When SECONDARY_EXEC_ENABLE_VIRT_EXCEPTIONS is set, vmx_vcpu_update_eptp() __vmwrites() EPTP_INDEX in altp2m_vcpu_destroy(). This means that when disabling altp2m on a domain after xc_altp2m_set_vcpu_enable_notify() has been successfully called, EPTP_INDEX ends up being stored as INVALID_ALTP2M. This makes it possible for vmx_vmexit_handler() to __vmread() the stale value after a subsequent call to xc_altp2m_set_vcpu_enable_notify(), and BUG_ON(idx >= MAX_ALTP2M).
Signed-off-by: Razvan Cojocaru <rcojoc...@bitdefender.com> --- Changes since V1: - Re-wrote the fix to affect the altp2m code instead of the code around the BUG_ON(). - Updated the patch description (and title - since the crash is really a host, not a domain, crash). --- xen/arch/x86/mm/altp2m.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/xen/arch/x86/mm/altp2m.c b/xen/arch/x86/mm/altp2m.c index 930bdc2..28c800b 100644 --- a/xen/arch/x86/mm/altp2m.c +++ b/xen/arch/x86/mm/altp2m.c @@ -58,8 +58,8 @@ altp2m_vcpu_destroy(struct vcpu *v) altp2m_vcpu_reset(v); - altp2m_vcpu_update_p2m(v); altp2m_vcpu_update_vmfunc_ve(v); + altp2m_vcpu_update_p2m(v); if ( v != current ) vcpu_unpause(v); -- 2.7.4 _______________________________________________ Xen-devel mailing list Xen-devel@lists.xenproject.org https://lists.xenproject.org/mailman/listinfo/xen-devel