[v2,2/3] hw/smbios: report error if table size is too large

Commit Message

Daniel P. Berrangé Sept. 23, 2020, 10:41 a.m. UTC

The SMBIOS 2.1 entry point uses a uint16 data type for reporting the
total length of the tables. If the user passes -smbios configuration to
QEMU that causes the table size to exceed this limit then various bad
behaviours result, including

 - firmware hangs in an infinite loop
 - firmware triggers a KVM crash on bad memory access
 - firmware silently discards user's SMBIOS data replacing it with
   a generic data set.

Limiting the size to 0xffff in QEMU avoids triggering most of these
problems. There is a remaining bug in SeaBIOS which tries to prepend its
own data for table 0, and does not check whether there is sufficient
space before attempting this.

Reviewed-by: Igor Mammedov <[e-mail]>
Reviewed-by: Philippe Mathieu-Daudé <[e-mail]>
Signed-off-by: Daniel P. Berrangé <[e-mail]>
---
 hw/smbios/smbios.c | 14 ++++++++++++++
 1 file changed, 14 insertions(+)

Patch

diff --git a/hw/smbios/smbios.c b/hw/smbios/smbios.c
index 8450fad285..3c87be6c91 100644
--- a/hw/smbios/smbios.c
+++ b/hw/smbios/smbios.c
@@ -365,6 +365,13 @@  static void smbios_register_config(void)
 
 opts_init(smbios_register_config);
 
+/*
+ * The SMBIOS 2.1 "structure table length" field in the
+ * entry point uses a 16-bit integer, so we're limited
+ * in total table size
+ */
+#define SMBIOS_21_MAX_TABLES_LEN 0xffff
+
 static void smbios_validate_table(MachineState *ms)
 {
     uint32_t expect_t4_count = smbios_legacy ?
@@ -375,6 +382,13 @@  static void smbios_validate_table(MachineState *ms)
                      expect_t4_count, smbios_type4_count);
         exit(1);
     }
+
+    if (smbios_ep_type == SMBIOS_ENTRY_POINT_21 &&
+        smbios_tables_len > SMBIOS_21_MAX_TABLES_LEN) {
+        error_report("SMBIOS 2.1 table length %zu exceeds %d",
+                     smbios_tables_len, SMBIOS_21_MAX_TABLES_LEN);
+        exit(1);
+    }
 }