Patch logs MCE AO, AR messages injected to guest or taken by QEMU itself.
We print the QEMU address for guest MCEs, helps on hypervisors that have
another source of MCE logging like mce log, and when they go missing.

For example we found these QEMU logs:

September 26th 2019, 17:36:02.309        Droplet-153258224: Guest MCE Memory 
Error at qemu addr 0x7f8ce14f5000 and guest 3d6f5000 addr of type BUS_MCEERR_AR 
injected   qemu-system-x86_64      amsN    ams3nodeNNNN

September 27th 2019, 06:25:03.234        Droplet-153258224: Guest MCE Memory 
Error at qemu addr 0x7f8ce14f5000 and guest 3d6f5000 addr of type BUS_MCEERR_AR 
injected   qemu-system-x86_64      amsN    ams3nodeNNNN

The first log had a corresponding mce log entry, the second didnt (logging
thresholds) we can infer from second entry same PA and mce type.

Signed-off-by: Mario Smarduch <msmard...@digitalocean.com>
---
 target/i386/kvm.c | 29 ++++++++++++++++++++++++-----
 1 file changed, 24 insertions(+), 5 deletions(-)

diff --git a/target/i386/kvm.c b/target/i386/kvm.c
index 11b9c85..7eeefac 100644
--- a/target/i386/kvm.c
+++ b/target/i386/kvm.c
@@ -592,9 +592,9 @@ static void kvm_mce_inject(X86CPU *cpu, hwaddr paddr, int 
code)
                        (MCM_ADDR_PHYS << 6) | 0xc, flags);
 }
 
-static void hardware_memory_error(void)
+static void hardware_memory_error(void *host_addr)
 {
-    fprintf(stderr, "Hardware memory error!\n");
+    error_report("QEMU got Hardware memory error at addr %p", host_addr);
     exit(1);
 }
 
@@ -618,15 +618,34 @@ void kvm_arch_on_sigbus_vcpu(CPUState *c, int code, void 
*addr)
             kvm_physical_memory_addr_from_host(c->kvm_state, addr, &paddr)) {
             kvm_hwpoison_page_add(ram_addr);
             kvm_mce_inject(cpu, paddr, code);
+
+            /*
+             * Use different logging severity based on error type.
+             * If there is additional MCE reporting on the hypervisor, QEMU VA
+             * could be another source to identify the PA and MCE details.
+             */
+            if (code == BUS_MCEERR_AR) {
+                error_report("Guest MCE Memory Error at QEMU addr %p and "
+                    "GUEST addr 0x%" HWADDR_PRIx " of type %s injected",
+                    addr, paddr, "BUS_MCEERR_AR");
+            } else {
+                 warn_report("Guest MCE Memory Error at QEMU addr %p and "
+                     "GUEST addr 0x%" HWADDR_PRIx " of type %s injected",
+                     addr, paddr, "BUS_MCEERR_AO");
+            }
+
             return;
         }
 
-        fprintf(stderr, "Hardware memory error for memory used by "
-                "QEMU itself instead of guest system!\n");
+        if (code == BUS_MCEERR_AO) {
+            warn_report("Hardware memory error at addr %p of type %s "
+                "for memory used by QEMU itself instead of guest system!",
+                 addr, "BUS_MCEERR_AO");
+        }
     }
 
     if (code == BUS_MCEERR_AR) {
-        hardware_memory_error();
+        hardware_memory_error(addr);
     }
 
     /* Hope we are lucky for AO MCE */
-- 
2.9.3


Reply via email to