From: Edward Cree <ecree@solarflare.com>
To: <linux-net-drivers@solarflare.com>, <davem@davemloft.net>
Cc: <netdev@vger.kernel.org>, <dahern@digitalocean.com>
Subject: [PATCH net-next 3/4] sfc: add statistics for ARFS
Date: Fri, 22 Nov 2019 17:57:27 +0000 [thread overview]
Message-ID: <964dd1b3-b26a-e5ee-7ac2-b4643206cb5f@solarflare.com> (raw)
In-Reply-To: <a41f9c29-db34-a2e4-1abd-bfe1a33b442e@solarflare.com>
Report the number of successful and failed insertions, and also the
current count of filters, to aid in tuning e.g. rps_flow_cnt.
Signed-off-by: Edward Cree <ecree@solarflare.com>
---
drivers/net/ethernet/sfc/ethtool.c | 6 ++++++
drivers/net/ethernet/sfc/net_driver.h | 4 ++++
drivers/net/ethernet/sfc/rx.c | 2 ++
3 files changed, 12 insertions(+)
diff --git a/drivers/net/ethernet/sfc/ethtool.c b/drivers/net/ethernet/sfc/ethtool.c
index 8db593fb9699..6a9347cd67f3 100644
--- a/drivers/net/ethernet/sfc/ethtool.c
+++ b/drivers/net/ethernet/sfc/ethtool.c
@@ -56,6 +56,9 @@ static u64 efx_get_atomic_stat(void *field)
#define EFX_ETHTOOL_UINT_CHANNEL_STAT(field) \
EFX_ETHTOOL_STAT(field, channel, n_##field, \
unsigned int, efx_get_uint_stat)
+#define EFX_ETHTOOL_UINT_CHANNEL_STAT_NO_N(field) \
+ EFX_ETHTOOL_STAT(field, channel, field, \
+ unsigned int, efx_get_uint_stat)
#define EFX_ETHTOOL_UINT_TXQ_STAT(field) \
EFX_ETHTOOL_STAT(tx_##field, tx_queue, field, \
@@ -87,6 +90,9 @@ static const struct efx_sw_stat_desc efx_sw_stat_desc[] = {
EFX_ETHTOOL_UINT_CHANNEL_STAT(rx_xdp_bad_drops),
EFX_ETHTOOL_UINT_CHANNEL_STAT(rx_xdp_tx),
EFX_ETHTOOL_UINT_CHANNEL_STAT(rx_xdp_redirect),
+ EFX_ETHTOOL_UINT_CHANNEL_STAT_NO_N(rfs_filter_count),
+ EFX_ETHTOOL_UINT_CHANNEL_STAT(rfs_succeeded),
+ EFX_ETHTOOL_UINT_CHANNEL_STAT(rfs_failed),
};
#define EFX_ETHTOOL_SW_STAT_COUNT ARRAY_SIZE(efx_sw_stat_desc)
diff --git a/drivers/net/ethernet/sfc/net_driver.h b/drivers/net/ethernet/sfc/net_driver.h
index 5b1b882f6c67..ccd480e699d3 100644
--- a/drivers/net/ethernet/sfc/net_driver.h
+++ b/drivers/net/ethernet/sfc/net_driver.h
@@ -444,6 +444,8 @@ enum efx_sync_events_state {
* @rfs_last_expiry: value of jiffies last time some accelerated RFS filters
* were checked for expiry
* @rfs_expire_index: next accelerated RFS filter ID to check for expiry
+ * @n_rfs_succeeded: number of successful accelerated RFS filter insertions
+ * @n_rfs_failed; number of failed accelerated RFS filter insertions
* @filter_work: Work item for efx_filter_rfs_expire()
* @rps_flow_id: Flow IDs of filters allocated for accelerated RFS,
* indexed by filter ID
@@ -497,6 +499,8 @@ struct efx_channel {
unsigned int rfs_filter_count;
unsigned int rfs_last_expiry;
unsigned int rfs_expire_index;
+ unsigned int n_rfs_succeeded;
+ unsigned int n_rfs_failed;
struct work_struct filter_work;
#define RPS_FLOW_ID_INVALID 0xFFFFFFFF
u32 *rps_flow_id;
diff --git a/drivers/net/ethernet/sfc/rx.c b/drivers/net/ethernet/sfc/rx.c
index 252a5f10596d..ef52b24ad9e7 100644
--- a/drivers/net/ethernet/sfc/rx.c
+++ b/drivers/net/ethernet/sfc/rx.c
@@ -1032,6 +1032,7 @@ static void efx_filter_rfs_work(struct work_struct *data)
req->spec.rem_host, ntohs(req->spec.rem_port),
req->spec.loc_host, ntohs(req->spec.loc_port),
req->rxq_index, req->flow_id, rc, arfs_id);
+ channel->n_rfs_succeeded++;
} else {
if (req->spec.ether_type == htons(ETH_P_IP))
netif_dbg(efx, rx_status, efx->net_dev,
@@ -1047,6 +1048,7 @@ static void efx_filter_rfs_work(struct work_struct *data)
req->spec.rem_host, ntohs(req->spec.rem_port),
req->spec.loc_host, ntohs(req->spec.loc_port),
req->rxq_index, req->flow_id, rc, arfs_id);
+ channel->n_rfs_failed++;
/* We're overloading the NIC's filter tables, so let's do a
* chunk of extra expiry work.
*/
next prev parent reply other threads:[~2019-11-22 17:57 UTC|newest]
Thread overview: 11+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-11-22 17:54 [PATCH net-next 0/4] sfc: ARFS expiry improvements Edward Cree
2019-11-22 17:57 ` [PATCH net-next 1/4] sfc: change ARFS expiry mechanism Edward Cree
2019-11-22 21:44 ` David Ahern
2019-11-22 17:57 ` [PATCH net-next 2/4] sfc: suppress MCDI errors from ARFS Edward Cree
2019-11-22 21:44 ` David Ahern
2019-11-22 17:57 ` Edward Cree [this message]
2019-11-23 17:45 ` [PATCH net-next] sfc: fix build without CONFIG_RFS_ACCEL Jakub Kicinski
2019-11-24 22:45 ` Jakub Kicinski
2019-11-25 9:59 ` Edward Cree
2019-11-22 17:57 ` [PATCH net-next 4/4] sfc: do ARFS expiry work occasionally even without NAPI poll Edward Cree
2019-11-23 1:56 ` [PATCH net-next 0/4] sfc: ARFS expiry improvements Jakub Kicinski
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=964dd1b3-b26a-e5ee-7ac2-b4643206cb5f@solarflare.com \
--to=ecree@solarflare.com \
--cc=dahern@digitalocean.com \
--cc=davem@davemloft.net \
--cc=linux-net-drivers@solarflare.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).