LoongArch: Fix shared cache size calculation

Current calculation of shared cache size is from the node (die) scope,
but we hope 'lscpu' to show the shared cache size of the whole package
for multi-die chips (e.g., Loongson-3C5000L, which contains 4 dies in
one package). So fix it by multiplying nodes_per_package.
Signed-off-by: Huacai Chen <chenhuacai@loongson.cn>

arch/loongarch/kernel/cacheinfo.c

@@ -4,8 +4,9 @@
  *
  * Copyright (C) 2020-2022 Loongson Technology Corporation Limited
  */
-#include <asm/cpu-info.h>
 #include <linux/cacheinfo.h>
+#include <asm/bootinfo.h>
+#include <asm/cpu-info.h>
 
 /* Populates leaf and increments to next leaf */
 #define populate_cache(cache, leaf, c_level, c_type)		\
@@ -17,6 +18,8 @@ do {								\
 	leaf->ways_of_associativity = c->cache.ways;		\
 	leaf->size = c->cache.linesz * c->cache.sets *		\
 		c->cache.ways;					\
+	if (leaf->level > 2)					\
+		leaf->size *= nodes_per_package;		\
 	leaf++;							\
 } while (0)
 
@@ -95,11 +98,15 @@ static void cache_cpumap_setup(unsigned int cpu)
 
 int populate_cache_leaves(unsigned int cpu)
 {
-	int level = 1;
+	int level = 1, nodes_per_package = 1;
 	struct cpuinfo_loongarch *c = &current_cpu_data;
 	struct cpu_cacheinfo *this_cpu_ci = get_cpu_cacheinfo(cpu);
 	struct cacheinfo *this_leaf = this_cpu_ci->info_list;
 
+	if (loongson_sysconf.nr_nodes > 1)
+		nodes_per_package = loongson_sysconf.cores_per_package
+					/ loongson_sysconf.cores_per_node;
+
 	if (c->icache.waysize) {
 		populate_cache(dcache, this_leaf, level, CACHE_TYPE_DATA);
 		populate_cache(icache, this_leaf, level++, CACHE_TYPE_INST);