KVM: SVM: Fix DEBUGCTL bugs
Sean Christopherson
maintainerv3 cover KVM: SVM: Fix DEBUGCTL bugs
Fix a long-lurking bug in SVM where KVM runs the guest with the host's DEBUGCTL if LBR virtualization is disabled. AMD CPUs rather stupidly context switch DEBUGCTL if and only if LBR virtualization is enabled (not just supported, but fully enabled).
The bug has gone unnoticed because until recently, the only bits that KVM would leave set were things like BTF, which are guest visible but won't cause functional problems unless guest software is being especially particular about #DBs.
The bug was exposed by the addition of BusLockTrap ("Detect" in the kernel), as the resulting #DBs due to split-lock accesses in guest userspace (lol Steam) get reflected into the guest by KVM.
Note, I don't love suppressing DEBUGCTL.BTF, but practically speaking that's likely the behavior that SVM guests have gotten the vast, vast majority of the time, and given that it's the behavior on Intel, it's (hopefully) a safe option for a fix, e.g. versus trying to add proper BTF virtualization on the fly.
v3:
- Suppress BTF, as KVM doesn't actually support it. [Ravi]
- Actually load the guest's DEBUGCTL (though amusingly, with BTF squashed, it's guaranteed to be '0' in this scenario). [Ravi]
v2:
- Load the guest's DEBUGCTL instead of simply zeroing it on VMRUN.
- Drop bits 5:3 from guest DEBUGCTL so that KVM doesn't let the guest unintentionally enable BusLockTrap (AMD repurposed bits). [Ravi]
- Collect a review. [Xiaoyao]
- Make bits 5:3 fully reserved, in a separate not-for-stable patch.
v1: https://lore.kernel.org/all/[email protected]
Sean Christopherson (6):
KVM: SVM: Drop DEBUGCTL[5:2] from guest's effective value
KVM: SVM: Suppress DEBUGCTL.BTF on AMD
KVM: x86: Snapshot the host's DEBUGCTL in common x86
KVM: SVM: Manually context switch DEBUGCTL if LBR virtualization is
disabled
KVM: x86: Snapshot the host's DEBUGCTL after disabling IRQs
KVM: SVM: Treat DEBUGCTL[5:2] as reserved
arch/x86/include/asm/kvm_host.h | 1 + arch/x86/kvm/svm/svm.c | 24 ++++++++++++++++++++++++ arch/x86/kvm/svm/svm.h | 2 +- arch/x86/kvm/vmx/vmx.c | 8 ++------ arch/x86/kvm/vmx/vmx.h | 2 -- arch/x86/kvm/x86.c | 2 ++ 6 files changed, 30 insertions(+), 9 deletions(-)
base-commit: fed48e2967f402f561d80075a20c5c9e16866e53
Signature · 1 line
2.48.1.711.g2feabab25a-goog
Sean Christopherson
maintainerPATCH v3 1/6 KVM: SVM: Drop DEBUGCTL[5:2] from guest's effective value
Drop bits 5:2 from the guest's effective DEBUGCTL value, as AMD changed the architectural behavior of the bits and broke backwards compatibility. On CPUs without BusLockTrap (or at least, in APMs from before ~2023), bits 5:2 controlled the behavior of external pins:
Performance-Monitoring/Breakpoint Pin-Control (PBi)—Bits 5:2, read/write. Software uses thesebits to control the type of information reported by the four external performance-monitoring/breakpoint pins on the processor. When a PBi bit is cleared to 0, the corresponding external pin (BPi) reports performance-monitor information. When a PBi bit is set to 1, the corresponding external pin (BPi) reports breakpoint information.
With the introduction of BusLockTrap, presumably to be compatible with Intel CPUs, AMD redefined bit 2 to be BLCKDB:
Bus Lock #DB Trap (BLCKDB)—Bit 2, read/write. Software sets this bit to enable generation of a #DB trap following successful execution of a bus lock when CPL is > 0.
and redefined bits 5:3 (and bit 6) as "6:3 Reserved MBZ".
Ideally, KVM would treat bits 5:2 as reserved. Defer that change to a feature cleanup to avoid breaking existing guest in LTS kernels. For now, drop the bits to retain backwards compatibility (of a sort).
Note, dropping bits 5:2 is still a guest-visible change, e.g. if the guest is enabling LBRs *and* the legacy PBi bits, then the state of the PBi bits is visible to the guest, whereas now the guest will always see '0'.
--- arch/x86/kvm/svm/svm.c | 12 ++++++++++++ arch/x86/kvm/svm/svm.h | 2 +- 2 files changed, 13 insertions(+), 1 deletion(-)
arch/x86/kvm/svm/svm.c +12 -0
@@ -3165,6 +3165,18 @@ static int svm_set_msr(struct kvm_vcpu *vcpu, struct msr_data *msr)kvm_pr_unimpl_wrmsr (vcpu, ecx, data);break ; } + +/* +* AMD changed the architectural behavior of bits 5:2. On CPUs +* without BusLockTrap, bits 5:2 control "external pins", but +* on CPUs that support BusLockDetect, bit 2 enables BusLockTrap +* and bits 5:3 are reserved-to-zero. Sadly, old KVM allowed +* the guest to set bits 5:2 despite not actually virtualizing +* Performance-Monitoring/Breakpoint external pins. Drop bits +* 5:2 for backwards compatibility. +*/ + data &= ~GENMASK (5 ,2 ); +if (data &DEBUGCTL_RESERVED_BITS )return 1 ;
arch/x86/kvm/svm/svm.h +1 -1
@@ -582,7 +582,7 @@ static inline bool is_vnmi_enabled(struct vcpu_svm *svm)/* svm.c */ #define MSR_INVALID 0xffffffffU -#define DEBUGCTL_RESERVED_BITS (~(0x3fULL)) +#define DEBUGCTL_RESERVED_BITS (~(DEBUGCTLMSR_BTF | DEBUGCTLMSR_LBR))extern bool dump_invalid_vmcb;
Signature · 1 line
2.48.1.711.g2feabab25a-goog
Sean Christopherson
maintainerPATCH v3 2/6 KVM: SVM: Suppress DEBUGCTL.BTF on AMD
In reply to Sean ChristophersonMark BTF as reserved in DEBUGCTL on AMD, as KVM doesn't actually support BTF, and fully enabling BTF virtualization is non-trivial due to interactions with the emulator, guest_debug, #DB interception, nested SVM, etc.
Don't inject #GP if the guest attempts to set BTF, as there's no way to communicate lack of support to the guest, and instead suppress the flag and treat the WRMSR as (partially) unsupported.
In short, make KVM behave the same on AMD and Intel (VMX already squashes BTF).
Note, due to other bugs in KVM's handling of DEBUGCTL, the only way BTF has "worked" in any capacity is if the guest simultaneously enables LBRs.
--- arch/x86/kvm/svm/svm.c | 9 +++++++++ arch/x86/kvm/svm/svm.h | 2 +- 2 files changed, 10 insertions(+), 1 deletion(-)
arch/x86/kvm/svm/svm.c +9 -0
@@ -3177,6 +3177,15 @@ static int svm_set_msr(struct kvm_vcpu *vcpu, struct msr_data *msr)*/ data &= ~GENMASK (5 ,2 ); +/* +* Suppress BTF as KVM doesn't virtualize BTF, but there's no +* way to communicate lack of support to the guest. +*/ +if (data &DEBUGCTLMSR_BTF ) { +kvm_pr_unimpl_wrmsr (vcpu,MSR_IA32_DEBUGCTLMSR , data); + data &= ~DEBUGCTLMSR_BTF ; + } +if (data &DEBUGCTL_RESERVED_BITS )return 1 ;
arch/x86/kvm/svm/svm.h +1 -1
@@ -582,7 +582,7 @@ static inline bool is_vnmi_enabled(struct vcpu_svm *svm)/* svm.c */ #define MSR_INVALID 0xffffffffU -#define DEBUGCTL_RESERVED_BITS (~(DEBUGCTLMSR_BTF | DEBUGCTLMSR_LBR)) +#define DEBUGCTL_RESERVED_BITS (~DEBUGCTLMSR_LBR)extern bool dump_invalid_vmcb;
Signature · 1 line
2.48.1.711.g2feabab25a-goog
Sean Christopherson
maintainerPATCH v3 3/6 KVM: x86: Snapshot the host's DEBUGCTL in common x86
In reply to Sean ChristophersonMove KVM's snapshot of DEBUGCTL to kvm_vcpu_arch and take the snapshot in common x86, so that SVM can also use the snapshot.
Opportunistically change the field to a u64. While bits 63:32 are reserved on AMD, not mentioned at all in Intel's SDM, and managed as an "unsigned long" by the kernel, DEBUGCTL is an MSR and therefore a 64-bit value.
--- arch/x86/include/asm/kvm_host.h | 1 + arch/x86/kvm/vmx/vmx.c | 8 ++------ arch/x86/kvm/vmx/vmx.h | 2 -- arch/x86/kvm/x86.c | 1 + 4 files changed, 4 insertions(+), 8 deletions(-)
arch/x86/include/asm/kvm_host.h +1 -0
@@ -781,6 +781,7 @@ struct kvm_vcpu_arch {u32 pkru;u32 hflags;u64 efer; +u64 host_debugctl;u64 apic_base;struct kvm_lapic *apic;/* kernel irqchip context */ bool load_eoi_exitmap_pending;
arch/x86/kvm/vmx/vmx.c +2 -6
@@ -1514,16 +1514,12 @@ void vmx_vcpu_load_vmcs(struct kvm_vcpu *vcpu, int cpu,*/ void vmx_vcpu_load (struct kvm_vcpu *vcpu,int cpu) { -struct vcpu_vmx *vmx =to_vmx (vcpu); -if (vcpu->scheduled_out && !kvm_pause_in_guest (vcpu->kvm))shrink_ple_window (vcpu);vmx_vcpu_load_vmcs (vcpu, cpu,NULL );vmx_vcpu_pi_load (vcpu, cpu); - - vmx->host_debugctlmsr =get_debugctlmsr (); }void vmx_vcpu_put (struct kvm_vcpu *vcpu)
@@ -7458,8 +7454,8 @@ fastpath_t vmx_vcpu_run(struct kvm_vcpu *vcpu, bool force_immediate_exit) }/* MSR_IA32_DEBUGCTLMSR is zeroed on vmexit. Restore it if needed */ -if (vmx->host_debugctlmsr) -update_debugctlmsr (vmx->host_debugctlmsr); +if (vcpu->arch.host_debugctl) +update_debugctlmsr (vcpu->arch.host_debugctl);#ifndef CONFIG_X86_64 /*
arch/x86/kvm/vmx/vmx.h +0 -2
@@ -340,8 +340,6 @@ struct vcpu_vmx {/* apic deadline value in host tsc */ u64 hv_deadline_tsc; -unsigned long host_debugctlmsr; -/* * Only bits masked by msr_ia32_feature_control_valid_bits can be set in * msr_ia32_feature_control. FEAT_CTL_LOCKED is always included
arch/x86/kvm/x86.c +1 -0
@@ -4991,6 +4991,7 @@ void kvm_arch_vcpu_load(struct kvm_vcpu *vcpu, int cpu)/* Save host pkru register if supported */ vcpu->arch.host_pkru =read_pkru (); + vcpu->arch.host_debugctl =get_debugctlmsr ();/* Apply any externally detected TSC adjustments (due to suspend) */ if (unlikely (vcpu->arch.tsc_offset_adjustment)) {
Signature · 1 line
2.48.1.711.g2feabab25a-goog
Sean Christopherson
maintainerPATCH v3 4/6 KVM: SVM: Manually context switch DEBUGCTL if LBR virtualization is disabled
In reply to Sean ChristophersonManually load the guest's DEBUGCTL prior to VMRUN (and restore the host's value on #VMEXIT) if it diverges from the host's value and LBR virtualization is disabled, as hardware only context switches DEBUGCTL if LBR virtualization is fully enabled. Running the guest with the host's value has likely been mildly problematic for quite some time, e.g. it will result in undesirable behavior if BTF diverges (with the caveat that KVM now suppresses guest BTF due to lack of support).
But the bug became fatal with the introduction of Bus Lock Trap ("Detect" in kernel paralance) support for AMD (commit 408eb7417a92 ("x86/bus_lock: Add support for AMD")), as a bus lock in the guest will trigger an unexpected #DB.
Note, suppressing the bus lock #DB, i.e. simply resuming the guest without injecting a #DB, is not an option. It wouldn't address the general issue with DEBUGCTL, e.g. for things like BTF, and there are other guest-visible side effects if BusLockTrap is left enabled.
If BusLockTrap is disabled, then DR6.BLD is reserved-to-1; any attempts to clear it by software are ignored. But if BusLockTrap is enabled, software can clear DR6.BLD:
Software enables bus lock trap by setting DebugCtl MSR[BLCKDB] (bit 2) to 1. When bus lock trap is enabled, ... The processor indicates that this #DB was caused by a bus lock by clearing DR6[BLD] (bit 11). DR6[11] previously had been defined to be always 1.
and clearing DR6.BLD is "sticky" in that it's not set (i.e. lowered) by other #DBs:
All other #DB exceptions leave DR6[BLD] unmodified
E.g. leaving BusLockTrap enable can confuse a legacy guest that writes '0' to reset DR6.
--- arch/x86/kvm/svm/svm.c | 14 ++++++++++++++ 1 file changed, 14 insertions(+)
arch/x86/kvm/svm/svm.c +14 -0
@@ -4274,6 +4274,16 @@ static __no_kcsan fastpath_t svm_vcpu_run(struct kvm_vcpu *vcpu,clgi ();kvm_load_guest_xsave_state (vcpu); +/* +* Hardware only context switches DEBUGCTL if LBR virtualization is +* enabled. Manually load DEBUGCTL if necessary (and restore it after +* VM-Exit), as running with the host's DEBUGCTL can negatively affect +* guest state and can even be fatal, e.g. due to Bus Lock Detect. +*/ +if (!(svm->vmcb->control.virt_ext &LBR_CTL_ENABLE_MASK ) && + vcpu->arch.host_debugctl != svm->vmcb->save.dbgctl) +update_debugctlmsr (svm->vmcb->save.dbgctl); +kvm_wait_lapic_expire (vcpu);/*
@@ -4301,6 +4311,10 @@ static __no_kcsan fastpath_t svm_vcpu_run(struct kvm_vcpu *vcpu,if (unlikely (svm->vmcb->control.exit_code ==SVM_EXIT_NMI ))kvm_before_interrupt (vcpu,KVM_HANDLING_NMI ); +if (!(svm->vmcb->control.virt_ext &LBR_CTL_ENABLE_MASK ) && + vcpu->arch.host_debugctl != svm->vmcb->save.dbgctl) +update_debugctlmsr (vcpu->arch.host_debugctl); +kvm_load_host_xsave_state (vcpu);stgi ();
Signature · 1 line
2.48.1.711.g2feabab25a-goog
Sean Christopherson
maintainerPATCH v3 5/6 KVM: x86: Snapshot the host's DEBUGCTL after disabling IRQs
In reply to Sean ChristophersonSnapshot the host's DEBUGCTL after disabling IRQs, as perf can toggle debugctl bits from IRQ context, e.g. when enabling/disabling events via smp_call_function_single(). Taking the snapshot (long) before IRQs are disabled could result in KVM effectively clobbering DEBUGCTL due to using a stale snapshot.
--- arch/x86/kvm/x86.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-)
arch/x86/kvm/x86.c +2 -1
@@ -4991,7 +4991,6 @@ void kvm_arch_vcpu_load(struct kvm_vcpu *vcpu, int cpu)/* Save host pkru register if supported */ vcpu->arch.host_pkru =read_pkru (); - vcpu->arch.host_debugctl =get_debugctlmsr ();/* Apply any externally detected TSC adjustments (due to suspend) */ if (unlikely (vcpu->arch.tsc_offset_adjustment)) {
@@ -10984,6 +10983,8 @@ static int vcpu_enter_guest(struct kvm_vcpu *vcpu)set_debugreg (0 ,7 ); } + vcpu->arch.host_debugctl =get_debugctlmsr (); +guest_timing_enter_irqoff ();for (;;) {
Signature · 1 line
2.48.1.711.g2feabab25a-goog
Sean Christopherson
maintainerPATCH v3 6/6 KVM: SVM: Treat DEBUGCTL[5:2] as reserved
In reply to Sean ChristophersonStop ignoring DEBUGCTL[5:2] on AMD CPUs and instead treat them as reserved. KVM has never properly virtualized AMD's legacy PBi bits, but did allow the guest (and host userspace) to set the bits. To avoid breaking guests when running on CPUs with BusLockTrap, which redefined bit 2 to BLCKDB and made bits 5:3 reserved, a previous KVM change ignored bits 5:3, e.g. so that legacy guest software wouldn't inadvertently enable BusLockTrap or hit a VMRUN failure due to setting reserved.
To allow for virtualizing BusLockTrap and whatever future features may use bits 5:3, treat bits 5:2 as reserved (and hope that doing so doesn't break any existing guests).
--- arch/x86/kvm/svm/svm.c | 11 ----------- 1 file changed, 11 deletions(-)
arch/x86/kvm/svm/svm.c +0 -11
@@ -3166,17 +3166,6 @@ static int svm_set_msr(struct kvm_vcpu *vcpu, struct msr_data *msr)break ; } -/* -* AMD changed the architectural behavior of bits 5:2. On CPUs -* without BusLockTrap, bits 5:2 control "external pins", but -* on CPUs that support BusLockDetect, bit 2 enables BusLockTrap -* and bits 5:3 are reserved-to-zero. Sadly, old KVM allowed -* the guest to set bits 5:2 despite not actually virtualizing -* Performance-Monitoring/Breakpoint external pins. Drop bits -* 5:2 for backwards compatibility. -*/ - data &= ~GENMASK (5 ,2 ); -/* * Suppress BTF as KVM doesn't virtualize BTF, but there's no * way to communicate lack of support to the guest.
Signature · 1 line
2.48.1.711.g2feabab25a-goog
Quoted 25 lines · Sean Christopherson
Fix a long-lurking bug in SVM where KVM runs the guest with the host's DEBUGCTL if LBR virtualization is disabled. AMD CPUs rather stupidly context switch DEBUGCTL if and only if LBR virtualization is enabled (not just supported, but fully enabled).
The bug has gone unnoticed because until recently, the only bits that KVM would leave set were things like BTF, which are guest visible but won't cause functional problems unless guest software is being especially particular about #DBs.
The bug was exposed by the addition of BusLockTrap ("Detect" in the kernel), as the resulting #DBs due to split-lock accesses in guest userspace (lol Steam) get reflected into the guest by KVM.
Note, I don't love suppressing DEBUGCTL.BTF, but practically speaking that's likely the behavior that SVM guests have gotten the vast, vast majority of the time, and given that it's the behavior on Intel, it's (hopefully) a safe option for a fix, e.g. versus trying to add proper BTF virtualization on the fly.
v3:
- Suppress BTF, as KVM doesn't actually support it. [Ravi]
- Actually load the guest's DEBUGCTL (though amusingly, with BTF squashed, it's guaranteed to be '0' in this scenario). [Ravi]
Quoted 8 lines
v2:
- Load the guest's DEBUGCTL instead of simply zeroing it on VMRUN.
- Drop bits 5:3 from guest DEBUGCTL so that KVM doesn't let the guest unintentionally enable BusLockTrap (AMD repurposed bits). [Ravi]
- Collect a review. [Xiaoyao]
- Make bits 5:3 fully reserved, in a separate not-for-stable patch.
For the series,
Thanks,
Ravi
Sean Christopherson
maintainerQuoted 33 lines · Ravi Bangoria
Quoted 32 lines · Sean Christopherson
Fix a long-lurking bug in SVM where KVM runs the guest with the host's DEBUGCTL if LBR virtualization is disabled. AMD CPUs rather stupidly context switch DEBUGCTL if and only if LBR virtualization is enabled (not just supported, but fully enabled).
The bug has gone unnoticed because until recently, the only bits that KVM would leave set were things like BTF, which are guest visible but won't cause functional problems unless guest software is being especially particular about #DBs.
The bug was exposed by the addition of BusLockTrap ("Detect" in the kernel), as the resulting #DBs due to split-lock accesses in guest userspace (lol Steam) get reflected into the guest by KVM.
Note, I don't love suppressing DEBUGCTL.BTF, but practically speaking that's likely the behavior that SVM guests have gotten the vast, vast majority of the time, and given that it's the behavior on Intel, it's (hopefully) a safe option for a fix, e.g. versus trying to add proper BTF virtualization on the fly.
v3:
- Suppress BTF, as KVM doesn't actually support it. [Ravi]
- Actually load the guest's DEBUGCTL (though amusingly, with BTF squashed, it's guaranteed to be '0' in this scenario). [Ravi]v2:
- Load the guest's DEBUGCTL instead of simply zeroing it on VMRUN.
- Drop bits 5:3 from guest DEBUGCTL so that KVM doesn't let the guest unintentionally enable BusLockTrap (AMD repurposed bits). [Ravi]
- Collect a review. [Xiaoyao]
- Make bits 5:3 fully reserved, in a separate not-for-stable patch.
Quoted 5 lines
Quoted 1 line
For the series,
Reviewed-and-tested-by Ravi Bangoria <[email protected]>
Thank you for all your help, much appreciated!
Sean Christopherson
maintainerQuoted 11 lines · Sean Christopherson
Fix a long-lurking bug in SVM where KVM runs the guest with the host's DEBUGCTL if LBR virtualization is disabled. AMD CPUs rather stupidly context switch DEBUGCTL if and only if LBR virtualization is enabled (not just supported, but fully enabled).
The bug has gone unnoticed because until recently, the only bits that KVM would leave set were things like BTF, which are guest visible but won't cause functional problems unless guest software is being especially particular about #DBs.
[...]
Applied 1-5 to kvm-x86 fixes (for 6.14). I'm going to hold off on making DEBUGCTL[5:2] reserved until at least 6.15.
[1/6] KVM: SVM: Drop DEBUGCTL[5:2] from guest's effective value
https://github.com/kvm-x86/linux/commit/ee89e8013383 [2/6] KVM: SVM: Suppress DEBUGCTL.BTF on AMD https://github.com/kvm-x86/linux/commit/d0eac42f5cec [3/6] KVM: x86: Snapshot the host's DEBUGCTL in common x86 https://github.com/kvm-x86/linux/commit/fb71c7959356 [4/6] KVM: SVM: Manually context switch DEBUGCTL if LBR virtualization is disabled https://github.com/kvm-x86/linux/commit/433265870ab3 [5/6] KVM: x86: Snapshot the host's DEBUGCTL after disabling IRQs https://github.com/kvm-x86/linux/commit/189ecdb3e112
[6/6] KVM: SVM: Treat DEBUGCTL[5:2] as reserved
(no commit info)
Quoted 25 lines · Sean Christopherson
Fix a long-lurking bug in SVM where KVM runs the guest with the host's DEBUGCTL if LBR virtualization is disabled. AMD CPUs rather stupidly context switch DEBUGCTL if and only if LBR virtualization is enabled (not just supported, but fully enabled).
The bug has gone unnoticed because until recently, the only bits that KVM would leave set were things like BTF, which are guest visible but won't cause functional problems unless guest software is being especially particular about #DBs.
The bug was exposed by the addition of BusLockTrap ("Detect" in the kernel), as the resulting #DBs due to split-lock accesses in guest userspace (lol Steam) get reflected into the guest by KVM.
Note, I don't love suppressing DEBUGCTL.BTF, but practically speaking that's likely the behavior that SVM guests have gotten the vast, vast majority of the time, and given that it's the behavior on Intel, it's (hopefully) a safe option for a fix, e.g. versus trying to add proper BTF virtualization on the fly.
v3:
- Suppress BTF, as KVM doesn't actually support it. [Ravi]
- Actually load the guest's DEBUGCTL (though amusingly, with BTF squashed, it's guaranteed to be '0' in this scenario). [Ravi]
Quoted 9 lines
v2:
- Load the guest's DEBUGCTL instead of simply zeroing it on VMRUN.
- Drop bits 5:3 from guest DEBUGCTL so that KVM doesn't let the guest unintentionally enable BusLockTrap (AMD repurposed bits). [Ravi]
- Collect a review. [Xiaoyao]
- Make bits 5:3 fully reserved, in a separate not-for-stable patch.
Hi,
Amusingly there is another DEBUGCTL issue, which I just got to the bottom of. (if I am not mistaken of course).
We currently don't let the guest set DEBUGCTL.FREEZE_WHILE_SMM and neither set it ourselves in GUEST_IA32_DEBUGCTL vmcs field, even when supported by the host (If I read the code correctly, I didn't verify this in runtime)
This means that the host #SMIs will interfere with the guest PMU.
In particular this causes the 'pmu' kvm-unit-test to fail, which is something that our CI caught.
I think that kvm should just set this bit, or even better, use the host value of this bit, and hide it from the guest, because the guest shouldn't know about host's smm, and we AFAIK don't really support freezing perfmon when the guest enters its own emulated SMM.
What do you think? I'll post patches if you think that this is a good idea. (A temp hack to set this bit always in GUEST_IA32_DEBUGCTL fixed the problem for me)
I also need to check if AMD also has this feature, or if this is Intel specific.
Best regards,
Maxim Levitsky
Quoted 20 lines
Sean Christopherson (6):
KVM: SVM: Drop DEBUGCTL[5:2] from guest's effective value
KVM: SVM: Suppress DEBUGCTL.BTF on AMD
KVM: x86: Snapshot the host's DEBUGCTL in common x86
KVM: SVM: Manually context switch DEBUGCTL if LBR virtualization is
disabled
KVM: x86: Snapshot the host's DEBUGCTL after disabling IRQs
KVM: SVM: Treat DEBUGCTL[5:2] as reservedarch/x86/include/asm/kvm_host.h | 1 + arch/x86/kvm/svm/svm.c | 24 ++++++++++++++++++++++++ arch/x86/kvm/svm/svm.h | 2 +- arch/x86/kvm/vmx/vmx.c | 8 ++------ arch/x86/kvm/vmx/vmx.h | 2 -- arch/x86/kvm/x86.c | 2 ++ 6 files changed, 30 insertions(+), 9 deletions(-)base-commit:
fed48e2967f402f561d80075a20c5c9e16866e53
Quoted 48 lines · Maxim Levitsky
Quoted 25 lines · Sean Christopherson
Fix a long-lurking bug in SVM where KVM runs the guest with the host's DEBUGCTL if LBR virtualization is disabled. AMD CPUs rather stupidly context switch DEBUGCTL if and only if LBR virtualization is enabled (not just supported, but fully enabled).
The bug has gone unnoticed because until recently, the only bits that KVM would leave set were things like BTF, which are guest visible but won't cause functional problems unless guest software is being especially particular about #DBs.
The bug was exposed by the addition of BusLockTrap ("Detect" in the kernel), as the resulting #DBs due to split-lock accesses in guest userspace (lol Steam) get reflected into the guest by KVM.
Note, I don't love suppressing DEBUGCTL.BTF, but practically speaking that's likely the behavior that SVM guests have gotten the vast, vast majority of the time, and given that it's the behavior on Intel, it's (hopefully) a safe option for a fix, e.g. versus trying to add proper BTF virtualization on the fly.
v3:
- Suppress BTF, as KVM doesn't actually support it. [Ravi]
- Actually load the guest's DEBUGCTL (though amusingly, with BTF squashed, it's guaranteed to be '0' in this scenario). [Ravi]Quoted 9 lines
v2:
- Load the guest's DEBUGCTL instead of simply zeroing it on VMRUN.
- Drop bits 5:3 from guest DEBUGCTL so that KVM doesn't let the guest unintentionally enable BusLockTrap (AMD repurposed bits). [Ravi]
- Collect a review. [Xiaoyao]
- Make bits 5:3 fully reserved, in a separate not-for-stable patch.Hi,
Amusingly there is another DEBUGCTL issue, which I just got to the bottom of. (if I am not mistaken of course).
We currently don't let the guest set DEBUGCTL.FREEZE_WHILE_SMM and neither set it ourselves in
GUEST_IA32_DEBUGCTLvmcs field, even when supported by the host (If I read the code correctly, I didn't verify this in runtime)This means that the host #SMIs will interfere with the guest PMU.
In particular this causes the 'pmu' kvm-unit-test to fail, which is something that our CI caught.
Quoted 8 lines
I think that kvm should just set this bit, or even better, use the host value of this bit, and hide it from the guest, because the guest shouldn't know about host's smm, and we AFAIK don't really support freezing perfmon when the guest enters its own emulated SMM.
What do you think? I'll post patches if you think that this is a good idea. (A temp hack to set this bit always in
GUEST_IA32_DEBUGCTLfixed the problem for me)I also need to check if AMD also has this feature, or if this is Intel specific.
Any update?
Best regards,
Maxim Levitsky
Quoted 23 lines
Best regards,
Maxim LevitskyQuoted 19 lines
Sean Christopherson (6):
KVM: SVM: Drop DEBUGCTL[5:2] from guest's effective value
KVM: SVM: Suppress DEBUGCTL.BTF on AMD
KVM: x86: Snapshot the host's DEBUGCTL in common x86
KVM: SVM: Manually context switch DEBUGCTL if LBR virtualization is
disabled
KVM: x86: Snapshot the host's DEBUGCTL after disabling IRQs
KVM: SVM: Treat DEBUGCTL[5:2] as reservedarch/x86/include/asm/kvm_host.h | 1 + arch/x86/kvm/svm/svm.c | 24 ++++++++++++++++++++++++ arch/x86/kvm/svm/svm.h | 2 +- arch/x86/kvm/vmx/vmx.c | 8 ++------ arch/x86/kvm/vmx/vmx.h | 2 -- arch/x86/kvm/x86.c | 2 ++ 6 files changed, 30 insertions(+), 9 deletions(-)base-commit:
fed48e2967f402f561d80075a20c5c9e16866e53
Sean Christopherson
maintainerQuoted 37 lines · Maxim Levitsky
Quoted 25 lines · Sean Christopherson
Fix a long-lurking bug in SVM where KVM runs the guest with the host's DEBUGCTL if LBR virtualization is disabled. AMD CPUs rather stupidly context switch DEBUGCTL if and only if LBR virtualization is enabled (not just supported, but fully enabled).
The bug has gone unnoticed because until recently, the only bits that KVM would leave set were things like BTF, which are guest visible but won't cause functional problems unless guest software is being especially particular about #DBs.
The bug was exposed by the addition of BusLockTrap ("Detect" in the kernel), as the resulting #DBs due to split-lock accesses in guest userspace (lol Steam) get reflected into the guest by KVM.
Note, I don't love suppressing DEBUGCTL.BTF, but practically speaking that's likely the behavior that SVM guests have gotten the vast, vast majority of the time, and given that it's the behavior on Intel, it's (hopefully) a safe option for a fix, e.g. versus trying to add proper BTF virtualization on the fly.
v3:
- Suppress BTF, as KVM doesn't actually support it. [Ravi]
- Actually load the guest's DEBUGCTL (though amusingly, with BTF squashed, it's guaranteed to be '0' in this scenario). [Ravi]Quoted 9 lines
v2:
- Load the guest's DEBUGCTL instead of simply zeroing it on VMRUN.
- Drop bits 5:3 from guest DEBUGCTL so that KVM doesn't let the guest unintentionally enable BusLockTrap (AMD repurposed bits). [Ravi]
- Collect a review. [Xiaoyao]
- Make bits 5:3 fully reserved, in a separate not-for-stable patch.
Quoted 8 lines
Hi,
Amusingly there is another DEBUGCTL issue, which I just got to the bottom of. (if I am not mistaken of course).
We currently don't let the guest set DEBUGCTL.FREEZE_WHILE_SMM and neither set it ourselves in
GUEST_IA32_DEBUGCTLvmcs field, even when supported by the host (If I read the code correctly, I didn't verify this in runtime)
Ugh, SMM. Yeah, KVM doesn't propagate DEBUGCTLMSR_FREEZE_IN_SMM to the guest value. KVM intercepts reads and writes to DEBUGCTL, so it should be easy enough to shove the bit in on writes, and drop it on reads.
Quoted 8 lines
This means that the host #SMIs will interfere with the guest PMU. In particular this causes the 'pmu' kvm-unit-test to fail, which is something that our CI caught.
I think that kvm should just set this bit, or even better, use the host value of this bit, and hide it from the guest, because the guest shouldn't know about host's smm, and we AFAIK don't really support freezing perfmon when the guest enters its own emulated SMM.
Agreed. Easy thing is to use the host's value, so that KVM doesn't need to check for its existence. I can't think of anything that would go sideways by freezing perfmon if the host happens to take an SMI.
Quoted 4 lines
What do you think? I'll post patches if you think that this is a good idea. (A temp hack to set this bit always in
GUEST_IA32_DEBUGCTLfixed the problem for me)I also need to check if AMD also has this feature, or if this is Intel specific.
Intel only. I assume/think/hope AMD's Host/Guest Only field in the event selector effectively hides SMM from the guest.
Quoted 63 lines · Sean Christopherson
Quoted 36 lines · Maxim Levitsky
Quoted 25 lines · Sean Christopherson
Fix a long-lurking bug in SVM where KVM runs the guest with the host's DEBUGCTL if LBR virtualization is disabled. AMD CPUs rather stupidly context switch DEBUGCTL if and only if LBR virtualization is enabled (not just supported, but fully enabled).
The bug has gone unnoticed because until recently, the only bits that KVM would leave set were things like BTF, which are guest visible but won't cause functional problems unless guest software is being especially particular about #DBs.
The bug was exposed by the addition of BusLockTrap ("Detect" in the kernel), as the resulting #DBs due to split-lock accesses in guest userspace (lol Steam) get reflected into the guest by KVM.
Note, I don't love suppressing DEBUGCTL.BTF, but practically speaking that's likely the behavior that SVM guests have gotten the vast, vast majority of the time, and given that it's the behavior on Intel, it's (hopefully) a safe option for a fix, e.g. versus trying to add proper BTF virtualization on the fly.
v3:
- Suppress BTF, as KVM doesn't actually support it. [Ravi]
- Actually load the guest's DEBUGCTL (though amusingly, with BTF squashed, it's guaranteed to be '0' in this scenario). [Ravi]Quoted 9 lines
v2:
- Load the guest's DEBUGCTL instead of simply zeroing it on VMRUN.
- Drop bits 5:3 from guest DEBUGCTL so that KVM doesn't let the guest unintentionally enable BusLockTrap (AMD repurposed bits). [Ravi]
- Collect a review. [Xiaoyao]
- Make bits 5:3 fully reserved, in a separate not-for-stable patch.Quoted 8 lines
Hi,
Amusingly there is another DEBUGCTL issue, which I just got to the bottom of. (if I am not mistaken of course).
We currently don't let the guest set DEBUGCTL.FREEZE_WHILE_SMM and neither set it ourselves in
GUEST_IA32_DEBUGCTLvmcs field, even when supported by the host (If I read the code correctly, I didn't verify this in runtime)Ugh, SMM. Yeah, KVM doesn't propagate
DEBUGCTLMSR_FREEZE_IN_SMMto the guest value. KVM intercepts reads and writes to DEBUGCTL, so it should be easy enough to shove the bit in on writes, and drop it on reads.Quoted 8 lines
This means that the host #SMIs will interfere with the guest PMU. In particular this causes the 'pmu' kvm-unit-test to fail, which is something that our CI caught.
I think that kvm should just set this bit, or even better, use the host value of this bit, and hide it from the guest, because the guest shouldn't know about host's smm, and we AFAIK don't really support freezing perfmon when the guest enters its own emulated SMM.
Agreed. Easy thing is to use the host's value, so that KVM doesn't need to check for its existence. I can't think of anything that would go sideways by freezing perfmon if the host happens to take an SMI.
Quoted 8 lines
Quoted 4 lines
What do you think? I'll post patches if you think that this is a good idea. (A temp hack to set this bit always in
GUEST_IA32_DEBUGCTLfixed the problem for me)I also need to check if AMD also has this feature, or if this is Intel specific.
Intel only. I assume/think/hope AMD's Host/Guest Only field in the event selector effectively hides SMM from the guest.
Hi,
I will post a patch soon then. I just got my hands on the CI machine where the test failed and yes, the machine receives about 8 #SMIs per second on each core. Oh well...
BTW pmu_counters_test selftest is also affected since it counts # of retired instructions. With #SMI getting in the way, the number of course soars.
It doesn't fail often at this rate but it does when the test test is done for sufficient number or times or you just get lucky.
Best regards,
Maxim Levitsky
Quoted 64 lines · Sean Christopherson
Quoted 37 lines · Maxim Levitsky
Quoted 25 lines · Sean Christopherson
Fix a long-lurking bug in SVM where KVM runs the guest with the host's DEBUGCTL if LBR virtualization is disabled. AMD CPUs rather stupidly context switch DEBUGCTL if and only if LBR virtualization is enabled (not just supported, but fully enabled).
The bug has gone unnoticed because until recently, the only bits that KVM would leave set were things like BTF, which are guest visible but won't cause functional problems unless guest software is being especially particular about #DBs.
The bug was exposed by the addition of BusLockTrap ("Detect" in the kernel), as the resulting #DBs due to split-lock accesses in guest userspace (lol Steam) get reflected into the guest by KVM.
Note, I don't love suppressing DEBUGCTL.BTF, but practically speaking that's likely the behavior that SVM guests have gotten the vast, vast majority of the time, and given that it's the behavior on Intel, it's (hopefully) a safe option for a fix, e.g. versus trying to add proper BTF virtualization on the fly.
v3:
- Suppress BTF, as KVM doesn't actually support it. [Ravi]
- Actually load the guest's DEBUGCTL (though amusingly, with BTF squashed, it's guaranteed to be '0' in this scenario). [Ravi]Quoted 9 lines
v2:
- Load the guest's DEBUGCTL instead of simply zeroing it on VMRUN.
- Drop bits 5:3 from guest DEBUGCTL so that KVM doesn't let the guest unintentionally enable BusLockTrap (AMD repurposed bits). [Ravi]
- Collect a review. [Xiaoyao]
- Make bits 5:3 fully reserved, in a separate not-for-stable patch.Quoted 8 lines
Hi,
Amusingly there is another DEBUGCTL issue, which I just got to the bottom of. (if I am not mistaken of course).
We currently don't let the guest set DEBUGCTL.FREEZE_WHILE_SMM and neither set it ourselves in
GUEST_IA32_DEBUGCTLvmcs field, even when supported by the host (If I read the code correctly, I didn't verify this in runtime)Ugh, SMM. Yeah, KVM doesn't propagate
DEBUGCTLMSR_FREEZE_IN_SMMto the guest value. KVM intercepts reads and writes to DEBUGCTL, so it should be easy enough to shove the bit in on writes, and drop it on reads.Quoted 8 lines
This means that the host #SMIs will interfere with the guest PMU. In particular this causes the 'pmu' kvm-unit-test to fail, which is something that our CI caught.
I think that kvm should just set this bit, or even better, use the host value of this bit, and hide it from the guest, because the guest shouldn't know about host's smm, and we AFAIK don't really support freezing perfmon when the guest enters its own emulated SMM.
Agreed. Easy thing is to use the host's value, so that KVM doesn't need to check for its existence. I can't think of anything that would go sideways by freezing perfmon if the host happens to take an SMI.
Quoted 7 lines
Quoted 4 lines
What do you think? I'll post patches if you think that this is a good idea. (A temp hack to set this bit always in
GUEST_IA32_DEBUGCTLfixed the problem for me)I also need to check if AMD also has this feature, or if this is Intel specific.
Intel only. I assume/think/hope AMD's Host/Guest Only field in the event selector effectively hides SMM from the guest.
Just using the GuestOnly bit does not hide SMM activity from guests. SMIs are generally intercepted (kvm_amd.intercept_smi defaults to true) and handled in the host context. So guest PMCs are isolated by a combination of having the GuestOnly bit set and the #VMEXITs resulting from SMI interception.
Quoted 67 lines · Sandipan Das
Quoted 66 lines · Sean Christopherson
Quoted 36 lines · Maxim Levitsky
Quoted 25 lines · Sean Christopherson
Fix a long-lurking bug in SVM where KVM runs the guest with the host's DEBUGCTL if LBR virtualization is disabled. AMD CPUs rather stupidly context switch DEBUGCTL if and only if LBR virtualization is enabled (not just supported, but fully enabled).
The bug has gone unnoticed because until recently, the only bits that KVM would leave set were things like BTF, which are guest visible but won't cause functional problems unless guest software is being especially particular about #DBs.
The bug was exposed by the addition of BusLockTrap ("Detect" in the kernel), as the resulting #DBs due to split-lock accesses in guest userspace (lol Steam) get reflected into the guest by KVM.
Note, I don't love suppressing DEBUGCTL.BTF, but practically speaking that's likely the behavior that SVM guests have gotten the vast, vast majority of the time, and given that it's the behavior on Intel, it's (hopefully) a safe option for a fix, e.g. versus trying to add proper BTF virtualization on the fly.
v3:
- Suppress BTF, as KVM doesn't actually support it. [Ravi]
- Actually load the guest's DEBUGCTL (though amusingly, with BTF squashed, it's guaranteed to be '0' in this scenario). [Ravi]Quoted 9 lines
v2:
- Load the guest's DEBUGCTL instead of simply zeroing it on VMRUN.
- Drop bits 5:3 from guest DEBUGCTL so that KVM doesn't let the guest unintentionally enable BusLockTrap (AMD repurposed bits). [Ravi]
- Collect a review. [Xiaoyao]
- Make bits 5:3 fully reserved, in a separate not-for-stable patch.Quoted 8 lines
Hi,
Amusingly there is another DEBUGCTL issue, which I just got to the bottom of. (if I am not mistaken of course).
We currently don't let the guest set DEBUGCTL.FREEZE_WHILE_SMM and neither set it ourselves in
GUEST_IA32_DEBUGCTLvmcs field, even when supported by the host (If I read the code correctly, I didn't verify this in runtime)Ugh, SMM. Yeah, KVM doesn't propagate
DEBUGCTLMSR_FREEZE_IN_SMMto the guest value. KVM intercepts reads and writes to DEBUGCTL, so it should be easy enough to shove the bit in on writes, and drop it on reads.Quoted 8 lines
This means that the host #SMIs will interfere with the guest PMU. In particular this causes the 'pmu' kvm-unit-test to fail, which is something that our CI caught.
I think that kvm should just set this bit, or even better, use the host value of this bit, and hide it from the guest, because the guest shouldn't know about host's smm, and we AFAIK don't really support freezing perfmon when the guest enters its own emulated SMM.
Agreed. Easy thing is to use the host's value, so that KVM doesn't need to check for its existence. I can't think of anything that would go sideways by freezing perfmon if the host happens to take an SMI.
Quoted 3 lines
What do you think? I'll post patches if you think that this is a good idea. (A temp hack to set this bit always in
GUEST_IA32_DEBUGCTLfixed the problem for me)
Quoted 7 lines
Quoted 4 lines
Quoted 1 line
I also need to check if AMD also has this feature, or if this is Intel specific.
Intel only. I assume/think/hope AMD's Host/Guest Only field in the event selector effectively hides SMM from the guest.
Just using the GuestOnly bit does not hide SMM activity from guests. SMIs are generally intercepted (kvm_amd.intercept_smi defaults to true)
Hi,
Actually this setting doesn't really work these days, at lesat not on my Zen2 machine (3070x).
Long ago I tested it, and despite loading the system with SMIs either via APIC or via 0xB2 ioport write, where in both cases I noticed significant slowdown of a VM, pinned on the receiving CPU I got no SMI VM exits.
BIOS likely has an option to override this setting.
I guess the reason is security, because with SVM,
one can effectively block the SMIs from being processed on the host.
Best regards,
Maxim Levitsky
Quoted 4 lines
and handled in the
host context. So guest PMCs are isolated by a combination of having the GuestOnly bit set and the #VMEXITs resulting from SMI interception.
Quoted 11 lines · Sean Christopherson
Fix a long-lurking bug in SVM where KVM runs the guest with the host's DEBUGCTL if LBR virtualization is disabled. AMD CPUs rather stupidly context switch DEBUGCTL if and only if LBR virtualization is enabled (not just supported, but fully enabled).
The bug has gone unnoticed because until recently, the only bits that KVM would leave set were things like BTF, which are guest visible but won't cause functional problems unless guest software is being especially particular about #DBs.
[...]
Applied patch 6 to kvm-x86 svm (1-5 already went into 6.15).
[6/6] KVM: SVM: Treat DEBUGCTL[5:2] as reserved
https://github.com/kvm-x86/linux/commit/5ecdb48dd918