* [PATCH v5 1/2] KVM: VMX: Bury all of the VMX preemption timer code under CONFIG_X86_64=y
2026-07-24 23:49 [PATCH v5 0/2] Compute safe bound for VMX preemption timer Jim Mattson
@ 2026-07-24 23:49 ` Jim Mattson
2026-07-28 5:16 ` Chao Gao
2026-07-28 7:13 ` Binbin Wu
2026-07-24 23:49 ` [PATCH v5 2/2] KVM: VMX: Cap VMX preemption timer to work around Intel erratum Jim Mattson
1 sibling, 2 replies; 6+ messages in thread
From: Jim Mattson @ 2026-07-24 23:49 UTC (permalink / raw)
To: seanjc, pbonzini; +Cc: chao.gao, kvm
From: Sean Christopherson <seanjc@google.com>
Double down on using the VMX preemption timer only for 64-bit kernels, and
bury the setup and runtime adjustment code, and all global variables, under
CONFIG_X86_64=y. This will allow addressing various Intel erratum without
running afoul of unused-but-set-variable and __udivdi3() warnings/errors on
32-bit kernels.
No functional change intended.
Signed-off-by: Sean Christopherson <seanjc@google.com>
---
arch/x86/kvm/vmx/vmx.c | 122 +++++++++++++++++++++++------------------
1 file changed, 69 insertions(+), 53 deletions(-)
diff --git a/arch/x86/kvm/vmx/vmx.c b/arch/x86/kvm/vmx/vmx.c
index e4b9ac7fed9f..a07faa066ef0 100644
--- a/arch/x86/kvm/vmx/vmx.c
+++ b/arch/x86/kvm/vmx/vmx.c
@@ -150,10 +150,12 @@ module_param(dump_invalid_vmcs, bool, 0644);
#define KVM_VMX_TSC_MULTIPLIER_MAX 0xffffffffffffffffULL
/* Guest_tsc -> host_tsc conversion requires 64-bit division. */
+#ifdef CONFIG_X86_64
static int __read_mostly cpu_preemption_timer_multi;
static bool __read_mostly enable_preemption_timer = 1;
-#ifdef CONFIG_X86_64
module_param_named(preemption_timer, enable_preemption_timer, bool, S_IRUGO);
+#else
+#define enable_preemption_timer false
#endif
extern bool __read_mostly allow_smaller_maxphyaddr;
@@ -7408,32 +7410,6 @@ static void vmx_refresh_guest_perf_global_control(struct kvm_vcpu *vcpu)
pmu->global_ctrl = vmcs_read64(GUEST_IA32_PERF_GLOBAL_CTRL);
}
-static void vmx_update_hv_timer(struct kvm_vcpu *vcpu, bool force_immediate_exit)
-{
- struct vcpu_vmx *vmx = to_vmx(vcpu);
- u64 tscl;
- u32 delta_tsc;
-
- if (force_immediate_exit) {
- vmcs_write32(VMX_PREEMPTION_TIMER_VALUE, 0);
- vmx->loaded_vmcs->hv_timer_soft_disabled = false;
- } else if (vmx->hv_deadline_tsc != -1) {
- tscl = rdtsc();
- if (vmx->hv_deadline_tsc > tscl)
- /* set_hv_timer ensures the delta fits in 32-bits */
- delta_tsc = (u32)((vmx->hv_deadline_tsc - tscl) >>
- cpu_preemption_timer_multi);
- else
- delta_tsc = 0;
-
- vmcs_write32(VMX_PREEMPTION_TIMER_VALUE, delta_tsc);
- vmx->loaded_vmcs->hv_timer_soft_disabled = false;
- } else if (!vmx->loaded_vmcs->hv_timer_soft_disabled) {
- vmcs_write32(VMX_PREEMPTION_TIMER_VALUE, -1);
- vmx->loaded_vmcs->hv_timer_soft_disabled = true;
- }
-}
-
void noinstr vmx_update_host_rsp(struct vcpu_vmx *vmx, unsigned long host_rsp)
{
if (unlikely(host_rsp != vmx->loaded_vmcs->host_state.rsp)) {
@@ -7519,6 +7495,8 @@ static noinstr void vmx_vcpu_enter_exit(struct kvm_vcpu *vcpu,
guest_state_exit_irqoff();
}
+static void vmx_update_hv_timer(struct kvm_vcpu *vcpu, bool force_immediate_exit);
+
fastpath_t vmx_vcpu_run(struct kvm_vcpu *vcpu, u64 run_flags)
{
bool force_immediate_exit = run_flags & KVM_RUN_FORCE_IMMEDIATE_EXIT;
@@ -8328,6 +8306,36 @@ static inline int u64_shl_div_u64(u64 a, unsigned int shift,
return 0;
}
+static __init void vmx_setup_preemption_timer(void)
+{
+ if (!cpu_has_vmx_preemption_timer())
+ enable_preemption_timer = false;
+
+ if (enable_preemption_timer) {
+ u64 use_timer_freq = 5000ULL * 1000 * 1000;
+
+ cpu_preemption_timer_multi =
+ vmx_misc_preemption_timer_rate(vmcs_config.misc);
+
+ if (tsc_khz)
+ use_timer_freq = (u64)tsc_khz * 1000;
+ use_timer_freq >>= cpu_preemption_timer_multi;
+
+ /*
+ * KVM "disables" the preemption timer by setting it to its max
+ * value. Don't use the timer if it might cause spurious exits
+ * at a rate faster than 0.1 Hz (of uninterrupted guest time).
+ */
+ if (use_timer_freq > 0xffffffffu / 10)
+ enable_preemption_timer = false;
+ }
+
+ if (!enable_preemption_timer) {
+ vt_x86_ops.set_hv_timer = NULL;
+ vt_x86_ops.cancel_hv_timer = NULL;
+ }
+}
+
int vmx_set_hv_timer(struct kvm_vcpu *vcpu, u64 guest_deadline_tsc,
bool *expired)
{
@@ -8372,6 +8380,39 @@ void vmx_cancel_hv_timer(struct kvm_vcpu *vcpu)
{
to_vmx(vcpu)->hv_deadline_tsc = -1;
}
+
+static void vmx_update_hv_timer(struct kvm_vcpu *vcpu, bool force_immediate_exit)
+{
+ struct vcpu_vmx *vmx = to_vmx(vcpu);
+ u64 tscl;
+ u32 delta_tsc;
+
+ if (force_immediate_exit) {
+ vmcs_write32(VMX_PREEMPTION_TIMER_VALUE, 0);
+ vmx->loaded_vmcs->hv_timer_soft_disabled = false;
+ } else if (vmx->hv_deadline_tsc != -1) {
+ tscl = rdtsc();
+ if (vmx->hv_deadline_tsc > tscl)
+ /* set_hv_timer ensures the delta fits in 32-bits */
+ delta_tsc = (u32)((vmx->hv_deadline_tsc - tscl) >>
+ cpu_preemption_timer_multi);
+ else
+ delta_tsc = 0;
+
+ vmcs_write32(VMX_PREEMPTION_TIMER_VALUE, delta_tsc);
+ vmx->loaded_vmcs->hv_timer_soft_disabled = false;
+ } else if (!vmx->loaded_vmcs->hv_timer_soft_disabled) {
+ vmcs_write32(VMX_PREEMPTION_TIMER_VALUE, -1);
+ vmx->loaded_vmcs->hv_timer_soft_disabled = true;
+ }
+}
+#else
+static __init void vmx_setup_preemption_timer(void) { }
+
+static void vmx_update_hv_timer(struct kvm_vcpu *vcpu, bool force_immediate_exit)
+{
+ BUILD_BUG_ON(1);
+}
#endif
void vmx_update_cpu_dirty_logging(struct kvm_vcpu *vcpu)
@@ -8734,32 +8775,7 @@ __init int vmx_hardware_setup(void)
if (!enable_ept || !enable_ept_ad_bits || !cpu_has_vmx_pml())
enable_pml = 0;
- if (!cpu_has_vmx_preemption_timer())
- enable_preemption_timer = false;
-
- if (enable_preemption_timer) {
- u64 use_timer_freq = 5000ULL * 1000 * 1000;
-
- cpu_preemption_timer_multi =
- vmx_misc_preemption_timer_rate(vmcs_config.misc);
-
- if (tsc_khz)
- use_timer_freq = (u64)tsc_khz * 1000;
- use_timer_freq >>= cpu_preemption_timer_multi;
-
- /*
- * KVM "disables" the preemption timer by setting it to its max
- * value. Don't use the timer if it might cause spurious exits
- * at a rate faster than 0.1 Hz (of uninterrupted guest time).
- */
- if (use_timer_freq > 0xffffffffu / 10)
- enable_preemption_timer = false;
- }
-
- if (!enable_preemption_timer) {
- vt_x86_ops.set_hv_timer = NULL;
- vt_x86_ops.cancel_hv_timer = NULL;
- }
+ vmx_setup_preemption_timer();
kvm_caps.supported_mce_cap |= MCG_LMCE_P;
kvm_caps.supported_mce_cap |= MCG_CMCI_P;
--
2.55.0.229.g6434b31f56-goog
^ permalink raw reply related [flat|nested] 6+ messages in thread* [PATCH v5 2/2] KVM: VMX: Cap VMX preemption timer to work around Intel erratum
2026-07-24 23:49 [PATCH v5 0/2] Compute safe bound for VMX preemption timer Jim Mattson
2026-07-24 23:49 ` [PATCH v5 1/2] KVM: VMX: Bury all of the VMX preemption timer code under CONFIG_X86_64=y Jim Mattson
@ 2026-07-24 23:49 ` Jim Mattson
2026-07-28 7:22 ` Binbin Wu
1 sibling, 1 reply; 6+ messages in thread
From: Jim Mattson @ 2026-07-24 23:49 UTC (permalink / raw)
To: seanjc, pbonzini; +Cc: chao.gao, kvm, Jim Mattson
Due to a widespread Intel erratum (e.g. EMR158), programming the
VMX-preemption timer with certain large values may cause the timer to
expire earlier than expected. The recommended workaround is to cap the
VMX-preemption timer value to strictly less than 2^25 * CPUID.15H:EBX[31:0]
/ CPUID.15H:EAX[31:0].
Calculate preemption_timer_limit during hardware setup based on CPUID 15H
when available, and return -ERANGE in vmx_set_hv_timer() if the shifted
delta_tsc reaches or exceeds preemption_timer_limit.
Use preemption_timer_limit - 1 instead of 0xffffffff when soft-disabling
the timer and when checking the timer frequency limit in hardware setup.
Reported-by: Sean Christopherson <seanjc@google.com>
Closes: https://lore.kernel.org/all/Zn9X0yFxZi_Mrlnt@google.com/
Suggested-by: Chao Gao <chao.gao@intel.com>
Assisted-by: Gemini:Gemini-Next
Reviewed-by: Chao Gao <chao.gao@intel.com>
Signed-off-by: Jim Mattson <jmattson@google.com>
---
arch/x86/kvm/vmx/vmx.c | 34 +++++++++++++++++++++++++++++-----
1 file changed, 29 insertions(+), 5 deletions(-)
diff --git a/arch/x86/kvm/vmx/vmx.c b/arch/x86/kvm/vmx/vmx.c
index a07faa066ef0..ebe83a641473 100644
--- a/arch/x86/kvm/vmx/vmx.c
+++ b/arch/x86/kvm/vmx/vmx.c
@@ -153,6 +153,7 @@ module_param(dump_invalid_vmcs, bool, 0644);
#ifdef CONFIG_X86_64
static int __read_mostly cpu_preemption_timer_multi;
static bool __read_mostly enable_preemption_timer = 1;
+static u64 __ro_after_init preemption_timer_limit;
module_param_named(preemption_timer, enable_preemption_timer, bool, S_IRUGO);
#else
#define enable_preemption_timer false
@@ -8306,6 +8307,26 @@ static inline int u64_shl_div_u64(u64 a, unsigned int shift,
return 0;
}
+/*
+ * Workaround for a widespread Intel erratum (e.g. EMR158) where the
+ * VMX-preemption timer may expire earlier than expected when programmed
+ * with large values. The workaround is to cap the timer value to strictly
+ * less than 2^25 * CPUID.15H:EBX / CPUID.15H:EAX.
+ */
+static __init u64 calc_preemption_timer_limit(void)
+{
+ u32 eax, ebx, ecx, edx;
+
+ if (cpuid_eax(0) < 0x15)
+ return 1ULL << 32;
+
+ cpuid(0x15, &eax, &ebx, &ecx, &edx);
+ if (!eax || !ebx)
+ return 1ULL << 32;
+
+ return min_t(u64, 1ULL << 32, ((u64)ebx << 25) / eax);
+}
+
static __init void vmx_setup_preemption_timer(void)
{
if (!cpu_has_vmx_preemption_timer())
@@ -8317,6 +8338,8 @@ static __init void vmx_setup_preemption_timer(void)
cpu_preemption_timer_multi =
vmx_misc_preemption_timer_rate(vmcs_config.misc);
+ preemption_timer_limit = calc_preemption_timer_limit();
+
if (tsc_khz)
use_timer_freq = (u64)tsc_khz * 1000;
use_timer_freq >>= cpu_preemption_timer_multi;
@@ -8326,7 +8349,8 @@ static __init void vmx_setup_preemption_timer(void)
* value. Don't use the timer if it might cause spurious exits
* at a rate faster than 0.1 Hz (of uninterrupted guest time).
*/
- if (use_timer_freq > 0xffffffffu / 10)
+ if (!preemption_timer_limit ||
+ use_timer_freq > (preemption_timer_limit - 1) / 10)
enable_preemption_timer = false;
}
@@ -8363,12 +8387,12 @@ int vmx_set_hv_timer(struct kvm_vcpu *vcpu, u64 guest_deadline_tsc,
return -ERANGE;
/*
- * If the delta tsc can't fit in the 32 bit after the multi shift,
- * we can't use the preemption timer.
+ * If the delta tsc exceeds the preemption timer limit after the
+ * multi shift, we can't use the preemption timer.
* It's possible that it fits on later vmentries, but checking
* on every vmentry is costly so we just use an hrtimer.
*/
- if (delta_tsc >> (cpu_preemption_timer_multi + 32))
+ if ((delta_tsc >> cpu_preemption_timer_multi) >= preemption_timer_limit)
return -ERANGE;
vmx->hv_deadline_tsc = tscl + delta_tsc;
@@ -8402,7 +8426,7 @@ static void vmx_update_hv_timer(struct kvm_vcpu *vcpu, bool force_immediate_exit
vmcs_write32(VMX_PREEMPTION_TIMER_VALUE, delta_tsc);
vmx->loaded_vmcs->hv_timer_soft_disabled = false;
} else if (!vmx->loaded_vmcs->hv_timer_soft_disabled) {
- vmcs_write32(VMX_PREEMPTION_TIMER_VALUE, -1);
+ vmcs_write32(VMX_PREEMPTION_TIMER_VALUE, preemption_timer_limit - 1);
vmx->loaded_vmcs->hv_timer_soft_disabled = true;
}
}
--
2.55.0.229.g6434b31f56-goog
^ permalink raw reply related [flat|nested] 6+ messages in thread