netdev.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Jiri Pirko <jiri@resnulli.us>
To: netdev@vger.kernel.org
Cc: davem@davemloft.net, idosch@mellanox.com, eladr@mellanox.com,
	yotamg@mellanox.com, nogahf@mellanox.com, arkadis@mellanox.com,
	ogerlitz@mellanox.com, roopa@cumulusnetworks.com,
	dsa@cumulusnetworks.com, nikolay@cumulusnetworks.com,
	andy@greyhouse.net, vivien.didelot@savoirfairelinux.com,
	andrew@lunn.ch, f.fainelli@gmail.com,
	alexander.h.duyck@intel.com, hannes@stressinduktion.org,
	kaber@trash.net
Subject: [patch net-next v3 09/12] ipv4: fib: Add sysctl to limit number of FIB dump retries
Date: Wed, 30 Nov 2016 11:09:03 +0100	[thread overview]
Message-ID: <1480500546-2544-10-git-send-email-jiri@resnulli.us> (raw)
In-Reply-To: <1480500546-2544-1-git-send-email-jiri@resnulli.us>

From: Ido Schimmel <idosch@mellanox.com>

When dumping the FIB tables in the next commit, the dump will be
considered invalid if notifications were sent in the FIB notification
chain mid-dump. In systems where routing changes are frequent, the dump
might need to be restarted multiple times.

Add sysctl to limit the number of FIB dump retries, thereby preventing
callers from looping for long periods of time.

Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: Jiri Pirko <jiri@mellanox.com>
---
 Documentation/networking/ip-sysctl.txt | 8 ++++++++
 include/net/netns/ipv4.h               | 1 +
 net/ipv4/fib_frontend.c                | 1 +
 net/ipv4/sysctl_net_ipv4.c             | 7 +++++++
 4 files changed, 17 insertions(+)

diff --git a/Documentation/networking/ip-sysctl.txt b/Documentation/networking/ip-sysctl.txt
index 5af48dd..5182b23 100644
--- a/Documentation/networking/ip-sysctl.txt
+++ b/Documentation/networking/ip-sysctl.txt
@@ -73,6 +73,14 @@ fib_multipath_use_neigh - BOOLEAN
 	0 - disabled
 	1 - enabled
 
+fib_dump_max_retries - INTEGER
+	Maximum number of retries until the FIB dump is aborted. For a
+	given net namespace, a FIB dump is considered invalid if
+	notifications were sent in the FIB notification chain mid-dump.
+	The dump will be retried until it is successful or maximum
+	number of retries has been reached.
+	Default: 5
+
 route/max_size - INTEGER
 	Maximum number of routes allowed in the kernel.  Increase
 	this when using large numbers of interfaces and/or routes.
diff --git a/include/net/netns/ipv4.h b/include/net/netns/ipv4.h
index f0cf5a1..71c4ca8 100644
--- a/include/net/netns/ipv4.h
+++ b/include/net/netns/ipv4.h
@@ -136,6 +136,7 @@ struct netns_ipv4 {
 	int sysctl_fib_multipath_use_neigh;
 #endif
 
+	int sysctl_fib_dump_max_retries;
 	unsigned int	fib_seq;	/* protected by rtnl_mutex */
 
 	atomic_t	rt_genid;
diff --git a/net/ipv4/fib_frontend.c b/net/ipv4/fib_frontend.c
index dbad5a1..43f7557 100644
--- a/net/ipv4/fib_frontend.c
+++ b/net/ipv4/fib_frontend.c
@@ -1219,6 +1219,7 @@ static int __net_init ip_fib_net_init(struct net *net)
 	int err;
 	size_t size = sizeof(struct hlist_head) * FIB_TABLE_HASHSZ;
 
+	net->ipv4.sysctl_fib_dump_max_retries = 5;
 	net->ipv4.fib_seq = 0;
 
 	/* Avoid false sharing : Use at least a full cache line */
diff --git a/net/ipv4/sysctl_net_ipv4.c b/net/ipv4/sysctl_net_ipv4.c
index 80bc36b..046147c 100644
--- a/net/ipv4/sysctl_net_ipv4.c
+++ b/net/ipv4/sysctl_net_ipv4.c
@@ -971,6 +971,13 @@ static struct ctl_table ipv4_net_table[] = {
 		.extra2		= &one,
 	},
 #endif
+	{
+		.procname	= "fib_dump_max_retries",
+		.data		= &init_net.ipv4.sysctl_fib_dump_max_retries,
+		.maxlen		= sizeof(int),
+		.mode		= 0644,
+		.proc_handler	= proc_dointvec,
+	},
 	{ }
 };
 
-- 
2.7.4

  parent reply	other threads:[~2016-11-30 10:09 UTC|newest]

Thread overview: 27+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2016-11-30 10:08 [patch net-next v3 00/12] ipv4: fib: Allow modules to dump FIB tables Jiri Pirko
2016-11-30 10:08 ` [patch net-next v3 01/12] ipv4: fib: Export free_fib_info() Jiri Pirko
2016-11-30 10:08 ` [patch net-next v3 02/12] ipv4: fib: Add fib_info_hold() helper Jiri Pirko
2016-11-30 10:08 ` [patch net-next v3 03/12] mlxsw: core: Create an ordered workqueue for FIB offload Jiri Pirko
2016-11-30 10:08 ` [patch net-next v3 04/12] mlxsw: spectrum_router: Implement FIB offload in deferred work Jiri Pirko
2016-11-30 10:08 ` [patch net-next v3 05/12] rocker: Create an ordered workqueue for FIB offload Jiri Pirko
2016-11-30 10:09 ` [patch net-next v3 06/12] rocker: Implement FIB offload in deferred work Jiri Pirko
2016-11-30 10:09 ` [patch net-next v3 07/12] ipv4: fib: Convert FIB notification chain to be atomic Jiri Pirko
2016-11-30 10:09 ` [patch net-next v3 08/12] ipv4: fib: Allow for consistent FIB dumping Jiri Pirko
2016-11-30 10:09 ` Jiri Pirko [this message]
2016-11-30 10:09 ` [patch net-next v3 10/12] ipv4: fib: Add an API to request a FIB dump Jiri Pirko
2016-11-30 10:09 ` [patch net-next v3 11/12] mlxsw: spectrum_router: Request a dump of FIB tables during init Jiri Pirko
2016-11-30 15:37   ` Hannes Frederic Sowa
2016-11-30 16:32     ` Ido Schimmel
2016-11-30 16:49       ` Hannes Frederic Sowa
2016-11-30 18:22         ` Ido Schimmel
2016-12-01 21:57           ` Hannes Frederic Sowa
2016-12-01 23:14             ` Ido Schimmel
2016-12-01 23:27               ` Hannes Frederic Sowa
2016-12-02  9:34                 ` Ido Schimmel
2016-12-01 20:04       ` David Miller
2016-12-01 20:40         ` Hannes Frederic Sowa
2016-12-01 20:54           ` Ido Schimmel
2016-12-01 21:09             ` Hannes Frederic Sowa
2016-12-01 21:21               ` Ido Schimmel
2016-12-01 21:09       ` Hannes Frederic Sowa
2016-11-30 10:09 ` [patch net-next v3 12/12] rocker: Register FIB notifier before creating ports Jiri Pirko

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=1480500546-2544-10-git-send-email-jiri@resnulli.us \
    --to=jiri@resnulli.us \
    --cc=alexander.h.duyck@intel.com \
    --cc=andrew@lunn.ch \
    --cc=andy@greyhouse.net \
    --cc=arkadis@mellanox.com \
    --cc=davem@davemloft.net \
    --cc=dsa@cumulusnetworks.com \
    --cc=eladr@mellanox.com \
    --cc=f.fainelli@gmail.com \
    --cc=hannes@stressinduktion.org \
    --cc=idosch@mellanox.com \
    --cc=kaber@trash.net \
    --cc=netdev@vger.kernel.org \
    --cc=nikolay@cumulusnetworks.com \
    --cc=nogahf@mellanox.com \
    --cc=ogerlitz@mellanox.com \
    --cc=roopa@cumulusnetworks.com \
    --cc=vivien.didelot@savoirfairelinux.com \
    --cc=yotamg@mellanox.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;
as well as URLs for NNTP newsgroup(s).