diff mbox

[RFC,14/18] pc: create apic_id_for_cpu() function (v3)

Message ID 1349270954-4657-15-git-send-email-ehabkost@redhat.com
State New
Headers show

Commit Message

Eduardo Habkost Oct. 3, 2012, 1:29 p.m. UTC 
Currently we need a way to calculate the Initial APIC ID using only the
CPU index (without needing a CPU object), as the NUMA fw_cfg data is
APIC-ID-based, and may include data for hotplug CPUs (that don't exist
yet), up to max_cpus.

Changes v2 -> v3:
  - Move the whole code to hw/pc.c, now only the PC code will need
    to handle topology-based APIC IDs

Changes v1 -> v2:
  - make function return value 'unsigned int' (it's not specific for the 8-bit
    xAPIC ID)
  - move implementation to cpu.c

Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
---
 hw/pc.c | 16 +++++++++++++++-
 1 file changed, 15 insertions(+), 1 deletion(-)
diff mbox

Patch

diff --git a/hw/pc.c b/hw/pc.c
index c64c218..4b30f38 100644
--- a/hw/pc.c
+++ b/hw/pc.c
@@ -568,6 +568,20 @@  static void pc_register_type(void)
 
 type_init(pc_register_type);
 
+/* Calculates initial APIC ID for a specific CPU index
+ *
+ * Currently we need to be able to calculate the APIC ID from the CPU index
+ * alone, as the QEMU<->Seabios interfaces have no concept of "CPU index",
+ * and the NUMA tables need the APIC ID of all CPUs up to max_cpus.
+ */
+static uint32_t apic_id_for_cpu(PC *pc, int cpu_index)
+{
+    /* right now APIC ID == CPU index. this will eventually change to use
+     * the CPU topology configuration properly
+     */
+    return cpu_index;
+}
+
 int e820_add_entry(uint64_t address, uint64_t length, uint32_t type)
 {
     int index = le32_to_cpu(e820_table.count);
@@ -935,7 +949,7 @@  void pc_cpus_init(PC *pc, const char *cpu_model)
     }
 
     for(i = 0; i < smp_cpus; i++) {
-        pc_new_cpu(pc, cpu_model, i);
+        pc_new_cpu(pc, cpu_model, apic_id_for_cpu(pc, i));
     }
 }