1 files changed, 164 insertions, 45 deletions

diff --git a/drivers/base/cacheinfo.c b/drivers/base/cacheinfo.c
index cbae8be1fe52..613410705a47 100644
--- a/drivers/base/cacheinfo.c
+++ b/drivers/base/cacheinfo.c
@@ -8,6 +8,7 @@
 #define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
 
 #include <linux/acpi.h>
+#include <linux/bitfield.h>
 #include <linux/bitops.h>
 #include <linux/cacheinfo.h>
 #include <linux/compiler.h>
@@ -58,7 +59,7 @@ bool last_level_cache_is_valid(unsigned int cpu)
 {
 	struct cacheinfo *llc;
 
-	if (!cache_leaves(cpu))
+	if (!cache_leaves(cpu) || !per_cpu_cacheinfo(cpu))
 		return false;
 
 	llc = per_cpu_cacheinfo_idx(cpu, cache_leaves(cpu) - 1);
@@ -183,6 +184,54 @@ static bool cache_node_is_unified(struct cacheinfo *this_leaf,
 	return of_property_read_bool(np, "cache-unified");
 }
 
+static bool match_cache_node(struct device_node *cpu,
+			     const struct device_node *cache_node)
+{
+	struct device_node *prev, *cache = of_find_next_cache_node(cpu);
+
+	while (cache) {
+		if (cache == cache_node) {
+			of_node_put(cache);
+			return true;
+		}
+
+		prev = cache;
+		cache = of_find_next_cache_node(cache);
+		of_node_put(prev);
+	}
+
+	return false;
+}
+
+#ifndef arch_compact_of_hwid
+#define arch_compact_of_hwid(_x)	(_x)
+#endif
+
+static void cache_of_set_id(struct cacheinfo *this_leaf,
+			    struct device_node *cache_node)
+{
+	struct device_node *cpu;
+	u32 min_id = ~0;
+
+	for_each_of_cpu_node(cpu) {
+		u64 id = of_get_cpu_hwid(cpu, 0);
+
+		id = arch_compact_of_hwid(id);
+		if (FIELD_GET(GENMASK_ULL(63, 32), id)) {
+			of_node_put(cpu);
+			return;
+		}
+
+		if (match_cache_node(cpu, cache_node))
+			min_id = min(min_id, id);
+	}
+
+	if (min_id != ~0) {
+		this_leaf->id = min_id;
+		this_leaf->attributes |= CACHE_ID;
+	}
+}
+
 static void cache_of_set_props(struct cacheinfo *this_leaf,
 			       struct device_node *np)
 {
@@ -198,33 +247,29 @@ static void cache_of_set_props(struct cacheinfo *this_leaf,
 	cache_get_line_size(this_leaf, np);
 	cache_nr_sets(this_leaf, np);
 	cache_associativity(this_leaf);
+	cache_of_set_id(this_leaf, np);
 }
 
 static int cache_setup_of_node(unsigned int cpu)
 {
-	struct device_node *np, *prev;
 	struct cacheinfo *this_leaf;
 	unsigned int index = 0;
 
-	np = of_cpu_device_node_get(cpu);
+	struct device_node *np __free(device_node) = of_cpu_device_node_get(cpu);
 	if (!np) {
 		pr_err("Failed to find cpu%d device node\n", cpu);
 		return -ENOENT;
 	}
 
 	if (!of_check_cache_nodes(np)) {
-		of_node_put(np);
 		return -ENOENT;
 	}
 
-	prev = np;
-
 	while (index < cache_leaves(cpu)) {
 		this_leaf = per_cpu_cacheinfo_idx(cpu, index);
 		if (this_leaf->level != 1) {
+			struct device_node *prev __free(device_node) = np;
 			np = of_find_next_cache_node(np);
-			of_node_put(prev);
-			prev = np;
 			if (!np)
 				break;
 		}
@@ -233,8 +278,6 @@ static int cache_setup_of_node(unsigned int cpu)
 		index++;
 	}
 
-	of_node_put(np);
-
 	if (index != cache_leaves(cpu)) /* not all OF nodes populated */
 		return -ENOENT;
 
@@ -243,17 +286,14 @@ static int cache_setup_of_node(unsigned int cpu)
 
 static bool of_check_cache_nodes(struct device_node *np)
 {
-	struct device_node *next;
-
 	if (of_property_present(np, "cache-size")   ||
 	    of_property_present(np, "i-cache-size") ||
 	    of_property_present(np, "d-cache-size") ||
 	    of_property_present(np, "cache-unified"))
 		return true;
 
-	next = of_find_next_cache_node(np);
+	struct device_node *next __free(device_node) = of_find_next_cache_node(np);
 	if (next) {
-		of_node_put(next);
 		return true;
 	}
 
@@ -264,11 +304,11 @@ static int of_count_cache_leaves(struct device_node *np)
 {
 	unsigned int leaves = 0;
 
-	if (of_property_read_bool(np, "cache-size"))
+	if (of_property_present(np, "cache-size"))
 		++leaves;
-	if (of_property_read_bool(np, "i-cache-size"))
+	if (of_property_present(np, "i-cache-size"))
 		++leaves;
-	if (of_property_read_bool(np, "d-cache-size"))
+	if (of_property_present(np, "d-cache-size"))
 		++leaves;
 
 	if (!leaves) {
@@ -287,12 +327,10 @@ static int of_count_cache_leaves(struct device_node *np)
 int init_of_cache_level(unsigned int cpu)
 {
 	struct cpu_cacheinfo *this_cpu_ci = get_cpu_cacheinfo(cpu);
-	struct device_node *np = of_cpu_device_node_get(cpu);
-	struct device_node *prev = NULL;
+	struct device_node *np __free(device_node) = of_cpu_device_node_get(cpu);
 	unsigned int levels = 0, leaves, level;
 
 	if (!of_check_cache_nodes(np)) {
-		of_node_put(np);
 		return -ENOENT;
 	}
 
@@ -300,30 +338,27 @@ int init_of_cache_level(unsigned int cpu)
 	if (leaves > 0)
 		levels = 1;
 
-	prev = np;
-	while ((np = of_find_next_cache_node(np))) {
-		of_node_put(prev);
-		prev = np;
+	while (1) {
+		struct device_node *prev __free(device_node) = np;
+		np = of_find_next_cache_node(np);
+		if (!np)
+			break;
+
 		if (!of_device_is_compatible(np, "cache"))
-			goto err_out;
+			return -EINVAL;
 		if (of_property_read_u32(np, "cache-level", &level))
-			goto err_out;
+			return -EINVAL;
 		if (level <= levels)
-			goto err_out;
+			return -EINVAL;
 
 		leaves += of_count_cache_leaves(np);
 		levels = level;
 	}
 
-	of_node_put(np);
 	this_cpu_ci->num_levels = levels;
 	this_cpu_ci->num_leaves = leaves;
 
 	return 0;
-
-err_out:
-	of_node_put(np);
-	return -EINVAL;
 }
 
 #else
@@ -382,9 +417,7 @@ static int cache_shared_cpu_map_setup(unsigned int cpu)
 
 		cpumask_set_cpu(cpu, &this_leaf->shared_cpu_map);
 		for_each_online_cpu(i) {
-			struct cpu_cacheinfo *sib_cpu_ci = get_cpu_cacheinfo(i);
-
-			if (i == cpu || !sib_cpu_ci->info_list)
+			if (i == cpu || !per_cpu_cacheinfo(i))
 				continue;/* skip if itself or no cacheinfo */
 			for (sib_index = 0; sib_index < cache_leaves(i); sib_index++) {
 				sib_leaf = per_cpu_cacheinfo_idx(i, sib_index);
@@ -424,10 +457,7 @@ static void cache_shared_cpu_map_remove(unsigned int cpu)
 	for (index = 0; index < cache_leaves(cpu); index++) {
 		this_leaf = per_cpu_cacheinfo_idx(cpu, index);
 		for_each_cpu(sibling, &this_leaf->shared_cpu_map) {
-			struct cpu_cacheinfo *sib_cpu_ci =
-						get_cpu_cacheinfo(sibling);
-
-			if (sibling == cpu || !sib_cpu_ci->info_list)
+			if (sibling == cpu || !per_cpu_cacheinfo(sibling))
 				continue;/* skip if itself or no cacheinfo */
 
 			for (sib_index = 0; sib_index < cache_leaves(sibling); sib_index++) {
@@ -478,11 +508,9 @@ int __weak populate_cache_leaves(unsigned int cpu)
 	return -ENOENT;
 }
 
-static inline
-int allocate_cache_info(int cpu)
+static inline int allocate_cache_info(int cpu)
 {
-	per_cpu_cacheinfo(cpu) = kcalloc(cache_leaves(cpu),
-					 sizeof(struct cacheinfo), GFP_ATOMIC);
+	per_cpu_cacheinfo(cpu) = kcalloc(cache_leaves(cpu), sizeof(struct cacheinfo), GFP_ATOMIC);
 	if (!per_cpu_cacheinfo(cpu)) {
 		cache_leaves(cpu) = 0;
 		return -ENOMEM;
@@ -554,7 +582,11 @@ static inline int init_level_allocate_ci(unsigned int cpu)
 	 */
 	ci_cacheinfo(cpu)->early_ci_levels = false;
 
-	if (cache_leaves(cpu) <= early_leaves)
+	/*
+	 * Some architectures (e.g., x86) do not use early initialization.
+	 * Allocate memory now in such case.
+	 */
+	if (cache_leaves(cpu) <= early_leaves && per_cpu_cacheinfo(cpu))
 		return 0;
 
 	kfree(per_cpu_cacheinfo(cpu));
@@ -898,24 +930,111 @@ err:
 	return rc;
 }
 
+static unsigned int cpu_map_shared_cache(bool online, unsigned int cpu,
+					 cpumask_t **map)
+{
+	struct cacheinfo *llc, *sib_llc;
+	unsigned int sibling;
+
+	if (!last_level_cache_is_valid(cpu))
+		return 0;
+
+	llc = per_cpu_cacheinfo_idx(cpu, cache_leaves(cpu) - 1);
+
+	if (llc->type != CACHE_TYPE_DATA && llc->type != CACHE_TYPE_UNIFIED)
+		return 0;
+
+	if (online) {
+		*map = &llc->shared_cpu_map;
+		return cpumask_weight(*map);
+	}
+
+	/* shared_cpu_map of offlined CPU will be cleared, so use sibling map */
+	for_each_cpu(sibling, &llc->shared_cpu_map) {
+		if (sibling == cpu || !last_level_cache_is_valid(sibling))
+			continue;
+		sib_llc = per_cpu_cacheinfo_idx(sibling, cache_leaves(sibling) - 1);
+		*map = &sib_llc->shared_cpu_map;
+		return cpumask_weight(*map);
+	}
+
+	return 0;
+}
+
+/*
+ * Calculate the size of the per-CPU data cache slice.  This can be
+ * used to estimate the size of the data cache slice that can be used
+ * by one CPU under ideal circumstances.  UNIFIED caches are counted
+ * in addition to DATA caches.  So, please consider code cache usage
+ * when use the result.
+ *
+ * Because the cache inclusive/non-inclusive information isn't
+ * available, we just use the size of the per-CPU slice of LLC to make
+ * the result more predictable across architectures.
+ */
+static void update_per_cpu_data_slice_size_cpu(unsigned int cpu)
+{
+	struct cpu_cacheinfo *ci;
+	struct cacheinfo *llc;
+	unsigned int nr_shared;
+
+	if (!last_level_cache_is_valid(cpu))
+		return;
+
+	ci = ci_cacheinfo(cpu);
+	llc = per_cpu_cacheinfo_idx(cpu, cache_leaves(cpu) - 1);
+
+	if (llc->type != CACHE_TYPE_DATA && llc->type != CACHE_TYPE_UNIFIED)
+		return;
+
+	nr_shared = cpumask_weight(&llc->shared_cpu_map);
+	if (nr_shared)
+		ci->per_cpu_data_slice_size = llc->size / nr_shared;
+}
+
+static void update_per_cpu_data_slice_size(bool cpu_online, unsigned int cpu,
+					   cpumask_t *cpu_map)
+{
+	unsigned int icpu;
+
+	for_each_cpu(icpu, cpu_map) {
+		if (!cpu_online && icpu == cpu)
+			continue;
+		update_per_cpu_data_slice_size_cpu(icpu);
+		setup_pcp_cacheinfo(icpu);
+	}
+}
+
 static int cacheinfo_cpu_online(unsigned int cpu)
 {
 	int rc = detect_cache_attributes(cpu);
+	cpumask_t *cpu_map;
 
 	if (rc)
 		return rc;
 	rc = cache_add_dev(cpu);
 	if (rc)
-		free_cache_attributes(cpu);
+		goto err;
+	if (cpu_map_shared_cache(true, cpu, &cpu_map))
+		update_per_cpu_data_slice_size(true, cpu, cpu_map);
+	return 0;
+err:
+	free_cache_attributes(cpu);
 	return rc;
 }
 
 static int cacheinfo_cpu_pre_down(unsigned int cpu)
 {
+	cpumask_t *cpu_map;
+	unsigned int nr_shared;
+
+	nr_shared = cpu_map_shared_cache(false, cpu, &cpu_map);
 	if (cpumask_test_and_clear_cpu(cpu, &cache_dev_map))
 		cpu_cache_sysfs_exit(cpu);
 
 	free_cache_attributes(cpu);
+	if (nr_shared > 1)
+		update_per_cpu_data_slice_size(false, cpu, cpu_map);
 	return 0;
 }