From: "Wei Hu (Xavier)" <huwei013@chinasoftinc.com>
To: <dev@dpdk.org>
Subject: [dpdk-dev] [PATCH 2/2] net/hns3: get Rx/Tx queue fbd in extend device statistics
Date: Tue, 21 Apr 2020 11:32:15 +0800 [thread overview]
Message-ID: <20200421033215.20584-3-huwei013@chinasoftinc.com> (raw)
In-Reply-To: <20200421033215.20584-1-huwei013@chinasoftinc.com>
From: Chengwen Feng <fengchengwen@huawei.com>
This patch adds getting Rx/Tx queue fbd information in extended device
statistics. The upper level application can get them by calling the
rte_eth_xstats_get API function.
The fbd registers of every Rx/Tx queue are very useful to identify the
Rx/Tx bottleneck.
1. The Rx queue fbd register is the number of the unprocessed buffer
descriptors which are waiting for driver to process;
2. The Tx queue fbd register is the number of the unprocessed buffer
descriptors which are waiting for network engine hardware to process.
As a result, we get the following output information in testpmd applcation
by using the command "show port xstats" as below:
rx_q0RX_QUEUE_FBD: 19
rx_q1RX_QUEUE_FBD: 18
tx_q0TX_QUEUE_FBD: 0
tx_q1TX_QUEUE_FBD: 0
Signed-off-by: Chengwen Feng <fengchengwen@huawei.com>
Signed-off-by: Wei Hu (Xavier) <xavier.huwei@huawei.com>
Signed-off-by: Huisong Li <lihuisong@huawei.com>
---
drivers/net/hns3/hns3_stats.c | 174 +++++++++++++++++-----------------
drivers/net/hns3/hns3_stats.h | 3 +-
2 files changed, 90 insertions(+), 87 deletions(-)
diff --git a/drivers/net/hns3/hns3_stats.c b/drivers/net/hns3/hns3_stats.c
index b3797ec53..2891875f4 100644
--- a/drivers/net/hns3/hns3_stats.c
+++ b/drivers/net/hns3/hns3_stats.c
@@ -13,6 +13,7 @@
#include "hns3_ethdev.h"
#include "hns3_rxtx.h"
#include "hns3_logs.h"
+#include "hns3_regs.h"
/* MAC statistics */
static const struct hns3_xstats_name_offset hns3_mac_strings[] = {
@@ -233,6 +234,16 @@ static const struct hns3_xstats_name_offset hns3_rx_bd_error_strings[] = {
HNS3_RX_BD_ERROR_STATS_FIELD_OFFSET(ol4_csum_erros)}
};
+/* The statistic of rx queue */
+static const struct hns3_xstats_name_offset hns3_rx_queue_strings[] = {
+ {"RX_QUEUE_FBD", HNS3_RING_RX_FBDNUM_REG}
+};
+
+/* The statistic of tx queue */
+static const struct hns3_xstats_name_offset hns3_tx_queue_strings[] = {
+ {"TX_QUEUE_FBD", HNS3_RING_TX_FBDNUM_REG}
+};
+
#define HNS3_NUM_MAC_STATS (sizeof(hns3_mac_strings) / \
sizeof(hns3_mac_strings[0]))
@@ -245,6 +256,12 @@ static const struct hns3_xstats_name_offset hns3_rx_bd_error_strings[] = {
#define HNS3_NUM_RX_BD_ERROR_XSTATS (sizeof(hns3_rx_bd_error_strings) / \
sizeof(hns3_rx_bd_error_strings[0]))
+#define HNS3_NUM_RX_QUEUE_STATS (sizeof(hns3_rx_queue_strings) / \
+ sizeof(hns3_rx_queue_strings[0]))
+
+#define HNS3_NUM_TX_QUEUE_STATS (sizeof(hns3_tx_queue_strings) / \
+ sizeof(hns3_tx_queue_strings[0]))
+
#define HNS3_FIX_NUM_STATS (HNS3_NUM_MAC_STATS + HNS3_NUM_ERROR_INT_XSTATS + \
HNS3_NUM_RESET_XSTATS)
@@ -541,13 +558,16 @@ static int
hns3_xstats_calc_num(struct rte_eth_dev *dev)
{
struct hns3_adapter *hns = dev->data->dev_private;
+ int bderr_stats = dev->data->nb_rx_queues * HNS3_NUM_RX_BD_ERROR_XSTATS;
+ int rx_queue_stats = dev->data->nb_rx_queues * HNS3_NUM_RX_QUEUE_STATS;
+ int tx_queue_stats = dev->data->nb_tx_queues * HNS3_NUM_TX_QUEUE_STATS;
if (hns->is_vf)
- return dev->data->nb_rx_queues * HNS3_NUM_RX_BD_ERROR_XSTATS +
- HNS3_NUM_RESET_XSTATS;
+ return bderr_stats + rx_queue_stats + tx_queue_stats +
+ HNS3_NUM_RESET_XSTATS;
else
- return dev->data->nb_rx_queues * HNS3_NUM_RX_BD_ERROR_XSTATS +
- HNS3_FIX_NUM_STATS;
+ return bderr_stats + rx_queue_stats + tx_queue_stats +
+ HNS3_FIX_NUM_STATS;
}
/*
@@ -573,6 +593,7 @@ hns3_dev_xstats_get(struct rte_eth_dev *dev, struct rte_eth_xstat *xstats,
struct hns3_mac_stats *mac_stats = &hw->mac_stats;
struct hns3_reset_stats *reset_stats = &hw->reset.stats;
struct hns3_rx_queue *rxq;
+ uint32_t reg_offset;
uint16_t i, j;
char *addr;
int count;
@@ -621,7 +642,7 @@ hns3_dev_xstats_get(struct rte_eth_dev *dev, struct rte_eth_xstat *xstats,
}
/* Get the Rx BD errors stats */
- for (j = 0; j != dev->data->nb_rx_queues; ++j) {
+ for (j = 0; j < dev->data->nb_rx_queues; j++) {
for (i = 0; i < HNS3_NUM_RX_BD_ERROR_XSTATS; i++) {
rxq = dev->data->rx_queues[j];
addr = (char *)rxq + hns3_rx_bd_error_strings[i].offset;
@@ -631,6 +652,30 @@ hns3_dev_xstats_get(struct rte_eth_dev *dev, struct rte_eth_xstat *xstats,
}
}
+ /* Get rx queue stats */
+ for (j = 0; j < dev->data->nb_rx_queues; j++) {
+ for (i = 0; i < HNS3_NUM_RX_QUEUE_STATS; i++) {
+ reg_offset = HNS3_TQP_REG_OFFSET +
+ HNS3_TQP_REG_SIZE * j;
+ xstats[count].value = hns3_read_dev(hw,
+ reg_offset + hns3_rx_queue_strings[i].offset);
+ xstats[count].id = count;
+ count++;
+ }
+ }
+
+ /* Get tx queue stats */
+ for (j = 0; j < dev->data->nb_tx_queues; j++) {
+ for (i = 0; i < HNS3_NUM_TX_QUEUE_STATS; i++) {
+ reg_offset = HNS3_TQP_REG_OFFSET +
+ HNS3_TQP_REG_SIZE * j;
+ xstats[count].value = hns3_read_dev(hw,
+ reg_offset + hns3_tx_queue_strings[i].offset);
+ xstats[count].id = count;
+ count++;
+ }
+ }
+
return count;
}
@@ -659,7 +704,7 @@ hns3_dev_xstats_get(struct rte_eth_dev *dev, struct rte_eth_xstat *xstats,
* is the number of entries filled in the stats table.
*/
int
-hns3_dev_xstats_get_names(__rte_unused struct rte_eth_dev *dev,
+hns3_dev_xstats_get_names(struct rte_eth_dev *dev,
struct rte_eth_xstat_name *xstats_names,
__rte_unused unsigned int size)
{
@@ -705,6 +750,24 @@ hns3_dev_xstats_get_names(__rte_unused struct rte_eth_dev *dev,
}
}
+ for (j = 0; j < dev->data->nb_rx_queues; j++) {
+ for (i = 0; i < HNS3_NUM_RX_QUEUE_STATS; i++) {
+ snprintf(xstats_names[count].name,
+ sizeof(xstats_names[count].name),
+ "rx_q%u%s", j, hns3_rx_queue_strings[i].name);
+ count++;
+ }
+ }
+
+ for (j = 0; j < dev->data->nb_tx_queues; j++) {
+ for (i = 0; i < HNS3_NUM_TX_QUEUE_STATS; i++) {
+ snprintf(xstats_names[count].name,
+ sizeof(xstats_names[count].name),
+ "tx_q%u%s", j, hns3_tx_queue_strings[i].name);
+ count++;
+ }
+ }
+
return count;
}
@@ -735,18 +798,13 @@ int
hns3_dev_xstats_get_by_id(struct rte_eth_dev *dev, const uint64_t *ids,
uint64_t *values, uint32_t size)
{
+ const uint32_t cnt_stats = hns3_xstats_calc_num(dev);
struct hns3_adapter *hns = dev->data->dev_private;
- struct hns3_pf *pf = &hns->pf;
+ struct rte_eth_xstat *values_copy;
struct hns3_hw *hw = &hns->hw;
- struct hns3_mac_stats *mac_stats = &hw->mac_stats;
- struct hns3_reset_stats *reset_stats = &hw->reset.stats;
- struct hns3_rx_queue *rxq;
- const uint32_t cnt_stats = hns3_xstats_calc_num(dev);
- uint64_t *values_copy;
+ uint32_t count_value;
uint64_t len;
- uint32_t count = 0;
- uint16_t i, j;
- char *addr;
+ uint32_t i;
int ret;
if (ids == NULL || size < cnt_stats)
@@ -759,7 +817,7 @@ hns3_dev_xstats_get_by_id(struct rte_eth_dev *dev, const uint64_t *ids,
return ret;
}
- len = cnt_stats * HNS3_VALUES_BYTES;
+ len = cnt_stats * sizeof(struct rte_eth_xstat);
values_copy = rte_zmalloc("hns3_xstats_values", len, 0);
if (values_copy == NULL) {
hns3_err(hw, "Failed to allocate %" PRIx64 " bytes needed "
@@ -767,36 +825,10 @@ hns3_dev_xstats_get_by_id(struct rte_eth_dev *dev, const uint64_t *ids,
return -ENOMEM;
}
- if (!hns->is_vf) {
- /* Get MAC name from hw->hw_xstats.mac_stats */
- for (i = 0; i < HNS3_NUM_MAC_STATS; i++) {
- addr = (char *)mac_stats + hns3_mac_strings[i].offset;
- values_copy[count] = *(uint64_t *)addr;
- count++;
- }
-
- for (i = 0; i < HNS3_NUM_ERROR_INT_XSTATS; i++) {
- addr = (char *)&pf->abn_int_stats +
- hns3_error_int_stats_strings[i].offset;
- values_copy[count] = *(uint64_t *)addr;
- count++;
- }
- }
-
- for (i = 0; i < HNS3_NUM_RESET_XSTATS; i++) {
- addr = (char *)reset_stats +
- hns3_reset_stats_strings[i].offset;
- values_copy[count] = *(uint64_t *)addr;
- count++;
- }
-
- for (j = 0; j != dev->data->nb_rx_queues; ++j) {
- for (i = 0; i < HNS3_NUM_RX_BD_ERROR_XSTATS; i++) {
- rxq = dev->data->rx_queues[j];
- addr = (char *)rxq + hns3_rx_bd_error_strings[i].offset;
- values_copy[count] = *(uint64_t *)addr;
- count++;
- }
+ count_value = hns3_dev_xstats_get(dev, values_copy, cnt_stats);
+ if (count_value != cnt_stats) {
+ rte_free(values_copy);
+ return -EINVAL;
}
for (i = 0; i < size; i++) {
@@ -806,7 +838,8 @@ hns3_dev_xstats_get_by_id(struct rte_eth_dev *dev, const uint64_t *ids,
rte_free(values_copy);
return -EINVAL;
}
- memcpy(&values[i], &values_copy[ids[i]], sizeof(values[i]));
+ memcpy(&values[i], &values_copy[ids[i]].value,
+ sizeof(values[i]));
}
rte_free(values_copy);
@@ -839,67 +872,38 @@ hns3_dev_xstats_get_names_by_id(struct rte_eth_dev *dev,
struct rte_eth_xstat_name *xstats_names,
const uint64_t *ids, uint32_t size)
{
+ const uint32_t cnt_stats = hns3_xstats_calc_num(dev);
struct hns3_adapter *hns = dev->data->dev_private;
- struct rte_eth_xstat_name *xstats_names_copy;
+ struct rte_eth_xstat_name *names_copy;
struct hns3_hw *hw = &hns->hw;
- const uint32_t cnt_stats = hns3_xstats_calc_num(dev);
- uint16_t count_name = 0;
- uint16_t i, j;
uint64_t len;
+ uint32_t i;
if (ids == NULL || xstats_names == NULL)
return cnt_stats;
len = cnt_stats * sizeof(struct rte_eth_xstat_name);
- xstats_names_copy = rte_zmalloc("hns3_xstats_names", len, 0);
- if (xstats_names_copy == NULL) {
+ names_copy = rte_zmalloc("hns3_xstats_names", len, 0);
+ if (names_copy == NULL) {
hns3_err(hw, "Failed to allocate %" PRIx64 " bytes needed "
"to store statistics names", len);
return -ENOMEM;
}
- if (!hns->is_vf) {
- for (i = 0; i < HNS3_NUM_MAC_STATS; i++) {
- snprintf(xstats_names_copy[count_name].name,
- sizeof(xstats_names_copy[count_name].name),
- "%s", hns3_mac_strings[i].name);
- count_name++;
- }
- for (i = 0; i < HNS3_NUM_ERROR_INT_XSTATS; i++) {
- snprintf(xstats_names_copy[count_name].name,
- sizeof(xstats_names_copy[count_name].name),
- "%s", hns3_error_int_stats_strings[i].name);
- count_name++;
- }
- }
- for (i = 0; i < HNS3_NUM_RESET_XSTATS; i++) {
- snprintf(xstats_names_copy[count_name].name,
- sizeof(xstats_names_copy[count_name].name),
- "%s", hns3_reset_stats_strings[i].name);
- count_name++;
- }
- for (j = 0; j != dev->data->nb_rx_queues; ++j) {
- for (i = 0; i < HNS3_NUM_RX_BD_ERROR_XSTATS; i++) {
- snprintf(xstats_names_copy[count_name].name,
- sizeof(xstats_names_copy[count_name].name),
- "rx_q%u%s", j,
- hns3_rx_bd_error_strings[i].name);
- count_name++;
- }
- }
+ (void)hns3_dev_xstats_get_names(dev, names_copy, cnt_stats);
for (i = 0; i < size; i++) {
if (ids[i] >= cnt_stats) {
hns3_err(hw, "ids[%d] (%" PRIx64 ") is invalid, "
"should < %u", i, ids[i], cnt_stats);
- rte_free(xstats_names_copy);
+ rte_free(names_copy);
return -EINVAL;
}
snprintf(xstats_names[i].name, sizeof(xstats_names[i].name),
- "%s", xstats_names_copy[ids[i]].name);
+ "%s", names_copy[ids[i]].name);
}
- rte_free(xstats_names_copy);
+ rte_free(names_copy);
return size;
}
diff --git a/drivers/net/hns3/hns3_stats.h b/drivers/net/hns3/hns3_stats.h
index 365bae083..0993c5f57 100644
--- a/drivers/net/hns3/hns3_stats.h
+++ b/drivers/net/hns3/hns3_stats.h
@@ -9,7 +9,6 @@
#define HNS3_MAC_CMD_NUM 21
#define HNS3_RD_FIRST_STATS_NUM 2
#define HNS3_RD_OTHER_STATS_NUM 4
-#define HNS3_VALUES_BYTES 8
/* TQP stats */
struct hns3_tqp_stats {
@@ -137,7 +136,7 @@ int hns3_stats_get(struct rte_eth_dev *dev, struct rte_eth_stats *rte_stats);
int hns3_dev_xstats_get(struct rte_eth_dev *dev, struct rte_eth_xstat *xstats,
unsigned int n);
int hns3_dev_xstats_reset(struct rte_eth_dev *dev);
-int hns3_dev_xstats_get_names(__rte_unused struct rte_eth_dev *dev,
+int hns3_dev_xstats_get_names(struct rte_eth_dev *dev,
struct rte_eth_xstat_name *xstats_names,
__rte_unused unsigned int size);
int hns3_dev_xstats_get_by_id(struct rte_eth_dev *dev,
--
2.23.0
next prev parent reply other threads:[~2020-04-21 3:32 UTC|newest]
Thread overview: 4+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-04-21 3:32 [dpdk-dev] [PATCH 0/2] updates for hns3 PMD driver Wei Hu (Xavier)
2020-04-21 3:32 ` [dpdk-dev] [PATCH 1/2] net/hns3: modify the format for firmware version Wei Hu (Xavier)
2020-04-21 3:32 ` Wei Hu (Xavier) [this message]
2020-04-21 17:21 ` [dpdk-dev] [PATCH 0/2] updates for hns3 PMD driver Ferruh Yigit
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=20200421033215.20584-3-huwei013@chinasoftinc.com \
--to=huwei013@chinasoftinc.com \
--cc=dev@dpdk.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).