netdev.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Alexander Duyck <alexander.duyck-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
To: netdev-u79uwXL29TY76Z2rM5mHXA@public.gmane.org,
	linux-kernel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
Cc: sridhar.samudrala-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org,
	edumazet-hpIqsD4AKlfQT0dZR+AlfA@public.gmane.org,
	davem-fT/PcQaiUtIeIZ0/mPfg9Q@public.gmane.org,
	linux-api-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
Subject: [net-next PATCH v2 1/8] net: Busy polling should ignore sender CPUs
Date: Thu, 23 Mar 2017 14:36:37 -0700	[thread overview]
Message-ID: <20170323213637.12615.11197.stgit@localhost.localdomain> (raw)
In-Reply-To: <20170323211820.12615.88907.stgit-bi+AKbBUZKY6gyzm1THtWbp2dZbC/Bob@public.gmane.org>

From: Alexander Duyck <alexander.h.duyck-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org>

This patch is a cleanup/fix for NAPI IDs following the changes that made it
so that sender_cpu and napi_id were doing a better job of sharing the same
location in the sk_buff.

One issue I found is that we weren't validating the napi_id as being valid
before we started trying to setup the busy polling.  This change corrects
that by using the MIN_NAPI_ID value that is now used in both allocating the
NAPI IDs, as well as validating them.

Signed-off-by: Alexander Duyck <alexander.h.duyck-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org>
---
 include/net/busy_poll.h |    9 +++++++--
 net/core/dev.c          |   13 +++++++++----
 2 files changed, 16 insertions(+), 6 deletions(-)

diff --git a/include/net/busy_poll.h b/include/net/busy_poll.h
index c0452de83086..3fcda9e70c3f 100644
--- a/include/net/busy_poll.h
+++ b/include/net/busy_poll.h
@@ -35,6 +35,12 @@
 extern unsigned int sysctl_net_busy_read __read_mostly;
 extern unsigned int sysctl_net_busy_poll __read_mostly;
 
+/*		0 - Reserved to indicate value not set
+ *     1..NR_CPUS - Reserved for sender_cpu
+ *  NR_CPUS+1..~0 - Region available for NAPI IDs
+ */
+#define MIN_NAPI_ID ((unsigned int)(NR_CPUS + 1))
+
 static inline bool net_busy_loop_on(void)
 {
 	return sysctl_net_busy_poll;
@@ -58,10 +64,9 @@ static inline unsigned long busy_loop_end_time(void)
 
 static inline bool sk_can_busy_loop(const struct sock *sk)
 {
-	return sk->sk_ll_usec && sk->sk_napi_id && !signal_pending(current);
+	return sk->sk_ll_usec && !signal_pending(current);
 }
 
-
 static inline bool busy_loop_timeout(unsigned long end_time)
 {
 	unsigned long now = busy_loop_us_clock();
diff --git a/net/core/dev.c b/net/core/dev.c
index 7869ae3837ca..ab337bf5bbf4 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -5066,15 +5066,20 @@ bool sk_busy_loop(struct sock *sk, int nonblock)
 	int (*napi_poll)(struct napi_struct *napi, int budget);
 	void *have_poll_lock = NULL;
 	struct napi_struct *napi;
+	unsigned int napi_id;
 	int rc;
 
 restart:
+	napi_id = READ_ONCE(sk->sk_napi_id);
+	if (napi_id < MIN_NAPI_ID)
+		return 0;
+
 	rc = false;
 	napi_poll = NULL;
 
 	rcu_read_lock();
 
-	napi = napi_by_id(sk->sk_napi_id);
+	napi = napi_by_id(napi_id);
 	if (!napi)
 		goto out;
 
@@ -5143,10 +5148,10 @@ static void napi_hash_add(struct napi_struct *napi)
 
 	spin_lock(&napi_hash_lock);
 
-	/* 0..NR_CPUS+1 range is reserved for sender_cpu use */
+	/* 0..NR_CPUS range is reserved for sender_cpu use */
 	do {
-		if (unlikely(++napi_gen_id < NR_CPUS + 1))
-			napi_gen_id = NR_CPUS + 1;
+		if (unlikely(++napi_gen_id < MIN_NAPI_ID))
+			napi_gen_id = MIN_NAPI_ID;
 	} while (napi_by_id(napi_gen_id));
 	napi->napi_id = napi_gen_id;
 

  parent reply	other threads:[~2017-03-23 21:36 UTC|newest]

Thread overview: 29+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2017-03-23 21:36 [net-next PATCH v2 0/8] Add busy poll support for epoll Alexander Duyck
     [not found] ` <20170323211820.12615.88907.stgit-bi+AKbBUZKY6gyzm1THtWbp2dZbC/Bob@public.gmane.org>
2017-03-23 21:36   ` Alexander Duyck [this message]
2017-03-23 22:05     ` [net-next PATCH v2 1/8] net: Busy polling should ignore sender CPUs Eric Dumazet
2017-03-23 21:36 ` [net-next PATCH v2 2/8] tcp: Record Rx hash and NAPI ID in tcp_child_process Alexander Duyck
     [not found]   ` <20170323213644.12615.27158.stgit-bi+AKbBUZKY6gyzm1THtWbp2dZbC/Bob@public.gmane.org>
2017-03-24  1:00     ` Eric Dumazet
2017-03-23 21:36 ` [net-next PATCH v2 3/8] net: Only define skb_mark_napi_id in one spot instead of two Alexander Duyck
     [not found]   ` <20170323213651.12615.62895.stgit-bi+AKbBUZKY6gyzm1THtWbp2dZbC/Bob@public.gmane.org>
2017-03-24  1:02     ` Eric Dumazet
2017-03-23 21:37 ` [net-next PATCH v2 4/8] net: Change return type of sk_busy_loop from bool to void Alexander Duyck
     [not found]   ` <20170323213715.12615.49246.stgit-bi+AKbBUZKY6gyzm1THtWbp2dZbC/Bob@public.gmane.org>
2017-03-24  1:02     ` Eric Dumazet
2017-03-23 21:37 ` [net-next PATCH v2 5/8] net: Track start of busy loop instead of when it should end Alexander Duyck
2017-03-24  1:24   ` Eric Dumazet
2017-03-24  3:42     ` Alexander Duyck
     [not found]       ` <CAKgT0Ue+UFOAhwxN-EOjWYZa9YZc2QaT5gvoLap70COs0rF7NA-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>
2017-03-24  4:27         ` Eric Dumazet
2017-03-24  4:38           ` Eric Dumazet
2017-03-24  5:55           ` Alexander Duyck
     [not found]             ` <CAKgT0UfQPLqHtZ_SjanbLoG6t5f36C4nsORuw6yjLT0xZrEoEQ-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>
2017-03-24 11:16               ` Eric Dumazet
     [not found]                 ` <1490354210.9687.44.camel-XN9IlZ5yJG9HTL0Zs8A6p+yfmBU6pStAUsxypvmhUTTZJqsBc5GL+g@public.gmane.org>
2017-03-24 15:48                   ` Alexander Duyck
2017-03-23 21:37 ` [net-next PATCH v2 6/8] net: Commonize busy polling code to focus on napi_id instead of socket Alexander Duyck
     [not found]   ` <20170323213749.12615.38165.stgit-bi+AKbBUZKY6gyzm1THtWbp2dZbC/Bob@public.gmane.org>
2017-03-23 22:25     ` Eric Dumazet
2017-03-23 21:37 ` [net-next PATCH v2 7/8] epoll: Add busy poll support to epoll with socket fds Alexander Duyck
2017-03-23 21:38 ` [net-next PATCH v2 8/8] net: Introduce SO_INCOMING_NAPI_ID Alexander Duyck
     [not found]   ` <20170323213802.12615.58216.stgit-bi+AKbBUZKY6gyzm1THtWbp2dZbC/Bob@public.gmane.org>
2017-03-23 22:22     ` Eric Dumazet
2017-03-23 22:43     ` Andy Lutomirski
2017-03-24  0:58       ` Alexander Duyck
     [not found]         ` <CAKgT0UcHJVycQ3+h09L2Ph=TVncqHPJ6dZpicUgBo7TaFTN7yw-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>
2017-03-24  4:47           ` Andy Lutomirski
2017-03-24  5:07             ` Eric Dumazet
2017-03-23 22:07 ` [net-next PATCH v2 0/8] Add busy poll support for epoll Alexei Starovoitov
     [not found]   ` <20170323220721.GA62356-+o4/htvd0TDFYCXBM6kdu7fOX0fSgVTm@public.gmane.org>
2017-03-23 22:38     ` Alexander Duyck
     [not found]       ` <CAKgT0Uctdf4N-w72EY8T5Zfw=QCSCgAesXdgOh-HUYdD=Aq9AA-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>
2017-03-23 22:49         ` Eric Dumazet

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=20170323213637.12615.11197.stgit@localhost.localdomain \
    --to=alexander.duyck-re5jqeeqqe8avxtiumwx3w@public.gmane.org \
    --cc=davem-fT/PcQaiUtIeIZ0/mPfg9Q@public.gmane.org \
    --cc=edumazet-hpIqsD4AKlfQT0dZR+AlfA@public.gmane.org \
    --cc=linux-api-u79uwXL29TY76Z2rM5mHXA@public.gmane.org \
    --cc=linux-kernel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org \
    --cc=netdev-u79uwXL29TY76Z2rM5mHXA@public.gmane.org \
    --cc=sridhar.samudrala-ral2JQCrhuEAvxtiuMwx3w@public.gmane.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).