[PATCH net v3] net: cbs: Fix software cbs to consider packet sending time

[PATCH net v3] net: cbs: Fix software cbs to consider packet sending time

[Zh-yuan Ye]

Currently the software CBS does not consider the packet sending time
when depleting the credits. It caused the throughput to be
Idleslope[kbps] * (Port transmit rate[kbps] / |Sendslope[kbps]|) where
Idleslope * (Port transmit rate / (Idleslope + |Sendslope|)) = Idleslope
is expected. In order to fix the issue above, this patch takes the time
when the packet sending completes into account by moving the anchor time
variable "last" ahead to the send completion time upon transmission and
adding wait when the next dequeue request comes before the send
completion time of the previous packet.

changelog:
V2->V3:
 - remove unnecessary whitespace cleanup
 - add the checks if port_rate is 0 before division

V1->V2:
 - combine variable "send_completed" into "last"
 - add the comment for estimate of the packet sending

Fixes: 585d763af09c ("net/sched: Introduce Credit Based Shaper (CBS) qdisc")
Signed-off-by: Zh-yuan Ye <ye.zh-yuan@socionext.com>
Reviewed-by: Vinicius Costa Gomes <vinicius.gomes@intel.com>
---
 net/sched/sch_cbs.c | 12 +++++++++++-
 1 file changed, 11 insertions(+), 1 deletion(-)

diff --git a/net/sched/sch_cbs.c b/net/sched/sch_cbs.c
index b2905b03a432..2eaac2ff380f 100644
--- a/net/sched/sch_cbs.c
+++ b/net/sched/sch_cbs.c
@@ -181,6 +181,11 @@ static struct sk_buff *cbs_dequeue_soft(struct Qdisc *sch)
 	s64 credits;
 	int len;
 
+	/* The previous packet is still being sent */
+	if (now < q->last) {
+		qdisc_watchdog_schedule_ns(&q->watchdog, q->last);
+		return NULL;
+	}
 	if (q->credits < 0) {
 		credits = timediff_to_credits(now - q->last, q->idleslope);
 
@@ -212,7 +217,12 @@ static struct sk_buff *cbs_dequeue_soft(struct Qdisc *sch)
 	credits += q->credits;
 
 	q->credits = max_t(s64, credits, q->locredit);
-	q->last = now;
+	/* Estimate of the transmission of the last byte of the packet in ns */
+	if (unlikely(atomic64_read(&q->port_rate) == 0))
+		q->last = now;
+	else
+		q->last = now + div64_s64(len * NSEC_PER_SEC,
+					  atomic64_read(&q->port_rate));
 
 	return skb;
 }
-- 
2.20.1

Re: [PATCH net v3] net: cbs: Fix software cbs to consider packet sending time

[Vinicius Costa Gomes]

Zh-yuan Ye <ye.zh-yuan@socionext.com> writes:

> Currently the software CBS does not consider the packet sending time
> when depleting the credits. It caused the throughput to be
> Idleslope[kbps] * (Port transmit rate[kbps] / |Sendslope[kbps]|) where
> Idleslope * (Port transmit rate / (Idleslope + |Sendslope|)) = Idleslope
> is expected. In order to fix the issue above, this patch takes the time
> when the packet sending completes into account by moving the anchor time
> variable "last" ahead to the send completion time upon transmission and
> adding wait when the next dequeue request comes before the send
> completion time of the previous packet.
>
> changelog:
> V2->V3:
>  - remove unnecessary whitespace cleanup
>  - add the checks if port_rate is 0 before division
>
> V1->V2:
>  - combine variable "send_completed" into "last"
>  - add the comment for estimate of the packet sending
>
> Fixes: 585d763af09c ("net/sched: Introduce Credit Based Shaper (CBS) qdisc")
> Signed-off-by: Zh-yuan Ye <ye.zh-yuan@socionext.com>
> Reviewed-by: Vinicius Costa Gomes <vinicius.gomes@intel.com>
> ---
>  net/sched/sch_cbs.c | 12 +++++++++++-
>  1 file changed, 11 insertions(+), 1 deletion(-)
>
> diff --git a/net/sched/sch_cbs.c b/net/sched/sch_cbs.c
> index b2905b03a432..2eaac2ff380f 100644
> --- a/net/sched/sch_cbs.c
> +++ b/net/sched/sch_cbs.c
> @@ -181,6 +181,11 @@ static struct sk_buff *cbs_dequeue_soft(struct Qdisc *sch)
>  	s64 credits;
>  	int len;
>  
> +	/* The previous packet is still being sent */
> +	if (now < q->last) {
> +		qdisc_watchdog_schedule_ns(&q->watchdog, q->last);
> +		return NULL;
> +	}
>  	if (q->credits < 0) {
>  		credits = timediff_to_credits(now - q->last, q->idleslope);
>  
> @@ -212,7 +217,12 @@ static struct sk_buff *cbs_dequeue_soft(struct Qdisc *sch)
>  	credits += q->credits;
>  
>  	q->credits = max_t(s64, credits, q->locredit);
> -	q->last = now;
> +	/* Estimate of the transmission of the last byte of the packet in ns */
> +	if (unlikely(atomic64_read(&q->port_rate) == 0))

Minor suggestion. I would only move 'atomic64_read()' to outside the
condition, so reading 'q->port_rate' is done only once.

It's looking good. When I saw the problems that the software mode
had with larger packets I should have thought of something like this.
Thanks for solving this.


Cheers,
-- 
Vinicius

Re: [PATCH net v3] net: cbs: Fix software cbs to consider packet sending time

[David Miller]

From: Zh-yuan Ye <ye.zh-yuan@socionext.com>
Date: Tue, 24 Mar 2020 17:28:25 +0900

> Currently the software CBS does not consider the packet sending time
> when depleting the credits. It caused the throughput to be
> Idleslope[kbps] * (Port transmit rate[kbps] / |Sendslope[kbps]|) where
> Idleslope * (Port transmit rate / (Idleslope + |Sendslope|)) = Idleslope
> is expected. In order to fix the issue above, this patch takes the time
> when the packet sending completes into account by moving the anchor time
> variable "last" ahead to the send completion time upon transmission and
> adding wait when the next dequeue request comes before the send
> completion time of the previous packet.
> 
> changelog:
> V2->V3:
>  - remove unnecessary whitespace cleanup
>  - add the checks if port_rate is 0 before division
> 
> V1->V2:
>  - combine variable "send_completed" into "last"
>  - add the comment for estimate of the packet sending
> 
> Fixes: 585d763af09c ("net/sched: Introduce Credit Based Shaper (CBS) qdisc")
> Signed-off-by: Zh-yuan Ye <ye.zh-yuan@socionext.com>
> Reviewed-by: Vinicius Costa Gomes <vinicius.gomes@intel.com>

Applied and queued up for -stable, thank you.