1 files changed, 199 insertions, 65 deletions

diff --git a/drivers/acpi/numa.c b/drivers/acpi/numa.c
index 64b98e82feb..24b5476449a 100644
--- a/drivers/acpi/numa.c
+++ b/drivers/acpi/numa.c
@@ -23,26 +23,68 @@
  *
  */
 #include <linux/module.h>
-#include <linux/config.h>
 #include <linux/init.h>
 #include <linux/kernel.h>
 #include <linux/types.h>
 #include <linux/errno.h>
 #include <linux/acpi.h>
-#include <acpi/acpi_bus.h>
-#include <acpi/acmacros.h>
+#include <linux/numa.h>
+
+#define PREFIX "ACPI: "
 
 #define ACPI_NUMA	0x80000000
 #define _COMPONENT	ACPI_NUMA
-ACPI_MODULE_NAME("numa")
+ACPI_MODULE_NAME("numa");
+
+static nodemask_t nodes_found_map = NODE_MASK_NONE;
+
+/* maps to convert between proximity domain and logical node ID */
+static int pxm_to_node_map[MAX_PXM_DOMAINS]
+			= { [0 ... MAX_PXM_DOMAINS - 1] = NUMA_NO_NODE };
+static int node_to_pxm_map[MAX_NUMNODES]
+			= { [0 ... MAX_NUMNODES - 1] = PXM_INVAL };
 
-extern int __init acpi_table_parse_madt_family(enum acpi_table_id id,
-					       unsigned long madt_size,
-					       int entry_id,
-					       acpi_madt_entry_handler handler,
-					       unsigned int max_entries);
+unsigned char acpi_srat_revision __initdata;
+
+int pxm_to_node(int pxm)
+{
+	if (pxm < 0)
+		return NUMA_NO_NODE;
+	return pxm_to_node_map[pxm];
+}
+
+int node_to_pxm(int node)
+{
+	if (node < 0)
+		return PXM_INVAL;
+	return node_to_pxm_map[node];
+}
+
+static void __acpi_map_pxm_to_node(int pxm, int node)
+{
+	if (pxm_to_node_map[pxm] == NUMA_NO_NODE || node < pxm_to_node_map[pxm])
+		pxm_to_node_map[pxm] = node;
+	if (node_to_pxm_map[node] == PXM_INVAL || pxm < node_to_pxm_map[node])
+		node_to_pxm_map[node] = pxm;
+}
+
+int acpi_map_pxm_to_node(int pxm)
+{
+	int node = pxm_to_node_map[pxm];
+
+	if (node == NUMA_NO_NODE) {
+		if (nodes_weight(nodes_found_map) >= MAX_NUMNODES)
+			return NUMA_NO_NODE;
+		node = first_unset_node(nodes_found_map);
+		__acpi_map_pxm_to_node(pxm, node);
+		node_set(node, nodes_found_map);
+	}
+
+	return node;
+}
 
-void __init acpi_table_print_srat_entry(acpi_table_entry_header * header)
+static void __init
+acpi_table_print_srat_entry(struct acpi_subtable_header *header)
 {
 
 	ACPI_FUNCTION_NAME("acpi_table_print_srat_entry");
@@ -52,40 +94,56 @@ void __init acpi_table_print_srat_entry(acpi_table_entry_header * header)
 
 	switch (header->type) {
 
-	case ACPI_SRAT_PROCESSOR_AFFINITY:
+	case ACPI_SRAT_TYPE_CPU_AFFINITY:
 #ifdef ACPI_DEBUG_OUTPUT
 		{
-			struct acpi_table_processor_affinity *p =
-			    (struct acpi_table_processor_affinity *)header;
+			struct acpi_srat_cpu_affinity *p =
+			    (struct acpi_srat_cpu_affinity *)header;
 			ACPI_DEBUG_PRINT((ACPI_DB_INFO,
 					  "SRAT Processor (id[0x%02x] eid[0x%02x]) in proximity domain %d %s\n",
-					  p->apic_id, p->lsapic_eid,
-					  p->proximity_domain,
-					  p->flags.
-					  enabled ? "enabled" : "disabled"));
+					  p->apic_id, p->local_sapic_eid,
+					  p->proximity_domain_lo,
+					  (p->flags & ACPI_SRAT_CPU_ENABLED)?
+					  "enabled" : "disabled"));
 		}
 #endif				/* ACPI_DEBUG_OUTPUT */
 		break;
 
-	case ACPI_SRAT_MEMORY_AFFINITY:
+	case ACPI_SRAT_TYPE_MEMORY_AFFINITY:
 #ifdef ACPI_DEBUG_OUTPUT
 		{
-			struct acpi_table_memory_affinity *p =
-			    (struct acpi_table_memory_affinity *)header;
+			struct acpi_srat_mem_affinity *p =
+			    (struct acpi_srat_mem_affinity *)header;
 			ACPI_DEBUG_PRINT((ACPI_DB_INFO,
-					  "SRAT Memory (0x%08x%08x length 0x%08x%08x type 0x%x) in proximity domain %d %s%s\n",
-					  p->base_addr_hi, p->base_addr_lo,
-					  p->length_hi, p->length_lo,
-					  p->memory_type, p->proximity_domain,
-					  p->flags.
-					  enabled ? "enabled" : "disabled",
-					  p->flags.
-					  hot_pluggable ? " hot-pluggable" :
-					  ""));
+					  "SRAT Memory (0x%lx length 0x%lx) in proximity domain %d %s%s%s\n",
+					  (unsigned long)p->base_address,
+					  (unsigned long)p->length,
+					  p->proximity_domain,
+					  (p->flags & ACPI_SRAT_MEM_ENABLED)?
+					  "enabled" : "disabled",
+					  (p->flags & ACPI_SRAT_MEM_HOT_PLUGGABLE)?
+					  " hot-pluggable" : "",
+					  (p->flags & ACPI_SRAT_MEM_NON_VOLATILE)?
+					  " non-volatile" : ""));
 		}
 #endif				/* ACPI_DEBUG_OUTPUT */
 		break;
 
+	case ACPI_SRAT_TYPE_X2APIC_CPU_AFFINITY:
+#ifdef ACPI_DEBUG_OUTPUT
+		{
+			struct acpi_srat_x2apic_cpu_affinity *p =
+			    (struct acpi_srat_x2apic_cpu_affinity *)header;
+			ACPI_DEBUG_PRINT((ACPI_DB_INFO,
+					  "SRAT Processor (x2apicid[0x%08x]) in"
+					  " proximity domain %d %s\n",
+					  p->apic_id,
+					  p->proximity_domain,
+					  (p->flags & ACPI_SRAT_CPU_ENABLED) ?
+					  "enabled" : "disabled"));
+		}
+#endif				/* ACPI_DEBUG_OUTPUT */
+		break;
 	default:
 		printk(KERN_WARNING PREFIX
 		       "Found unsupported SRAT entry (type = 0x%x)\n",
@@ -94,31 +152,81 @@ void __init acpi_table_print_srat_entry(acpi_table_entry_header * header)
 	}
 }
 
-static int __init acpi_parse_slit(unsigned long phys_addr, unsigned long size)
+/*
+ * A lot of BIOS fill in 10 (= no distance) everywhere. This messes
+ * up the NUMA heuristics which wants the local node to have a smaller
+ * distance than the others.
+ * Do some quick checks here and only use the SLIT if it passes.
+ */
+static int __init slit_valid(struct acpi_table_slit *slit)
+{
+	int i, j;
+	int d = slit->locality_count;
+	for (i = 0; i < d; i++) {
+		for (j = 0; j < d; j++)  {
+			u8 val = slit->entry[d*i + j];
+			if (i == j) {
+				if (val != LOCAL_DISTANCE)
+					return 0;
+			} else if (val <= LOCAL_DISTANCE)
+				return 0;
+		}
+	}
+	return 1;
+}
+
+static int __init acpi_parse_slit(struct acpi_table_header *table)
 {
 	struct acpi_table_slit *slit;
-	u32 localities;
 
-	if (!phys_addr || !size)
+	if (!table)
 		return -EINVAL;
 
-	slit = (struct acpi_table_slit *)__va(phys_addr);
-
-	/* downcast just for %llu vs %lu for i386/ia64  */
-	localities = (u32) slit->localities;
+	slit = (struct acpi_table_slit *)table;
 
+	if (!slit_valid(slit)) {
+		printk(KERN_INFO "ACPI: SLIT table looks invalid. Not used.\n");
+		return -EINVAL;
+	}
 	acpi_numa_slit_init(slit);
 
 	return 0;
 }
 
+void __init __weak
+acpi_numa_x2apic_affinity_init(struct acpi_srat_x2apic_cpu_affinity *pa)
+{
+	printk(KERN_WARNING PREFIX
+	       "Found unsupported x2apic [0x%08x] SRAT entry\n", pa->apic_id);
+	return;
+}
+
+
 static int __init
-acpi_parse_processor_affinity(acpi_table_entry_header * header,
+acpi_parse_x2apic_affinity(struct acpi_subtable_header *header,
+			   const unsigned long end)
+{
+	struct acpi_srat_x2apic_cpu_affinity *processor_affinity;
+
+	processor_affinity = (struct acpi_srat_x2apic_cpu_affinity *)header;
+	if (!processor_affinity)
+		return -EINVAL;
+
+	acpi_table_print_srat_entry(header);
+
+	/* let architecture-dependent part to do it */
+	acpi_numa_x2apic_affinity_init(processor_affinity);
+
+	return 0;
+}
+
+static int __init
+acpi_parse_processor_affinity(struct acpi_subtable_header *header,
 			      const unsigned long end)
 {
-	struct acpi_table_processor_affinity *processor_affinity;
+	struct acpi_srat_cpu_affinity *processor_affinity;
 
-	processor_affinity = (struct acpi_table_processor_affinity *)header;
+	processor_affinity = (struct acpi_srat_cpu_affinity *)header;
 	if (!processor_affinity)
 		return -EINVAL;
 
@@ -130,69 +238,85 @@ acpi_parse_processor_affinity(acpi_table_entry_header * header,
 	return 0;
 }
 
+static int __initdata parsed_numa_memblks;
+
 static int __init
-acpi_parse_memory_affinity(acpi_table_entry_header * header,
+acpi_parse_memory_affinity(struct acpi_subtable_header * header,
 			   const unsigned long end)
 {
-	struct acpi_table_memory_affinity *memory_affinity;
+	struct acpi_srat_mem_affinity *memory_affinity;
 
-	memory_affinity = (struct acpi_table_memory_affinity *)header;
+	memory_affinity = (struct acpi_srat_mem_affinity *)header;
 	if (!memory_affinity)
 		return -EINVAL;
 
 	acpi_table_print_srat_entry(header);
 
 	/* let architecture-dependent part to do it */
-	acpi_numa_memory_affinity_init(memory_affinity);
-
+	if (!acpi_numa_memory_affinity_init(memory_affinity))
+		parsed_numa_memblks++;
 	return 0;
 }
 
-static int __init acpi_parse_srat(unsigned long phys_addr, unsigned long size)
+static int __init acpi_parse_srat(struct acpi_table_header *table)
 {
 	struct acpi_table_srat *srat;
-
-	if (!phys_addr || !size)
+	if (!table)
 		return -EINVAL;
 
-	srat = (struct acpi_table_srat *)__va(phys_addr);
+	srat = (struct acpi_table_srat *)table;
+	acpi_srat_revision = srat->header.revision;
+
+	/* Real work done in acpi_table_parse_srat below. */
 
 	return 0;
 }
 
-int __init
-acpi_table_parse_srat(enum acpi_srat_entry_id id,
-		      acpi_madt_entry_handler handler, unsigned int max_entries)
+static int __init
+acpi_table_parse_srat(enum acpi_srat_type id,
+		      acpi_tbl_entry_handler handler, unsigned int max_entries)
 {
-	return acpi_table_parse_madt_family(ACPI_SRAT,
+	return acpi_table_parse_entries(ACPI_SIG_SRAT,
 					    sizeof(struct acpi_table_srat), id,
 					    handler, max_entries);
 }
 
 int __init acpi_numa_init(void)
 {
-	int result;
+	int cnt = 0;
 
-	/* SRAT: Static Resource Affinity Table */
-	result = acpi_table_parse(ACPI_SRAT, acpi_parse_srat);
+	/*
+	 * Should not limit number with cpu num that is from NR_CPUS or nr_cpus=
+	 * SRAT cpu entries could have different order with that in MADT.
+	 * So go over all cpu entries in SRAT to get apicid to node mapping.
+	 */
 
-	if (result > 0) {
-		result = acpi_table_parse_srat(ACPI_SRAT_PROCESSOR_AFFINITY,
-					       acpi_parse_processor_affinity,
-					       NR_CPUS);
-		result = acpi_table_parse_srat(ACPI_SRAT_MEMORY_AFFINITY, acpi_parse_memory_affinity, NR_NODE_MEMBLKS);	// IA64 specific
+	/* SRAT: Static Resource Affinity Table */
+	if (!acpi_table_parse(ACPI_SIG_SRAT, acpi_parse_srat)) {
+		acpi_table_parse_srat(ACPI_SRAT_TYPE_X2APIC_CPU_AFFINITY,
+				     acpi_parse_x2apic_affinity, 0);
+		acpi_table_parse_srat(ACPI_SRAT_TYPE_CPU_AFFINITY,
+				     acpi_parse_processor_affinity, 0);
+		cnt = acpi_table_parse_srat(ACPI_SRAT_TYPE_MEMORY_AFFINITY,
+					    acpi_parse_memory_affinity,
+					    NR_NODE_MEMBLKS);
 	}
 
 	/* SLIT: System Locality Information Table */
-	result = acpi_table_parse(ACPI_SLIT, acpi_parse_slit);
+	acpi_table_parse(ACPI_SIG_SLIT, acpi_parse_slit);
 
 	acpi_numa_arch_fixup();
+
+	if (cnt < 0)
+		return cnt;
+	else if (!parsed_numa_memblks)
+		return -ENOENT;
 	return 0;
 }
 
-int acpi_get_pxm(acpi_handle h)
+static int acpi_get_pxm(acpi_handle h)
 {
-	unsigned long pxm;
+	unsigned long long pxm;
 	acpi_status status;
 	acpi_handle handle;
 	acpi_handle phandle = h;
@@ -201,10 +325,20 @@ int acpi_get_pxm(acpi_handle h)
 		handle = phandle;
 		status = acpi_evaluate_integer(handle, "_PXM", NULL, &pxm);
 		if (ACPI_SUCCESS(status))
-			return (int)pxm;
+			return pxm;
 		status = acpi_get_parent(handle, &phandle);
 	} while (ACPI_SUCCESS(status));
 	return -1;
 }
 
-EXPORT_SYMBOL(acpi_get_pxm);
+int acpi_get_node(acpi_handle handle)
+{
+	int pxm;
+
+	pxm = acpi_get_pxm(handle);
+	if (pxm < 0 || pxm >= MAX_PXM_DOMAINS)
+		return NUMA_NO_NODE;
+
+	return acpi_map_pxm_to_node(pxm);
+}
+EXPORT_SYMBOL(acpi_get_node);