- 23 Mar, 2023 3 commits
-
-
Santosh Shukla authored
Add support for SVM's Virtual NMIs implementation, which adds proper tracking of virtual NMI blocking, and an intr_ctrl flag that software can set to mark a virtual NMI as pending. Pending virtual NMIs are serviced by hardware if/when virtual NMIs become unblocked, i.e. act more or less like real NMIs. Introduce two new kvm_x86_ops callbacks so to support SVM's vNMI, as KVM needs to treat a pending vNMI as partially injected. Specifically, if two NMIs (for L1) arrive concurrently in KVM's software model, KVM's ABI is to inject one and pend the other. Without vNMI, KVM manually tracks the pending NMI and uses NMI windows to detect when the NMI should be injected. With vNMI, the pending NMI is simply stuffed into the VMCB and handed off to hardware. This means that KVM needs to be able to set a vNMI pending on-demand, and also query if a vNMI is pending, e.g. to honor the "at most one NMI pending" rule and to preserve all NMIs across save and restore. Warn if KVM attempts to open an NMI window when vNMI is fully enabled, as the above logic should prevent KVM from ever getting to kvm_check_and_inject_events() with two NMIs pending _in software_, and the "at most one NMI pending" logic should prevent having an NMI pending in hardware and an NMI pending in software if NMIs are also blocked, i.e. if KVM can't immediately inject the second NMI. Signed-off-by:
Santosh Shukla <Santosh.Shukla@amd.com> Co-developed-by:
Maxim Levitsky <mlevitsk@redhat.com> Signed-off-by:
Sean Christopherson <seanjc@google.com>
-
Sean Christopherson authored
Use the asynchronous NMI queue to handle pending NMIs coming in from userspace during KVM_SET_VCPU_EVENTS so that all of KVM's logic for handling multiple NMIs goes through process_nmi(). This will simplify supporting SVM's upcoming "virtual NMI" functionality, which will need changes KVM manages pending NMIs. Signed-off-by: -
Santosh Shukla authored
Add defines for three new bits in VMVC::int_ctrl that are part of SVM's Virtual NMI (vNMI) support: V_NMI_PENDING_MASK(11) - Virtual NMI is pending V_NMI_BLOCKING_MASK(12) - Virtual NMI is masked V_NMI_ENABLE_MASK(26) - Enable NMI virtualization To "inject" an NMI, the hypervisor (KVM) sets V_NMI_PENDING. When the CPU services the pending vNMI, hardware clears V_NMI_PENDING and sets V_NMI_BLOCKING, e.g. to indicate that the vCPU is handling an NMI. Hardware clears V_NMI_BLOCKING upon successful execution of IRET, or if a VM-Exit occurs while delivering the virtual NMI. Reviewed-by:
-
- 22 Mar, 2023 8 commits
-
-
Sean Christopherson authored
The existing X86_FEATURE_VNMI is a synthetic feature flag that exists purely to maintain /proc/cpuinfo's ABI, the "real" Intel vNMI feature flag is tracked as VMX_FEATURE_VIRTUAL_NMIS, as the feature is enumerated through VMX MSRs, not CPUID. AMD is also gaining virtual NMI support, but in true VMX vs. SVM form, enumerates support through CPUID, i.e. wants to add real feature flag for vNMI. Redefine the syntheic X86_FEATURE_VNMI to AMD's real CPUID bit to avoid having both X86_FEATURE_VNMI and e.g. X86_FEATURE_AMD_VNMI. Signed-off-by: -
Sean Christopherson authored
Save all pending NMIs in KVM_GET_VCPU_EVENTS, and queue KVM_REQ_NMI if one or more NMIs are pending after KVM_SET_VCPU_EVENTS in order to re-evaluate pending NMIs with respect to NMI blocking. KVM allows multiple NMIs to be pending in order to faithfully emulate bare metal handling of simultaneous NMIs (on bare metal, truly simultaneous NMIs are impossible, i.e. one will always arrive first and be consumed). Support for simultaneous NMIs botched the save/restore though. KVM only saves one pending NMI, but allows userspace to restore 255 pending NMIs as kvm_vcpu_events.nmi.pending is a u8, and KVM's internal state is stored in an unsigned int. Fixes: 7460fb4a ("KVM: Fix simultaneous NMIs") Signed-off-by:
-
Sean Christopherson authored
Tweak the code and comment that deals with concurrent NMIs to explicitly call out that x86 allows exactly one pending NMI, but that KVM needs to temporarily allow two pending NMIs in order to workaround the fact that the target vCPU cannot immediately recognize an incoming NMI, unlike bare metal. No functional change intended. Signed-off-by:
-
Sean Christopherson authored
Don't raise KVM_REQ_EVENT if no NMIs are pending at the end of process_nmi(). Finishing process_nmi() without a pending NMI will become much more likely when KVM gains support for AMD's vNMI, which allows pending vNMIs in hardware, i.e. doesn't require explicit injection. Signed-off-by:
-
Maxim Levitsky authored
SEV-ES guests don't use IRET interception for the detection of an end of a NMI. Therefore it makes sense to create a wrapper to avoid repeating the check for the SEV-ES. No functional change is intended. Suggested-by:
-
Maxim Levitsky authored
If L1 doesn't intercept interrupts, then KVM will use vmcb02's V_IRQ to detect an interrupt window for L1 IRQs. On a subsequent nested VM-Exit, KVM might need to copy the current V_IRQ from vmcb02 to vmcb01 to continue waiting for an interrupt window, i.e. if there is still a pending IRQ for L1. Raise KVM_REQ_EVENT on nested exit if L1 isn't intercepting IRQs to ensure that KVM will re-enable interrupt window detection if needed. Note that this is a theoretical bug because KVM already raises KVM_REQ_EVENT on each nested VM exit, because the nested VM exit resets RFLAGS and kvm_set_rflags() raises the KVM_REQ_EVENT unconditionally. Explicitly raise KVM_REQ_EVENT for the interrupt window case to avoid having an unnecessary dependency on kvm_set_rflags(), and to document the scenario. Signed-off-by:
-
Santosh Shukla authored
Disable intercept of virtual interrupts (used to detect interrupt windows) if the saved host (L1) RFLAGS.IF is '0', as the effective RFLAGS.IF for L1 interrupts will never be set while L2 is running (L2's RFLAGS.IF doesn't affect L1 IRQs when virtual interrupts are enabled). Suggested-by:
-
Santosh Shukla authored
Don't sync vmcb02 V_IRQ back to vmcb12 if KVM (L0) is intercepting virtual interrupts in order to request an interrupt window, as KVM has usurped vmcb02's int_ctl. If an interrupt window opens before the next VM-Exit, svm_clear_vintr() will restore vmcb12's int_ctl. If no window opens, V_IRQ will be correctly preserved in vmcb12's int_ctl (because it was never recognized while L2 was running). Suggested-by:
-
- 16 Mar, 2023 9 commits
-
-
Thomas Huth authored
All kvm_arch_vm_ioctl() implementations now only deal with "int" types as return values, so we can change the return type of these functions to use "int" instead of "long". Signed-off-by:
Thomas Huth <thuth@redhat.com> Acked-by:
Anup Patel <anup@brainfault.org> Message-Id: <20230208140105.655814-7-thuth@redhat.com> Signed-off-by:
Paolo Bonzini <pbonzini@redhat.com>
-
Thomas Huth authored
KVM functions use "long" return values for functions that are wired up to "struct file_operations", but otherwise use "int" return values for functions that can return 0/-errno in order to avoid unintentional divergences between 32-bit and 64-bit kernels. Some code still uses "long" in unnecessary spots, though, which can cause a little bit of confusion and unnecessary size casts. Let's change these spots to use "int" types, too. Signed-off-by:
-
Thomas Huth authored
In case of success, this function returns the amount of handled bytes. However, this does not work for large values: The function is called from kvm_arch_vm_ioctl() (which still returns a long), which in turn is called from kvm_vm_ioctl() in virt/kvm/kvm_main.c. And that function stores the return value in an "int r" variable. So the upper 32-bits of the "long" return value are lost there. KVM ioctl functions should only return "int" values, so let's limit the amount of bytes that can be requested here to INT_MAX to avoid the problem with the truncated return value. We can then also change the return type of the function to "int" to make it clearer that it is not possible to return a "long" here. Fixes: f0376edb ("KVM: arm64: Add ioctl to fetch/store tags in a guest") Signed-off-by:
Cornelia Huck <cohuck@redhat.com> Reviewed-by:
Gavin Shan <gshan@redhat.com> Reviewed-by:
Steven Price <steven.price@arm.com> Message-Id: <20230208140105.655814-5-thuth@redhat.com> Signed-off-by:
-
Thomas Huth authored
The KVM_GET_NR_MMU_PAGES ioctl is quite questionable on 64-bit hosts since it fails to return the full 64 bits of the value that can be set with the corresponding KVM_SET_NR_MMU_PAGES call. Its "long" return value is truncated into an "int" in the kvm_arch_vm_ioctl() function. Since this ioctl also never has been used by userspace applications (QEMU, Google's internal VMM, kvmtool and CrosVM have been checked), it's likely the best if we remove this badly designed ioctl before anybody really tries to use it. Signed-off-by:
-
Thomas Huth authored
These two functions only return normal integers, so it does not make sense to declare the return type as "long" here. Reviewed-by:
Claudio Imbrenda <imbrenda@linux.ibm.com> Signed-off-by:
-
Thomas Huth authored
Most functions that are related to kvm_arch_vm_ioctl() already use "int" as return type to pass error values back to the caller. Some outlier functions use "long" instead for no good reason (they do not really require long values here). Let's standardize on "int" here to avoid casting the values back and forth between the two types. Reviewed-by:
Nicholas Piggin <npiggin@gmail.com> Signed-off-by:
-
Emanuele Giuseppe Esposito authored
FLUSH_L1D was already added in 11e34e64, but the feature is not visible to userspace yet. The bit definition: CPUID.(EAX=7,ECX=0):EDX[bit 28] If the feature is supported by the host, kvm should support it too so that userspace can choose whether to expose it to the guest or not. One disadvantage of not exposing it is that the guest will report a non existing vulnerability in /sys/devices/system/cpu/vulnerabilities/mmio_stale_data because the mitigation is present only if the guest supports (FLUSH_L1D and MD_CLEAR) or FB_CLEAR. Signed-off-by:
Emanuele Giuseppe Esposito <eesposit@redhat.com> Message-Id: <20230201132905.549148-4-eesposit@redhat.com> Signed-off-by:
-
Emanuele Giuseppe Esposito authored
Expose IA32_FLUSH_CMD to the guest if the guest CPUID enumerates support for this MSR. As with IA32_PRED_CMD, permission for unintercepted writes to this MSR will be granted to the guest after the first non-zero write. Signed-off-by:
-
Emanuele Giuseppe Esposito authored
Expose IA32_FLUSH_CMD to the guest if the guest CPUID enumerates support for this MSR. As with IA32_PRED_CMD, permission for unintercepted writes to this MSR will be granted to the guest after the first non-zero write. Co-developed-by:
Jim Mattson <jmattson@google.com> Signed-off-by:
-
- 14 Mar, 2023 20 commits
-
-
Sean Christopherson authored
Rename enable_evmcs to __kvm_is_using_evmcs to match its wrapper, and to avoid confusion with enabling eVMCS for nested virtualization, i.e. have "enable eVMCS" be reserved for "enable eVMCS support for L1". No functional change intended. Signed-off-by:
-
Sean Christopherson authored
Wrap enable_evmcs in a helper and stub it out when CONFIG_HYPERV=n in order to eliminate the static branch nop placeholders. clang-14 is clever enough to elide the nop, but gcc-12 is not. Stubbing out the key reduces the size of kvm-intel.ko by ~7.5% (200KiB) when compiled with gcc-12 (there are a _lot_ of VMCS accesses throughout KVM). Reviewed-by:
Vitaly Kuznetsov <vkuznets@redhat.com> Signed-off-by:
-
Sean Christopherson authored
Move the macros that define the set of VMCS controls that are supported by eVMCS1 from hyperv.h to hyperv.c, i.e. make them "private". The macros should never be consumed directly by KVM at-large since the "final" set of supported controls depends on guest CPUID. No functional change intended. Reviewed-by:
-
Lai Jiangshan authored
Drop FNAME(is_self_change_mapping) and instead rely on kvm_mmu_hugepage_adjust() to adjust the hugepage accordingly. Prior to commit 4cd071d1 ("KVM: x86/mmu: Move calls to thp_adjust() down a level"), the hugepage adjustment was done before allocating new shadow pages, i.e. failed to restrict the hugepage sizes if a new shadow page resulted in account_shadowed() changing the disallowed hugepage tracking. Removing FNAME(is_self_change_mapping) fixes a bug reported by Huang Hang where KVM unnecessarily forces a 4KiB page. FNAME(is_self_change_mapping) has a defect in that it blindly disables _all_ hugepage mappings rather than trying to reduce the size of the hugepage. If the guest is writing to a 1GiB page and the 1GiB is self-referential but a 2MiB page is not, then KVM can and should create a 2MiB mapping. Add a comment above the call to kvm_mmu_hugepage_adjust() to call out the new dependency on adjusting the hugepage size after walking indirect PTEs. Reported-by:
Huang Hang <hhuang@linux.alibaba.com> Signed-off-by:
Lai Jiangshan <jiangshan.ljs@antgroup.com> Link: https://lore.kernel.org/r/20221213125538.81209-1-jiangshanlai@gmail.com [sean: rework changelog after separating out the emulator change] Signed-off-by:
-
Lai Jiangshan authored
Move the detection of write #PF to shadow pages, i.e. a fault on a write to a page table that is being shadowed by KVM that is used to translate the write itself, from FNAME(is_self_change_mapping) to FNAME(fetch). There is no need to detect the self-referential write before kvm_faultin_pfn() as KVM does not consume EMULTYPE_WRITE_PF_TO_SP for accesses that resolve to "error or no-slot" pfns, i.e. KVM doesn't allow retrying MMIO accesses or writes to read-only memslots. Detecting the EMULTYPE_WRITE_PF_TO_SP scenario in FNAME(fetch) will allow dropping FNAME(is_self_change_mapping) entirely, as the hugepage interaction can be deferred to kvm_mmu_hugepage_adjust(). Cc: Huang Hang <hhuang@linux.alibaba.com> Signed-off-by:
-
Sean Christopherson authored
Use a new EMULTYPE flag, EMULTYPE_WRITE_PF_TO_SP, to track page faults on self-changing writes to shadowed page tables instead of propagating that information to the emulator via a semi-persistent vCPU flag. Using a flag in "struct kvm_vcpu_arch" is confusing, especially as implemented, as it's not at all obvious that clearing the flag only when emulation actually occurs is correct. E.g. if KVM sets the flag and then retries the fault without ever getting to the emulator, the flag will be left set for future calls into the emulator. But because the flag is consumed if and only if both EMULTYPE_PF and EMULTYPE_ALLOW_RETRY_PF are set, and because EMULTYPE_ALLOW_RETRY_PF is deliberately not set for direct MMUs, emulated MMIO, or while L2 is active, KVM avoids false positives on a stale flag since FNAME(page_fault) is guaranteed to be run and refresh the flag before it's ultimately consumed by the tail end of reexecute_instruction(). Signed-off-by:
-
Vipin Sharma authored
Add missing KVM_EXIT_* reasons in KVM selftests from include/uapi/linux/kvm.h Signed-off-by:
Vipin Sharma <vipinsh@google.com> Message-Id: <20230204014547.583711-5-vipinsh@google.com> Signed-off-by:
-
Sean Christopherson authored
Add and use a macro to generate the KVM exit reason strings array instead of relying on developers to correctly copy+paste+edit each string. Signed-off-by:
-
Vipin Sharma authored
Print what KVM exit reason a test was expecting and what it actually got int TEST_ASSERT_KVM_EXIT_REASON(). Signed-off-by:
-
Vipin Sharma authored
Make TEST_ASSERT_KVM_EXIT_REASON() macro and replace all exit reason test assert statements with it. No functional changes intended. Signed-off-by:
David Matlack <dmatlack@google.com> Message-Id: <20230204014547.583711-2-vipinsh@google.com> Signed-off-by:
-
David Woodhouse authored
When kvm_xen_evtchn_send() takes the slow path because the shinfo GPC needs to be revalidated, it used to violate the SRCU vs. kvm->lock locking rules and potentially cause a deadlock. Now that lockdep is learning to catch such things, make sure that code path is exercised by the selftest. Link: https://lore.kernel.org/all/20230113124606.10221-2-dwmw2@infradead.orgSigned-off-by:
David Woodhouse <dwmw@amazon.co.uk> Signed-off-by:
-
David Woodhouse authored
The xen_shinfo_test started off with very few iterations, and the numbers we used in GUEST_SYNC() were precisely mapped to the RUNSTATE_xxx values anyway to start with. It has since grown quite a few more tests, and it's kind of awful to be handling them all as bare numbers. Especially when I want to add a new test in the middle. Define an enum for the test stages, and use it both in the guest code and the host switch statement. No functional change, if I can count to 24. Signed-off-by:
-
Sean Christopherson authored
Add wrappers to do hypercalls using VMCALL/VMMCALL and Xen's register ABI (as opposed to full Xen-style hypercalls through a hypervisor provided page). Using the common helpers dedups a pile of code, and uses the native hypercall instruction when running on AMD. Signed-off-by:
-
Sean Christopherson authored
Extract the guts of kvm_hypercall() to a macro so that Xen hypercalls, which have a different register ABI, can reuse the VMCALL vs. VMMCALL logic. No functional change intended. Signed-off-by:
-
Sean Christopherson authored
WARN if generating a GATag given a VM ID and vCPU ID doesn't yield the same IDs when pulling the IDs back out of the tag. Don't bother adding error handling to callers, this is very much a paranoid sanity check as KVM fully controls the VM ID and is supposed to reject too-big vCPU IDs. Signed-off-by:
Suravee Suthikulpanit <suravee.suthikulpanit@amd.com> Tested-by:
-
Suravee Suthikulpanit authored
Define AVIC_VCPU_ID_MASK based on AVIC_PHYSICAL_MAX_INDEX, i.e. the mask that effectively controls the largest guest physical APIC ID supported by x2AVIC, instead of hardcoding the number of bits to 8 (and the number of VM bits to 24). The AVIC GATag is programmed into the AMD IOMMU IRTE to provide a reference back to KVM in case the IOMMU cannot inject an interrupt into a non-running vCPU. In such a case, the IOMMU notifies software by creating a GALog entry with the corresponded GATag, and KVM then uses the GATag to find the correct VM+vCPU to kick. Dropping bit 8 from the GATag results in kicking the wrong vCPU when targeting vCPUs with x2APIC ID > 255. Fixes: 4d1d7942 ("KVM: SVM: Introduce logic to (de)activate x2AVIC mode") Cc: stable@vger.kernel.org Reported-by:
Alejandro Jimenez <alejandro.j.jimenez@oracle.com> Signed-off-by:
-
Sean Christopherson authored
Define the "physical table max index mask" as bits 8:0, not 9:0. x2AVIC currently supports a max of 512 entries, i.e. the max index is 511, and the inputs to GENMASK_ULL() are inclusive. The bug is benign as bit 9 is reserved and never set by KVM, i.e. KVM is just clearing bits that are guaranteed to be zero. Note, as of this writing, APM "Rev. 3.39-October 2022" incorrectly states that bits 11:8 are reserved in Table B-1. VMCB Layout, Control Area. I.e. that table wasn't updated when x2AVIC support was added. Opportunistically fix the comment for the max AVIC ID to align with the code, and clean up comment formatting too. Fixes: 4d1d7942 ("KVM: SVM: Introduce logic to (de)activate x2AVIC mode") Cc: stable@vger.kernel.org Cc: Alejandro Jimenez <alejandro.j.jimenez@oracle.com> Cc: Suravee Suthikulpanit <suravee.suthikulpanit@amd.com> Signed-off-by:
-
Paolo Bonzini authored
Right now, if KVM memory stress tests are run with hugetlb sources but hugetlb is not available (either in the kernel or because /proc/sys/vm/nr_hugepages is 0) the test will fail with a memory allocation error. This makes it impossible to add tests that default to hugetlb-backed memory, because on a machine with a default configuration they will fail. Therefore, check HugePages_Total as well and, if zero, direct the user to enable hugepages in procfs. Furthermore, return KSFT_SKIP whenever hugetlb is not available. Signed-off-by: -
Rong Tao authored
Code indentation should use tabs where possible and miss a '*'. Signed-off-by:
Rong Tao <rongtao@cestc.cn> Message-Id: <tencent_A492CB3F9592578451154442830EA1B02C07@qq.com> Signed-off-by:
-
Rong Tao authored
Code indentation should use tabs where possible. Signed-off-by:
-
