diff options
| author |   Steffen Klassert <steffen.klassert@secunet.com> | 2015-09-29 11:40:49 +0200 |
|---|---|---|
| committer | Steffen Klassert <steffen.klassert@secunet.com> | 2015-09-29 11:44:16 +0200 |
| commit | c386578f1cdb4dac230395a951f88027f64346e3 (patch) | |
| tree | 8e8442a599588ae8876e2d0dc0975602ab0dbd69 | |
| parent | net: help compiler generate better code in eth_get_headlen (diff) | |
| download | linux-dev-c386578f1cdb4dac230395a951f88027f64346e3.tar.xz linux-dev-c386578f1cdb4dac230395a951f88027f64346e3.zip | |
xfrm: Let the flowcache handle its size by default.
The xfrm flowcache size is limited by the flowcache limit
(4096 * number of online cpus) and the xfrm garbage collector
threshold (2 * 32768), whatever is reached first. This means
that we can hit the garbage collector limit only on systems
with more than 16 cpus. On such systems we simply refuse
new allocations if we reach the limit, so new flows are dropped.
On syslems with 16 or less cpus, we hit the flowcache limit.
In this case, we shrink the flow cache instead of refusing new
flows.
We increase the xfrm garbage collector threshold to INT_MAX
to get the same behaviour, independent of the number of cpus.
The xfrm garbage collector threshold can still be set below
the flowcache limit to reduce the memory usage of the flowcache.
Tested-by: Dan Streetman <dan.streetman@canonical.com>
Signed-off-by: Steffen Klassert <steffen.klassert@secunet.com>
| -rw-r--r-- | Documentation/networking/ip-sysctl.txt | 6 | ||||
| -rw-r--r-- | net/ipv4/xfrm4_policy.c | 2 | ||||
| -rw-r--r-- | net/ipv6/xfrm6_policy.c | 2 |
3 files changed, 6 insertions, 4 deletions
diff --git a/Documentation/networking/ip-sysctl.txt b/Documentation/networking/ip-sysctl.txt index ebe94f2cab98..260f30b2def6 100644 --- a/Documentation/networking/ip-sysctl.txt +++ b/Documentation/networking/ip-sysctl.txt @@ -1199,7 +1199,8 @@ tag - INTEGER xfrm4_gc_thresh - INTEGER The threshold at which we will start garbage collecting for IPv4 destination cache entries. At twice this value the system will - refuse new allocations. + refuse new allocations. The value must be set below the flowcache + limit (4096 * number of online cpus) to take effect. igmp_link_local_mcast_reports - BOOLEAN Enable IGMP reports for link local multicast groups in the @@ -1645,7 +1646,8 @@ ratelimit - INTEGER xfrm6_gc_thresh - INTEGER The threshold at which we will start garbage collecting for IPv6 destination cache entries. At twice this value the system will - refuse new allocations. + refuse new allocations. The value must be set below the flowcache + limit (4096 * number of online cpus) to take effect. IPv6 Update by: diff --git a/net/ipv4/xfrm4_policy.c b/net/ipv4/xfrm4_policy.c index 0304d1680ca2..75e8d48c03fb 100644 --- a/net/ipv4/xfrm4_policy.c +++ b/net/ipv4/xfrm4_policy.c @@ -246,7 +246,7 @@ static struct dst_ops xfrm4_dst_ops = { .destroy = xfrm4_dst_destroy, .ifdown = xfrm4_dst_ifdown, .local_out = __ip_local_out, - .gc_thresh = 32768, + .gc_thresh = INT_MAX, }; static struct xfrm_policy_afinfo xfrm4_policy_afinfo = { diff --git a/net/ipv6/xfrm6_policy.c b/net/ipv6/xfrm6_policy.c index 30caa289c5db..2fad59320b6c 100644 --- a/net/ipv6/xfrm6_policy.c +++ b/net/ipv6/xfrm6_policy.c @@ -287,7 +287,7 @@ static struct dst_ops xfrm6_dst_ops = { .destroy = xfrm6_dst_destroy, .ifdown = xfrm6_dst_ifdown, .local_out = __ip6_local_out, - .gc_thresh = 32768, + .gc_thresh = INT_MAX, }; static struct xfrm_policy_afinfo xfrm6_policy_afinfo = { |