[PATCH v8 -tip 18/26] sched: Add a per-thread core scheduling interface

["Joel Fernandes (Google)" <joel@joelfernandes.org>]

Add a per-thread core scheduling interface which allows a thread to share a
core with another thread, or have a core exclusively for itself.

ChromeOS uses core-scheduling to securely enable hyperthreading.  This cuts
down the keypress latency in Google docs from 150ms to 50ms while improving
the camera streaming frame rate by ~3%.

Tested-by: Julien Desfossez <jdesfossez@digitalocean.com>
Signed-off-by: Joel Fernandes (Google) <joel@joelfernandes.org>
---
 include/linux/sched.h            |  2 ++
 include/uapi/linux/prctl.h       |  3 ++
 kernel/sched/core.c              | 51 +++++++++++++++++++++++++++++---
 kernel/sys.c                     |  3 ++
 tools/include/uapi/linux/prctl.h |  3 ++
 5 files changed, 58 insertions(+), 4 deletions(-)

diff --git a/include/linux/sched.h b/include/linux/sched.h
index c6034c00846a..4cb76575afa8 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -2078,11 +2078,13 @@ void sched_core_unsafe_enter(void);
 void sched_core_unsafe_exit(void);
 bool sched_core_wait_till_safe(unsigned long ti_check);
 bool sched_core_kernel_protected(void);
+int sched_core_share_pid(pid_t pid);
 #else
 #define sched_core_unsafe_enter(ignore) do { } while (0)
 #define sched_core_unsafe_exit(ignore) do { } while (0)
 #define sched_core_wait_till_safe(ignore) do { } while (0)
 #define sched_core_kernel_protected(ignore) do { } while (0)
+#define sched_core_share_pid(pid_t pid) do { } while (0)
 #endif
 
 #endif
diff --git a/include/uapi/linux/prctl.h b/include/uapi/linux/prctl.h
index c334e6a02e5f..217b0482aea1 100644
--- a/include/uapi/linux/prctl.h
+++ b/include/uapi/linux/prctl.h
@@ -248,4 +248,7 @@ struct prctl_mm_map {
 #define PR_SET_IO_FLUSHER		57
 #define PR_GET_IO_FLUSHER		58
 
+/* Request the scheduler to share a core */
+#define PR_SCHED_CORE_SHARE		59
+
 #endif /* _LINUX_PRCTL_H */
diff --git a/kernel/sched/core.c b/kernel/sched/core.c
index 30a9e4cb5ce1..a0678614a056 100644
--- a/kernel/sched/core.c
+++ b/kernel/sched/core.c
@@ -310,6 +310,7 @@ static int __sched_core_stopper(void *data)
 }
 
 static DEFINE_MUTEX(sched_core_mutex);
+static DEFINE_MUTEX(sched_core_tasks_mutex);
 static int sched_core_count;
 
 static void __sched_core_enable(void)
@@ -3588,8 +3589,9 @@ int sched_fork(unsigned long clone_flags, struct task_struct *p)
 	RB_CLEAR_NODE(&p->core_node);
 
 	/*
-	 * Tag child via per-task cookie only if parent is tagged via per-task
-	 * cookie. This is independent of, but can be additive to the CGroup tagging.
+	 * If parent is tagged via per-task cookie, tag the child (either with
+	 * the parent's cookie, or a new one). The final cookie is calculated
+	 * by concatenating the per-task cookie with that of the CGroup's.
 	 */
 	if (current->core_task_cookie) {
 
@@ -9301,7 +9303,7 @@ static int sched_core_share_tasks(struct task_struct *t1, struct task_struct *t2
 	unsigned long cookie;
 	int ret = -ENOMEM;
 
-	mutex_lock(&sched_core_mutex);
+	mutex_lock(&sched_core_tasks_mutex);
 
 	/*
 	 * NOTE: sched_core_get() is done by sched_core_alloc_task_cookie() or
@@ -9400,10 +9402,51 @@ static int sched_core_share_tasks(struct task_struct *t1, struct task_struct *t2
 
 	ret = 0;
 out_unlock:
-	mutex_unlock(&sched_core_mutex);
+	mutex_unlock(&sched_core_tasks_mutex);
 	return ret;
 }
 
+/* Called from prctl interface: PR_SCHED_CORE_SHARE */
+int sched_core_share_pid(pid_t pid)
+{
+	struct task_struct *task;
+	int err;
+
+	if (pid == 0) { /* Recent current task's cookie. */
+		/* Resetting a cookie requires privileges. */
+		if (current->core_task_cookie)
+			if (!capable(CAP_SYS_ADMIN))
+				return -EPERM;
+		task = NULL;
+	} else {
+		rcu_read_lock();
+		task = pid ? find_task_by_vpid(pid) : current;
+		if (!task) {
+			rcu_read_unlock();
+			return -ESRCH;
+		}
+
+		get_task_struct(task);
+
+		/*
+		 * Check if this process has the right to modify the specified
+		 * process. Use the regular "ptrace_may_access()" checks.
+		 */
+		if (!ptrace_may_access(task, PTRACE_MODE_READ_REALCREDS)) {
+			rcu_read_unlock();
+			err = -EPERM;
+			goto out_put;
+		}
+		rcu_read_unlock();
+	}
+
+	err = sched_core_share_tasks(current, task);
+out_put:
+	if (task)
+		put_task_struct(task);
+	return err;
+}
+
 /* CGroup interface */
 static u64 cpu_core_tag_read_u64(struct cgroup_subsys_state *css, struct cftype *cft)
 {
diff --git a/kernel/sys.c b/kernel/sys.c
index 6401880dff74..17911b8680b1 100644
--- a/kernel/sys.c
+++ b/kernel/sys.c
@@ -2530,6 +2530,9 @@ SYSCALL_DEFINE5(prctl, int, option, unsigned long, arg2, unsigned long, arg3,
 
 		error = (current->flags & PR_IO_FLUSHER) == PR_IO_FLUSHER;
 		break;
+	case PR_SCHED_CORE_SHARE:
+		error = sched_core_share_pid(arg2);
+		break;
 	default:
 		error = -EINVAL;
 		break;
diff --git a/tools/include/uapi/linux/prctl.h b/tools/include/uapi/linux/prctl.h
index 07b4f8131e36..9318f643c4b3 100644
--- a/tools/include/uapi/linux/prctl.h
+++ b/tools/include/uapi/linux/prctl.h
@@ -238,4 +238,7 @@ struct prctl_mm_map {
 #define PR_SET_IO_FLUSHER		57
 #define PR_GET_IO_FLUSHER		58
 
+/* Request the scheduler to share a core */
+#define PR_SCHED_CORE_SHARE		59
+
 #endif /* _LINUX_PRCTL_H */
-- 
2.29.0.rc1.297.gfa9743e501-goog