid. Fixes: 95a3d445 ("Switch kvmclock data to a PER_CPU variable") Signed-off-by:
Zelin Deng <zelin.deng@linux.alibaba.com> Cc: <stable@vger.kernel.org> Message-Id: <1632892429-101194-3-git-send-email-zelin.deng@linux.alibaba.com> Signed-off-by:
-
