KVM: svm: Avoid NULL pointer dereference in svm_hv_update_vp_id()
Subsystems (4)
Vitaly Kuznetsov
maintainerPATCH 1/1 KVM: svm: Avoid NULL pointer dereference in svm_hv_update_vp_id()
Hyper-V context is allocated dynamically when Hyper-V features are enabled on a vCPU but svm_hv_update_vp_id() is called unconditionally from svm_vcpu_run(), this leads to dereferencing to_hv_vcpu(vcpu) which can be NULL. Use kvm_hv_get_vpindex() wrapper to avoid the problem.
---
- The patch introducing the issue is currently in kvm/queue.
--- arch/x86/kvm/svm/svm_onhyperv.h | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-)
arch/x86/kvm/svm/svm_onhyperv.h +3 -2
@@ -99,9 +99,10 @@ static inline void svm_hv_update_vp_id(struct vmcb *vmcb, {struct hv_enlightenments *hve = (struct hv_enlightenments *)vmcb->control.reserved_sw; +u32 vp_index =kvm_hv_get_vpindex (vcpu); -if (hve->hv_vp_id !=to_hv_vcpu (vcpu)->vp_index) { - hve->hv_vp_id =to_hv_vcpu (vcpu)->vp_index; +if (hve->hv_vp_id != vp_index) { + hve->hv_vp_id = vp_index;vmcb_mark_dirty (vmcb,VMCB_HV_NESTED_ENLIGHTENMENTS ); } }
Signature · 1 line
2.31.1
Paolo Bonzini
maintainerQuoted 21 lines · Vitaly Kuznetsov
Hyper-V context is allocated dynamically when Hyper-V features are enabled on a vCPU but
svm_hv_update_vp_id()is called unconditionally fromsvm_vcpu_run(), this leads to dereferencingto_hv_vcpu(vcpu)which can be NULL. Usekvm_hv_get_vpindex()wrapper to avoid the problem.Fixes 4ba0d72aaa32 ("KVM: SVM: hyper-v: Direct Virtual Flush support")Signed-off-by Vitaly Kuznetsov <[email protected]>---
- The patch introducing the issue is currently in kvm/queue.--- arch/x86/kvm/svm/svm_onhyperv.h | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-)arch/x86/kvm/svm/svm_onhyperv.h +0
-0@@ -99,9 +99,10 @@ static inline void svm_hv_update_vp_id(struct vmcb *vmcb, { struct hv_enlightenments *hve =
arch/x86/kvm/svm/svm_onhyperv.h static inline void svm_hv_update_vp_id(struct vmcb *vmcb, · line 108
Quoted 10 lines
arch/x86/kvm/svm/svm_onhyperv.h +3
-2(struct hv_enlightenments *)vmcb->control.reserved_sw; +u32 vp_index =kvm_hv_get_vpindex (vcpu); -if (hve->hv_vp_id !=to_hv_vcpu (vcpu)->vp_index) { - hve->hv_vp_id =to_hv_vcpu (vcpu)->vp_index; +if (hve->hv_vp_id != vp_index) { + hve->hv_vp_id = vp_index;vmcb_mark_dirty (vmcb,VMCB_HV_NESTED_ENLIGHTENMENTS ); } }
Squashed, thanks.
Paolo