summaryrefslogtreecommitdiffstats
path: root/kernel
diff options
context:
space:
mode:
Diffstat (limited to 'kernel')
-rw-r--r--kernel/bpf/local_storage.c5
-rw-r--r--kernel/bpf/sockmap.c91
-rw-r--r--kernel/bpf/verifier.c10
-rw-r--r--kernel/bpf/xskmap.c10
-rw-r--r--kernel/cgroup/cgroup.c25
-rw-r--r--kernel/dma/Kconfig3
-rw-r--r--kernel/events/core.c17
-rw-r--r--kernel/locking/test-ww_mutex.c10
-rw-r--r--kernel/power/suspend.c6
-rw-r--r--kernel/sched/core.c2
-rw-r--r--kernel/sched/deadline.c2
-rw-r--r--kernel/sched/fair.c104
-rw-r--r--kernel/sched/sched.h3
13 files changed, 222 insertions, 66 deletions
diff --git a/kernel/bpf/local_storage.c b/kernel/bpf/local_storage.c
index 22ad967d1e5f..830d7f095748 100644
--- a/kernel/bpf/local_storage.c
+++ b/kernel/bpf/local_storage.c
@@ -129,7 +129,7 @@ static int cgroup_storage_update_elem(struct bpf_map *map, void *_key,
struct bpf_cgroup_storage *storage;
struct bpf_storage_buffer *new;
- if (flags & BPF_NOEXIST)
+ if (flags != BPF_ANY && flags != BPF_EXIST)
return -EINVAL;
storage = cgroup_storage_lookup((struct bpf_cgroup_storage_map *)map,
@@ -195,6 +195,9 @@ static struct bpf_map *cgroup_storage_map_alloc(union bpf_attr *attr)
if (attr->key_size != sizeof(struct bpf_cgroup_storage_key))
return ERR_PTR(-EINVAL);
+ if (attr->value_size == 0)
+ return ERR_PTR(-EINVAL);
+
if (attr->value_size > PAGE_SIZE)
return ERR_PTR(-E2BIG);
diff --git a/kernel/bpf/sockmap.c b/kernel/bpf/sockmap.c
index 488ef9663c01..0a0f2ec75370 100644
--- a/kernel/bpf/sockmap.c
+++ b/kernel/bpf/sockmap.c
@@ -132,6 +132,7 @@ struct smap_psock {
struct work_struct gc_work;
struct proto *sk_proto;
+ void (*save_unhash)(struct sock *sk);
void (*save_close)(struct sock *sk, long timeout);
void (*save_data_ready)(struct sock *sk);
void (*save_write_space)(struct sock *sk);
@@ -143,6 +144,7 @@ static int bpf_tcp_recvmsg(struct sock *sk, struct msghdr *msg, size_t len,
static int bpf_tcp_sendmsg(struct sock *sk, struct msghdr *msg, size_t size);
static int bpf_tcp_sendpage(struct sock *sk, struct page *page,
int offset, size_t size, int flags);
+static void bpf_tcp_unhash(struct sock *sk);
static void bpf_tcp_close(struct sock *sk, long timeout);
static inline struct smap_psock *smap_psock_sk(const struct sock *sk)
@@ -184,6 +186,7 @@ static void build_protos(struct proto prot[SOCKMAP_NUM_CONFIGS],
struct proto *base)
{
prot[SOCKMAP_BASE] = *base;
+ prot[SOCKMAP_BASE].unhash = bpf_tcp_unhash;
prot[SOCKMAP_BASE].close = bpf_tcp_close;
prot[SOCKMAP_BASE].recvmsg = bpf_tcp_recvmsg;
prot[SOCKMAP_BASE].stream_memory_read = bpf_tcp_stream_read;
@@ -217,6 +220,7 @@ static int bpf_tcp_init(struct sock *sk)
return -EBUSY;
}
+ psock->save_unhash = sk->sk_prot->unhash;
psock->save_close = sk->sk_prot->close;
psock->sk_proto = sk->sk_prot;
@@ -305,30 +309,12 @@ static struct smap_psock_map_entry *psock_map_pop(struct sock *sk,
return e;
}
-static void bpf_tcp_close(struct sock *sk, long timeout)
+static void bpf_tcp_remove(struct sock *sk, struct smap_psock *psock)
{
- void (*close_fun)(struct sock *sk, long timeout);
struct smap_psock_map_entry *e;
struct sk_msg_buff *md, *mtmp;
- struct smap_psock *psock;
struct sock *osk;
- lock_sock(sk);
- rcu_read_lock();
- psock = smap_psock_sk(sk);
- if (unlikely(!psock)) {
- rcu_read_unlock();
- release_sock(sk);
- return sk->sk_prot->close(sk, timeout);
- }
-
- /* The psock may be destroyed anytime after exiting the RCU critial
- * section so by the time we use close_fun the psock may no longer
- * be valid. However, bpf_tcp_close is called with the sock lock
- * held so the close hook and sk are still valid.
- */
- close_fun = psock->save_close;
-
if (psock->cork) {
free_start_sg(psock->sock, psock->cork, true);
kfree(psock->cork);
@@ -379,6 +365,42 @@ static void bpf_tcp_close(struct sock *sk, long timeout)
kfree(e);
e = psock_map_pop(sk, psock);
}
+}
+
+static void bpf_tcp_unhash(struct sock *sk)
+{
+ void (*unhash_fun)(struct sock *sk);
+ struct smap_psock *psock;
+
+ rcu_read_lock();
+ psock = smap_psock_sk(sk);
+ if (unlikely(!psock)) {
+ rcu_read_unlock();
+ if (sk->sk_prot->unhash)
+ sk->sk_prot->unhash(sk);
+ return;
+ }
+ unhash_fun = psock->save_unhash;
+ bpf_tcp_remove(sk, psock);
+ rcu_read_unlock();
+ unhash_fun(sk);
+}
+
+static void bpf_tcp_close(struct sock *sk, long timeout)
+{
+ void (*close_fun)(struct sock *sk, long timeout);
+ struct smap_psock *psock;
+
+ lock_sock(sk);
+ rcu_read_lock();
+ psock = smap_psock_sk(sk);
+ if (unlikely(!psock)) {
+ rcu_read_unlock();
+ release_sock(sk);
+ return sk->sk_prot->close(sk, timeout);
+ }
+ close_fun = psock->save_close;
+ bpf_tcp_remove(sk, psock);
rcu_read_unlock();
release_sock(sk);
close_fun(sk, timeout);
@@ -2097,8 +2119,12 @@ static int sock_map_update_elem(struct bpf_map *map,
return -EINVAL;
}
+ /* ULPs are currently supported only for TCP sockets in ESTABLISHED
+ * state.
+ */
if (skops.sk->sk_type != SOCK_STREAM ||
- skops.sk->sk_protocol != IPPROTO_TCP) {
+ skops.sk->sk_protocol != IPPROTO_TCP ||
+ skops.sk->sk_state != TCP_ESTABLISHED) {
fput(socket->file);
return -EOPNOTSUPP;
}
@@ -2453,6 +2479,16 @@ static int sock_hash_update_elem(struct bpf_map *map,
return -EINVAL;
}
+ /* ULPs are currently supported only for TCP sockets in ESTABLISHED
+ * state.
+ */
+ if (skops.sk->sk_type != SOCK_STREAM ||
+ skops.sk->sk_protocol != IPPROTO_TCP ||
+ skops.sk->sk_state != TCP_ESTABLISHED) {
+ fput(socket->file);
+ return -EOPNOTSUPP;
+ }
+
lock_sock(skops.sk);
preempt_disable();
rcu_read_lock();
@@ -2543,10 +2579,22 @@ const struct bpf_map_ops sock_hash_ops = {
.map_check_btf = map_check_no_btf,
};
+static bool bpf_is_valid_sock_op(struct bpf_sock_ops_kern *ops)
+{
+ return ops->op == BPF_SOCK_OPS_PASSIVE_ESTABLISHED_CB ||
+ ops->op == BPF_SOCK_OPS_ACTIVE_ESTABLISHED_CB;
+}
BPF_CALL_4(bpf_sock_map_update, struct bpf_sock_ops_kern *, bpf_sock,
struct bpf_map *, map, void *, key, u64, flags)
{
WARN_ON_ONCE(!rcu_read_lock_held());
+
+ /* ULPs are currently supported only for TCP sockets in ESTABLISHED
+ * state. This checks that the sock ops triggering the update is
+ * one indicating we are (or will be soon) in an ESTABLISHED state.
+ */
+ if (!bpf_is_valid_sock_op(bpf_sock))
+ return -EOPNOTSUPP;
return sock_map_ctx_update_elem(bpf_sock, map, key, flags);
}
@@ -2565,6 +2613,9 @@ BPF_CALL_4(bpf_sock_hash_update, struct bpf_sock_ops_kern *, bpf_sock,
struct bpf_map *, map, void *, key, u64, flags)
{
WARN_ON_ONCE(!rcu_read_lock_held());
+
+ if (!bpf_is_valid_sock_op(bpf_sock))
+ return -EOPNOTSUPP;
return sock_hash_ctx_update_elem(bpf_sock, map, key, flags);
}
diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c
index bb07e74b34a2..465952a8e465 100644
--- a/kernel/bpf/verifier.c
+++ b/kernel/bpf/verifier.c
@@ -2896,6 +2896,15 @@ static int adjust_scalar_min_max_vals(struct bpf_verifier_env *env,
u64 umin_val, umax_val;
u64 insn_bitness = (BPF_CLASS(insn->code) == BPF_ALU64) ? 64 : 32;
+ if (insn_bitness == 32) {
+ /* Relevant for 32-bit RSH: Information can propagate towards
+ * LSB, so it isn't sufficient to only truncate the output to
+ * 32 bits.
+ */
+ coerce_reg_to_size(dst_reg, 4);
+ coerce_reg_to_size(&src_reg, 4);
+ }
+
smin_val = src_reg.smin_value;
smax_val = src_reg.smax_value;
umin_val = src_reg.umin_value;
@@ -3131,7 +3140,6 @@ static int adjust_scalar_min_max_vals(struct bpf_verifier_env *env,
if (BPF_CLASS(insn->code) != BPF_ALU64) {
/* 32-bit ALU ops are (32,32)->32 */
coerce_reg_to_size(dst_reg, 4);
- coerce_reg_to_size(&src_reg, 4);
}
__reg_deduce_bounds(dst_reg);
diff --git a/kernel/bpf/xskmap.c b/kernel/bpf/xskmap.c
index 9f8463afda9c..47147c9e184d 100644
--- a/kernel/bpf/xskmap.c
+++ b/kernel/bpf/xskmap.c
@@ -192,11 +192,8 @@ static int xsk_map_update_elem(struct bpf_map *map, void *key, void *value,
sock_hold(sock->sk);
old_xs = xchg(&m->xsk_map[i], xs);
- if (old_xs) {
- /* Make sure we've flushed everything. */
- synchronize_net();
+ if (old_xs)
sock_put((struct sock *)old_xs);
- }
sockfd_put(sock);
return 0;
@@ -212,11 +209,8 @@ static int xsk_map_delete_elem(struct bpf_map *map, void *key)
return -EINVAL;
old_xs = xchg(&m->xsk_map[k], NULL);
- if (old_xs) {
- /* Make sure we've flushed everything. */
- synchronize_net();
+ if (old_xs)
sock_put((struct sock *)old_xs);
- }
return 0;
}
diff --git a/kernel/cgroup/cgroup.c b/kernel/cgroup/cgroup.c
index aae10baf1902..4a3dae2a8283 100644
--- a/kernel/cgroup/cgroup.c
+++ b/kernel/cgroup/cgroup.c
@@ -2836,11 +2836,12 @@ restart:
}
/**
- * cgroup_save_control - save control masks of a subtree
+ * cgroup_save_control - save control masks and dom_cgrp of a subtree
* @cgrp: root of the target subtree
*
- * Save ->subtree_control and ->subtree_ss_mask to the respective old_
- * prefixed fields for @cgrp's subtree including @cgrp itself.
+ * Save ->subtree_control, ->subtree_ss_mask and ->dom_cgrp to the
+ * respective old_ prefixed fields for @cgrp's subtree including @cgrp
+ * itself.
*/
static void cgroup_save_control(struct cgroup *cgrp)
{
@@ -2850,6 +2851,7 @@ static void cgroup_save_control(struct cgroup *cgrp)
cgroup_for_each_live_descendant_pre(dsct, d_css, cgrp) {
dsct->old_subtree_control = dsct->subtree_control;
dsct->old_subtree_ss_mask = dsct->subtree_ss_mask;
+ dsct->old_dom_cgrp = dsct->dom_cgrp;
}
}
@@ -2875,11 +2877,12 @@ static void cgroup_propagate_control(struct cgroup *cgrp)
}
/**
- * cgroup_restore_control - restore control masks of a subtree
+ * cgroup_restore_control - restore control masks and dom_cgrp of a subtree
* @cgrp: root of the target subtree
*
- * Restore ->subtree_control and ->subtree_ss_mask from the respective old_
- * prefixed fields for @cgrp's subtree including @cgrp itself.
+ * Restore ->subtree_control, ->subtree_ss_mask and ->dom_cgrp from the
+ * respective old_ prefixed fields for @cgrp's subtree including @cgrp
+ * itself.
*/
static void cgroup_restore_control(struct cgroup *cgrp)
{
@@ -2889,6 +2892,7 @@ static void cgroup_restore_control(struct cgroup *cgrp)
cgroup_for_each_live_descendant_post(dsct, d_css, cgrp) {
dsct->subtree_control = dsct->old_subtree_control;
dsct->subtree_ss_mask = dsct->old_subtree_ss_mask;
+ dsct->dom_cgrp = dsct->old_dom_cgrp;
}
}
@@ -3196,6 +3200,8 @@ static int cgroup_enable_threaded(struct cgroup *cgrp)
{
struct cgroup *parent = cgroup_parent(cgrp);
struct cgroup *dom_cgrp = parent->dom_cgrp;
+ struct cgroup *dsct;
+ struct cgroup_subsys_state *d_css;
int ret;
lockdep_assert_held(&cgroup_mutex);
@@ -3225,12 +3231,13 @@ static int cgroup_enable_threaded(struct cgroup *cgrp)
*/
cgroup_save_control(cgrp);
- cgrp->dom_cgrp = dom_cgrp;
+ cgroup_for_each_live_descendant_pre(dsct, d_css, cgrp)
+ if (dsct == cgrp || cgroup_is_threaded(dsct))
+ dsct->dom_cgrp = dom_cgrp;
+
ret = cgroup_apply_control(cgrp);
if (!ret)
parent->nr_threaded_children++;
- else
- cgrp->dom_cgrp = cgrp;
cgroup_finalize_control(cgrp, ret);
return ret;
diff --git a/kernel/dma/Kconfig b/kernel/dma/Kconfig
index 9bd54304446f..1b1d63b3634b 100644
--- a/kernel/dma/Kconfig
+++ b/kernel/dma/Kconfig
@@ -23,6 +23,9 @@ config ARCH_HAS_SYNC_DMA_FOR_CPU
bool
select NEED_DMA_MAP_STATE
+config ARCH_HAS_SYNC_DMA_FOR_CPU_ALL
+ bool
+
config DMA_DIRECT_OPS
bool
depends on HAS_DMA
diff --git a/kernel/events/core.c b/kernel/events/core.c
index c80549bf82c6..5a97f34bc14c 100644
--- a/kernel/events/core.c
+++ b/kernel/events/core.c
@@ -3935,6 +3935,12 @@ int perf_event_read_local(struct perf_event *event, u64 *value,
goto out;
}
+ /* If this is a pinned event it must be running on this CPU */
+ if (event->attr.pinned && event->oncpu != smp_processor_id()) {
+ ret = -EBUSY;
+ goto out;
+ }
+
/*
* If the event is currently on this CPU, its either a per-task event,
* or local to this CPU. Furthermore it means its ACTIVE (otherwise
@@ -8308,6 +8314,8 @@ void perf_tp_event(u16 event_type, u64 count, void *record, int entry_size,
goto unlock;
list_for_each_entry_rcu(event, &ctx->event_list, event_entry) {
+ if (event->cpu != smp_processor_id())
+ continue;
if (event->attr.type != PERF_TYPE_TRACEPOINT)
continue;
if (event->attr.config != entry->type)
@@ -9425,9 +9433,7 @@ static void free_pmu_context(struct pmu *pmu)
if (pmu->task_ctx_nr > perf_invalid_context)
return;
- mutex_lock(&pmus_lock);
free_percpu(pmu->pmu_cpu_context);
- mutex_unlock(&pmus_lock);
}
/*
@@ -9683,12 +9689,8 @@ EXPORT_SYMBOL_GPL(perf_pmu_register);
void perf_pmu_unregister(struct pmu *pmu)
{
- int remove_device;
-
mutex_lock(&pmus_lock);
- remove_device = pmu_bus_running;
list_del_rcu(&pmu->entry);
- mutex_unlock(&pmus_lock);
/*
* We dereference the pmu list under both SRCU and regular RCU, so
@@ -9700,13 +9702,14 @@ void perf_pmu_unregister(struct pmu *pmu)
free_percpu(pmu->pmu_disable_count);
if (pmu->type >= PERF_TYPE_MAX)
idr_remove(&pmu_idr, pmu->type);
- if (remove_device) {
+ if (pmu_bus_running) {
if (pmu->nr_addr_filters)
device_remove_file(pmu->dev, &dev_attr_nr_addr_filters);
device_del(pmu->dev);
put_device(pmu->dev);
}
free_pmu_context(pmu);
+ mutex_unlock(&pmus_lock);
}
EXPORT_SYMBOL_GPL(perf_pmu_unregister);
diff --git a/kernel/locking/test-ww_mutex.c b/kernel/locking/test-ww_mutex.c
index 0be047dbd897..65a3b7e55b9f 100644
--- a/kernel/locking/test-ww_mutex.c
+++ b/kernel/locking/test-ww_mutex.c
@@ -260,7 +260,7 @@ static void test_cycle_work(struct work_struct *work)
{
struct test_cycle *cycle = container_of(work, typeof(*cycle), work);
struct ww_acquire_ctx ctx;
- int err;
+ int err, erra = 0;
ww_acquire_init(&ctx, &ww_class);
ww_mutex_lock(&cycle->a_mutex, &ctx);
@@ -270,17 +270,19 @@ static void test_cycle_work(struct work_struct *work)
err = ww_mutex_lock(cycle->b_mutex, &ctx);
if (err == -EDEADLK) {
+ err = 0;
ww_mutex_unlock(&cycle->a_mutex);
ww_mutex_lock_slow(cycle->b_mutex, &ctx);
- err = ww_mutex_lock(&cycle->a_mutex, &ctx);
+ erra = ww_mutex_lock(&cycle->a_mutex, &ctx);
}
if (!err)
ww_mutex_unlock(cycle->b_mutex);
- ww_mutex_unlock(&cycle->a_mutex);
+ if (!erra)
+ ww_mutex_unlock(&cycle->a_mutex);
ww_acquire_fini(&ctx);
- cycle->result = err;
+ cycle->result = err ?: erra;
}
static int __test_cycle(unsigned int nthreads)
diff --git a/kernel/power/suspend.c b/kernel/power/suspend.c
index 5342f6fc022e..0bd595a0b610 100644
--- a/kernel/power/suspend.c
+++ b/kernel/power/suspend.c
@@ -63,6 +63,12 @@ static DECLARE_SWAIT_QUEUE_HEAD(s2idle_wait_head);
enum s2idle_states __read_mostly s2idle_state;
static DEFINE_RAW_SPINLOCK(s2idle_lock);
+bool pm_suspend_via_s2idle(void)
+{
+ return mem_sleep_current == PM_SUSPEND_TO_IDLE;
+}
+EXPORT_SYMBOL_GPL(pm_suspend_via_s2idle);
+
void s2idle_set_ops(const struct platform_s2idle_ops *ops)
{
lock_system_sleep();
diff --git a/kernel/sched/core.c b/kernel/sched/core.c
index 625bc9897f62..ad97f3ba5ec5 100644
--- a/kernel/sched/core.c
+++ b/kernel/sched/core.c
@@ -1167,7 +1167,7 @@ void set_task_cpu(struct task_struct *p, unsigned int new_cpu)
if (task_cpu(p) != new_cpu) {
if (p->sched_class->migrate_task_rq)
- p->sched_class->migrate_task_rq(p);
+ p->sched_class->migrate_task_rq(p, new_cpu);
p->se.nr_migrations++;
rseq_migrate(p);
perf_event_task_migrate(p);
diff --git a/kernel/sched/deadline.c b/kernel/sched/deadline.c
index 997ea7b839fa..91e4202b0634 100644
--- a/kernel/sched/deadline.c
+++ b/kernel/sched/deadline.c
@@ -1607,7 +1607,7 @@ out:
return cpu;
}
-static void migrate_task_rq_dl(struct task_struct *p)
+static void migrate_task_rq_dl(struct task_struct *p, int new_cpu __maybe_unused)
{
struct rq *rq;
diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
index f808ddf2a868..7fc4a371bdd2 100644
--- a/kernel/sched/fair.c
+++ b/kernel/sched/fair.c
@@ -1392,6 +1392,17 @@ bool should_numa_migrate_memory(struct task_struct *p, struct page * page,
int last_cpupid, this_cpupid;
this_cpupid = cpu_pid_to_cpupid(dst_cpu, current->pid);
+ last_cpupid = page_cpupid_xchg_last(page, this_cpupid);
+
+ /*
+ * Allow first faults or private faults to migrate immediately early in
+ * the lifetime of a task. The magic number 4 is based on waiting for
+ * two full passes of the "multi-stage node selection" test that is
+ * executed below.
+ */
+ if ((p->numa_preferred_nid == -1 || p->numa_scan_seq <= 4) &&
+ (cpupid_pid_unset(last_cpupid) || cpupid_match_pid(p, last_cpupid)))
+ return true;
/*
* Multi-stage node selection is used in conjunction with a periodic
@@ -1410,7 +1421,6 @@ bool should_numa_migrate_memory(struct task_struct *p, struct page * page,
* This quadric squishes small probabilities, making it less likely we
* act on an unlikely task<->page relation.
*/
- last_cpupid = page_cpupid_xchg_last(page, this_cpupid);
if (!cpupid_pid_unset(last_cpupid) &&
cpupid_to_nid(last_cpupid) != dst_nid)
return false;
@@ -1514,6 +1524,21 @@ struct task_numa_env {
static void task_numa_assign(struct task_numa_env *env,
struct task_struct *p, long imp)
{
+ struct rq *rq = cpu_rq(env->dst_cpu);
+
+ /* Bail out if run-queue part of active NUMA balance. */
+ if (xchg(&rq->numa_migrate_on, 1))
+ return;
+
+ /*
+ * Clear previous best_cpu/rq numa-migrate flag, since task now
+ * found a better CPU to move/swap.
+ */
+ if (env->best_cpu != -1) {
+ rq = cpu_rq(env->best_cpu);
+ WRITE_ONCE(rq->numa_migrate_on, 0);
+ }
+
if (env->best_task)
put_task_struct(env->best_task);
if (p)
@@ -1553,6 +1578,13 @@ static bool load_too_imbalanced(long src_load, long dst_load,
}
/*
+ * Maximum NUMA importance can be 1998 (2*999);
+ * SMALLIMP @ 30 would be close to 1998/64.
+ * Used to deter task migration.
+ */
+#define SMALLIMP 30
+
+/*
* This checks if the overall compute and NUMA accesses of the system would
* be improved if the source tasks was migrated to the target dst_cpu taking
* into account that it might be best if task running on the dst_cpu should
@@ -1569,6 +1601,9 @@ static void task_numa_compare(struct task_numa_env *env,
long moveimp = imp;
int dist = env->dist;
+ if (READ_ONCE(dst_rq->numa_migrate_on))
+ return;
+
rcu_read_lock();
cur = task_rcu_dereference(&dst_rq->curr);
if (cur && ((cur->flags & PF_EXITING) || is_idle_task(cur)))
@@ -1582,7 +1617,7 @@ static void task_numa_compare(struct task_numa_env *env,
goto unlock;
if (!cur) {
- if (maymove || imp > env->best_imp)
+ if (maymove && moveimp >= env->best_imp)
goto assign;
else
goto unlock;
@@ -1625,16 +1660,22 @@ static void task_numa_compare(struct task_numa_env *env,
task_weight(cur, env->dst_nid, dist);
}
- if (imp <= env->best_imp)
- goto unlock;
-
if (maymove && moveimp > imp && moveimp > env->best_imp) {
- imp = moveimp - 1;
+ imp = moveimp;
cur = NULL;
goto assign;
}
/*
+ * If the NUMA importance is less than SMALLIMP,
+ * task migration might only result in ping pong
+ * of tasks and also hurt performance due to cache
+ * misses.
+ */
+ if (imp < SMALLIMP || imp <= env->best_imp + SMALLIMP / 2)
+ goto unlock;
+
+ /*
* In the overloaded case, try and keep the load balanced.
*/
load = task_h_load(env->p) - task_h_load(cur);
@@ -1710,6 +1751,7 @@ static int task_numa_migrate(struct task_struct *p)
.best_cpu = -1,
};
struct sched_domain *sd;
+ struct rq *best_rq;
unsigned long taskweight, groupweight;
int nid, ret, dist;
long taskimp, groupimp;
@@ -1805,20 +1847,17 @@ static int task_numa_migrate(struct task_struct *p)
if (env.best_cpu == -1)
return -EAGAIN;
- /*
- * Reset the scan period if the task is being rescheduled on an
- * alternative node to recheck if the tasks is now properly placed.
- */
- p->numa_scan_period = task_scan_start(p);
-
+ best_rq = cpu_rq(env.best_cpu);
if (env.best_task == NULL) {
ret = migrate_task_to(p, env.best_cpu);
+ WRITE_ONCE(best_rq->numa_migrate_on, 0);
if (ret != 0)
trace_sched_stick_numa(p, env.src_cpu, env.best_cpu);
return ret;
}
ret = migrate_swap(p, env.best_task, env.best_cpu, env.src_cpu);
+ WRITE_ONCE(best_rq->numa_migrate_on, 0);
if (ret != 0)
trace_sched_stick_numa(p, env.src_cpu, task_cpu(env.best_task));
@@ -2596,6 +2635,39 @@ void task_tick_numa(struct rq *rq, struct task_struct *curr)
}
}
+static void update_scan_period(struct task_struct *p, int new_cpu)
+{
+ int src_nid = cpu_to_node(task_cpu(p));
+ int dst_nid = cpu_to_node(new_cpu);
+
+ if (!static_branch_likely(&sched_numa_balancing))
+ return;
+
+ if (!p->mm || !p->numa_faults || (p->flags & PF_EXITING))
+ return;
+
+ if (src_nid == dst_nid)
+ return;
+
+ /*
+ * Allow resets if faults have been trapped before one scan
+ * has completed. This is most likely due to a new task that
+ * is pulled cross-node due to wakeups or load balancing.
+ */
+ if (p->numa_scan_seq) {
+ /*
+ * Avoid scan adjustments if moving to the preferred
+ * node or if the task was not previously running on
+ * the preferred node.
+ */
+ if (dst_nid == p->numa_preferred_nid ||
+ (p->numa_preferred_nid != -1 && src_nid != p->numa_preferred_nid))
+ return;
+ }
+
+ p->numa_scan_period = task_scan_start(p);
+}
+
#else
static void task_tick_numa(struct rq *rq, struct task_struct *curr)
{
@@ -2609,6 +2681,10 @@ static inline void account_numa_dequeue(struct rq *rq, struct task_struct *p)
{
}
+static inline void update_scan_period(struct task_struct *p, int new_cpu)
+{
+}
+
#endif /* CONFIG_NUMA_BALANCING */
static void
@@ -6275,7 +6351,7 @@ static void detach_entity_cfs_rq(struct sched_entity *se);
* cfs_rq_of(p) references at time of call are still valid and identify the
* previous CPU. The caller guarantees p->pi_lock or task_rq(p)->lock is held.
*/
-static void migrate_task_rq_fair(struct task_struct *p)
+static void migrate_task_rq_fair(struct task_struct *p, int new_cpu)
{
/*
* As blocked tasks retain absolute vruntime the migration needs to
@@ -6328,6 +6404,8 @@ static void migrate_task_rq_fair(struct task_struct *p)
/* We have migrated, no longer consider this task hot */
p->se.exec_start = 0;
+
+ update_scan_period(p, new_cpu);
}
static void task_dead_fair(struct task_struct *p)
diff --git a/kernel/sched/sched.h b/kernel/sched/sched.h
index 4a2e8cae63c4..455fa330de04 100644
--- a/kernel/sched/sched.h
+++ b/kernel/sched/sched.h
@@ -783,6 +783,7 @@ struct rq {
#ifdef CONFIG_NUMA_BALANCING
unsigned int nr_numa_running;
unsigned int nr_preferred_running;
+ unsigned int numa_migrate_on;
#endif
#define CPU_LOAD_IDX_MAX 5
unsigned long cpu_load[CPU_LOAD_IDX_MAX];
@@ -1523,7 +1524,7 @@ struct sched_class {
#ifdef CONFIG_SMP
int (*select_task_rq)(struct task_struct *p, int task_cpu, int sd_flag, int flags);
- void (*migrate_task_rq)(struct task_struct *p);
+ void (*migrate_task_rq)(struct task_struct *p, int new_cpu);
void (*task_woken)(struct rq *this_rq, struct task_struct *task);
generated by cgit v1.2.3 (git 2.39.1) at 2024-12-21 08:37:53 +0100