x86/virt/tdx: Fix lockdep assertion failure in cache flush for kexec
Subsystems (3)
PATCH v2 1/1 x86/virt/tdx: Fix lockdep assertion failure in cache flush for kexec
TDX can leave the cache in an incoherent state for the memory it uses. During kexec the kernel does a WBINVD for each CPU before memory gets reused in the second kernel.
There were two considerations for where this WBINVD should happen. In order to handle cases where the cache might get into an incoherent state while the kexec is in the initial stages, it is needed to do this later in the kexec path, when the kexecing CPU stops all remote CPUs. However, the later kexec process is sensitive to existing races. So to avoid perturbing that operation, it is better to do it earlier.
The existing solution is to track the need for the kexec time WBINVD generically (i.e., not just for TDX) in a per-cpu var. The late invocation only happens if the earlier TDX specific logic in tdx_cpu_flush_cache_for_kexec() didn’t take care of the work. This earlier WBINVD logic was built into KVM’s existing syscore ops shutdown() handler, which is called earlier in the kexec path.
However, this accidentally added it to KVM’s unload path as well (also the "error path" when bringing up TDX during KVM module load), which uses the same internal functions. This makes some sense too, though, because if KVM is getting unloaded, TDX cache affecting operations will likely cease. So it is a good point to do the work before KVM is unloaded and won't have a chance to handle the shutdown operation in the future.
Unfortunately this KVM unload invocation triggers a lockdep warning in tdx_cpu_flush_cache_for_kexec():
IS_ENABLED(CONFIG_PREEMPT_COUNT) && __lockdep_enabled && (preempt_count() == 0 && this_cpu_read(hardirqs_enabled)) WARNING: arch/x86/virt/vmx/tdx/tdx.c:1875 at tdx_cpu_flush_cache_for_kexec+0x36/0x60, CPU#0: cpuhp/0/22 ... Call Trace: <TASK> vt_disable_virtualization_cpu+0x1c/0x30 [kvm_intel] kvm_arch_disable_virtualization_cpu+0x12/0x80 [kvm] kvm_offline_cpu+0x24/0x40 [kvm] cpuhp_invoke_callback+0x1b0/0x740 ...
Since tdx_cpu_flush_cache_for_kexec() is doing WBINVD on a specific CPU, it has an assert for preemption being disabled. This works fine for the kexec time invocation, but the KVM unload path calls this as part of a CPUHP callback for which, despite always executing on the target CPU, preemption is not disabled.
It might be better to add the earlier invocation logic to a dedicated arch/x86 TDX syscore shutdown() handler, but to make the fix more backport friendly just adjust the lockdep assert in the tdx_cpu_flush_cache_for_kexec().
The real requirement is tdx_cpu_flush_cache_for_kexec() must be done on the same CPU. It's OK that it can be preempted in the middle as long as it won't be rescheduled to another CPU.
Remove the too strong lockdep_assert_preemption_disabled(), and change this_cpu_{read|write}() to __this_cpu_{read|write}() which provide the more proper check (when CONFIG_DEBUG_PREEMPT is true), which checks all conditions that the context cannot be moved to another CPU to run in the middle.
---
v1 -> v2:
- Collect tags - Thanks Nikolay, Rick and Sean!.
- Add the actual lockdep warn splat - Rick, Sean
v1: https://lore.kernel.org/lkml/[email protected]/
--- arch/x86/virt/vmx/tdx/tdx.c | 6 ++---- 1 file changed, 2 insertions(+), 4 deletions(-)
arch/x86/virt/vmx/tdx/tdx.c +2 -4
@@ -1872,9 +1872,7 @@ EXPORT_SYMBOL_FOR_KVM(tdh_phymem_page_wbinvd_hkid);#ifdef CONFIG_KEXEC_CORE void tdx_cpu_flush_cache_for_kexec (void ) { -lockdep_assert_preemption_disabled (); - -if (!this_cpu_read (cache_state_incoherent)) +if (!__this_cpu_read (cache_state_incoherent))return ;/*
@@ -1883,7 +1881,7 @@ void tdx_cpu_flush_cache_for_kexec(void)* there should be no more SEAMCALLs on this CPU. */ wbinvd (); - this_cpu_write(cache_state_incoherent, false); + __this_cpu_write(cache_state_incoherent, false); }EXPORT_SYMBOL_FOR_KVM (tdx_cpu_flush_cache_for_kexec);#endif
base-commit: 0f409eaea53e49932cf92a761de66345c9a4b4be
Signature · 1 line
2.53.0
Quoted 61 lines · Kai Huang
TDX can leave the cache in an incoherent state for the memory it uses. During kexec the kernel does a WBINVD for each CPU before memory gets reused in the second kernel.
There were two considerations for where this WBINVD should happen. In order to handle cases where the cache might get into an incoherent state while the kexec is in the initial stages, it is needed to do this later in the kexec path, when the kexecing CPU stops all remote CPUs. However, the later kexec process is sensitive to existing races. So to avoid perturbing that operation, it is better to do it earlier.
The existing solution is to track the need for the kexec time WBINVD generically (i.e., not just for TDX) in a per-cpu var. The late invocation only happens if the earlier TDX specific logic in
tdx_cpu_flush_cache_for_kexec()didn’t take care of the work. This earlier WBINVD logic was built into KVM’s existing syscore opsshutdown()handler, which is called earlier in the kexec path.However, this accidentally added it to KVM’s unload path as well (also the "error path" when bringing up TDX during KVM module load), which uses the same internal functions. This makes some sense too, though, because if KVM is getting unloaded, TDX cache affecting operations will likely cease. So it is a good point to do the work before KVM is unloaded and won't have a chance to handle the shutdown operation in the future.
Unfortunately this KVM unload invocation triggers a lockdep warning in
tdx_cpu_flush_cache_for_kexec():IS_ENABLED(CONFIG_PREEMPT_COUNT) && __lockdep_enabled && (preempt_count() == 0 && this_cpu_read(hardirqs_enabled)) WARNING: arch/x86/virt/vmx/tdx/tdx.c:1875 at tdx_cpu_flush_cache_for_kexec+0x36/0x60, CPU#0: cpuhp/0/22 ... Call Trace: <TASK> vt_disable_virtualization_cpu+0x1c/0x30 [kvm_intel] kvm_arch_disable_virtualization_cpu+0x12/0x80 [kvm] kvm_offline_cpu+0x24/0x40 [kvm] cpuhp_invoke_callback+0x1b0/0x740 ...Since
tdx_cpu_flush_cache_for_kexec()is doing WBINVD on a specific CPU, it has an assert for preemption being disabled. This works fine for the kexec time invocation, but the KVM unload path calls this as part of a CPUHP callback for which, despite always executing on the target CPU, preemption is not disabled.It might be better to add the earlier invocation logic to a dedicated
arch/x86TDX syscoreshutdown()handler, but to make the fix more backport friendly just adjust the lockdep assert in thetdx_cpu_flush_cache_for_kexec().The real requirement is
tdx_cpu_flush_cache_for_kexec()must be done on the same CPU. It's OK that it can be preempted in the middle as long as it won't be rescheduled to another CPU.Remove the too strong
lockdep_assert_preemption_disabled(), and change this_cpu_{read|write}() to __this_cpu_{read|write}() which provide the more proper check (whenCONFIG_DEBUG_PREEMPTis true), which checks all conditions that the context cannot be moved to another CPU to run in the middle.
Quoted 8 lines
Fixes 61221d07e815 ("KVM/TDX: Explicitly do WBINVD when no more TDX SEAMCALLs")Reported-by Vishal Verma <[email protected]>Tested-by Vishal Verma <[email protected]>Reviewed-by Nikolay Borisov <[email protected]>Reviewed-by Rick Edgecombe <[email protected]>Signed-off-by Kai Huang <[email protected]>
Signature · 1 line
Kiryl Shutsemau / Kirill A. Shutemov
Quoted 2 lines
Thanks!