netdev.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Alexander Duyck <aduyck@mirantis.com>
To: netdev@vger.kernel.org, intel-wired-lan@lists.osuosl.org
Cc: jeffrey.t.kirsher@intel.com
Subject: [PATCH v2] ixgbevf: Limit lowest interrupt rate for adaptive interrupt moderation to 12K
Date: Tue, 29 Sep 2015 13:11:15 -0700	[thread overview]
Message-ID: <20150929201022.1388.91691.stgit@ahduyck-vm-fedora22> (raw)
In-Reply-To: <20150929195346.1125.48492.stgit@ahduyck-vm-fedora22>

This patch is the ixgbevf version of commit 8ac34f10a5ea4 "ixgbe: Limit
lowest interrupt rate for adaptive interrupt moderation to 12K"

The same logic applies here as well as the same results since a netperf
test will starve for memory in the time from one Tx interrupt to the next.
As a result the ixgbevf driver underperformed when compared to vhost_net.

Signed-off-by: Alexander Duyck <aduyck@mirantis.com>
---

v2: Fixed comment describing lowest interrupt rate as 8000 ints/s

 drivers/net/ethernet/intel/ixgbevf/ethtool.c      |    2 +-
 drivers/net/ethernet/intel/ixgbevf/ixgbevf.h      |    3 +--
 drivers/net/ethernet/intel/ixgbevf/ixgbevf_main.c |    6 +++---
 3 files changed, 5 insertions(+), 6 deletions(-)

diff --git a/drivers/net/ethernet/intel/ixgbevf/ethtool.c b/drivers/net/ethernet/intel/ixgbevf/ethtool.c
index d3e5f5b37999..c48aef613b0a 100644
--- a/drivers/net/ethernet/intel/ixgbevf/ethtool.c
+++ b/drivers/net/ethernet/intel/ixgbevf/ethtool.c
@@ -774,7 +774,7 @@ static int ixgbevf_set_coalesce(struct net_device *netdev,
 		adapter->tx_itr_setting = ec->tx_coalesce_usecs;
 
 	if (adapter->tx_itr_setting == 1)
-		tx_itr_param = IXGBE_10K_ITR;
+		tx_itr_param = IXGBE_12K_ITR;
 	else
 		tx_itr_param = adapter->tx_itr_setting;
 
diff --git a/drivers/net/ethernet/intel/ixgbevf/ixgbevf.h b/drivers/net/ethernet/intel/ixgbevf/ixgbevf.h
index 04c7ec8446e0..8b437b54cd98 100644
--- a/drivers/net/ethernet/intel/ixgbevf/ixgbevf.h
+++ b/drivers/net/ethernet/intel/ixgbevf/ixgbevf.h
@@ -326,8 +326,7 @@ static inline bool ixgbevf_qv_disable(struct ixgbevf_q_vector *q_vector)
 #define IXGBE_MIN_RSC_ITR	24
 #define IXGBE_100K_ITR		40
 #define IXGBE_20K_ITR		200
-#define IXGBE_10K_ITR		400
-#define IXGBE_8K_ITR		500
+#define IXGBE_12K_ITR		336
 
 /* Helper macros to switch between ints/sec and what the register uses.
  * And yes, it's the same math going both ways.  The lowest value
diff --git a/drivers/net/ethernet/intel/ixgbevf/ixgbevf_main.c b/drivers/net/ethernet/intel/ixgbevf/ixgbevf_main.c
index 35da2d74e73e..1551b5dfde7e 100644
--- a/drivers/net/ethernet/intel/ixgbevf/ixgbevf_main.c
+++ b/drivers/net/ethernet/intel/ixgbevf/ixgbevf_main.c
@@ -1136,7 +1136,7 @@ static void ixgbevf_configure_msix(struct ixgbevf_adapter *adapter)
 		if (q_vector->tx.ring && !q_vector->rx.ring) {
 			/* Tx only vector */
 			if (adapter->tx_itr_setting == 1)
-				q_vector->itr = IXGBE_10K_ITR;
+				q_vector->itr = IXGBE_12K_ITR;
 			else
 				q_vector->itr = adapter->tx_itr_setting;
 		} else {
@@ -1194,7 +1194,7 @@ static void ixgbevf_update_itr(struct ixgbevf_q_vector *q_vector,
 	/* simple throttle rate management
 	 *    0-20MB/s lowest (100000 ints/s)
 	 *   20-100MB/s low   (20000 ints/s)
-	 *  100-1249MB/s bulk (8000 ints/s)
+	 *  100-1249MB/s bulk (12000 ints/s)
 	 */
 	/* what was last interrupt timeslice? */
 	timepassed_us = q_vector->itr >> 2;
@@ -1245,7 +1245,7 @@ static void ixgbevf_set_itr(struct ixgbevf_q_vector *q_vector)
 		break;
 	case bulk_latency:
 	default:
-		new_itr = IXGBE_8K_ITR;
+		new_itr = IXGBE_12K_ITR;
 		break;
 	}
 

      reply	other threads:[~2015-09-29 20:11 UTC|newest]

Thread overview: 2+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-09-29 19:53 [PATCH] ixgbevf: Limit lowest interrupt rate for adaptive interrupt moderation to 12K Alexander Duyck
2015-09-29 20:11 ` Alexander Duyck [this message]

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20150929201022.1388.91691.stgit@ahduyck-vm-fedora22 \
    --to=aduyck@mirantis.com \
    --cc=intel-wired-lan@lists.osuosl.org \
    --cc=jeffrey.t.kirsher@intel.com \
    --cc=netdev@vger.kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).