LoongArch: Increase max supported CPUs up to 2048

Increase max supported CPUs up to 2048, including:
1. Increase CSR.CPUID register's effective width;
2. Define MAX_CORE_PIC (a.k.a. max physical ID) to 2048;
3. Allow NR_CPUS (a.k.a. max logical ID) to be as large as 2048;
4. Introduce acpi_numa_x2apic_affinity_init() to handle ACPI SRAT
   for CPUID >= 256.

Note: The reason of increasing to 2048 rather than 4096/8192 is because
      the IPI hardware can only support 2048 as a maximum.

Reviewed-by: Yanteng Si <si.yanteng@linux.dev>
Signed-off-by: Huacai Chen <chenhuacai@loongson.cn>

Author: chenhuacai

arch/loongarch/Kconfig

@@ -496,10 +496,10 @@ config HOTPLUG_CPU
 	  Say N if you want to disable CPU hotplug.
 
 config NR_CPUS
-	int "Maximum number of CPUs (2-256)"
-	range 2 256
+	int "Maximum number of CPUs (2-2048)"
+	range 2 2048
+	default "2048"
 	depends on SMP
-	default "64"
 	help
 	  This allows you to specify the maximum number of CPUs which this
 	  kernel will support.

arch/loongarch/include/asm/acpi.h

@@ -33,7 +33,7 @@ static inline bool acpi_has_cpu_in_madt(void)
 	return true;
 }
 
-#define MAX_CORE_PIC 256
+#define MAX_CORE_PIC 2048
 
 extern struct list_head acpi_wakeup_device_list;
 extern struct acpi_madt_core_pic acpi_core_pic[MAX_CORE_PIC];

arch/loongarch/include/asm/loongarch.h

@@ -411,8 +411,8 @@
 
 /* Config CSR registers */
 #define LOONGARCH_CSR_CPUID		0x20	/* CPU core id */
-#define  CSR_CPUID_COREID_WIDTH		9
-#define  CSR_CPUID_COREID		_ULCAST_(0x1ff)
+#define  CSR_CPUID_COREID_WIDTH		11
+#define  CSR_CPUID_COREID		_ULCAST_(0x7ff)
 
 #define LOONGARCH_CSR_PRCFG1		0x21	/* Config1 */
 #define  CSR_CONF1_VSMAX_SHIFT		12

arch/loongarch/kernel/acpi.c

@@ -244,11 +244,6 @@ void __init acpi_boot_table_init(void)
 
 #ifdef CONFIG_ACPI_NUMA
 
-static __init int setup_node(int pxm)
-{
-	return acpi_map_pxm_to_node(pxm);
-}
-
 void __init numa_set_distance(int from, int to, int distance)
 {
 	if ((u8)distance != distance || (from == to && distance != LOCAL_DISTANCE)) {
@@ -280,7 +275,41 @@ acpi_numa_processor_affinity_init(struct acpi_srat_cpu_affinity *pa)
 		pxm |= (pa->proximity_domain_hi[1] << 16);
 		pxm |= (pa->proximity_domain_hi[2] << 24);
 	}
-	node = setup_node(pxm);
+	node = acpi_map_pxm_to_node(pxm);
+	if (node < 0) {
+		pr_err("SRAT: Too many proximity domains %x\n", pxm);
+		bad_srat();
+		return;
+	}
+
+	if (pa->apic_id >= CONFIG_NR_CPUS) {
+		pr_info("SRAT: PXM %u -> CPU 0x%02x -> Node %u skipped apicid that is too big\n",
+				pxm, pa->apic_id, node);
+		return;
+	}
+
+	early_numa_add_cpu(pa->apic_id, node);
+
+	set_cpuid_to_node(pa->apic_id, node);
+	node_set(node, numa_nodes_parsed);
+	pr_info("SRAT: PXM %u -> CPU 0x%02x -> Node %u\n", pxm, pa->apic_id, node);
+}
+
+void __init
+acpi_numa_x2apic_affinity_init(struct acpi_srat_x2apic_cpu_affinity *pa)
+{
+	int pxm, node;
+
+	if (srat_disabled())
+		return;
+	if (pa->header.length < sizeof(struct acpi_srat_x2apic_cpu_affinity)) {
+		bad_srat();
+		return;
+	}
+	if ((pa->flags & ACPI_SRAT_CPU_ENABLED) == 0)
+		return;
+	pxm = pa->proximity_domain;
+	node = acpi_map_pxm_to_node(pxm);
 	if (node < 0) {
 		pr_err("SRAT: Too many proximity domains %x\n", pxm);
 		bad_srat();