[PATCHv2 1/2] bpf: Add support for absolute value BPF timers

[Tero Kristo <tero.kristo@xxxxxxxxxxxxxxx>]

Add a new flag BPF_F_TIMER_ABS that can be passed to bpf_timer_start()
to start an absolute value timer instead of the default relative value.
This makes the timer expire at an exact point in time, instead of a time
with latencies induced by both the BPF and timer subsystems.

Suggested-by: Artem Bityutskiy <artem.bityutskiy@xxxxxxxxxxxxxxx>
Signed-off-by: Tero Kristo <tero.kristo@xxxxxxxxxxxxxxx>
---
v2:
 - added suggested by from Artem
 - copied the change from include/uapi/linux/bpf.h to tools/include also
   
 include/uapi/linux/bpf.h       | 15 +++++++++++++++
 kernel/bpf/helpers.c           | 11 +++++++++--
 tools/include/uapi/linux/bpf.h | 15 +++++++++++++++
 3 files changed, 39 insertions(+), 2 deletions(-)

diff --git a/include/uapi/linux/bpf.h b/include/uapi/linux/bpf.h
index 464ca3f01fe7..7f5b71847984 100644
--- a/include/uapi/linux/bpf.h
+++ b/include/uapi/linux/bpf.h
@@ -4951,6 +4951,12 @@ union bpf_attr {
  *		different maps if key/value layout matches across maps.
  *		Every bpf_timer_set_callback() can have different callback_fn.
  *
+ *		*flags* can be one of:
+ *
+ *		**BPF_F_TIMER_ABS**
+ *			Start the timer in absolute expire value instead of the
+ *			default relative one.
+ *
  *	Return
  *		0 on success.
  *		**-EINVAL** if *timer* was not initialized with bpf_timer_init() earlier
@@ -7050,4 +7056,13 @@ struct bpf_core_relo {
 	enum bpf_core_relo_kind kind;
 };
 
+/*
+ * Flags to control bpf_timer_start() behaviour.
+ *     - BPF_F_TIMER_ABS: Timeout passed is absolute time, by default it is
+ *       relative to current time.
+ */
+enum {
+	BPF_F_TIMER_ABS = (1ULL << 0),
+};
+
 #endif /* _UAPI__LINUX_BPF_H__ */
diff --git a/kernel/bpf/helpers.c b/kernel/bpf/helpers.c
index af30c6cbd65d..924849d89828 100644
--- a/kernel/bpf/helpers.c
+++ b/kernel/bpf/helpers.c
@@ -1253,10 +1253,11 @@ BPF_CALL_3(bpf_timer_start, struct bpf_timer_kern *, timer, u64, nsecs, u64, fla
 {
 	struct bpf_hrtimer *t;
 	int ret = 0;
+	enum hrtimer_mode mode;
 
 	if (in_nmi())
 		return -EOPNOTSUPP;
-	if (flags)
+	if (flags > BPF_F_TIMER_ABS)
 		return -EINVAL;
 	__bpf_spin_lock_irqsave(&timer->lock);
 	t = timer->timer;
@@ -1264,7 +1265,13 @@ BPF_CALL_3(bpf_timer_start, struct bpf_timer_kern *, timer, u64, nsecs, u64, fla
 		ret = -EINVAL;
 		goto out;
 	}
-	hrtimer_start(&t->timer, ns_to_ktime(nsecs), HRTIMER_MODE_REL_SOFT);
+
+	if (flags & BPF_F_TIMER_ABS)
+		mode = HRTIMER_MODE_ABS_SOFT;
+	else
+		mode = HRTIMER_MODE_REL_SOFT;
+
+	hrtimer_start(&t->timer, ns_to_ktime(nsecs), mode);
 out:
 	__bpf_spin_unlock_irqrestore(&timer->lock);
 	return ret;
diff --git a/tools/include/uapi/linux/bpf.h b/tools/include/uapi/linux/bpf.h
index 464ca3f01fe7..7f5b71847984 100644
--- a/tools/include/uapi/linux/bpf.h
+++ b/tools/include/uapi/linux/bpf.h
@@ -4951,6 +4951,12 @@ union bpf_attr {
  *		different maps if key/value layout matches across maps.
  *		Every bpf_timer_set_callback() can have different callback_fn.
  *
+ *		*flags* can be one of:
+ *
+ *		**BPF_F_TIMER_ABS**
+ *			Start the timer in absolute expire value instead of the
+ *			default relative one.
+ *
  *	Return
  *		0 on success.
  *		**-EINVAL** if *timer* was not initialized with bpf_timer_init() earlier
@@ -7050,4 +7056,13 @@ struct bpf_core_relo {
 	enum bpf_core_relo_kind kind;
 };
 
+/*
+ * Flags to control bpf_timer_start() behaviour.
+ *     - BPF_F_TIMER_ABS: Timeout passed is absolute time, by default it is
+ *       relative to current time.
+ */
+enum {
+	BPF_F_TIMER_ABS = (1ULL << 0),
+};
+
 #endif /* _UAPI__LINUX_BPF_H__ */
-- 
2.25.1