summaryrefslogtreecommitdiff
path: root/kernel/sched
diff options
context:
space:
mode:
authorPeter Oskolkov <posk@google.com>2020-09-23 16:36:16 -0700
committerPeter Zijlstra <peterz@infradead.org>2020-09-25 14:23:27 +0200
commit2a36ab717e8fe678d98f81c14a0b124712719840 (patch)
treea4956fc970342eb2853b9e739009058504290676 /kernel/sched
parent233e7aca4c8a2c764f556bba9644c36154017e7f (diff)
rseq/membarrier: Add MEMBARRIER_CMD_PRIVATE_EXPEDITED_RSEQ
This patchset is based on Google-internal RSEQ work done by Paul Turner and Andrew Hunter. When working with per-CPU RSEQ-based memory allocations, it is sometimes important to make sure that a global memory location is no longer accessed from RSEQ critical sections. For example, there can be two per-CPU lists, one is "active" and accessed per-CPU, while another one is inactive and worked on asynchronously "off CPU" (e.g. garbage collection is performed). Then at some point the two lists are swapped, and a fast RCU-like mechanism is required to make sure that the previously active list is no longer accessed. This patch introduces such a mechanism: in short, membarrier() syscall issues an IPI to a CPU, restarting a potentially active RSEQ critical section on the CPU. Signed-off-by: Peter Oskolkov <posk@google.com> Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Acked-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com> Link: https://lkml.kernel.org/r/20200923233618.2572849-1-posk@google.com
Diffstat (limited to 'kernel/sched')
-rw-r--r--kernel/sched/membarrier.c136
1 files changed, 106 insertions, 30 deletions
diff --git a/kernel/sched/membarrier.c b/kernel/sched/membarrier.c
index 168479a7d61b..e23e74d52db5 100644
--- a/kernel/sched/membarrier.c
+++ b/kernel/sched/membarrier.c
@@ -18,6 +18,14 @@
#define MEMBARRIER_PRIVATE_EXPEDITED_SYNC_CORE_BITMASK 0
#endif
+#ifdef CONFIG_RSEQ
+#define MEMBARRIER_CMD_PRIVATE_EXPEDITED_RSEQ_BITMASK \
+ (MEMBARRIER_CMD_PRIVATE_EXPEDITED_RSEQ \
+ | MEMBARRIER_CMD_REGISTER_PRIVATE_EXPEDITED_RSEQ_BITMASK)
+#else
+#define MEMBARRIER_CMD_PRIVATE_EXPEDITED_RSEQ_BITMASK 0
+#endif
+
#define MEMBARRIER_CMD_BITMASK \
(MEMBARRIER_CMD_GLOBAL | MEMBARRIER_CMD_GLOBAL_EXPEDITED \
| MEMBARRIER_CMD_REGISTER_GLOBAL_EXPEDITED \
@@ -30,6 +38,11 @@ static void ipi_mb(void *info)
smp_mb(); /* IPIs should be serializing but paranoid. */
}
+static void ipi_rseq(void *info)
+{
+ rseq_preempt(current);
+}
+
static void ipi_sync_rq_state(void *info)
{
struct mm_struct *mm = (struct mm_struct *) info;
@@ -129,19 +142,27 @@ static int membarrier_global_expedited(void)
return 0;
}
-static int membarrier_private_expedited(int flags)
+static int membarrier_private_expedited(int flags, int cpu_id)
{
- int cpu;
cpumask_var_t tmpmask;
struct mm_struct *mm = current->mm;
+ smp_call_func_t ipi_func = ipi_mb;
- if (flags & MEMBARRIER_FLAG_SYNC_CORE) {
+ if (flags == MEMBARRIER_FLAG_SYNC_CORE) {
if (!IS_ENABLED(CONFIG_ARCH_HAS_MEMBARRIER_SYNC_CORE))
return -EINVAL;
if (!(atomic_read(&mm->membarrier_state) &
MEMBARRIER_STATE_PRIVATE_EXPEDITED_SYNC_CORE_READY))
return -EPERM;
+ } else if (flags == MEMBARRIER_FLAG_RSEQ) {
+ if (!IS_ENABLED(CONFIG_RSEQ))
+ return -EINVAL;
+ if (!(atomic_read(&mm->membarrier_state) &
+ MEMBARRIER_STATE_PRIVATE_EXPEDITED_RSEQ_READY))
+ return -EPERM;
+ ipi_func = ipi_rseq;
} else {
+ WARN_ON_ONCE(flags);
if (!(atomic_read(&mm->membarrier_state) &
MEMBARRIER_STATE_PRIVATE_EXPEDITED_READY))
return -EPERM;
@@ -156,35 +177,59 @@ static int membarrier_private_expedited(int flags)
*/
smp_mb(); /* system call entry is not a mb. */
- if (!zalloc_cpumask_var(&tmpmask, GFP_KERNEL))
+ if (cpu_id < 0 && !zalloc_cpumask_var(&tmpmask, GFP_KERNEL))
return -ENOMEM;
cpus_read_lock();
- rcu_read_lock();
- for_each_online_cpu(cpu) {
+
+ if (cpu_id >= 0) {
struct task_struct *p;
- /*
- * Skipping the current CPU is OK even through we can be
- * migrated at any point. The current CPU, at the point
- * where we read raw_smp_processor_id(), is ensured to
- * be in program order with respect to the caller
- * thread. Therefore, we can skip this CPU from the
- * iteration.
- */
- if (cpu == raw_smp_processor_id())
- continue;
- p = rcu_dereference(cpu_rq(cpu)->curr);
- if (p && p->mm == mm)
- __cpumask_set_cpu(cpu, tmpmask);
+ if (cpu_id >= nr_cpu_ids || !cpu_online(cpu_id))
+ goto out;
+ if (cpu_id == raw_smp_processor_id())
+ goto out;
+ rcu_read_lock();
+ p = rcu_dereference(cpu_rq(cpu_id)->curr);
+ if (!p || p->mm != mm) {
+ rcu_read_unlock();
+ goto out;
+ }
+ rcu_read_unlock();
+ } else {
+ int cpu;
+
+ rcu_read_lock();
+ for_each_online_cpu(cpu) {
+ struct task_struct *p;
+
+ /*
+ * Skipping the current CPU is OK even through we can be
+ * migrated at any point. The current CPU, at the point
+ * where we read raw_smp_processor_id(), is ensured to
+ * be in program order with respect to the caller
+ * thread. Therefore, we can skip this CPU from the
+ * iteration.
+ */
+ if (cpu == raw_smp_processor_id())
+ continue;
+ p = rcu_dereference(cpu_rq(cpu)->curr);
+ if (p && p->mm == mm)
+ __cpumask_set_cpu(cpu, tmpmask);
+ }
+ rcu_read_unlock();
}
- rcu_read_unlock();
preempt_disable();
- smp_call_function_many(tmpmask, ipi_mb, NULL, 1);
+ if (cpu_id >= 0)
+ smp_call_function_single(cpu_id, ipi_func, NULL, 1);
+ else
+ smp_call_function_many(tmpmask, ipi_func, NULL, 1);
preempt_enable();
- free_cpumask_var(tmpmask);
+out:
+ if (cpu_id < 0)
+ free_cpumask_var(tmpmask);
cpus_read_unlock();
/*
@@ -283,11 +328,18 @@ static int membarrier_register_private_expedited(int flags)
set_state = MEMBARRIER_STATE_PRIVATE_EXPEDITED,
ret;
- if (flags & MEMBARRIER_FLAG_SYNC_CORE) {
+ if (flags == MEMBARRIER_FLAG_SYNC_CORE) {
if (!IS_ENABLED(CONFIG_ARCH_HAS_MEMBARRIER_SYNC_CORE))
return -EINVAL;
ready_state =
MEMBARRIER_STATE_PRIVATE_EXPEDITED_SYNC_CORE_READY;
+ } else if (flags == MEMBARRIER_FLAG_RSEQ) {
+ if (!IS_ENABLED(CONFIG_RSEQ))
+ return -EINVAL;
+ ready_state =
+ MEMBARRIER_STATE_PRIVATE_EXPEDITED_RSEQ_READY;
+ } else {
+ WARN_ON_ONCE(flags);
}
/*
@@ -299,6 +351,8 @@ static int membarrier_register_private_expedited(int flags)
return 0;
if (flags & MEMBARRIER_FLAG_SYNC_CORE)
set_state |= MEMBARRIER_STATE_PRIVATE_EXPEDITED_SYNC_CORE;
+ if (flags & MEMBARRIER_FLAG_RSEQ)
+ set_state |= MEMBARRIER_STATE_PRIVATE_EXPEDITED_RSEQ;
atomic_or(set_state, &mm->membarrier_state);
ret = sync_runqueues_membarrier_state(mm);
if (ret)
@@ -310,8 +364,15 @@ static int membarrier_register_private_expedited(int flags)
/**
* sys_membarrier - issue memory barriers on a set of threads
- * @cmd: Takes command values defined in enum membarrier_cmd.
- * @flags: Currently needs to be 0. For future extensions.
+ * @cmd: Takes command values defined in enum membarrier_cmd.
+ * @flags: Currently needs to be 0 for all commands other than
+ * MEMBARRIER_CMD_PRIVATE_EXPEDITED_RSEQ: in the latter
+ * case it can be MEMBARRIER_CMD_FLAG_CPU, indicating that @cpu_id
+ * contains the CPU on which to interrupt (= restart)
+ * the RSEQ critical section.
+ * @cpu_id: if @flags == MEMBARRIER_CMD_FLAG_CPU, indicates the cpu on which
+ * RSEQ CS should be interrupted (@cmd must be
+ * MEMBARRIER_CMD_PRIVATE_EXPEDITED_RSEQ).
*
* If this system call is not implemented, -ENOSYS is returned. If the
* command specified does not exist, not available on the running
@@ -337,10 +398,21 @@ static int membarrier_register_private_expedited(int flags)
* smp_mb() X O O
* sys_membarrier() O O O
*/
-SYSCALL_DEFINE2(membarrier, int, cmd, int, flags)
+SYSCALL_DEFINE3(membarrier, int, cmd, unsigned int, flags, int, cpu_id)
{
- if (unlikely(flags))
- return -EINVAL;
+ switch (cmd) {
+ case MEMBARRIER_CMD_PRIVATE_EXPEDITED_RSEQ:
+ if (unlikely(flags && flags != MEMBARRIER_CMD_FLAG_CPU))
+ return -EINVAL;
+ break;
+ default:
+ if (unlikely(flags))
+ return -EINVAL;
+ }
+
+ if (!(flags & MEMBARRIER_CMD_FLAG_CPU))
+ cpu_id = -1;
+
switch (cmd) {
case MEMBARRIER_CMD_QUERY:
{
@@ -362,13 +434,17 @@ SYSCALL_DEFINE2(membarrier, int, cmd, int, flags)
case MEMBARRIER_CMD_REGISTER_GLOBAL_EXPEDITED:
return membarrier_register_global_expedited();
case MEMBARRIER_CMD_PRIVATE_EXPEDITED:
- return membarrier_private_expedited(0);
+ return membarrier_private_expedited(0, cpu_id);
case MEMBARRIER_CMD_REGISTER_PRIVATE_EXPEDITED:
return membarrier_register_private_expedited(0);
case MEMBARRIER_CMD_PRIVATE_EXPEDITED_SYNC_CORE:
- return membarrier_private_expedited(MEMBARRIER_FLAG_SYNC_CORE);
+ return membarrier_private_expedited(MEMBARRIER_FLAG_SYNC_CORE, cpu_id);
case MEMBARRIER_CMD_REGISTER_PRIVATE_EXPEDITED_SYNC_CORE:
return membarrier_register_private_expedited(MEMBARRIER_FLAG_SYNC_CORE);
+ case MEMBARRIER_CMD_PRIVATE_EXPEDITED_RSEQ:
+ return membarrier_private_expedited(MEMBARRIER_FLAG_RSEQ, cpu_id);
+ case MEMBARRIER_CMD_REGISTER_PRIVATE_EXPEDITED_RSEQ:
+ return membarrier_register_private_expedited(MEMBARRIER_FLAG_RSEQ);
default:
return -EINVAL;
}