- 28 Sep, 2020 40 commits
-
-
Alexander Graf authored
We will introduce the concept of MSRs that may not be handled in kernel space soon. Some MSRs are directly passed through to the guest, effectively making them handled by KVM from user space's point of view. This patch introduces all logic required to ensure that MSRs that user space wants trapped are not marked as direct access for guests. Signed-off-by:
Alexander Graf <graf@amazon.com> Message-Id: <20200925143422.21718-7-graf@amazon.com> [Replace "_idx" with "_slot". - Paolo] Signed-off-by:
Paolo Bonzini <pbonzini@redhat.com>
-
Alexander Graf authored
We will introduce the concept of MSRs that may not be handled in kernel space soon. Some MSRs are directly passed through to the guest, effectively making them handled by KVM from user space's point of view. This patch introduces all logic required to ensure that MSRs that user space wants trapped are not marked as direct access for guests. Signed-off-by:
-
Aaron Lewis authored
Prepare vmx and svm for a subsequent change that ensures the MSR permission bitmap is set to allow an MSR that userspace is tracking to force a vmx_vmexit in the guest. Signed-off-by:
Aaron Lewis <aaronlewis@google.com> Reviewed-by:
Oliver Upton <oupton@google.com> [agraf: rebase, adapt SVM scheme to nested changes that came in between] Signed-off-by:
-
Alexander Graf authored
In the following commits we will add pieces of MSR filtering. To ensure that code compiles even with the feature half-merged, let's add a few stubs and struct definitions before the real patches start. Signed-off-by:
-
Alexander Graf authored
MSRs are weird. Some of them are normal control registers, such as EFER. Some however are registers that really are model specific, not very interesting to virtualization workloads, and not performance critical. Others again are really just windows into package configuration. Out of these MSRs, only the first category is necessary to implement in kernel space. Rarely accessed MSRs, MSRs that should be fine tunes against certain CPU models and MSRs that contain information on the package level are much better suited for user space to process. However, over time we have accumulated a lot of MSRs that are not the first category, but still handled by in-kernel KVM code. This patch adds a generic interface to handle WRMSR and RDMSR from user space. With this, any future MSR that is part of the latter categories can be handled in user space. Furthermore, it allows us to replace the existing "ignore_msrs" logic with something that applies per-VM rather than on the full system. That way you can run productive VMs in parallel to experimental ones where you don't care about proper MSR handling. Signed-off-by:
Jim Mattson <jmattson@google.com> Message-Id: <20200925143422.21718-3-graf@amazon.com> Signed-off-by:
-
Alexander Graf authored
When we find an MSR that we can not handle, bubble up that error code as MSR error return code. Follow up patches will use that to expose the fact that an MSR is not handled by KVM to user space. Suggested-by:
-
Sean Christopherson authored
Rename "index" to "slot" in struct vmx_uret_msr to align with the terminology used by common x86's kvm_user_return_msrs, and to avoid conflating "MSR's ECX index" with "MSR's index into an array". No functional change intended. Signed-off-by:
Sean Christopherson <sean.j.christopherson@intel.com> Message-Id: <20200923180409.32255-16-sean.j.christopherson@intel.com> Signed-off-by:
-
Sean Christopherson authored
Rename "vmx_msr_index" to "vmx_uret_msrs_list" to associate it with the uret MSRs array, and to avoid conflating "MSR's ECX index" with "MSR's index into an array". Similarly, don't use "slot" in the name as that terminology is claimed by the common x86 "user_return_msrs" mechanism. No functional change intended. Signed-off-by:
-
Sean Christopherson authored
Add "uret" to vmx_set_guest_msr() to explicitly associate it with the guest_uret_msrs array, and to differentiate it from vmx_set_msr() as well as VMX's load/store MSRs. No functional change intended. Signed-off-by:
-
Sean Christopherson authored
Rename "find_msr_entry" to scope it to VMX and to associate it with guest_uret_msrs. Drop the "entry" so that the function name pairs with the existing __vmx_find_uret_msr(), which intentionally uses a double underscore prefix instead of appending "index" or "slot" as those names are already claimed by other pieces of the user return MSR stack. No functional change intended. Signed-off-by:
-
Sean Christopherson authored
Add vmx_setup_uret_msr() to wrap the lookup and manipulation of the uret MSRs array during setup_msrs(). In addition to consolidating code, this eliminates move_msr_up(), which while being a very literally description of the function, isn't exacly helpful in understanding the net effect of the code. No functional change intended. Signed-off-by:
-
Sean Christopherson authored
Move checking for the existence of MSR_EFER in the uret MSR array into update_transition_efer() so that the lookup and manipulation of the array in setup_msrs() occur back-to-back. This paves the way toward adding a helper to wrap the lookup and manipulation. To avoid unnecessary overhead, defer the lookup until the uret array would actually be modified in update_transition_efer(). EFER obviously exists on CPUs that support the dedicated VMCS fields for switching EFER, and EFER must exist for the guest and host EFER.NX value to diverge, i.e. there is no danger of attempting to read/write EFER when it doesn't exist. Signed-off-by:
-
Sean Christopherson authored
Check for RDTSCP support prior to checking if MSR_TSC_AUX is in the uret MSRs array so that the array lookup and manipulation are back-to-back. This paves the way toward adding a helper to wrap the lookup and manipulation. No functional change intended. Signed-off-by:
-
Sean Christopherson authored
Rename "__find_msr_index" to scope it to VMX, associate it with guest_uret_msrs, and to avoid conflating "MSR's ECX index" with "MSR's array index". Similarly, don't use "slot" in the name so as to avoid colliding the common x86's half of "user_return_msrs" (the slot in kvm_user_return_msrs is not the same slot in guest_uret_msrs). No functional change intended. Signed-off-by:
-
Sean Christopherson authored
Add "uret" to "guest_msrs_ready" to explicitly associate it with the "guest_uret_msrs" array, and replace "ready" with "loaded" to more precisely reflect what it tracks, e.g. "ready" could be interpreted as meaning ready for processing (setup_msrs() has run), which is wrong. "loaded" also aligns with the similar "guest_state_loaded" field. No functional change intended. Signed-off-by:
-
Sean Christopherson authored
Add "uret" into the name of "save_nmsrs" to explicitly associate it with the guest_uret_msrs array, and replace "save" with "active" (for lack of a better word) to better describe what is being tracked. While "save" is more or less accurate when viewed as a literal description of the field, e.g. it holds the number of MSRs that were saved into the array the last time setup_msrs() was invoked, it can easily be misinterpreted by the reader, e.g. as meaning the number of MSRs that were saved from hardware at some point in the past, or as the number of MSRs that need to be saved at some point in the future, both of which are wrong. No functional change intended. Signed-off-by:
-
Sean Christopherson authored
Rename vcpu_vmx.nsmrs to vcpu_vmx.nr_uret_msrs to explicitly associate it with the guest_uret_msrs array. No functional change intended. Signed-off-by:
-
Sean Christopherson authored
Rename struct "shared_msr_entry" to "vmx_uret_msr" to align with x86's rename of "shared_msrs" to "user_return_msrs", and to call out that the struct is specific to VMX, i.e. not part of the generic "shared_msrs" framework. Abbreviate "user_return" as "uret" to keep line lengths marginally sane and code more or less readable. No functional change intended. Signed-off-by:
-
Sean Christopherson authored
Add "loadstore" to vmx_find_msr_index() to differentiate it from the so called shared MSRs helpers (which will soon be renamed), and replace "index" with "slot" to better convey that the helper returns slot in the array, not the MSR index (the value that gets stuffed into ECX). No functional change intended. Signed-off-by:
-
Sean Christopherson authored
Add "MAX" to the LOADSTORE and so called SHARED MSR defines to make it more clear that the define controls the array size, as opposed to the actual number of valid entries that are in the array. No functional change intended. Signed-off-by:
-
Sean Christopherson authored
Rename the "shared_msrs" mechanism, which is used to defer restoring MSRs that are only consumed when running in userspace, to a more banal but less likely to be confusing "user_return_msrs". The "shared" nomenclature is confusing as it's not obvious who is sharing what, e.g. reasonable interpretations are that the guest value is shared by vCPUs in a VM, or that the MSR value is shared/common to guest and host, both of which are wrong. "shared" is also misleading as the MSR value (in hardware) is not guaranteed to be shared/reused between VMs (if that's indeed the correct interpretation of the name), as the ability to share values between VMs is simply a side effect (albiet a very nice side effect) of deferring restoration of the host value until returning from userspace. "user_return" avoids the above confusion by describing the mechanism itself instead of its effects. Signed-off-by:
-
Sean Christopherson authored
Move initialization of 'struct kvm_mmu' fields into alloc_mmu_pages() to consolidate code, and rename the helper to __kvm_mmu_create(). No functional change intended. Signed-off-by:
-
Sean Christopherson authored
Read vmcs.EXIT_QUALIFICATION and vmcs.VM_EXIT_INTR_INFO only if the VM-Exit is being reflected to L1 now that they are no longer passed directly to the kvm_nested_vmexit tracepoint. No functional change intended. Signed-off-by:
-
Sean Christopherson authored
Use the newly introduced TRACE_EVENT_KVM_EXIT to define the guts of kvm_nested_vmexit so that it captures and prints the same information as kvm_exit. This has the bonus side effect of fixing the interrupt info and error code printing for the case where they're invalid, e.g. if the exit was a failed VM-Entry. This also sets the stage for retrieving EXIT_QUALIFICATION and VM_EXIT_INTR_INFO in nested_vmx_reflect_vmexit() if and only if the VM-Exit is being routed to L1. Signed-off-by:
-
Sean Christopherson authored
Macrofy the definition of kvm_exit so that the definition can be reused verbatim by kvm_nested_vmexit. No functional change intended. Signed-off-by:
-
Sean Christopherson authored
Extend the kvm_exit tracepoint to align it with kvm_nested_vmexit in terms of what information is captured. On SVM, add interrupt info and error code, while on VMX it add IDT vectoring and error code. This sets the stage for macrofying the kvm_exit tracepoint definition so that it can be reused for kvm_nested_vmexit without loss of information. Opportunistically stuff a zero for VM_EXIT_INTR_INFO if the VM-Enter failed, as the field is guaranteed to be invalid. Note, it'd be possible to further filter the interrupt/exception fields based on the VM-Exit reason, but the helper is intended only for tracepoints, i.e. an extra VMREAD or two is a non-issue, the failed VM-Enter case is just low hanging fruit. Signed-off-by:
-
Sean Christopherson authored
Add a helper, is_exception_with_error_code(), to provide the simple but difficult to read code of checking for a valid exception with an error code given a vmcs.VM_EXIT_INTR_INFO value. The helper will gain another user, vmx_get_exit_info(), in a future patch. Signed-off-by:
-
Sean Christopherson authored
Use kvm_rip_read() to read the guest's RIP for the nested VM-Exit tracepoint instead of having the caller pass in an argument. Params that are passed into a tracepoint are evaluated even if the tracepoint is disabled, i.e. passing in RIP for VMX incurs a VMREAD and retpoline to retrieve a value that may never be used, e.g. if the exit is due to a hardware interrupt. Signed-off-by:
-
Sean Christopherson authored
Add RIP to the kvm_entry tracepoint to help debug if the kvm_exit tracepoint is disabled or if VM-Enter fails, in which case the kvm_exit tracepoint won't be hit. Read RIP from within the tracepoint itself to avoid a potential VMREAD and retpoline if the guest's RIP isn't available. Signed-off-by:
-
Sean Christopherson authored
WARN if KVM attempts to switch to the currently loaded VMCS. Now that nested_vmx_free_vcpu() doesn't blindly call vmx_switch_vmcs(), all paths that lead to vmx_switch_vmcs() are implicitly guarded by guest vs. host mode, e.g. KVM should never emulate VMX instructions when guest mode is active, and nested_vmx_vmexit() should never be called when host mode is active. Signed-off-by:
-
Sean Christopherson authored
Remove the explicit switch to vmcs01 and the call to free_nested() in nested_vmx_free_vcpu(). free_nested(), which is called unconditionally by vmx_leave_nested(), ensures vmcs01 is loaded prior to freeing vmcs02 and friends. Signed-off-by:
-
Sean Christopherson authored
Add a WARN in free_nested() to ensure vmcs01 is loaded prior to freeing vmcs02 and friends, and explicitly switch to vmcs01 if it's not. KVM is supposed to keep is_guest_mode() and loaded_vmcs==vmcs02 synchronized, but bugs happen and freeing vmcs02 while it's in use will escalate a KVM error to a use-after-free and potentially crash the kernel. Do the WARN and switch even in the !vmxon case to help detect latent bugs. free_nested() is not a hot path, and the check is cheap. Signed-off-by:
-
Sean Christopherson authored
Move free_nested() down below vmx_switch_vmcs() so that a future patch can do an "emergency" invocation of vmx_switch_vmcs() if vmcs01 is not the loaded VMCS when freeing nested resources. No functional change intended. Signed-off-by:
-
Sean Christopherson authored
Call guest_state_valid() directly instead of querying emulation_required when checking if L1 is attempting VM-Enter with invalid guest state. If emulate_invalid_guest_state is false, KVM will fixup segment regs to avoid emulation and will never set emulation_required, i.e. KVM will incorrectly miss the associated consistency checks because the nested path stuffs segments directly into vmcs02. Opportunsitically add Consistency Check tracing to make future debug suck a little less. Fixes: 2bb8cafe ("KVM: vVMX: signal failure for nested VMEntry if emulation_required") Fixes: 3184a995 ("KVM: nVMX: fix vmentry failure code when L2 state would require emulation") Cc: stable@vger.kernel.org Signed-off-by:
-
Sean Christopherson authored
Reload vmcs01 when bailing from nested_vmx_enter_non_root_mode() as KVM expects vmcs01 to be loaded when is_guest_mode() is false. Fixes: 671ddc70 ("KVM: nVMX: Don't leak L1 MMIO regions to L2") Cc: stable@vger.kernel.org Cc: Dan Cross <dcross@google.com> Cc: Jim Mattson <jmattson@google.com> Cc: Peter Shier <pshier@google.com> Signed-off-by:
-
Sean Christopherson authored
Explicitly reset the segment cache after stuffing guest segment regs in prepare_vmcs02_rare(). Although the cache is reset when switching to vmcs02, there is nothing that prevents KVM from re-populating the cache prior to writing vmcs02 with vmcs12's values. E.g. if the vCPU is preempted after switching to vmcs02 but before prepare_vmcs02_rare(), kvm_arch_vcpu_put() will dereference GUEST_SS_AR_BYTES via .get_cpl() and cache the stale vmcs02 value. While the current code base only caches stale data in the preemption case, it's theoretically possible future code could read a segment register during the nested flow itself, i.e. this isn't technically illegal behavior in kvm_arch_vcpu_put(), although it did introduce the bug. This manifests as an unexpected nested VM-Enter failure when running with unrestricted guest disabled if the above preemption case coincides with L1 switching L2's CPL, e.g. when switching from a L2 vCPU at CPL3 to to a L2 vCPU at CPL0. stack_segment_valid() will see the new SS_SEL but the old SS_AR_BYTES and incorrectly mark the guest state as invalid due to SS.dpl != SS.rpl. Don't bother updating the cache even though prepare_vmcs02_rare() writes every segment. With unrestricted guest, guest segments are almost never read, let alone L2 guest segments. On the other hand, populating the cache requires a large number of memory writes, i.e. it's unlikely to be a net win. Updating the cache would be a win when unrestricted guest is not supported, as guest_state_valid() will immediately cache all segment registers. But, nested virtualization without unrestricted guest is dirt slow, saving some VMREADs won't change that, and every CPU manufactured in the last decade supports unrestricted guest. In other words, the extra (minor) complexity isn't worth the trouble. Note, kvm_arch_vcpu_put() may see stale data when querying guest CPL depending on when preemption occurs. This is "ok" in that the usage is imperfect by nature, i.e. it's used heuristically to improve performance but doesn't affect functionality. kvm_arch_vcpu_put() could be "fixed" by also disabling preemption while loading segments, but that's pointless and misleading as reading state from kvm_sched_{in,out}() is guaranteed to see stale data in one form or another. E.g. even if all the usage of regs_avail is fixed to call kvm_register_mark_available() after the associated state is set, the individual state might still be stale with respect to the overall vCPU state. I.e. making functional decisions in an asynchronous hook is doomed from the get go. Thankfully KVM doesn't do that. Fixes: de63ad4c ("KVM: X86: implement the logic for spinlock optimization") Cc: stable@vger.kernel.org Signed-off-by: -
Sean Christopherson authored
Use bools to track write and user faults throughout the page fault paths and down into mmu_set_spte(). The actual usage is purely boolean, but that's not obvious without digging into all paths as the current code uses a mix of bools (TDP and try_async_pf) and ints (shadow paging and mmu_set_spte()). No true functional change intended (although the pgprintk() will now print 0/1 instead of 0/PFERR_WRITE_MASK). Signed-off-by:
-
Sean Christopherson authored
Move the "ITLB multi-hit workaround enabled" check into the callers of disallowed_hugepage_adjust() to make it more obvious that the helper is specific to the workaround, and to be consistent with the accounting, i.e. account_huge_nx_page() is called if and only if the workaround is enabled. No functional change intended. Signed-off-by:
-
Sean Christopherson authored
Rename 'hlevel', which presumably stands for 'host level', to simply 'level' in FNAME(fetch). The variable hasn't tracked the host level for quite some time. Signed-off-by:
-
Sean Christopherson authored
Condition the accounting of a disallowed huge NX page on the original requested level of the page being greater than the current iterator level. This does two things: accounts the page if and only if a huge page was actually disallowed, and accounts the shadow page if and only if it was the level at which the huge page was disallowed. For the latter case, the previous logic would account all shadow pages used to create the translation for the forced small page, e.g. even PML4, which can't be a huge page on current hardware, would be accounted as having been a disallowed huge page when using 5-level EPT. The overzealous accounting is purely a performance issue, i.e. the recovery thread will spuriously zap shadow pages, but otherwise the bad behavior is harmless. Cc: Junaid Shahid <junaids@google.com> Fixes: b8e8c830 ("kvm: mmu: ITLB_MULTIHIT mitigation") Signed-off-by:
-
