[HVM] Currently in xentrace, MMIO is handled just like other page faults. The

tracing parameters are error code and the guest virtual address.
However, if we are to analyze MMIO, tracing gpa makes more sense.

This small pactch replaces the va with gpa when tracing MMIO.

Signed-off-by: Qing He <qing.he@intel.com>

diff --git a/xen/arch/x86/hvm/vmx/vmx.c b/xen/arch/x86/hvm/vmx/vmx.c
index cc06e922c950c7bedccda9d2914376d48c152e76..918d7c5587e587e7eb613fea0b5d7e2de4a5c8d5 100644 (file)
--- a/xen/arch/x86/hvm/vmx/vmx.c
+++ b/xen/arch/x86/hvm/vmx/vmx.c
@@ -811,6 +811,8 @@ static int vmx_do_page_fault(unsigned long va, struct cpu_user_regs *regs)
             return 1;
         }
         TRACE_VMEXIT (2,2);
+        /* in the case of MMIO, we are more interested in gpa than in va */
+        TRACE_VMEXIT (4,gpa);
         handle_mmio(va, gpa);
         return 1;
     }