linux/net
Jon Maxwell af6d10345c ipv6: remove max_size check inline with ipv4
In ip6_dst_gc() replace:

  if (entries > gc_thresh)

With:

  if (entries > ops->gc_thresh)

Sending Ipv6 packets in a loop via a raw socket triggers an issue where a
route is cloned by ip6_rt_cache_alloc() for each packet sent. This quickly
consumes the Ipv6 max_size threshold which defaults to 4096 resulting in
these warnings:

[1]   99.187805] dst_alloc: 7728 callbacks suppressed
[2] Route cache is full: consider increasing sysctl net.ipv6.route.max_size.
.
.
[300] Route cache is full: consider increasing sysctl net.ipv6.route.max_size.

When this happens the packet is dropped and sendto() gets a network is
unreachable error:

remaining pkt 200557 errno 101
remaining pkt 196462 errno 101
.
.
remaining pkt 126821 errno 101

Implement David Aherns suggestion to remove max_size check seeing that Ipv6
has a GC to manage memory usage. Ipv4 already does not check max_size.

Here are some memory comparisons for Ipv4 vs Ipv6 with the patch:

Test by running 5 instances of a program that sends UDP packets to a raw
socket 5000000 times. Compare Ipv4 and Ipv6 performance with a similar
program.

Ipv4:

Before test:

MemFree:        29427108 kB
Slab:             237612 kB

ip6_dst_cache       1912   2528    256   32    2 : tunables    0    0    0
xfrm_dst_cache         0      0    320   25    2 : tunables    0    0    0
ip_dst_cache        2881   3990    192   42    2 : tunables    0    0    0

During test:

MemFree:        29417608 kB
Slab:             247712 kB

ip6_dst_cache       1912   2528    256   32    2 : tunables    0    0    0
xfrm_dst_cache         0      0    320   25    2 : tunables    0    0    0
ip_dst_cache       44394  44394    192   42    2 : tunables    0    0    0

After test:

MemFree:        29422308 kB
Slab:             238104 kB

ip6_dst_cache       1912   2528    256   32    2 : tunables    0    0    0
xfrm_dst_cache         0      0    320   25    2 : tunables    0    0    0
ip_dst_cache        3048   4116    192   42    2 : tunables    0    0    0

Ipv6 with patch:

Errno 101 errors are not observed anymore with the patch.

Before test:

MemFree:        29422308 kB
Slab:             238104 kB

ip6_dst_cache       1912   2528    256   32    2 : tunables    0    0    0
xfrm_dst_cache         0      0    320   25    2 : tunables    0    0    0
ip_dst_cache        3048   4116    192   42    2 : tunables    0    0    0

During Test:

MemFree:        29431516 kB
Slab:             240940 kB

ip6_dst_cache      11980  12064    256   32    2 : tunables    0    0    0
xfrm_dst_cache         0      0    320   25    2 : tunables    0    0    0
ip_dst_cache        3048   4116    192   42    2 : tunables    0    0    0

After Test:

MemFree:        29441816 kB
Slab:             238132 kB

ip6_dst_cache       1902   2432    256   32    2 : tunables    0    0    0
xfrm_dst_cache         0      0    320   25    2 : tunables    0    0    0
ip_dst_cache        3048   4116    192   42    2 : tunables    0    0    0

Tested-by: Andrea Mayer <andrea.mayer@uniroma2.it>
Signed-off-by: Jon Maxwell <jmaxwell37@gmail.com>
Reviewed-by: David Ahern <dsahern@kernel.org>
Link: https://lore.kernel.org/r/20230112012532.311021-1-jmaxwell37@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2023-01-13 20:59:14 -08:00
..
6lowpan
9p xen: branch for v6.2-rc4 2023-01-12 17:02:20 -06:00
802 treewide: Convert del_timer*() to timer_shutdown*() 2022-12-25 13:38:09 -08:00
8021q net: Remove the obsolte u64_stats_fetch_*_irq() users (net). 2022-10-28 20:13:54 -07:00
appletalk
atm driver core: make struct class.dev_uevent() take a const * 2022-11-24 17:12:15 +01:00
ax25 ax25: af_ax25: Remove unnecessary (void*) conversions 2022-11-16 13:31:03 +00:00
batman-adv Networking changes for 6.2. 2022-12-13 15:47:48 -08:00
bluetooth Networking changes for 6.2. 2022-12-13 15:47:48 -08:00
bpf New Feature: 2022-12-17 14:06:53 -06:00
bpfilter
bridge treewide: Convert del_timer*() to timer_shutdown*() 2022-12-25 13:38:09 -08:00
caif caif: don't assume iov_iter type 2023-01-13 20:44:20 -08:00
can Networking changes for 6.2. 2022-12-13 15:47:48 -08:00
ceph Treewide: Stop corrupting socket's task_frag 2022-12-19 17:28:49 -08:00
core ipv6: remove max_size check inline with ipv4 2023-01-13 20:59:14 -08:00
dcb net: dcb: move getapptrust to separate function 2022-11-15 15:27:43 +01:00
dccp Merge git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net 2022-11-29 13:04:52 -08:00
devlink devlink: keep the instance mutex alive until references are gone 2023-01-11 20:49:32 -08:00
dns_resolver cred: Do not default to init_cred in prepare_kernel_cred() 2022-11-01 10:04:52 -07:00
dsa net: dsa: microchip: ptp: move pdelay_rsp correction field to tail tag 2023-01-13 08:40:41 +00:00
ethernet net: ethernet: use sysfs_emit() to instead of scnprintf() 2022-12-07 20:02:44 -08:00
ethtool ethtool: add tx aggregation parameters 2023-01-13 10:23:52 +00:00
hsr hsr: Use a single struct for self_node. 2022-12-01 20:26:22 -08:00
ieee802154 Merge tag 'ieee802154-for-net-next-2022-12-05' of git://git.kernel.org/pub/scm/linux/kernel/git/sschmidt/wpan-next 2022-12-07 17:33:26 -08:00
ife
ipv4 Including fixes from bpf, wifi, and netfilter. 2023-01-05 12:40:50 -08:00
ipv6 ipv6: remove max_size check inline with ipv4 2023-01-13 20:59:14 -08:00
iucv
kcm kcm: close race conditions on sk_receive_queue 2022-11-15 12:42:26 +01:00
key Merge branch 'master' of git://git.kernel.org/pub/scm/linux/kernel/git/klassert/ipsec-next 2022-11-29 20:50:51 -08:00
l2tp Merge git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net 2022-11-29 13:04:52 -08:00
l3mdev
lapb
llc
mac80211 treewide: Convert del_timer*() to timer_shutdown*() 2022-12-25 13:38:09 -08:00
mac802154 Merge git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net 2022-12-08 18:19:59 -08:00
mctp mctp: Remove device type check at unregister 2022-12-19 17:20:22 -08:00
mpls net: Remove the obsolte u64_stats_fetch_*_irq() users (net). 2022-10-28 20:13:54 -07:00
mptcp mptcp: add statistics for mptcp socket in use 2023-01-09 07:30:50 +00:00
ncsi net/ncsi: Silence runtime memcpy() false positive warning 2022-12-06 17:29:14 -08:00
netfilter netlink: add macro for checking dump ctx size 2023-01-05 22:13:39 -08:00
netlabel
netlink Networking changes for 6.2. 2022-12-13 15:47:48 -08:00
netrom
nfc nfc: Fix potential resource leaks 2022-12-26 09:09:23 +00:00
nsh
openvswitch net: openvswitch: release vport resources on failure 2022-12-21 17:48:12 -08:00
packet Networking changes for 6.2. 2022-12-13 15:47:48 -08:00
phonet
psample
qrtr
rds rds: remove redundant variable total_payload_len 2022-11-04 10:09:50 +00:00
rfkill driver core: make struct class.dev_uevent() take a const * 2022-11-24 17:12:15 +01:00
rose rose: Fix NULL pointer dereference in rose_send_frame() 2022-11-02 11:57:30 +00:00
rxrpc rxrpc: Fix incoming call setup race 2023-01-07 09:30:26 +00:00
sched net: sched: disallow noqueue for qdisc classes 2023-01-10 18:19:32 -08:00
sctp Networking changes for 6.2. 2022-12-13 15:47:48 -08:00
smc use less confusing names for iov_iter direction initializers 2022-11-25 13:01:55 -05:00
strparser
sunrpc nfsd-6.2 fixes: 2023-01-10 15:03:06 -06:00
switchdev
tipc tipc: fix unexpected link reset due to discovery messages 2023-01-06 12:53:10 +00:00
tls Networking changes for 6.2. 2022-12-13 15:47:48 -08:00
unix unix: Fix race in SOCK_SEQPACKET's unix_dgram_sendmsg() 2022-12-15 11:35:18 +01:00
vmw_vsock vsock: return errors other than -ENOMEM to socket 2023-01-12 12:53:54 +01:00
wireless Driver Core changes for 6.2-rc1 2022-12-16 03:54:54 -08:00
x25 net/x25: Fix skb leak in x25_lapb_receive_frame() 2022-11-15 20:22:19 -08:00
xdp bpf: Expand map key argument of bpf_redirect_map to u64 2022-11-15 09:00:27 -08:00
xfrm Treewide: Stop corrupting socket's task_frag 2022-12-19 17:28:49 -08:00
compat.c use less confusing names for iov_iter direction initializers 2022-11-25 13:01:55 -05:00
devres.c
Kconfig
Kconfig.debug net: make NET_(DEV|NS)_REFCNT_TRACKER depend on NET 2022-09-20 14:23:56 -07:00
Makefile devlink: move code to a dedicated directory 2023-01-05 22:12:00 -08:00
socket.c sock: add tracepoint for send recv length 2023-01-13 10:25:10 +00:00
sysctl_net.c