[PATCH] sched: dynamic config sd_flags if described in DT

[Qing Wang]

From: Wang Qing <wangqing@xxxxxxxx>

The device tree can describe cache topology by "next-level-cache",
Prefer configuring SD_SHARE_PKG_RESOURCES from DT instead of default value.

Signed-off-by: Wang Qing <wangqing@xxxxxxxx>
---
 arch/arm/kernel/smp.c         |  1 +
 arch/arm64/kernel/smp.c       |  1 +
 arch/riscv/kernel/smpboot.c   |  1 +
 drivers/base/arch_topology.c  | 59 +++++++++++++++++++++++++++++++++++++++++++
 include/linux/arch_topology.h |  2 ++
 kernel/sched/topology.c       | 10 ++++++++
 6 files changed, 74 insertions(+)

diff --git a/arch/arm/kernel/smp.c b/arch/arm/kernel/smp.c
index 73fc645..62bbd9a
--- a/arch/arm/kernel/smp.c
+++ b/arch/arm/kernel/smp.c
@@ -508,6 +508,7 @@ void __init smp_prepare_cpus(unsigned int max_cpus)
 	unsigned int ncores = num_possible_cpus();
 
 	init_cpu_topology();
+	init_cpu_cache_topology();
 
 	smp_store_cpu_info(smp_processor_id());
 
diff --git a/arch/arm64/kernel/smp.c b/arch/arm64/kernel/smp.c
index 27df5c1..94cf649
--- a/arch/arm64/kernel/smp.c
+++ b/arch/arm64/kernel/smp.c
@@ -723,6 +723,7 @@ void __init smp_prepare_cpus(unsigned int max_cpus)
 	unsigned int this_cpu;
 
 	init_cpu_topology();
+	init_cpu_cache_topology();
 
 	this_cpu = smp_processor_id();
 	store_cpu_topology(this_cpu);
diff --git a/arch/riscv/kernel/smpboot.c b/arch/riscv/kernel/smpboot.c
index 622f226..4f5a8b7
--- a/arch/riscv/kernel/smpboot.c
+++ b/arch/riscv/kernel/smpboot.c
@@ -41,6 +41,7 @@ static DECLARE_COMPLETION(cpu_running);
 void __init smp_prepare_boot_cpu(void)
 {
 	init_cpu_topology();
+	init_cpu_cache_topology();
 #ifdef CONFIG_RISCV_ERRATA_ALTERNATIVE
 	apply_boot_alternatives();
 #endif
diff --git a/drivers/base/arch_topology.c b/drivers/base/arch_topology.c
index 9761541..127f540
--- a/drivers/base/arch_topology.c
+++ b/drivers/base/arch_topology.c
@@ -609,6 +609,65 @@ static int __init parse_dt_topology(void)
 #endif
 
 /*
+ * cpu cache topology table
+ */
+#define MAX_CACHE_LEVEL 7
+struct device_node *cache_topology[NR_CPUS][MAX_CACHE_LEVEL];
+
+void init_cpu_cache_topology(void)
+{
+	struct device_node *node_cpu, *node_cache;
+	int cpu, level;
+
+	for_each_possible_cpu(cpu) {
+		node_cpu = of_get_cpu_node(cpu, NULL);
+		if (!node_cpu)
+			continue;
+
+		level = 0;
+		node_cache = node_cpu;
+		while (level < MAX_CACHE_LEVEL) {
+			node_cache = of_parse_phandle(node_cache, "next-level-cache", 0);
+			if (!node_cache)
+				break;
+
+			cache_topology[cpu][level++] = node_cache;
+		}
+		of_node_put(node_cpu);
+	}
+}
+
+int cpus_share_self_cache(const struct cpumask *cpu_map)
+{
+	int cache_level, cpu_id;
+	int first, last;
+	int id = cpumask_first(cpu_map);
+	int size = cpumask_weight(cpu_map);
+
+	for (cache_level = 0; cache_level < MAX_CACHE_LEVEL; cache_level++) {
+		if (!cache_topology[cpu][cache_level])
+			return -1;
+
+		first = -1;
+		last = id;
+		for (cpu_id = 0; cpu_id < NR_CPUS; cpu_id++) {
+			if (cache_topology[cpu][cache_level] == cache_topology[cpu_id][cache_level]) {
+				if (cpu_id < id || cpu_id >= id + size)
+					return 0;
+
+				first = (first == -1)?cpu_id:first;
+				last = cpu_id;
+			}
+		}
+
+		if (first == id && last == id + size)
+			return 1;
+	}
+
+	return 0;
+}
+
+/*
  * cpu topology table
  */
 struct cpu_topology cpu_topology[NR_CPUS];
diff --git a/include/linux/arch_topology.h b/include/linux/arch_topology.h
index cce6136b..862e584
--- a/include/linux/arch_topology.h
+++ b/include/linux/arch_topology.h
@@ -82,6 +82,8 @@ extern struct cpu_topology cpu_topology[NR_CPUS];
 #define topology_cluster_cpumask(cpu)	(&cpu_topology[cpu].cluster_sibling)
 #define topology_llc_cpumask(cpu)	(&cpu_topology[cpu].llc_sibling)
 void init_cpu_topology(void);
+void init_cpu_cache_topology(void);
+int cpus_share_self_cache(const struct cpumask *cpu_map);
 void store_cpu_topology(unsigned int cpuid);
 const struct cpumask *cpu_coregroup_mask(int cpu);
 const struct cpumask *cpu_clustergroup_mask(int cpu);
diff --git a/kernel/sched/topology.c b/kernel/sched/topology.c
index d201a70..8264e2d
--- a/kernel/sched/topology.c
+++ b/kernel/sched/topology.c
@@ -1527,6 +1527,7 @@ sd_init(struct sched_domain_topology_level *tl,
 	struct sched_domain *sd = *per_cpu_ptr(sdd->sd, cpu);
 	int sd_id, sd_weight, sd_flags = 0;
 	struct cpumask *sd_span;
+	int ret;
 
 #ifdef CONFIG_NUMA
 	/*
@@ -1539,6 +1540,15 @@ sd_init(struct sched_domain_topology_level *tl,
 
 	if (tl->sd_flags)
 		sd_flags = (*tl->sd_flags)();
+
+#ifdef CONFIG_GENERIC_ARCH_TOPOLOGY
+	ret = cpus_share_self_cache(cpu_map);
+	if (ret == 1)
+		sd_flags |= SD_SHARE_PKG_RESOURCES;
+	else if (ret == 0)
+		sd_flags &= ~SD_SHARE_PKG_RESOURCES;
+#endif
+
 	if (WARN_ONCE(sd_flags & ~TOPOLOGY_SD_FLAGS,
 			"wrong sd_flags in topology description\n"))
 		sd_flags &= TOPOLOGY_SD_FLAGS;
-- 
2.7.4