sched, cpuset: customize sched domains, core (1d3504fc) · Commits · 戴 / test

include/asm-ia64/topology.h

+1 −1

Original line number	Original line	Diff line number	Diff line
	@@ -93,7 +93,7 @@ void build_cpu_to_node_map(void);
	.cache_nice_tries = 2, \		.cache_nice_tries = 2, \
	.busy_idx = 3, \		.busy_idx = 3, \
	.idle_idx = 2, \		.idle_idx = 2, \
	.newidle_idx = 0, /* unused */ \		.newidle_idx = 2, \
	.wake_idx = 1, \		.wake_idx = 1, \
	.forkexec_idx = 1, \		.forkexec_idx = 1, \
	.flags = SD_LOAD_BALANCE \		.flags = SD_LOAD_BALANCE \

include/asm-sh/topology.h

+1 −1

Original line number	Original line	Diff line number	Diff line
	@@ -16,7 +16,7 @@
	.cache_nice_tries = 2, \		.cache_nice_tries = 2, \
	.busy_idx = 3, \		.busy_idx = 3, \
	.idle_idx = 2, \		.idle_idx = 2, \
	.newidle_idx = 0, \		.newidle_idx = 2, \
	.wake_idx = 1, \		.wake_idx = 1, \
	.forkexec_idx = 1, \		.forkexec_idx = 1, \
	.flags = SD_LOAD_BALANCE \		.flags = SD_LOAD_BALANCE \

include/asm-x86/topology.h

+1 −1

Original line number	Original line	Diff line number	Diff line
	@@ -147,7 +147,7 @@ extern unsigned long node_remap_size[];

	# define SD_CACHE_NICE_TRIES 2		# define SD_CACHE_NICE_TRIES 2
	# define SD_IDLE_IDX 2		# define SD_IDLE_IDX 2
	# define SD_NEWIDLE_IDX 0		# define SD_NEWIDLE_IDX 2
	# define SD_FORKEXEC_IDX 1		# define SD_FORKEXEC_IDX 1

	#endif		#endif

include/linux/sched.h

+22 −1

Original line number	Original line	Diff line number	Diff line
	@@ -704,6 +704,7 @@ enum cpu_idle_type {
	#define SD_POWERSAVINGS_BALANCE 256 /* Balance for power savings */		#define SD_POWERSAVINGS_BALANCE 256 /* Balance for power savings */
	#define SD_SHARE_PKG_RESOURCES 512 /* Domain members share cpu pkg resources */		#define SD_SHARE_PKG_RESOURCES 512 /* Domain members share cpu pkg resources */
	#define SD_SERIALIZE 1024 /* Only a single load balancing instance */		#define SD_SERIALIZE 1024 /* Only a single load balancing instance */
			#define SD_WAKE_IDLE_FAR 2048 /* Gain latency sacrificing cache hit */

	#define BALANCE_FOR_MC_POWER \		#define BALANCE_FOR_MC_POWER \
	(sched_smt_power_savings ? SD_POWERSAVINGS_BALANCE : 0)		(sched_smt_power_savings ? SD_POWERSAVINGS_BALANCE : 0)
	@@ -733,6 +734,24 @@ struct sched_group {
	u32 reciprocal_cpu_power;		u32 reciprocal_cpu_power;
	};		};

			enum sched_domain_level {
			SD_LV_NONE = 0,
			SD_LV_SIBLING,
			SD_LV_MC,
			SD_LV_CPU,
			SD_LV_NODE,
			SD_LV_ALLNODES,
			SD_LV_MAX
			};

			struct sched_domain_attr {
			int relax_domain_level;
			};

			#define SD_ATTR_INIT (struct sched_domain_attr) { \
			.relax_domain_level = -1, \
			}

	struct sched_domain {		struct sched_domain {
	/* These fields must be setup */		/* These fields must be setup */
	struct sched_domain parent; / top domain must be null terminated */		struct sched_domain parent; / top domain must be null terminated */
	@@ -750,6 +769,7 @@ struct sched_domain {
	unsigned int wake_idx;		unsigned int wake_idx;
	unsigned int forkexec_idx;		unsigned int forkexec_idx;
	int flags; /* See SD_* */		int flags; /* See SD_* */
			enum sched_domain_level level;

	/* Runtime fields. */		/* Runtime fields. */
	unsigned long last_balance; /* init to jiffies. units in jiffies */		unsigned long last_balance; /* init to jiffies. units in jiffies */
	@@ -789,7 +809,8 @@ struct sched_domain {
	#endif		#endif
	};		};

	extern void partition_sched_domains(int ndoms_new, cpumask_t *doms_new);		extern void partition_sched_domains(int ndoms_new, cpumask_t *doms_new,
			struct sched_domain_attr *dattr_new);
	extern int arch_reinit_sched_domains(void);		extern int arch_reinit_sched_domains(void);

	#endif /* CONFIG_SMP */		#endif /* CONFIG_SMP */

kernel/cpuset.c

+60 −1

Original line number	Original line	Diff line number	Diff line
	@@ -98,6 +98,9 @@ struct cpuset {
	/* partition number for rebuild_sched_domains() */		/* partition number for rebuild_sched_domains() */
	int pn;		int pn;

			/* for custom sched domain */
			int relax_domain_level;

	/* used for walking a cpuset heirarchy */		/* used for walking a cpuset heirarchy */
	struct list_head stack_list;		struct list_head stack_list;
	};		};
	@@ -478,6 +481,16 @@ static int cpusets_overlap(struct cpuset a, struct cpuset b)
	return cpus_intersects(a->cpus_allowed, b->cpus_allowed);		return cpus_intersects(a->cpus_allowed, b->cpus_allowed);
	}		}

			static void
			update_domain_attr(struct sched_domain_attr dattr, struct cpuset c)
			{
			if (!dattr)
			return;
			if (dattr->relax_domain_level < c->relax_domain_level)
			dattr->relax_domain_level = c->relax_domain_level;
			return;
			}

	/*		/*
	* rebuild_sched_domains()		* rebuild_sched_domains()
	*		*
	@@ -553,12 +566,14 @@ static void rebuild_sched_domains(void)
	int csn; /* how many cpuset ptrs in csa so far */		int csn; /* how many cpuset ptrs in csa so far */
	int i, j, k; /* indices for partition finding loops */		int i, j, k; /* indices for partition finding loops */
	cpumask_t doms; / resulting partition; i.e. sched domains */		cpumask_t doms; / resulting partition; i.e. sched domains */
			struct sched_domain_attr dattr; / attributes for custom domains */
	int ndoms; /* number of sched domains in result */		int ndoms; /* number of sched domains in result */
	int nslot; /* next empty doms[] cpumask_t slot */		int nslot; /* next empty doms[] cpumask_t slot */

	q = NULL;		q = NULL;
	csa = NULL;		csa = NULL;
	doms = NULL;		doms = NULL;
			dattr = NULL;

	/* Special case for the 99% of systems with one, full, sched domain */		/* Special case for the 99% of systems with one, full, sched domain */
	if (is_sched_load_balance(&top_cpuset)) {		if (is_sched_load_balance(&top_cpuset)) {
	@@ -566,6 +581,11 @@ static void rebuild_sched_domains(void)
	doms = kmalloc(sizeof(cpumask_t), GFP_KERNEL);		doms = kmalloc(sizeof(cpumask_t), GFP_KERNEL);
	if (!doms)		if (!doms)
	goto rebuild;		goto rebuild;
			dattr = kmalloc(sizeof(struct sched_domain_attr), GFP_KERNEL);
			if (dattr) {
			*dattr = SD_ATTR_INIT;
			update_domain_attr(dattr, &top_cpuset);
			}
	*doms = top_cpuset.cpus_allowed;		*doms = top_cpuset.cpus_allowed;
	goto rebuild;		goto rebuild;
	}		}
	@@ -622,6 +642,7 @@ restart:
	doms = kmalloc(ndoms * sizeof(cpumask_t), GFP_KERNEL);		doms = kmalloc(ndoms * sizeof(cpumask_t), GFP_KERNEL);
	if (!doms)		if (!doms)
	goto rebuild;		goto rebuild;
			dattr = kmalloc(ndoms * sizeof(struct sched_domain_attr), GFP_KERNEL);

	for (nslot = 0, i = 0; i < csn; i++) {		for (nslot = 0, i = 0; i < csn; i++) {
	struct cpuset *a = csa[i];		struct cpuset *a = csa[i];
	@@ -644,12 +665,15 @@ restart:
	}		}

	cpus_clear(*dp);		cpus_clear(*dp);
			if (dattr)
			*(dattr + nslot) = SD_ATTR_INIT;
	for (j = i; j < csn; j++) {		for (j = i; j < csn; j++) {
	struct cpuset *b = csa[j];		struct cpuset *b = csa[j];

	if (apn == b->pn) {		if (apn == b->pn) {
	cpus_or(dp, dp, b->cpus_allowed);		cpus_or(dp, dp, b->cpus_allowed);
	b->pn = -1;		b->pn = -1;
			update_domain_attr(dattr, b);
	}		}
	}		}
	nslot++;		nslot++;
	@@ -660,7 +684,7 @@ restart:
	rebuild:		rebuild:
	/* Have scheduler rebuild sched domains */		/* Have scheduler rebuild sched domains */
	get_online_cpus();		get_online_cpus();
	partition_sched_domains(ndoms, doms);		partition_sched_domains(ndoms, doms, dattr);
	put_online_cpus();		put_online_cpus();

	done:		done:
	@@ -668,6 +692,7 @@ done:
	kfifo_free(q);		kfifo_free(q);
	kfree(csa);		kfree(csa);
	/* Don't kfree(doms) -- partition_sched_domains() does that. */		/* Don't kfree(doms) -- partition_sched_domains() does that. */
			/* Don't kfree(dattr) -- partition_sched_domains() does that. */
	}		}

	static inline int started_after_time(struct task_struct *t1,		static inline int started_after_time(struct task_struct *t1,
	@@ -1011,6 +1036,21 @@ static int update_memory_pressure_enabled(struct cpuset cs, char buf)
	return 0;		return 0;
	}		}

			static int update_relax_domain_level(struct cpuset cs, char buf)
			{
			int val = simple_strtol(buf, NULL, 10);

			if (val < 0)
			val = -1;

			if (val != cs->relax_domain_level) {
			cs->relax_domain_level = val;
			rebuild_sched_domains();
			}

			return 0;
			}

	/*		/*
	* update_flag - read a 0 or a 1 in a file and update associated flag		* update_flag - read a 0 or a 1 in a file and update associated flag
	* bit: the bit to update (CS_CPU_EXCLUSIVE, CS_MEM_EXCLUSIVE,		* bit: the bit to update (CS_CPU_EXCLUSIVE, CS_MEM_EXCLUSIVE,
	@@ -1202,6 +1242,7 @@ typedef enum {
	FILE_CPU_EXCLUSIVE,		FILE_CPU_EXCLUSIVE,
	FILE_MEM_EXCLUSIVE,		FILE_MEM_EXCLUSIVE,
	FILE_SCHED_LOAD_BALANCE,		FILE_SCHED_LOAD_BALANCE,
			FILE_SCHED_RELAX_DOMAIN_LEVEL,
	FILE_MEMORY_PRESSURE_ENABLED,		FILE_MEMORY_PRESSURE_ENABLED,
	FILE_MEMORY_PRESSURE,		FILE_MEMORY_PRESSURE,
	FILE_SPREAD_PAGE,		FILE_SPREAD_PAGE,
	@@ -1256,6 +1297,9 @@ static ssize_t cpuset_common_file_write(struct cgroup *cont,
	case FILE_SCHED_LOAD_BALANCE:		case FILE_SCHED_LOAD_BALANCE:
	retval = update_flag(CS_SCHED_LOAD_BALANCE, cs, buffer);		retval = update_flag(CS_SCHED_LOAD_BALANCE, cs, buffer);
	break;		break;
			case FILE_SCHED_RELAX_DOMAIN_LEVEL:
			retval = update_relax_domain_level(cs, buffer);
			break;
	case FILE_MEMORY_MIGRATE:		case FILE_MEMORY_MIGRATE:
	retval = update_flag(CS_MEMORY_MIGRATE, cs, buffer);		retval = update_flag(CS_MEMORY_MIGRATE, cs, buffer);
	break;		break;
	@@ -1354,6 +1398,9 @@ static ssize_t cpuset_common_file_read(struct cgroup *cont,
	case FILE_SCHED_LOAD_BALANCE:		case FILE_SCHED_LOAD_BALANCE:
	*s++ = is_sched_load_balance(cs) ? '1' : '0';		*s++ = is_sched_load_balance(cs) ? '1' : '0';
	break;		break;
			case FILE_SCHED_RELAX_DOMAIN_LEVEL:
			s += sprintf(s, "%d", cs->relax_domain_level);
			break;
	case FILE_MEMORY_MIGRATE:		case FILE_MEMORY_MIGRATE:
	*s++ = is_memory_migrate(cs) ? '1' : '0';		*s++ = is_memory_migrate(cs) ? '1' : '0';
	break;		break;
	@@ -1424,6 +1471,13 @@ static struct cftype cft_sched_load_balance = {
	.private = FILE_SCHED_LOAD_BALANCE,		.private = FILE_SCHED_LOAD_BALANCE,
	};		};

			static struct cftype cft_sched_relax_domain_level = {
			.name = "sched_relax_domain_level",
			.read = cpuset_common_file_read,
			.write = cpuset_common_file_write,
			.private = FILE_SCHED_RELAX_DOMAIN_LEVEL,
			};

	static struct cftype cft_memory_migrate = {		static struct cftype cft_memory_migrate = {
	.name = "memory_migrate",		.name = "memory_migrate",
	.read = cpuset_common_file_read,		.read = cpuset_common_file_read,
	@@ -1475,6 +1529,9 @@ static int cpuset_populate(struct cgroup_subsys ss, struct cgroup cont)
	return err;		return err;
	if ((err = cgroup_add_file(cont, ss, &cft_sched_load_balance)) < 0)		if ((err = cgroup_add_file(cont, ss, &cft_sched_load_balance)) < 0)
	return err;		return err;
			if ((err = cgroup_add_file(cont, ss,
			&cft_sched_relax_domain_level)) < 0)
			return err;
	if ((err = cgroup_add_file(cont, ss, &cft_memory_pressure)) < 0)		if ((err = cgroup_add_file(cont, ss, &cft_memory_pressure)) < 0)
	return err;		return err;
	if ((err = cgroup_add_file(cont, ss, &cft_spread_page)) < 0)		if ((err = cgroup_add_file(cont, ss, &cft_spread_page)) < 0)
	@@ -1559,6 +1616,7 @@ static struct cgroup_subsys_state *cpuset_create(
	nodes_clear(cs->mems_allowed);		nodes_clear(cs->mems_allowed);
	cs->mems_generation = cpuset_mems_generation++;		cs->mems_generation = cpuset_mems_generation++;
	fmeter_init(&cs->fmeter);		fmeter_init(&cs->fmeter);
			cs->relax_domain_level = -1;

	cs->parent = parent;		cs->parent = parent;
	number_of_cpusets++;		number_of_cpusets++;
	@@ -1631,6 +1689,7 @@ int __init cpuset_init(void)
	fmeter_init(&top_cpuset.fmeter);		fmeter_init(&top_cpuset.fmeter);
	top_cpuset.mems_generation = cpuset_mems_generation++;		top_cpuset.mems_generation = cpuset_mems_generation++;
	set_bit(CS_SCHED_LOAD_BALANCE, &top_cpuset.flags);		set_bit(CS_SCHED_LOAD_BALANCE, &top_cpuset.flags);
			top_cpuset.relax_domain_level = -1;

	err = register_filesystem(&cpuset_fs_type);		err = register_filesystem(&cpuset_fs_type);
	if (err < 0)		if (err < 0)

Admin message