x86: allow limiting the max C-state sub-state

Allow limiting the max C-state sub-state by appending to the max_cstate
command-line parameter. E.g. max_cstate=1,0
The limit only applies to the highest legal C-state. For example:
 max_cstate = 1, max_csubstate = 0 ==> C0, C1 okay, but not C1E
 max_cstate = 1, max_csubstate = 1 ==> C0, C1 and C1E okay, but not C2
 max_cstate = 2, max_csubstate = 0 ==> C0, C1, C1E, C2 okay, but not C3
 max_cstate = 2, max_csubstate = 1 ==> C0, C1, C1E, C2 okay, but not C3

Signed-off-by: Ross Lagerwall <ross.lagerwall@xxxxxxxxxx>
Signed-off-by: Jan Beulich <jbeulich@xxxxxxxx>
---
v2: Explicitly log "unlimited". Pass NULL in the 2nd simple_strtoul()
    invocation.

--- a/docs/misc/xen-command-line.pandoc
+++ b/docs/misc/xen-command-line.pandoc
@@ -1374,9 +1374,11 @@ Specify the maximum number of CPUs that
 This option is ignored in **pv-shim** mode.
 
 ### max_cstate (x86)
-> `= <integer>`
+> `= <integer>[,<integer>]`
 
-Specify the deepest C-state CPUs are permitted to be placed in.
+Specify the deepest C-state CPUs are permitted to be placed in, and
+optionally the maximum sub C-state to be used used.  The latter only applies
+to the highest permitted C-state.
 
 ### max_gsi_irqs (x86)
 > `= <integer>`
--- a/xen/arch/x86/acpi/cpu_idle.c
+++ b/xen/arch/x86/acpi/cpu_idle.c
@@ -104,7 +104,17 @@ bool lapic_timer_init(void)
 
 void (*__read_mostly pm_idle_save)(void);
 unsigned int max_cstate __read_mostly = UINT_MAX;
-integer_param("max_cstate", max_cstate);
+unsigned int max_csubstate __read_mostly = UINT_MAX;
+
+static int __init parse_cstate(const char *s)
+{
+    max_cstate = simple_strtoul(s, &s, 0);
+    if ( *s == ',' )
+        max_csubstate = simple_strtoul(s + 1, NULL, 0);
+    return 0;
+}
+custom_param("max_cstate", parse_cstate);
+
 static bool __read_mostly local_apic_timer_c2_ok;
 boolean_param("lapic_timer_c2_ok", local_apic_timer_c2_ok);
 
@@ -347,7 +357,13 @@ static void dump_cx(unsigned char key)
 
     printk("'%c' pressed -> printing ACPI Cx structures\n", key);
     if ( max_cstate < UINT_MAX )
+    {
         printk("max state: C%u\n", max_cstate);
+        if ( max_csubstate < UINT_MAX )
+            printk("max sub-state: %u\n", max_csubstate);
+        else
+            printk("max sub-state: unlimited\n");
+    }
     else
         printk("max state: unlimited\n");
     for_each_present_cpu ( cpu )
@@ -592,7 +608,13 @@ static void acpi_processor_idle(void)
 
         do {
             cx = &power->states[next_state];
-        } while ( cx->type > max_state && --next_state );
+        } while ( (cx->type > max_state ||
+                   cx->entry_method == ACPI_CSTATE_EM_NONE ||
+                   (cx->entry_method == ACPI_CSTATE_EM_FFH &&
+                    cx->type == max_cstate &&
+                    (cx->address & MWAIT_SUBSTATE_MASK) > max_csubstate)) &&
+                  --next_state );
+            cx = &power->states[next_state];
         if ( next_state )
         {
             if ( cx->type == ACPI_STATE_C3 && power->flags.bm_check &&
--- a/xen/arch/x86/cpu/mwait-idle.c
+++ b/xen/arch/x86/cpu/mwait-idle.c
@@ -731,7 +731,9 @@ static void mwait_idle(void)
 
 		do {
 			cx = &power->states[next_state];
-		} while (cx->type > max_state && --next_state);
+		} while ((cx->type > max_state || (cx->type == max_cstate &&
+			  MWAIT_HINT2SUBSTATE(cx->address) > max_csubstate)) &&
+			 --next_state);
 		if (!next_state)
 			cx = NULL;
 		else if (tb_init_done)
--- a/xen/include/xen/acpi.h
+++ b/xen/include/xen/acpi.h
@@ -141,13 +141,21 @@ void acpi_unregister_gsi (u32 gsi);
 
 #ifdef	CONFIG_ACPI_CSTATE
 /*
- * Set highest legal C-state
- * 0: C0 okay, but not C1
- * 1: C1 okay, but not C2
- * 2: C2 okay, but not C3 etc.
+ * max_cstate sets the highest legal C-state.
+ * max_cstate = 0: C0 okay, but not C1
+ * max_cstate = 1: C1 okay, but not C2
+ * max_cstate = 2: C2 okay, but not C3 etc.
+
+ * max_csubstate sets the highest legal C-state sub-state. Only applies to the
+ * highest legal C-state.
+ * max_cstate = 1, max_csubstate = 0 ==> C0, C1 okay, but not C1E
+ * max_cstate = 1, max_csubstate = 1 ==> C0, C1 and C1E okay, but not C2
+ * max_cstate = 2, max_csubstate = 0 ==> C0, C1, C1E, C2 okay, but not C3
+ * max_cstate = 2, max_csubstate = 1 ==> C0, C1, C1E, C2 okay, but not C3
  */
 
 extern unsigned int max_cstate;
+extern unsigned int max_csubstate;
 
 static inline unsigned int acpi_get_cstate_limit(void)
 {