* Patch "net/mlx5e: Added BW check for DIM decision mechanism" has been added to the 4.9-stable tree
@ 2017-06-29 16:59 gregkh
0 siblings, 0 replies; only message in thread
From: gregkh @ 2017-06-29 16:59 UTC (permalink / raw)
To: talgi; +Cc: stable, stable-commits
This is a note to let you know that I've just added the patch titled
net/mlx5e: Added BW check for DIM decision mechanism
to the 4.9-stable tree which can be found at:
http://www.kernel.org/git/?p=linux/kernel/git/stable/stable-queue.git;a=summary
The filename of the patch is:
net-mlx5e-added-bw-check-for-dim-decision-mechanism.patch
and it can be found in the queue-4.9 subdirectory.
If you, or anyone else, feels it should not be added to the stable tree,
please let <stable@vger.kernel.org> know about it.
>From foo@baz Thu Jun 29 18:57:46 CEST 2017
From: Tal Gilboa <talgi@mellanox.com>
Date: Mon, 15 May 2017 14:13:16 +0300
Subject: net/mlx5e: Added BW check for DIM decision mechanism
From: Tal Gilboa <talgi@mellanox.com>
[ Upstream commit c3164d2fc48fd4fa0477ab658b644559c3fe9073 ]
DIM (Dynamically-tuned Interrupt Moderation) is a mechanism designed for
changing the channel interrupt moderation values in order to reduce CPU
overhead for all traffic types.
Until now only interrupt and packet rate were sampled.
We found a scenario on which we get a false indication since a change in
DIM caused more aggregation and reduced packet rate while increasing BW.
We now regard a change as succesfull iff:
current_BW > (prev_BW + threshold) or
current_BW ~= prev_BW and current_PR > (prev_PR + threshold) or
current_BW ~= prev_BW and current_PR ~= prev_PR and
current_IR < (prev_IR - threshold)
Where BW = Bandwidth, PR = Packet rate and IR = Interrupt rate
Improvements (ConnectX-4Lx 25GbE, single RX queue, LRO off)
--------------------------------------------------
packet size | before[Mb/s] | after[Mb/s] | gain |
2B | 343.4 | 359.4 | 4.5% |
16B | 2739.7 | 2814.8 | 2.7% |
64B | 9739 | 10185.3 | 4.5% |
Fixes: cb3c7fd4f839 ("net/mlx5e: Support adaptive RX coalescing")
Signed-off-by: Tal Gilboa <talgi@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
---
drivers/net/ethernet/mellanox/mlx5/core/en.h | 2 +
drivers/net/ethernet/mellanox/mlx5/core/en_rx_am.c | 41 +++++++++++----------
2 files changed, 24 insertions(+), 19 deletions(-)
--- a/drivers/net/ethernet/mellanox/mlx5/core/en.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en.h
@@ -283,12 +283,14 @@ struct mlx5e_dma_info {
struct mlx5e_rx_am_stats {
int ppms; /* packets per msec */
+ int bpms; /* bytes per msec */
int epms; /* events per msec */
};
struct mlx5e_rx_am_sample {
ktime_t time;
unsigned int pkt_ctr;
+ unsigned int byte_ctr;
u16 event_ctr;
};
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_rx_am.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_rx_am.c
@@ -183,28 +183,27 @@ static void mlx5e_am_exit_parking(struct
mlx5e_am_step(am);
}
+#define IS_SIGNIFICANT_DIFF(val, ref) \
+ (((100 * abs((val) - (ref))) / (ref)) > 10) /* more than 10% difference */
+
static int mlx5e_am_stats_compare(struct mlx5e_rx_am_stats *curr,
struct mlx5e_rx_am_stats *prev)
{
- int diff;
-
- if (!prev->ppms)
- return curr->ppms ? MLX5E_AM_STATS_BETTER :
- MLX5E_AM_STATS_SAME;
-
- diff = curr->ppms - prev->ppms;
- if (((100 * abs(diff)) / prev->ppms) > 10) /* more than 10% diff */
- return (diff > 0) ? MLX5E_AM_STATS_BETTER :
- MLX5E_AM_STATS_WORSE;
-
- if (!prev->epms)
- return curr->epms ? MLX5E_AM_STATS_WORSE :
+ if (!prev->bpms)
+ return curr->bpms ? MLX5E_AM_STATS_BETTER :
MLX5E_AM_STATS_SAME;
- diff = curr->epms - prev->epms;
- if (((100 * abs(diff)) / prev->epms) > 10) /* more than 10% diff */
- return (diff < 0) ? MLX5E_AM_STATS_BETTER :
- MLX5E_AM_STATS_WORSE;
+ if (IS_SIGNIFICANT_DIFF(curr->bpms, prev->bpms))
+ return (curr->bpms > prev->bpms) ? MLX5E_AM_STATS_BETTER :
+ MLX5E_AM_STATS_WORSE;
+
+ if (IS_SIGNIFICANT_DIFF(curr->ppms, prev->ppms))
+ return (curr->ppms > prev->ppms) ? MLX5E_AM_STATS_BETTER :
+ MLX5E_AM_STATS_WORSE;
+
+ if (IS_SIGNIFICANT_DIFF(curr->epms, prev->epms))
+ return (curr->epms < prev->epms) ? MLX5E_AM_STATS_BETTER :
+ MLX5E_AM_STATS_WORSE;
return MLX5E_AM_STATS_SAME;
}
@@ -266,6 +265,7 @@ static void mlx5e_am_sample(struct mlx5e
{
s->time = ktime_get();
s->pkt_ctr = rq->stats.packets;
+ s->byte_ctr = rq->stats.bytes;
s->event_ctr = rq->cq.event_ctr;
}
@@ -278,12 +278,15 @@ static void mlx5e_am_calc_stats(struct m
/* u32 holds up to 71 minutes, should be enough */
u32 delta_us = ktime_us_delta(end->time, start->time);
unsigned int npkts = end->pkt_ctr - start->pkt_ctr;
+ unsigned int nbytes = end->byte_ctr - start->byte_ctr;
if (!delta_us)
return;
- curr_stats->ppms = (npkts * USEC_PER_MSEC) / delta_us;
- curr_stats->epms = (MLX5E_AM_NEVENTS * USEC_PER_MSEC) / delta_us;
+ curr_stats->ppms = DIV_ROUND_UP(npkts * USEC_PER_MSEC, delta_us);
+ curr_stats->bpms = DIV_ROUND_UP(nbytes * USEC_PER_MSEC, delta_us);
+ curr_stats->epms = DIV_ROUND_UP(MLX5E_AM_NEVENTS * USEC_PER_MSEC,
+ delta_us);
}
void mlx5e_rx_am_work(struct work_struct *work)
Patches currently in stable-queue which might be from talgi@mellanox.com are
queue-4.9/net-mlx5e-added-bw-check-for-dim-decision-mechanism.patch
queue-4.9/net-mlx5e-fix-wrong-indications-in-dim-due-to-counter-wraparound.patch
^ permalink raw reply [flat|nested] only message in thread
only message in thread, other threads:[~2017-06-29 17:00 UTC | newest]
Thread overview: (only message) (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2017-06-29 16:59 Patch "net/mlx5e: Added BW check for DIM decision mechanism" has been added to the 4.9-stable tree gregkh
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.