summaryrefslogtreecommitdiff
path: root/net/core/sysctl_net_core.c
diff options
context:
space:
mode:
Diffstat (limited to 'net/core/sysctl_net_core.c')
-rw-r--r--net/core/sysctl_net_core.c141
1 files changed, 84 insertions, 57 deletions
diff --git a/net/core/sysctl_net_core.c b/net/core/sysctl_net_core.c
index 2079000691e2..8d4decb2606f 100644
--- a/net/core/sysctl_net_core.c
+++ b/net/core/sysctl_net_core.c
@@ -28,12 +28,14 @@
#include <net/rps.h>
#include "dev.h"
+#include "net-sysfs.h"
static int int_3600 = 3600;
static int min_sndbuf = SOCK_MIN_SNDBUF;
static int min_rcvbuf = SOCK_MIN_RCVBUF;
static int max_skb_frags = MAX_SKB_FRAGS;
static int min_mem_pcpu_rsv = SK_MEMORY_PCPU_RESERVE;
+static int netdev_budget_usecs_min = 2 * USEC_PER_SEC / HZ;
static int net_msg_warn; /* Unused, but still a sysctl */
@@ -51,82 +53,88 @@ int sysctl_devconf_inherit_init_net __read_mostly;
EXPORT_SYMBOL(sysctl_devconf_inherit_init_net);
#if IS_ENABLED(CONFIG_NET_FLOW_LIMIT) || IS_ENABLED(CONFIG_RPS)
-static void dump_cpumask(void *buffer, size_t *lenp, loff_t *ppos,
- struct cpumask *mask)
+static int dump_cpumask(void *buffer, size_t *lenp, loff_t *ppos,
+ struct cpumask *mask)
{
- char kbuf[128];
+ char *kbuf;
int len;
if (*ppos || !*lenp) {
*lenp = 0;
- return;
+ return 0;
+ }
+
+ /* CPUs are displayed as a hex bitmap + a comma between each groups of 8
+ * nibbles (except the last one which has a newline instead).
+ * Guesstimate the buffer size at the group granularity level.
+ */
+ len = min(DIV_ROUND_UP(nr_cpumask_bits, 32) * (8 + 1), *lenp);
+ kbuf = kmalloc(len, GFP_KERNEL);
+ if (!kbuf) {
+ *lenp = 0;
+ return -ENOMEM;
}
- len = min(sizeof(kbuf) - 1, *lenp);
len = scnprintf(kbuf, len, "%*pb", cpumask_pr_args(mask));
if (!len) {
*lenp = 0;
- return;
+ goto free_buf;
}
- if (len < *lenp)
- kbuf[len++] = '\n';
+ /* scnprintf writes a trailing null char not counted in the returned
+ * length, override it with a newline.
+ */
+ kbuf[len++] = '\n';
memcpy(buffer, kbuf, len);
*lenp = len;
*ppos += len;
+
+free_buf:
+ kfree(kbuf);
+ return 0;
}
#endif
#ifdef CONFIG_RPS
-static struct cpumask *rps_default_mask_cow_alloc(struct net *net)
-{
- struct cpumask *rps_default_mask;
-
- if (net->core.rps_default_mask)
- return net->core.rps_default_mask;
-
- rps_default_mask = kzalloc(cpumask_size(), GFP_KERNEL);
- if (!rps_default_mask)
- return NULL;
+DEFINE_MUTEX(rps_default_mask_mutex);
- /* pairs with READ_ONCE in rx_queue_default_mask() */
- WRITE_ONCE(net->core.rps_default_mask, rps_default_mask);
- return rps_default_mask;
-}
-
-static int rps_default_mask_sysctl(struct ctl_table *table, int write,
+static int rps_default_mask_sysctl(const struct ctl_table *table, int write,
void *buffer, size_t *lenp, loff_t *ppos)
{
struct net *net = (struct net *)table->data;
+ struct cpumask *mask;
int err = 0;
- rtnl_lock();
+ mutex_lock(&rps_default_mask_mutex);
+ mask = net->core.rps_default_mask;
if (write) {
- struct cpumask *rps_default_mask = rps_default_mask_cow_alloc(net);
-
+ if (!mask) {
+ mask = kzalloc(cpumask_size(), GFP_KERNEL);
+ net->core.rps_default_mask = mask;
+ }
err = -ENOMEM;
- if (!rps_default_mask)
+ if (!mask)
goto done;
- err = cpumask_parse(buffer, rps_default_mask);
+ err = cpumask_parse(buffer, mask);
if (err)
goto done;
- err = rps_cpumask_housekeeping(rps_default_mask);
+ err = rps_cpumask_housekeeping(mask);
if (err)
goto done;
} else {
- dump_cpumask(buffer, lenp, ppos,
- net->core.rps_default_mask ? : cpu_none_mask);
+ err = dump_cpumask(buffer, lenp, ppos,
+ mask ?: cpu_none_mask);
}
done:
- rtnl_unlock();
+ mutex_unlock(&rps_default_mask_mutex);
return err;
}
-static int rps_sock_flow_sysctl(struct ctl_table *table, int write,
+static int rps_sock_flow_sysctl(const struct ctl_table *table, int write,
void *buffer, size_t *lenp, loff_t *ppos)
{
unsigned int orig_size, size;
@@ -184,7 +192,7 @@ static int rps_sock_flow_sysctl(struct ctl_table *table, int write,
if (orig_sock_table) {
static_branch_dec(&rps_needed);
static_branch_dec(&rfs_needed);
- kvfree_rcu_mightsleep(orig_sock_table);
+ kvfree_rcu(orig_sock_table, rcu);
}
}
}
@@ -198,7 +206,7 @@ static int rps_sock_flow_sysctl(struct ctl_table *table, int write,
#ifdef CONFIG_NET_FLOW_LIMIT
static DEFINE_MUTEX(flow_limit_update_mutex);
-static int flow_limit_cpu_sysctl(struct ctl_table *table, int write,
+static int flow_limit_cpu_sysctl(const struct ctl_table *table, int write,
void *buffer, size_t *lenp, loff_t *ppos)
{
struct sd_flow_limit *cur;
@@ -222,7 +230,7 @@ static int flow_limit_cpu_sysctl(struct ctl_table *table, int write,
lockdep_is_held(&flow_limit_update_mutex));
if (cur && !cpumask_test_cpu(i, mask)) {
RCU_INIT_POINTER(sd->flow_limit, NULL);
- kfree_rcu_mightsleep(cur);
+ kfree_rcu(cur, rcu);
} else if (!cur && cpumask_test_cpu(i, mask)) {
cur = kzalloc_node(len, GFP_KERNEL,
cpu_to_node(i));
@@ -231,7 +239,7 @@ static int flow_limit_cpu_sysctl(struct ctl_table *table, int write,
ret = -ENOMEM;
goto write_unlock;
}
- cur->num_buckets = netdev_flow_limit_table_len;
+ cur->log_buckets = ilog2(netdev_flow_limit_table_len);
rcu_assign_pointer(sd->flow_limit, cur);
}
}
@@ -247,7 +255,7 @@ write_unlock:
}
rcu_read_unlock();
- dump_cpumask(buffer, lenp, ppos, mask);
+ ret = dump_cpumask(buffer, lenp, ppos, mask);
}
done:
@@ -255,7 +263,7 @@ done:
return ret;
}
-static int flow_limit_table_len_sysctl(struct ctl_table *table, int write,
+static int flow_limit_table_len_sysctl(const struct ctl_table *table, int write,
void *buffer, size_t *lenp, loff_t *ppos)
{
unsigned int old, *ptr;
@@ -277,7 +285,7 @@ static int flow_limit_table_len_sysctl(struct ctl_table *table, int write,
#endif /* CONFIG_NET_FLOW_LIMIT */
#ifdef CONFIG_NET_SCHED
-static int set_default_qdisc(struct ctl_table *table, int write,
+static int set_default_qdisc(const struct ctl_table *table, int write,
void *buffer, size_t *lenp, loff_t *ppos)
{
char id[IFNAMSIZ];
@@ -296,14 +304,14 @@ static int set_default_qdisc(struct ctl_table *table, int write,
}
#endif
-static int proc_do_dev_weight(struct ctl_table *table, int write,
+static int proc_do_dev_weight(const struct ctl_table *table, int write,
void *buffer, size_t *lenp, loff_t *ppos)
{
static DEFINE_MUTEX(dev_weight_mutex);
int ret, weight;
mutex_lock(&dev_weight_mutex);
- ret = proc_dointvec(table, write, buffer, lenp, ppos);
+ ret = proc_dointvec_minmax(table, write, buffer, lenp, ppos);
if (!ret && write) {
weight = READ_ONCE(weight_p);
WRITE_ONCE(net_hotdata.dev_rx_weight, weight * dev_weight_rx_bias);
@@ -314,7 +322,7 @@ static int proc_do_dev_weight(struct ctl_table *table, int write,
return ret;
}
-static int proc_do_rss_key(struct ctl_table *table, int write,
+static int proc_do_rss_key(const struct ctl_table *table, int write,
void *buffer, size_t *lenp, loff_t *ppos)
{
struct ctl_table fake_table;
@@ -327,7 +335,7 @@ static int proc_do_rss_key(struct ctl_table *table, int write,
}
#ifdef CONFIG_BPF_JIT
-static int proc_dointvec_minmax_bpf_enable(struct ctl_table *table, int write,
+static int proc_dointvec_minmax_bpf_enable(const struct ctl_table *table, int write,
void *buffer, size_t *lenp,
loff_t *ppos)
{
@@ -360,7 +368,7 @@ static int proc_dointvec_minmax_bpf_enable(struct ctl_table *table, int write,
# ifdef CONFIG_HAVE_EBPF_JIT
static int
-proc_dointvec_minmax_bpf_restricted(struct ctl_table *table, int write,
+proc_dointvec_minmax_bpf_restricted(const struct ctl_table *table, int write,
void *buffer, size_t *lenp, loff_t *ppos)
{
if (!capable(CAP_SYS_ADMIN))
@@ -371,7 +379,7 @@ proc_dointvec_minmax_bpf_restricted(struct ctl_table *table, int write,
# endif /* CONFIG_HAVE_EBPF_JIT */
static int
-proc_dolongvec_minmax_bpf_restricted(struct ctl_table *table, int write,
+proc_dolongvec_minmax_bpf_restricted(const struct ctl_table *table, int write,
void *buffer, size_t *lenp, loff_t *ppos)
{
if (!capable(CAP_SYS_ADMIN))
@@ -396,6 +404,7 @@ static struct ctl_table net_core_table[] = {
.maxlen = sizeof(int),
.mode = 0644,
.proc_handler = proc_do_dev_weight,
+ .extra1 = SYSCTL_ONE,
},
{
.procname = "dev_weight_rx_bias",
@@ -403,6 +412,7 @@ static struct ctl_table net_core_table[] = {
.maxlen = sizeof(int),
.mode = 0644,
.proc_handler = proc_do_dev_weight,
+ .extra1 = SYSCTL_ONE,
},
{
.procname = "dev_weight_tx_bias",
@@ -410,6 +420,7 @@ static struct ctl_table net_core_table[] = {
.maxlen = sizeof(int),
.mode = 0644,
.proc_handler = proc_do_dev_weight,
+ .extra1 = SYSCTL_ONE,
},
{
.procname = "netdev_max_backlog",
@@ -491,15 +502,6 @@ static struct ctl_table net_core_table[] = {
.mode = 0644,
.proc_handler = proc_dointvec,
},
- {
- .procname = "tstamp_allow_data",
- .data = &sysctl_tstamp_allow_data,
- .maxlen = sizeof(int),
- .mode = 0644,
- .proc_handler = proc_dointvec_minmax,
- .extra1 = SYSCTL_ZERO,
- .extra2 = SYSCTL_ONE
- },
#ifdef CONFIG_RPS
{
.procname = "rps_sock_flow_entries",
@@ -577,7 +579,7 @@ static struct ctl_table net_core_table[] = {
.maxlen = sizeof(unsigned int),
.mode = 0644,
.proc_handler = proc_dointvec_minmax,
- .extra1 = SYSCTL_ZERO,
+ .extra1 = &netdev_budget_usecs_min,
},
{
.procname = "fb_tunnels_only_for_init_net",
@@ -665,6 +667,31 @@ static struct ctl_table netns_core_table[] = {
.extra2 = SYSCTL_ONE,
.proc_handler = proc_dou8vec_minmax,
},
+ {
+ .procname = "txq_reselection_ms",
+ .data = &init_net.core.sysctl_txq_reselection,
+ .maxlen = sizeof(int),
+ .mode = 0644,
+ .proc_handler = proc_dointvec_ms_jiffies,
+ },
+ {
+ .procname = "tstamp_allow_data",
+ .data = &init_net.core.sysctl_tstamp_allow_data,
+ .maxlen = sizeof(u8),
+ .mode = 0644,
+ .proc_handler = proc_dou8vec_minmax,
+ .extra1 = SYSCTL_ZERO,
+ .extra2 = SYSCTL_ONE
+ },
+ {
+ .procname = "bypass_prot_mem",
+ .data = &init_net.core.sysctl_bypass_prot_mem,
+ .maxlen = sizeof(u8),
+ .mode = 0644,
+ .proc_handler = proc_dou8vec_minmax,
+ .extra1 = SYSCTL_ZERO,
+ .extra2 = SYSCTL_ONE
+ },
/* sysctl_core_net_init() will set the values after this
* to readonly in network namespaces
*/
generated by cgit (git 2.34.1) at 2025-12-22 00:57:57 +0000