Netdev List
 help / color / mirror / Atom feed
From: Pei Lee Ling <pei.lee.ling@intel.com>
To: mkubecek@suse.cz
Cc: netdev@vger.kernel.org, shubham.das@intel.com,
	siddaraju.dh@intel.com, balaji.chintalapalle@intel.com,
	pei.lee.ling@intel.com
Subject: [RFC PATCH ethtool-next] ethtool: Add clear FEC stats Support
Date: Mon, 21 Sep 2026 11:25:48 -0500	[thread overview]
Message-ID: <20260921162548.2376757-1-pei.lee.ling@intel.com> (raw)

In high-speed links such as 25G, 100G, and 200G, FEC counters continuously
accumulate over time. When troubleshooting intermittent physical-layer
issues, users need to establish a clean baseline without resetting the
entire device interface. Therefore, this dedicated FEC-stats-clear
operation avoids unnecessary operational disruptions.

Introducing a fec-clear command would also help to standardize behavior 
across different network drivers, providing a more consistent and predictable
user experience.

To clear-fec-stats:
ethtool --clear-fec-stats <ETH1>

Signed-off-by: Pei Lee Ling <pei.lee.ling@intel.com>
---
 ethtool.c                              |  5 ++++
 netlink/extapi.h                       |  2 ++
 netlink/fec.c                          | 40 ++++++++++++++++++++++++++
 uapi/linux/ethtool_netlink_generated.h |  1 +
 4 files changed, 48 insertions(+)

diff --git a/ethtool.c b/ethtool.c
index 3076c12..92d641f 100644
--- a/ethtool.c
+++ b/ethtool.c
@@ -6389,6 +6389,11 @@ static const struct option args[] = {
 		.help	= "Set FEC settings",
 		.xhelp	= "		[ encoding auto|off|rs|baser|llrs [...] ]\n"
 	},
+	{
+		.opts   = "--clear-fec-stats",
+		.nlfunc = nl_cfecstats,
+		.help   = "Clear FEC statistics",
+	},
 	{
 		.opts	= "-Q|--per-queue",
 		.func	= do_perqueue,
diff --git a/netlink/extapi.h b/netlink/extapi.h
index b0e4588..272677e 100644
--- a/netlink/extapi.h
+++ b/netlink/extapi.h
@@ -43,6 +43,7 @@ int nl_cable_test_tdr(struct cmd_context *ctx);
 int nl_gtunnels(struct cmd_context *ctx);
 int nl_gfec(struct cmd_context *ctx);
 int nl_sfec(struct cmd_context *ctx);
+int nl_cfecstats(struct cmd_context *ctx);
 bool nl_gstats_chk(struct cmd_context *ctx);
 int nl_gstats(struct cmd_context *ctx);
 int nl_gmodule(struct cmd_context *ctx);
@@ -124,6 +125,7 @@ nl_get_eeprom_page(struct cmd_context *ctx __maybe_unused,
 #define nl_gtunnels		NULL
 #define nl_gfec			NULL
 #define nl_sfec			NULL
+#define nl_cfecstats		NULL
 #define nl_gstats_chk		NULL
 #define nl_gstats		NULL
 #define nl_getmodule		NULL
diff --git a/netlink/fec.c b/netlink/fec.c
index c014749..60630d5 100644
--- a/netlink/fec.c
+++ b/netlink/fec.c
@@ -438,3 +438,43 @@ int nl_sfec(struct cmd_context *ctx)
 	else
 		return nlctx->exit_code ?: 83;
 }
+
+/* FEC_STATS_CLEAR */
+
+int nl_cfecstats(struct cmd_context *ctx)
+{
+	struct nl_context *nlctx = ctx->nlctx;
+	struct nl_msg_buff *msgbuff;
+	struct nl_socket *nlsk;
+	u32 flags = 0;
+	int ret;
+
+	if (netlink_cmd_check(ctx, ETHTOOL_MSG_FEC_SET, false))
+		return -EOPNOTSUPP;
+
+	nlctx->devname = ctx->devname;
+	nlsk = nlctx->ethnl_socket;
+	msgbuff = &nlsk->msgbuff;
+
+	ret = msg_init(nlctx, msgbuff, ETHTOOL_MSG_FEC_SET,
+		       NLM_F_REQUEST | NLM_F_ACK);
+	if (ret < 0)
+		return 2;
+
+	if (ethnla_fill_header(msgbuff, ETHTOOL_A_FEC_HEADER,
+			       ctx->devname, 0))
+		return -EMSGSIZE;
+
+	if (ethnla_put_u8(msgbuff, ETHTOOL_A_FEC_STATS_CLEAR, flags))
+		return -EMSGSIZE;
+
+	ret = nlsock_sendmsg(nlsk, NULL);
+	if (ret < 0)
+		return 83;
+
+	ret = nlsock_process_reply(nlsk, nomsg_reply_cb, nlctx);
+	if (ret == 0)
+		return 0;
+	else
+		return nlctx->exit_code ?: 83;
+}
diff --git a/uapi/linux/ethtool_netlink_generated.h b/uapi/linux/ethtool_netlink_generated.h
index 572c5f9..2e66531 100644
--- a/uapi/linux/ethtool_netlink_generated.h
+++ b/uapi/linux/ethtool_netlink_generated.h
@@ -595,6 +595,7 @@ enum {
 	ETHTOOL_A_FEC_AUTO,
 	ETHTOOL_A_FEC_ACTIVE,
 	ETHTOOL_A_FEC_STATS,
+	ETHTOOL_A_FEC_STATS_CLEAR,
 
 	__ETHTOOL_A_FEC_CNT,
 	ETHTOOL_A_FEC_MAX = (__ETHTOOL_A_FEC_CNT - 1)
-- 
2.43.0


                 reply	other threads:[~2026-09-21 16:26 UTC|newest]

Thread overview: [no followups] expand[flat|nested]  mbox.gz  Atom feed

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=20260921162548.2376757-1-pei.lee.ling@intel.com \
    --to=pei.lee.ling@intel.com \
    --cc=balaji.chintalapalle@intel.com \
    --cc=mkubecek@suse.cz \
    --cc=netdev@vger.kernel.org \
    --cc=shubham.das@intel.com \
    --cc=siddaraju.dh@intel.com \
    /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