netns xfrm: per-netns sysctls

Make
	net.core.xfrm_aevent_etime
	net.core.xfrm_acq_expires
	net.core.xfrm_aevent_rseqth
	net.core.xfrm_larval_drop

sysctls per-netns.

For that make net_core_path[] global, register it to prevent two
/proc/net/core antries and change initcall position -- xfrm_init() is called
from fs_initcall, so this one should be fs_initcall at least.

Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

6 files changed, 105 insertions, 56 deletions

diff --git a/net/core/sysctl_net_core.c b/net/core/sysctl_net_core.c
index 2bc0384b0448..83d3398559ea 100644
--- a/net/core/sysctl_net_core.c
+++ b/net/core/sysctl_net_core.c
@@ -12,7 +12,6 @@
 #include <linux/netdevice.h>
 #include <linux/init.h>
 #include <net/sock.h>
-#include <net/xfrm.h>
 
 static struct ctl_table net_core_table[] = {
 #ifdef CONFIG_NET
@@ -89,40 +88,6 @@ static struct ctl_table net_core_table[] = {
                 .mode           = 0644,
                 .proc_handler   = proc_dointvec
         },
-#ifdef CONFIG_XFRM
-        {
-                .ctl_name       = NET_CORE_AEVENT_ETIME,
-                .procname       = "xfrm_aevent_etime",
-                .data           = &sysctl_xfrm_aevent_etime,
-                .maxlen         = sizeof(u32),
-                .mode           = 0644,
-                .proc_handler   = proc_dointvec
-        },
-        {
-                .ctl_name       = NET_CORE_AEVENT_RSEQTH,
-                .procname       = "xfrm_aevent_rseqth",
-                .data           = &sysctl_xfrm_aevent_rseqth,
-                .maxlen         = sizeof(u32),
-                .mode           = 0644,
-                .proc_handler   = proc_dointvec
-        },
-        {
-                .ctl_name       = CTL_UNNUMBERED,
-                .procname       = "xfrm_larval_drop",
-                .data           = &sysctl_xfrm_larval_drop,
-                .maxlen         = sizeof(int),
-                .mode           = 0644,
-                .proc_handler   = proc_dointvec
-        },
-        {
-                .ctl_name       = CTL_UNNUMBERED,
-                .procname       = "xfrm_acq_expires",
-                .data           = &sysctl_xfrm_acq_expires,
-                .maxlen         = sizeof(int),
-                .mode           = 0644,
-                .proc_handler   = proc_dointvec
-        },
-#endif /* CONFIG_XFRM */
 #endif /* CONFIG_NET */
         {
                 .ctl_name       = NET_CORE_BUDGET,
@@ -155,7 +120,7 @@ static struct ctl_table netns_core_table[] = {
         { .ctl_name = 0 }
 };
 
-static __net_initdata struct ctl_path net_core_path[] = {
+__net_initdata struct ctl_path net_core_path[] = {
         { .procname = "net", .ctl_name = CTL_NET, },
         { .procname = "core", .ctl_name = NET_CORE, },
         { },
@@ -207,8 +172,11 @@ static __net_initdata struct pernet_operations sysctl_core_ops = {
 
 static __init int sysctl_core_init(void)
 {
+        static struct ctl_table empty[1];
+
+        register_sysctl_paths(net_core_path, empty);
         register_net_sysctl_rotable(net_core_path, net_core_table);
         return register_pernet_subsys(&sysctl_core_ops);
 }
 
-__initcall(sysctl_core_init);
+fs_initcall(sysctl_core_init);
diff --git a/net/xfrm/Makefile b/net/xfrm/Makefile
index 0f439a72ccab..c631047e1b27 100644
--- a/net/xfrm/Makefile
+++ b/net/xfrm/Makefile
@@ -3,8 +3,8 @@
 #
 
 obj-$(CONFIG_XFRM) := xfrm_policy.o xfrm_state.o xfrm_hash.o \
-                      xfrm_input.o xfrm_output.o xfrm_algo.o
+                      xfrm_input.o xfrm_output.o xfrm_algo.o \
+                      xfrm_sysctl.o
 obj-$(CONFIG_XFRM_STATISTICS) += xfrm_proc.o
 obj-$(CONFIG_XFRM_USER) += xfrm_user.o
 obj-$(CONFIG_XFRM_IPCOMP) += xfrm_ipcomp.o
-
diff --git a/net/xfrm/xfrm_policy.c b/net/xfrm/xfrm_policy.c
index 38822b34ba7d..393cc65dbfa4 100644
--- a/net/xfrm/xfrm_policy.c
+++ b/net/xfrm/xfrm_policy.c
@@ -34,8 +34,6 @@
 
 #include "xfrm_hash.h"
 
-int sysctl_xfrm_larval_drop __read_mostly = 1;
-
 DEFINE_MUTEX(xfrm_cfg_mutex);
 EXPORT_SYMBOL(xfrm_cfg_mutex);
 
@@ -1671,7 +1669,7 @@ restart:
 
                 if (unlikely(nx<0)) {
                         err = nx;
-                        if (err == -EAGAIN && sysctl_xfrm_larval_drop) {
+                        if (err == -EAGAIN && net->xfrm.sysctl_larval_drop) {
                                 /* EREMOTE tells the caller to generate
                                  * a one-shot blackhole route.
                                  */
@@ -2504,8 +2502,13 @@ static int __net_init xfrm_net_init(struct net *net)
         rv = xfrm_policy_init(net);
         if (rv < 0)
                 goto out_policy;
+        rv = xfrm_sysctl_init(net);
+        if (rv < 0)
+                goto out_sysctl;
         return 0;
 
+out_sysctl:
+        xfrm_policy_fini(net);
 out_policy:
         xfrm_state_fini(net);
 out_state:
@@ -2516,6 +2519,7 @@ out_statistics:
 
 static void __net_exit xfrm_net_exit(struct net *net)
 {
+        xfrm_sysctl_fini(net);
         xfrm_policy_fini(net);
         xfrm_state_fini(net);
         xfrm_statistics_fini(net);
diff --git a/net/xfrm/xfrm_state.c b/net/xfrm/xfrm_state.c
index 662e47b0bcc3..2fd57f8f77c1 100644
--- a/net/xfrm/xfrm_state.c
+++ b/net/xfrm/xfrm_state.c
@@ -24,14 +24,6 @@
 
 #include "xfrm_hash.h"
 
-u32 sysctl_xfrm_aevent_etime __read_mostly = XFRM_AE_ETIME;
-EXPORT_SYMBOL(sysctl_xfrm_aevent_etime);
-
-u32 sysctl_xfrm_aevent_rseqth __read_mostly = XFRM_AE_SEQT_SIZE;
-EXPORT_SYMBOL(sysctl_xfrm_aevent_rseqth);
-
-u32 sysctl_xfrm_acq_expires __read_mostly = 30;
-
 /* Each xfrm_state may be linked to two tables:
 
    1. Hash table by (spi,daddr,ah/esp) to find SA by SPI. (input,ctl)
@@ -851,8 +843,8 @@ xfrm_state_find(xfrm_address_t *daddr, xfrm_address_t *saddr,
                                 h = xfrm_spi_hash(net, &x->id.daddr, x->id.spi, x->id.proto, family);
                                 hlist_add_head(&x->byspi, net->xfrm.state_byspi+h);
                         }
-                        x->lft.hard_add_expires_seconds = sysctl_xfrm_acq_expires;
-                        x->timer.expires = jiffies + sysctl_xfrm_acq_expires*HZ;
+                        x->lft.hard_add_expires_seconds = net->xfrm.sysctl_acq_expires;
+                        x->timer.expires = jiffies + net->xfrm.sysctl_acq_expires*HZ;
                         add_timer(&x->timer);
                         net->xfrm.state_num++;
                         xfrm_hash_grow_check(net, x->bydst.next != NULL);
@@ -1040,9 +1032,9 @@ static struct xfrm_state *__find_acq_core(struct net *net, unsigned short family
                 x->props.family = family;
                 x->props.mode = mode;
                 x->props.reqid = reqid;
-                x->lft.hard_add_expires_seconds = sysctl_xfrm_acq_expires;
+                x->lft.hard_add_expires_seconds = net->xfrm.sysctl_acq_expires;
                 xfrm_state_hold(x);
-                x->timer.expires = jiffies + sysctl_xfrm_acq_expires*HZ;
+                x->timer.expires = jiffies + net->xfrm.sysctl_acq_expires*HZ;
                 add_timer(&x->timer);
                 list_add(&x->km.all, &net->xfrm.state_all);
                 hlist_add_head(&x->bydst, net->xfrm.state_bydst+h);
diff --git a/net/xfrm/xfrm_sysctl.c b/net/xfrm/xfrm_sysctl.c
new file mode 100644
index 000000000000..2e6ffb66f06f
--- /dev/null
+++ b/net/xfrm/xfrm_sysctl.c
@@ -0,0 +1,85 @@
+#include <linux/sysctl.h>
+#include <net/net_namespace.h>
+#include <net/xfrm.h>
+
+static void __xfrm_sysctl_init(struct net *net)
+{
+        net->xfrm.sysctl_aevent_etime = XFRM_AE_ETIME;
+        net->xfrm.sysctl_aevent_rseqth = XFRM_AE_SEQT_SIZE;
+        net->xfrm.sysctl_larval_drop = 1;
+        net->xfrm.sysctl_acq_expires = 30;
+}
+
+#ifdef CONFIG_SYSCTL
+static struct ctl_table xfrm_table[] = {
+        {
+                .ctl_name       = NET_CORE_AEVENT_ETIME,
+                .procname       = "xfrm_aevent_etime",
+                .maxlen         = sizeof(u32),
+                .mode           = 0644,
+                .proc_handler   = proc_dointvec
+        },
+        {
+                .ctl_name       = NET_CORE_AEVENT_RSEQTH,
+                .procname       = "xfrm_aevent_rseqth",
+                .maxlen         = sizeof(u32),
+                .mode           = 0644,
+                .proc_handler   = proc_dointvec
+        },
+        {
+                .ctl_name       = CTL_UNNUMBERED,
+                .procname       = "xfrm_larval_drop",
+                .maxlen         = sizeof(int),
+                .mode           = 0644,
+                .proc_handler   = proc_dointvec
+        },
+        {
+                .ctl_name       = CTL_UNNUMBERED,
+                .procname       = "xfrm_acq_expires",
+                .maxlen         = sizeof(int),
+                .mode           = 0644,
+                .proc_handler   = proc_dointvec
+        },
+        {}
+};
+
+int __net_init xfrm_sysctl_init(struct net *net)
+{
+        struct ctl_table *table;
+
+        __xfrm_sysctl_init(net);
+
+        table = kmemdup(xfrm_table, sizeof(xfrm_table), GFP_KERNEL);
+        if (!table)
+                goto out_kmemdup;
+        table[0].data = &net->xfrm.sysctl_aevent_etime;
+        table[1].data = &net->xfrm.sysctl_aevent_rseqth;
+        table[2].data = &net->xfrm.sysctl_larval_drop;
+        table[3].data = &net->xfrm.sysctl_acq_expires;
+
+        net->xfrm.sysctl_hdr = register_net_sysctl_table(net, net_core_path, table);
+        if (!net->xfrm.sysctl_hdr)
+                goto out_register;
+        return 0;
+
+out_register:
+        kfree(table);
+out_kmemdup:
+        return -ENOMEM;
+}
+
+void xfrm_sysctl_fini(struct net *net)
+{
+        struct ctl_table *table;
+
+        table = net->xfrm.sysctl_hdr->ctl_table_arg;
+        unregister_net_sysctl_table(net->xfrm.sysctl_hdr);
+        kfree(table);
+}
+#else
+int __net_init xfrm_sysctl_init(struct net *net)
+{
+        __xfrm_sysctl_init(net);
+        return 0;
+}
+#endif
diff --git a/net/xfrm/xfrm_user.c b/net/xfrm/xfrm_user.c
index b7240d5b77ad..38ffaf33312e 100644
--- a/net/xfrm/xfrm_user.c
+++ b/net/xfrm/xfrm_user.c
@@ -368,9 +368,9 @@ static struct xfrm_state *xfrm_state_construct(struct net *net,
                 goto error;
 
         x->km.seq = p->seq;
-        x->replay_maxdiff = sysctl_xfrm_aevent_rseqth;
+        x->replay_maxdiff = net->xfrm.sysctl_aevent_rseqth;
         /* sysctl_xfrm_aevent_etime is in 100ms units */
-        x->replay_maxage = (sysctl_xfrm_aevent_etime*HZ)/XFRM_AE_ETH_M;
+        x->replay_maxage = (net->xfrm.sysctl_aevent_etime*HZ)/XFRM_AE_ETH_M;
         x->preplay.bitmap = 0;
         x->preplay.seq = x->replay.seq+x->replay_maxdiff;
         x->preplay.oseq = x->replay.oseq +x->replay_maxdiff;