[PATCH v7 04/13] xen/cpufreq: add missing default: case for x86 vendor

[Penny Zheng <Penny.Zheng@xxxxxxx>]

Since we are missing default case for x86 vendor, there is possibility (i.e.
new vendor introduced) that we will return successfully while missing the
whole cpufreq driver initialization process.
Move "ret = -ENOENTRY" forward to cover default case for x86 vendor, and
add error log

Signed-off-by: Penny Zheng <Penny.Zheng@xxxxxxx>
---
v6 -> v7:
- new commit
---
 xen/arch/x86/acpi/cpufreq/cpufreq.c | 8 ++++++--
 1 file changed, 6 insertions(+), 2 deletions(-)

diff --git a/xen/arch/x86/acpi/cpufreq/cpufreq.c 
b/xen/arch/x86/acpi/cpufreq/cpufreq.c
index d18735c7ae..e227376bab 100644
--- a/xen/arch/x86/acpi/cpufreq/cpufreq.c
+++ b/xen/arch/x86/acpi/cpufreq/cpufreq.c
@@ -131,11 +131,11 @@ static int __init cf_check cpufreq_driver_init(void)
 
     if ( cpufreq_controller == FREQCTL_xen )
     {
+        ret = -ENOENT;
+
         switch ( boot_cpu_data.x86_vendor )
         {
         case X86_VENDOR_INTEL:
-            ret = -ENOENT;
-
             for ( unsigned int i = 0; i < cpufreq_xen_cnt; i++ )
             {
                 switch ( cpufreq_xen_opts[i] )
@@ -162,6 +162,10 @@ static int __init cf_check cpufreq_driver_init(void)
         case X86_VENDOR_HYGON:
             ret = IS_ENABLED(CONFIG_AMD) ? powernow_register_driver() : 
-ENODEV;
             break;
+
+        default:
+            printk(XENLOG_ERR "Cpufreq: unsupported x86 vendor\n");
+            break;
         }
     }
 
-- 
2.34.1