From: isubramanian@apm.com (Iyappan Subramanian)
To: linux-arm-kernel@lists.infradead.org
Subject: [PATCH 3/6] drivers: net: xgene: fix ununiform latency across queues
Date: Mon, 9 May 2016 17:04:13 -0700 [thread overview]
Message-ID: <1462838656-22043-4-git-send-email-isubramanian@apm.com> (raw)
In-Reply-To: <1462838656-22043-1-git-send-email-isubramanian@apm.com>
This patch addresses ununiform latency across queues by adding
more queues to match with, upto number of CPU cores.
Also, number of interrupts are increased and the channel numbers
are reordered.
Signed-off-by: Iyappan Subramanian <isubramanian@apm.com>
Tested-by: Toan Le <toanle@apm.com>
---
drivers/net/ethernet/apm/xgene/xgene_enet_main.c | 30 ++++++++++++++++++------
drivers/net/ethernet/apm/xgene/xgene_enet_main.h | 8 +++----
2 files changed, 27 insertions(+), 11 deletions(-)
diff --git a/drivers/net/ethernet/apm/xgene/xgene_enet_main.c b/drivers/net/ethernet/apm/xgene/xgene_enet_main.c
index ed90664..e75d1a9 100644
--- a/drivers/net/ethernet/apm/xgene/xgene_enet_main.c
+++ b/drivers/net/ethernet/apm/xgene/xgene_enet_main.c
@@ -1247,6 +1247,13 @@ static int xgene_enet_get_irqs(struct xgene_enet_pdata *pdata)
for (i = 0; i < max_irqs; i++) {
ret = platform_get_irq(pdev, i);
if (ret <= 0) {
+ if (pdata->phy_mode == PHY_INTERFACE_MODE_XGMII) {
+ max_irqs = i;
+ pdata->rxq_cnt = max_irqs / 2;
+ pdata->txq_cnt = max_irqs / 2;
+ pdata->cq_cnt = max_irqs / 2;
+ break;
+ }
dev_err(dev, "Unable to get ENET IRQ\n");
ret = ret ? : -ENXIO;
return ret;
@@ -1450,19 +1457,28 @@ static void xgene_enet_setup_ops(struct xgene_enet_pdata *pdata)
pdata->port_ops = &xgene_xgport_ops;
pdata->cle_ops = &xgene_cle3in_ops;
pdata->rm = RM0;
- pdata->rxq_cnt = XGENE_NUM_RX_RING;
- pdata->txq_cnt = XGENE_NUM_TX_RING;
- pdata->cq_cnt = XGENE_NUM_TXC_RING;
+ if (!pdata->rxq_cnt) {
+ pdata->rxq_cnt = XGENE_NUM_RX_RING;
+ pdata->txq_cnt = XGENE_NUM_TX_RING;
+ pdata->cq_cnt = XGENE_NUM_TXC_RING;
+ }
break;
}
if (pdata->enet_id == XGENE_ENET1) {
switch (pdata->port_id) {
case 0:
- pdata->cpu_bufnum = START_CPU_BUFNUM_0;
- pdata->eth_bufnum = START_ETH_BUFNUM_0;
- pdata->bp_bufnum = START_BP_BUFNUM_0;
- pdata->ring_num = START_RING_NUM_0;
+ if (pdata->phy_mode == PHY_INTERFACE_MODE_XGMII) {
+ pdata->cpu_bufnum = X2_START_CPU_BUFNUM_0;
+ pdata->eth_bufnum = X2_START_ETH_BUFNUM_0;
+ pdata->bp_bufnum = X2_START_BP_BUFNUM_0;
+ pdata->ring_num = START_RING_NUM_0;
+ } else {
+ pdata->cpu_bufnum = START_CPU_BUFNUM_0;
+ pdata->eth_bufnum = START_ETH_BUFNUM_0;
+ pdata->bp_bufnum = START_BP_BUFNUM_0;
+ pdata->ring_num = START_RING_NUM_0;
+ }
break;
case 1:
if (pdata->phy_mode == PHY_INTERFACE_MODE_XGMII) {
diff --git a/drivers/net/ethernet/apm/xgene/xgene_enet_main.h b/drivers/net/ethernet/apm/xgene/xgene_enet_main.h
index 0a2887b..cc40c30 100644
--- a/drivers/net/ethernet/apm/xgene/xgene_enet_main.h
+++ b/drivers/net/ethernet/apm/xgene/xgene_enet_main.h
@@ -49,10 +49,10 @@
#define XGENE_ENET_MSS 1448
#define XGENE_MIN_ENET_FRAME_SIZE 60
-#define XGENE_MAX_ENET_IRQ 8
-#define XGENE_NUM_RX_RING 4
-#define XGENE_NUM_TX_RING 4
-#define XGENE_NUM_TXC_RING 4
+#define XGENE_MAX_ENET_IRQ 16
+#define XGENE_NUM_RX_RING 8
+#define XGENE_NUM_TX_RING 8
+#define XGENE_NUM_TXC_RING 8
#define START_CPU_BUFNUM_0 0
#define START_ETH_BUFNUM_0 2
--
1.9.1
next prev parent reply other threads:[~2016-05-10 0:04 UTC|newest]
Thread overview: 8+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-05-10 0:04 [PATCH 0/6] drivers: net: xgene: Bug fixes Iyappan Subramanian
2016-05-10 0:04 ` [PATCH 1/6] drivers: net: xgene: fix IPv4 forward crash Iyappan Subramanian
2016-05-10 0:04 ` [PATCH 2/6] drivers: net: xgene: fix sharing of irqs Iyappan Subramanian
2016-05-10 0:04 ` Iyappan Subramanian [this message]
2016-05-10 0:04 ` [PATCH 4/6] drivers: net: xgene: fix statistics counters race condition Iyappan Subramanian
2016-05-10 0:04 ` [PATCH 5/6] drivers: net: xgene: Using static MSS values Iyappan Subramanian
2016-05-10 19:53 ` David Miller
2016-05-10 0:04 ` [PATCH 6/6] drivers: net: xgene: fix register offset Iyappan Subramanian
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=1462838656-22043-4-git-send-email-isubramanian@apm.com \
--to=isubramanian@apm.com \
--cc=linux-arm-kernel@lists.infradead.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).