Oliver Upton fa808ed4e1 KVM: arm64: Ensure a VMID is allocated before programming VTTBR_EL2 ... Vladimir reports that a race condition to attach a VMID to a stage-2 MMU sometimes results in a vCPU entering the guest with a VMID of 0: | CPU1 | CPU2 | | | | kvm_arch_vcpu_ioctl_run | | vcpu_load <= load VTTBR_EL2 | | kvm_vmid->id = 0 | | | kvm_arch_vcpu_ioctl_run | | vcpu_load <= load VTTBR_EL2 | | with kvm_vmid->id = 0| | kvm_arm_vmid_update <= allocates fresh | | kvm_vmid->id and | | reload VTTBR_EL2 | | | | | kvm_arm_vmid_update <= observes that kvm_vmid->id | | already allocated, | | skips reload VTTBR_EL2 Oh yeah, it's as bad as it looks. Remember that VHE loads the stage-2 MMU eagerly but a VMID only gets attached to the MMU later on in the KVM_RUN loop. Even in the "best case" where VTTBR_EL2 correctly gets reprogrammed before entering the EL1&0 regime, there is a period of time where hardware is configured with VMID 0. That's completely insane. So, rather than decorating the 'late' binding with another hack, just allocate the damn thing up front. Attaching a VMID from vcpu_load() is still rollover safe since (surprise!) it'll always get called after a vCPU was preempted. Excuse me while I go find a brown paper bag. Cc: stable@vger.kernel.org Fixes: 934bf871f0 ("KVM: arm64: Load the stage-2 MMU context in kvm_vcpu_load_vhe()") Reported-by: Vladimir Murzin <vladimir.murzin@arm.com> Signed-off-by: Oliver Upton <oliver.upton@linux.dev> Link: https://lore.kernel.org/r/20250219220737.130842-1-oliver.upton@linux.dev Signed-off-by: Marc Zyngier <maz@kernel.org>		2025-02-20 16:29:28 +00:00
..
hyp	KVM/arm64 fixes for 6.14, take #2	2025-02-14 18:32:47 -05:00
vgic	KVM: arm64: vgic: Hoist SGI/PPI alloc from vgic_init() to kvm_create_vgic()	2025-02-13 18:03:54 +00:00
.gitignore
arch_timer.c	KVM: arm64: timer: Drop warning on failed interrupt signalling	2025-02-13 18:03:54 +00:00
arm.c	KVM: arm64: Ensure a VMID is allocated before programming VTTBR_EL2	2025-02-20 16:29:28 +00:00
at.c	arm64 updates for 6.14	2025-01-20 21:21:49 -08:00
debug.c
emulate-nested.c	Merge branch kvm-arm64/nv-timers into kvmarm-master/next	2025-01-17 11:04:53 +00:00
fpsimd.c	KVM: arm64: Simplify warning in kvm_arch_vcpu_load_fp()	2025-02-13 17:55:13 +00:00
guest.c
handle_exit.c
hypercalls.c
inject_fault.c
Kconfig
Makefile
mmio.c
mmu.c	Merge branch kvm-arm64/misc-6.14 into kvmarm-master/next	2025-01-17 11:06:50 +00:00
nested.c	KVM: arm64: Fix nested S2 MMU structures reallocation	2025-02-04 15:02:16 +00:00
pauth.c
pkvm.c
pmu-emul.c
pmu.c
psci.c
ptdump.c
pvtime.c
reset.c
stacktrace.c
sys_regs.c	KVM/arm64 fixes for 6.14, take #1	2025-02-04 11:14:53 -05:00
sys_regs.h
trace_arm.h
trace_handle_exit.h
trace.h
trng.c
va_layout.c
vgic-sys-reg-v3.c
vmid.c	KVM: arm64: Ensure a VMID is allocated before programming VTTBR_EL2	2025-02-20 16:29:28 +00:00