sched/cputime: Add vtime guest task state

Record guest as a VTIME state instead of guessing it from VTIME_SYS and
PF_VCPU. This is going to simplify the cputime read side especially as
its state machine is going to further expand in order to fully support
kcpustat on nohz_full.

Signed-off-by: Frederic Weisbecker <[email protected]>
Signed-off-by: Peter Zijlstra (Intel) <[email protected]>
Cc: Jacek Anaszewski <[email protected]>
Cc: Linus Torvalds <[email protected]>
Cc: Pavel Machek <[email protected]>
Cc: Peter Zijlstra <[email protected]>
Cc: Rafael J . Wysocki <[email protected]>
Cc: Rik van Riel <[email protected]>
Cc: Thomas Gleixner <[email protected]>
Cc: Viresh Kumar <[email protected]>
Cc: Wanpeng Li <[email protected]>
Cc: Yauheni Kaliuta <[email protected]>
Link: https://lkml.kernel.org/r/[email protected]
Signed-off-by: Ingo Molnar <[email protected]>

2 files changed, 13 insertions, 7 deletions

diff --git a/include/linux/sched.h b/include/linux/sched.h
index 4ae19be2c126..988c4da00c31 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -255,6 +255,8 @@ enum vtime_state {
 	VTIME_SYS,
 	/* Task runs in userspace in a CPU with VTIME active: */
 	VTIME_USER,
+	/* Task runs as guests in a CPU with VTIME active: */
+	VTIME_GUEST,
 };
 
 struct vtime {
diff --git a/kernel/sched/cputime.c b/kernel/sched/cputime.c
index 2e885e870aa1..34086afc3518 100644
--- a/kernel/sched/cputime.c
+++ b/kernel/sched/cputime.c
@@ -733,7 +733,7 @@ static void __vtime_account_kernel(struct task_struct *tsk,
 				   struct vtime *vtime)
 {
 	/* We might have scheduled out from guest path */
-	if (tsk->flags & PF_VCPU)
+	if (vtime->state == VTIME_GUEST)
 		vtime_account_guest(tsk, vtime);
 	else
 		vtime_account_system(tsk, vtime);
@@ -788,6 +788,7 @@ void vtime_guest_enter(struct task_struct *tsk)
 	write_seqcount_begin(&vtime->seqcount);
 	vtime_account_system(tsk, vtime);
 	tsk->flags |= PF_VCPU;
+	vtime->state = VTIME_GUEST;
 	write_seqcount_end(&vtime->seqcount);
 }
 EXPORT_SYMBOL_GPL(vtime_guest_enter);
@@ -799,6 +800,7 @@ void vtime_guest_exit(struct task_struct *tsk)
 	write_seqcount_begin(&vtime->seqcount);
 	vtime_account_guest(tsk, vtime);
 	tsk->flags &= ~PF_VCPU;
+	vtime->state = VTIME_SYS;
 	write_seqcount_end(&vtime->seqcount);
 }
 EXPORT_SYMBOL_GPL(vtime_guest_exit);
@@ -826,6 +828,8 @@ void vtime_task_switch_generic(struct task_struct *prev)
 	write_seqcount_begin(&vtime->seqcount);
 	if (is_idle_task(current))
 		vtime->state = VTIME_IDLE;
+	else if (current->flags & PF_VCPU)
+		vtime->state = VTIME_GUEST;
 	else
 		vtime->state = VTIME_SYS;
 	vtime->starttime = sched_clock();
@@ -860,7 +864,7 @@ u64 task_gtime(struct task_struct *t)
 		seq = read_seqcount_begin(&vtime->seqcount);
 
 		gtime = t->gtime;
-		if (vtime->state == VTIME_SYS && t->flags & PF_VCPU)
+		if (vtime->state == VTIME_GUEST)
 			gtime += vtime->gtime + vtime_delta(vtime);
 
 	} while (read_seqcount_retry(&vtime->seqcount, seq));
@@ -898,13 +902,13 @@ void task_cputime(struct task_struct *t, u64 *utime, u64 *stime)
 		delta = vtime_delta(vtime);
 
 		/*
-		 * Task runs either in user or kernel space, add pending nohz time to
-		 * the right place.
+		 * Task runs either in user (including guest) or kernel space,
+		 * add pending nohz time to the right place.
 		 */
-		if (vtime->state == VTIME_USER || t->flags & PF_VCPU)
-			*utime += vtime->utime + delta;
-		else if (vtime->state == VTIME_SYS)
+		if (vtime->state == VTIME_SYS)
 			*stime += vtime->stime + delta;
+		else
+			*utime += vtime->utime + delta;
 	} while (read_seqcount_retry(&vtime->seqcount, seq));
 }
 #endif /* CONFIG_VIRT_CPU_ACCOUNTING_GEN */