Commit 9ab948a9 authored by David Ahern's avatar David Ahern Committed by David S. Miller

ipv4: Allow amount of dirty memory from fib resizing to be controllable

fib_trie implementation calls synchronize_rcu when a certain amount of
pages are dirty from freed entries. The number of pages was determined
experimentally in 2009 (commit c3059477).

At the current setting, synchronize_rcu is called often -- 51 times in a
second in one test with an average of an 8 msec delay adding a fib entry.
The total impact is a lot of slow down modifying the fib. This is seen
in the output of 'time' - the difference between real time and sys+user.
For example, using 720,022 single path routes and 'ip -batch'[1]:

    $ time ./ip -batch ipv4/routes-1-hops
    real    0m14.214s
    user    0m2.513s
    sys     0m6.783s

So roughly 35% of the actual time to install the routes is from the ip
command getting scheduled out, most notably due to synchronize_rcu (this
is observed using 'perf sched timehist').

This patch makes the amount of dirty memory configurable between 64k where
the synchronize_rcu is called often (small, low end systems that are memory
sensitive) to 64M where synchronize_rcu is called rarely during a large
FIB change (for high end systems with lots of memory). The default is 512kB
which corresponds to the current setting of 128 pages with a 4kB page size.

As an example, at 16MB the worst interval shows 4 calls to synchronize_rcu
in a second blocking for up to 30 msec in a single instance, and a total
of almost 100 msec across the 4 calls in the second. The trade off is
allowing FIB entries to consume more memory in a given time window but
but with much better fib insertion rates (~30% increase in prefixes/sec).
With this patch and net.ipv4.fib_sync_mem set to 16MB, the same batch
file runs in:

    $ time ./ip -batch ipv4/routes-1-hops
    real    0m9.692s
    user    0m2.491s
    sys     0m6.769s

So the dead time is reduced to about 1/2 second or <5% of the real time.

[1] 'ip' modified to not request ACK messages which improves route
    insertion times by about 20%
Signed-off-by: default avatarDavid Ahern <dsahern@gmail.com>
Signed-off-by: default avatarDavid S. Miller <davem@davemloft.net>
parent 12132768
......@@ -81,6 +81,11 @@ fib_multipath_hash_policy - INTEGER
0 - Layer 3
1 - Layer 4
fib_sync_mem - UNSIGNED INTEGER
Amount of dirty memory from fib entries that can be backlogged before
synchronize_rcu is forced.
Default: 512kB Minimum: 64kB Maximum: 64MB
ip_forward_update_priority - INTEGER
Whether to update SKB priority from "TOS" field in IPv4 header after it
is forwarded. The new SKB priority is mapped from TOS field value
......
......@@ -38,6 +38,10 @@
#define IPV4_MAX_PMTU 65535U /* RFC 2675, Section 5.1 */
#define IPV4_MIN_MTU 68 /* RFC 791 */
extern unsigned int sysctl_fib_sync_mem;
extern unsigned int sysctl_fib_sync_mem_min;
extern unsigned int sysctl_fib_sync_mem_max;
struct sock;
struct inet_skb_parm {
......
......@@ -183,14 +183,16 @@ struct trie {
};
static struct key_vector *resize(struct trie *t, struct key_vector *tn);
static size_t tnode_free_size;
static unsigned int tnode_free_size;
/*
* synchronize_rcu after call_rcu for that many pages; it should be especially
* useful before resizing the root node with PREEMPT_NONE configs; the value was
* obtained experimentally, aiming to avoid visible slowdown.
* synchronize_rcu after call_rcu for outstanding dirty memory; it should be
* especially useful before resizing the root node with PREEMPT_NONE configs;
* the value was obtained experimentally, aiming to avoid visible slowdown.
*/
static const int sync_pages = 128;
unsigned int sysctl_fib_sync_mem = 512 * 1024;
unsigned int sysctl_fib_sync_mem_min = 64 * 1024;
unsigned int sysctl_fib_sync_mem_max = 64 * 1024 * 1024;
static struct kmem_cache *fn_alias_kmem __ro_after_init;
static struct kmem_cache *trie_leaf_kmem __ro_after_init;
......@@ -504,7 +506,7 @@ static void tnode_free(struct key_vector *tn)
tn = container_of(head, struct tnode, rcu)->kv;
}
if (tnode_free_size >= PAGE_SIZE * sync_pages) {
if (tnode_free_size >= sysctl_fib_sync_mem) {
tnode_free_size = 0;
synchronize_rcu();
}
......
......@@ -549,6 +549,15 @@ static struct ctl_table ipv4_table[] = {
.mode = 0644,
.proc_handler = proc_doulongvec_minmax,
},
{
.procname = "fib_sync_mem",
.data = &sysctl_fib_sync_mem,
.maxlen = sizeof(sysctl_fib_sync_mem),
.mode = 0644,
.proc_handler = proc_douintvec_minmax,
.extra1 = &sysctl_fib_sync_mem_min,
.extra2 = &sysctl_fib_sync_mem_max,
},
{ }
};
......
Markdown is supported
0%
or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment