nagai/kernel
3
0
Fork 0
forked from mirrors/linux
Code Pull requests Activity

Namespaceify min_pmtu sysctl

Browse source 
This patch enables the sysctl min_pmtu to be configured per net
namespace.

Signed-off-by: xu xin <xu.xin16@zte.com.cn>
Signed-off-by: David S. Miller <davem@davemloft.net>
This commit is contained in:
xu xin 2022-01-04 10:59:34 +00:00 committed by David S. Miller
parent 7d714ff14d
commit 1de6b15a43
2 changed files with 39 additions and 16 deletions
repo.diff.show_diff_stats Download patch file Download diff file Expand all files Collapse all files

2
include/net/netns/ipv4.h
View file

@ -85,6 +85,8 @@ struct netns_ipv4 {
int sysctl_icmp_ratelimit; int sysctl_icmp_ratelimit;
int sysctl_icmp_ratemask; int sysctl_icmp_ratemask;
u32 ip_rt_min_pmtu;
struct local_ports ip_local_ports; struct local_ports ip_local_ports;
u8 sysctl_tcp_ecn; u8 sysctl_tcp_ecn;

53
net/ipv4/route.c
View file

@ -110,6 +110,8 @@
#define RT_GC_TIMEOUT (300*HZ) #define RT_GC_TIMEOUT (300*HZ)
#define DEFAULT_MIN_PMTU (512 + 20 + 20)
static int ip_rt_max_size; static int ip_rt_max_size;
static int ip_rt_redirect_number __read_mostly = 9; static int ip_rt_redirect_number __read_mostly = 9;
static int ip_rt_redirect_load __read_mostly = HZ / 50; static int ip_rt_redirect_load __read_mostly = HZ / 50;
@ -117,7 +119,6 @@ static int ip_rt_redirect_silence __read_mostly = ((HZ / 50) << (9 + 1));
static int ip_rt_error_cost __read_mostly = HZ; static int ip_rt_error_cost __read_mostly = HZ;
static int ip_rt_error_burst __read_mostly = 5 * HZ; static int ip_rt_error_burst __read_mostly = 5 * HZ;
static int ip_rt_mtu_expires __read_mostly = 10 * 60 * HZ; static int ip_rt_mtu_expires __read_mostly = 10 * 60 * HZ;
static u32 ip_rt_min_pmtu __read_mostly = 512 + 20 + 20;
static int ip_rt_min_advmss __read_mostly = 256; static int ip_rt_min_advmss __read_mostly = 256;
static int ip_rt_gc_timeout __read_mostly = RT_GC_TIMEOUT; static int ip_rt_gc_timeout __read_mostly = RT_GC_TIMEOUT;
@ -1018,9 +1019,9 @@ static void __ip_rt_update_pmtu(struct rtable *rt, struct flowi4 *fl4, u32 mtu)
if (old_mtu < mtu) if (old_mtu < mtu)
return; return;
if (mtu < ip_rt_min_pmtu) { if (mtu < net->ipv4.ip_rt_min_pmtu) {
lock = true; lock = true;
mtu = min(old_mtu, ip_rt_min_pmtu); mtu = min(old_mtu, net->ipv4.ip_rt_min_pmtu);
} }
if (rt->rt_pmtu == mtu && !lock && if (rt->rt_pmtu == mtu && !lock &&
@ -3541,14 +3542,6 @@ static struct ctl_table ipv4_route_table[] = {
.mode = 0644, .mode = 0644,
.proc_handler = proc_dointvec_jiffies, .proc_handler = proc_dointvec_jiffies,
}, },
{
.procname = "min_pmtu",
.data = &ip_rt_min_pmtu,
.maxlen = sizeof(int),
.mode = 0644,
.proc_handler = proc_dointvec_minmax,
.extra1 = &ip_min_valid_pmtu,
},
{ {
.procname = "min_adv_mss", .procname = "min_adv_mss",
.data = &ip_rt_min_advmss, .data = &ip_rt_min_advmss,
@ -3561,13 +3554,21 @@ static struct ctl_table ipv4_route_table[] = {
static const char ipv4_route_flush_procname[] = "flush"; static const char ipv4_route_flush_procname[] = "flush";
static struct ctl_table ipv4_route_flush_table[] = { static struct ctl_table ipv4_route_netns_table[] = {
{ {
.procname = ipv4_route_flush_procname, .procname = ipv4_route_flush_procname,
.maxlen = sizeof(int), .maxlen = sizeof(int),
.mode = 0200, .mode = 0200,
.proc_handler = ipv4_sysctl_rtcache_flush, .proc_handler = ipv4_sysctl_rtcache_flush,
}, },
{
.procname = "min_pmtu",
.data = &init_net.ipv4.ip_rt_min_pmtu,
.maxlen = sizeof(int),
.mode = 0644,
.proc_handler = proc_dointvec_minmax,
.extra1 = &ip_min_valid_pmtu,
},
{ }, { },
}; };
@ -3575,9 +3576,11 @@ static __net_init int sysctl_route_net_init(struct net *net)
{ {
struct ctl_table *tbl; struct ctl_table *tbl;
tbl = ipv4_route_flush_table; tbl = ipv4_route_netns_table;
if (!net_eq(net, &init_net)) { if (!net_eq(net, &init_net)) {
tbl = kmemdup(tbl, sizeof(ipv4_route_flush_table), GFP_KERNEL); int i;
tbl = kmemdup(tbl, sizeof(ipv4_route_netns_table), GFP_KERNEL);
if (!tbl) if (!tbl)
goto err_dup; goto err_dup;
@ -3586,6 +3589,12 @@ static __net_init int sysctl_route_net_init(struct net *net)
if (tbl[0].procname != ipv4_route_flush_procname) if (tbl[0].procname != ipv4_route_flush_procname)
tbl[0].procname = NULL; tbl[0].procname = NULL;
} }
/* Update the variables to point into the current struct net
* except for the first element flush
*/
for (i = 1; i < ARRAY_SIZE(ipv4_route_netns_table) - 1; i++)
tbl[i].data += (void *)net - (void *)&init_net;
} }
tbl[0].extra1 = net; tbl[0].extra1 = net;
@ -3595,7 +3604,7 @@ static __net_init int sysctl_route_net_init(struct net *net)
return 0; return 0;
err_reg: err_reg:
if (tbl != ipv4_route_flush_table) if (tbl != ipv4_route_netns_table)
kfree(tbl); kfree(tbl);
err_dup: err_dup:
return -ENOMEM; return -ENOMEM;
@ -3607,7 +3616,7 @@ static __net_exit void sysctl_route_net_exit(struct net *net)
tbl = net->ipv4.route_hdr->ctl_table_arg; tbl = net->ipv4.route_hdr->ctl_table_arg;
unregister_net_sysctl_table(net->ipv4.route_hdr); unregister_net_sysctl_table(net->ipv4.route_hdr);
BUG_ON(tbl == ipv4_route_flush_table); BUG_ON(tbl == ipv4_route_netns_table);
kfree(tbl); kfree(tbl);
} }
@ -3617,6 +3626,17 @@ static __net_initdata struct pernet_operations sysctl_route_ops = {
}; };
#endif #endif
static __net_init int netns_ip_rt_init(struct net *net)
{
/* Set default value for namespaceified sysctls */
net->ipv4.ip_rt_min_pmtu = DEFAULT_MIN_PMTU;
return 0;
}
static struct pernet_operations __net_initdata ip_rt_ops = {
.init = netns_ip_rt_init,
};
static __net_init int rt_genid_init(struct net *net) static __net_init int rt_genid_init(struct net *net)
{ {
atomic_set(&net->ipv4.rt_genid, 0); atomic_set(&net->ipv4.rt_genid, 0);
@ -3722,6 +3742,7 @@ int __init ip_rt_init(void)
#ifdef CONFIG_SYSCTL #ifdef CONFIG_SYSCTL
register_pernet_subsys(&sysctl_route_ops); register_pernet_subsys(&sysctl_route_ops);
#endif #endif
register_pernet_subsys(&ip_rt_ops);
register_pernet_subsys(&rt_genid_ops); register_pernet_subsys(&rt_genid_ops);
register_pernet_subsys(&ipv4_inetpeer_ops); register_pernet_subsys(&ipv4_inetpeer_ops);
return 0; return 0;