mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: Joe Damato <jdamato@fastly.com>
To: netdev@vger.kernel.org
Cc: edumazet@google.com, amritha.nambiar@intel.com,
	sridhar.samudrala@intel.com, sdf@fomichev.me, bjorn@rivosinc.com,
	hch@infradead.org, willy@infradead.org,
	willemdebruijn.kernel@gmail.com, skhawaja@google.com,
	kuba@kernel.org, Joe Damato <jdamato@fastly.com>,
	Martin Karsten <mkarsten@uwaterloo.ca>,
	"David S. Miller" <davem@davemloft.net>,
	Paolo Abeni <pabeni@redhat.com>, Jiri Pirko <jiri@resnulli.us>,
	Sebastian Andrzej Siewior <bigeasy@linutronix.de>,
	Lorenzo Bianconi <lorenzo@kernel.org>,
	Breno Leitao <leitao@debian.org>,
	Johannes Berg <johannes.berg@intel.com>,
	Alexander Lobakin <aleksander.lobakin@intel.com>,
	linux-kernel@vger.kernel.org (open list)
Subject: [PATCH net-next 1/5] net: napi: Make napi_defer_hard_irqs per-NAPI
Date: Thu, 29 Aug 2024 13:11:57 +0000	[thread overview]
Message-ID: <20240829131214.169977-2-jdamato@fastly.com> (raw)
In-Reply-To: <20240829131214.169977-1-jdamato@fastly.com>

Allow per-NAPI defer_hard_irqs setting.

The existing sysfs parameter is respected; writes to sysfs will write to
all NAPI structs for the device and the net_device defer_hard_irq field.
Reads from sysfs will read from the net_device field.

sysfs code was updated to guard against what appears to be a potential
overflow as the field is an int, but the value passed in is an unsigned
long.

The ability to set defer_hard_irqs on specific NAPI instances will be
added in a later commit, via netdev-genl.

Signed-off-by: Joe Damato <jdamato@fastly.com>
Reviewed-by: Martin Karsten <mkarsten@uwaterloo.ca>
Tested-by: Martin Karsten <mkarsten@uwaterloo.ca>
---
 include/linux/netdevice.h | 23 +++++++++++++++++++++++
 net/core/dev.c            | 29 ++++++++++++++++++++++++++---
 net/core/net-sysfs.c      |  5 ++++-
 3 files changed, 53 insertions(+), 4 deletions(-)

diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index fce70990b209..7d53380da4c0 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -371,6 +371,7 @@ struct napi_struct {
 	struct list_head	rx_list; /* Pending GRO_NORMAL skbs */
 	int			rx_count; /* length of rx_list */
 	unsigned int		napi_id;
+	int			defer_hard_irqs;
 	struct hrtimer		timer;
 	struct task_struct	*thread;
 	/* control-path-only fields follow */
@@ -534,6 +535,28 @@ static inline void napi_schedule_irqoff(struct napi_struct *n)
 		__napi_schedule_irqoff(n);
 }
 
+/**
+ * napi_get_defer_hard_irqs - get the NAPI's defer_hard_irqs
+ * @n: napi struct to get the defer_hard_irqs field from
+ *
+ * Returns the per-NAPI value of the defar_hard_irqs field.
+ */
+int napi_get_defer_hard_irqs(const struct napi_struct *n);
+
+/**
+ * napi_set_defer_hard_irqs - set the defer_hard_irqs for a napi
+ * @n: napi_struct to set the defer_hard_irqs field
+ * @defer: the value the field should be set to
+ */
+void napi_set_defer_hard_irqs(struct napi_struct *n, int defer);
+
+/**
+ * netdev_set_defer_hard_irqs - set defer_hard_irqs for all NAPIs of a netdev
+ * @netdev: the net_device for which all NAPIs will have their defer_hard_irqs set
+ * @defer: the defer_hard_irqs value to set
+ */
+void netdev_set_defer_hard_irqs(struct net_device *netdev, int defer);
+
 /**
  * napi_complete_done - NAPI processing complete
  * @n: NAPI context
diff --git a/net/core/dev.c b/net/core/dev.c
index 63987b8b7c85..f7baff0da057 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -6212,6 +6212,28 @@ void __napi_schedule_irqoff(struct napi_struct *n)
 }
 EXPORT_SYMBOL(__napi_schedule_irqoff);
 
+int napi_get_defer_hard_irqs(const struct napi_struct *n)
+{
+	return READ_ONCE(n->defer_hard_irqs);
+}
+EXPORT_SYMBOL_GPL(napi_get_defer_hard_irqs);
+
+void napi_set_defer_hard_irqs(struct napi_struct *n, int defer)
+{
+	WRITE_ONCE(n->defer_hard_irqs, defer);
+}
+EXPORT_SYMBOL_GPL(napi_set_defer_hard_irqs);
+
+void netdev_set_defer_hard_irqs(struct net_device *netdev, int defer)
+{
+	struct napi_struct *napi;
+
+	WRITE_ONCE(netdev->napi_defer_hard_irqs, defer);
+	list_for_each_entry(napi, &netdev->napi_list, dev_list)
+		napi_set_defer_hard_irqs(napi, defer);
+}
+EXPORT_SYMBOL_GPL(netdev_set_defer_hard_irqs);
+
 bool napi_complete_done(struct napi_struct *n, int work_done)
 {
 	unsigned long flags, val, new, timeout = 0;
@@ -6230,7 +6252,7 @@ bool napi_complete_done(struct napi_struct *n, int work_done)
 	if (work_done) {
 		if (n->gro_bitmask)
 			timeout = READ_ONCE(n->dev->gro_flush_timeout);
-		n->defer_hard_irqs_count = READ_ONCE(n->dev->napi_defer_hard_irqs);
+		n->defer_hard_irqs_count = napi_get_defer_hard_irqs(n);
 	}
 	if (n->defer_hard_irqs_count > 0) {
 		n->defer_hard_irqs_count--;
@@ -6368,7 +6390,7 @@ static void busy_poll_stop(struct napi_struct *napi, void *have_poll_lock,
 	bpf_net_ctx = bpf_net_ctx_set(&__bpf_net_ctx);
 
 	if (flags & NAPI_F_PREFER_BUSY_POLL) {
-		napi->defer_hard_irqs_count = READ_ONCE(napi->dev->napi_defer_hard_irqs);
+		napi->defer_hard_irqs_count = napi_get_defer_hard_irqs(napi);
 		timeout = READ_ONCE(napi->dev->gro_flush_timeout);
 		if (napi->defer_hard_irqs_count && timeout) {
 			hrtimer_start(&napi->timer, ns_to_ktime(timeout), HRTIMER_MODE_REL_PINNED);
@@ -6650,6 +6672,7 @@ void netif_napi_add_weight(struct net_device *dev, struct napi_struct *napi,
 	INIT_HLIST_NODE(&napi->napi_hash_node);
 	hrtimer_init(&napi->timer, CLOCK_MONOTONIC, HRTIMER_MODE_REL_PINNED);
 	napi->timer.function = napi_watchdog;
+	napi_set_defer_hard_irqs(napi, READ_ONCE(dev->napi_defer_hard_irqs));
 	init_gro_hash(napi);
 	napi->skb = NULL;
 	INIT_LIST_HEAD(&napi->rx_list);
@@ -11032,7 +11055,7 @@ void netdev_sw_irq_coalesce_default_on(struct net_device *dev)
 
 	if (!IS_ENABLED(CONFIG_PREEMPT_RT)) {
 		dev->gro_flush_timeout = 20000;
-		dev->napi_defer_hard_irqs = 1;
+		netdev_set_defer_hard_irqs(dev, 1);
 	}
 }
 EXPORT_SYMBOL_GPL(netdev_sw_irq_coalesce_default_on);
diff --git a/net/core/net-sysfs.c b/net/core/net-sysfs.c
index 0e2084ce7b75..8272f0144d81 100644
--- a/net/core/net-sysfs.c
+++ b/net/core/net-sysfs.c
@@ -425,7 +425,10 @@ NETDEVICE_SHOW_RW(gro_flush_timeout, fmt_ulong);
 
 static int change_napi_defer_hard_irqs(struct net_device *dev, unsigned long val)
 {
-	WRITE_ONCE(dev->napi_defer_hard_irqs, val);
+	if (val > S32_MAX)
+		return -EINVAL;
+
+	netdev_set_defer_hard_irqs(dev, (int)val);
 	return 0;
 }
 
-- 
2.25.1


  reply	other threads:[~2024-08-29 13:13 UTC|newest]

Thread overview: 48+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-08-29 13:11 [PATCH net-next 0/5] Add support for per-NAPI config via netlink Joe Damato
2024-08-29 13:11 ` Joe Damato [this message]
2024-08-29 13:46   ` [PATCH net-next 1/5] net: napi: Make napi_defer_hard_irqs per-NAPI Eric Dumazet
2024-08-29 22:05   ` Jakub Kicinski
2024-08-30  9:14     ` Joe Damato
2024-08-30 20:21       ` Jakub Kicinski
2024-08-30 20:23         ` Joe Damato
2024-08-30  8:36   ` Simon Horman
2024-08-30  9:11     ` Joe Damato
2024-08-30 16:50   ` kernel test robot
2024-08-29 13:11 ` [PATCH net-next 2/5] netdev-genl: Dump napi_defer_hard_irqs Joe Damato
2024-08-29 22:08   ` Jakub Kicinski
2024-08-30  9:10     ` Joe Damato
2024-08-30 20:28       ` Jakub Kicinski
2024-08-30 20:31         ` Joe Damato
2024-08-30 21:22           ` Jakub Kicinski
2024-08-29 13:11 ` [PATCH net-next 3/5] net: napi: Make gro_flush_timeout per-NAPI Joe Damato
2024-08-29 13:48   ` Eric Dumazet
2024-08-29 13:57     ` Joe Damato
2024-08-29 15:28     ` Joe Damato
2024-08-29 15:31       ` Eric Dumazet
2024-08-29 15:39         ` Joe Damato
2024-08-30 16:18   ` kernel test robot
2024-08-30 16:18   ` kernel test robot
2024-08-29 13:12 ` [PATCH net-next 4/5] netdev-genl: Dump gro_flush_timeout Joe Damato
2024-08-29 22:09   ` Jakub Kicinski
2024-08-30  9:17     ` Joe Damato
2024-08-29 13:12 ` [PATCH net-next 5/5] netdev-genl: Support setting per-NAPI config values Joe Damato
2024-08-29 22:31   ` Jakub Kicinski
2024-08-30 10:43     ` Joe Damato
2024-08-30 21:22       ` Jakub Kicinski
2024-08-31 17:27         ` Joe Damato
2024-09-03  0:49           ` Jakub Kicinski
2024-09-02 16:56         ` Joe Damato
2024-09-03  1:02           ` Jakub Kicinski
2024-09-03 19:04             ` Samiullah Khawaja
2024-09-03 19:40               ` Jakub Kicinski
2024-09-03 21:58                 ` Samiullah Khawaja
2024-09-05  9:20                   ` Joe Damato
2024-09-08 15:54                 ` Joe Damato
2024-09-04 23:40           ` Stanislav Fomichev
2024-09-04 23:54             ` Jakub Kicinski
2024-09-05  9:32               ` Joe Damato
2024-09-08 15:57               ` Joe Damato
2024-09-09 23:03                 ` Jakub Kicinski
2024-09-05  9:30             ` Joe Damato
2024-09-05 16:56               ` Stanislav Fomichev
2024-09-05 17:05                 ` Joe Damato

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=20240829131214.169977-2-jdamato@fastly.com \
    --to=jdamato@fastly.com \
    --cc=aleksander.lobakin@intel.com \
    --cc=amritha.nambiar@intel.com \
    --cc=bigeasy@linutronix.de \
    --cc=bjorn@rivosinc.com \
    --cc=davem@davemloft.net \
    --cc=edumazet@google.com \
    --cc=hch@infradead.org \
    --cc=jiri@resnulli.us \
    --cc=johannes.berg@intel.com \
    --cc=kuba@kernel.org \
    --cc=leitao@debian.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=lorenzo@kernel.org \
    --cc=mkarsten@uwaterloo.ca \
    --cc=netdev@vger.kernel.org \
    --cc=pabeni@redhat.com \
    --cc=sdf@fomichev.me \
    --cc=skhawaja@google.com \
    --cc=sridhar.samudrala@intel.com \
    --cc=willemdebruijn.kernel@gmail.com \
    --cc=willy@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

all inboxes | Powered by JetHome®