OpenCloudOS-Kernel/net/core/drop_monitor.c

1692 lines
39 KiB
C
Raw Normal View History

// SPDX-License-Identifier: GPL-2.0-only
/*
* Monitoring code for network dropped packet alerts
*
* Copyright (C) 2009 Neil Horman <nhorman@tuxdriver.com>
*/
#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
#include <linux/netdevice.h>
#include <linux/etherdevice.h>
#include <linux/string.h>
#include <linux/if_arp.h>
#include <linux/inetdevice.h>
#include <linux/inet.h>
#include <linux/interrupt.h>
#include <linux/netpoll.h>
#include <linux/sched.h>
#include <linux/delay.h>
#include <linux/types.h>
#include <linux/workqueue.h>
#include <linux/netlink.h>
#include <linux/net_dropmon.h>
#include <linux/percpu.h>
#include <linux/timer.h>
#include <linux/bitops.h>
include cleanup: Update gfp.h and slab.h includes to prepare for breaking implicit slab.h inclusion from percpu.h percpu.h is included by sched.h and module.h and thus ends up being included when building most .c files. percpu.h includes slab.h which in turn includes gfp.h making everything defined by the two files universally available and complicating inclusion dependencies. percpu.h -> slab.h dependency is about to be removed. Prepare for this change by updating users of gfp and slab facilities include those headers directly instead of assuming availability. As this conversion needs to touch large number of source files, the following script is used as the basis of conversion. http://userweb.kernel.org/~tj/misc/slabh-sweep.py The script does the followings. * Scan files for gfp and slab usages and update includes such that only the necessary includes are there. ie. if only gfp is used, gfp.h, if slab is used, slab.h. * When the script inserts a new include, it looks at the include blocks and try to put the new include such that its order conforms to its surrounding. It's put in the include block which contains core kernel includes, in the same order that the rest are ordered - alphabetical, Christmas tree, rev-Xmas-tree or at the end if there doesn't seem to be any matching order. * If the script can't find a place to put a new include (mostly because the file doesn't have fitting include block), it prints out an error message indicating which .h file needs to be added to the file. The conversion was done in the following steps. 1. The initial automatic conversion of all .c files updated slightly over 4000 files, deleting around 700 includes and adding ~480 gfp.h and ~3000 slab.h inclusions. The script emitted errors for ~400 files. 2. Each error was manually checked. Some didn't need the inclusion, some needed manual addition while adding it to implementation .h or embedding .c file was more appropriate for others. This step added inclusions to around 150 files. 3. The script was run again and the output was compared to the edits from #2 to make sure no file was left behind. 4. Several build tests were done and a couple of problems were fixed. e.g. lib/decompress_*.c used malloc/free() wrappers around slab APIs requiring slab.h to be added manually. 5. The script was run on all .h files but without automatically editing them as sprinkling gfp.h and slab.h inclusions around .h files could easily lead to inclusion dependency hell. Most gfp.h inclusion directives were ignored as stuff from gfp.h was usually wildly available and often used in preprocessor macros. Each slab.h inclusion directive was examined and added manually as necessary. 6. percpu.h was updated not to include slab.h. 7. Build test were done on the following configurations and failures were fixed. CONFIG_GCOV_KERNEL was turned off for all tests (as my distributed build env didn't work with gcov compiles) and a few more options had to be turned off depending on archs to make things build (like ipr on powerpc/64 which failed due to missing writeq). * x86 and x86_64 UP and SMP allmodconfig and a custom test config. * powerpc and powerpc64 SMP allmodconfig * sparc and sparc64 SMP allmodconfig * ia64 SMP allmodconfig * s390 SMP allmodconfig * alpha SMP allmodconfig * um on x86_64 SMP allmodconfig 8. percpu.h modifications were reverted so that it could be applied as a separate patch and serve as bisection point. Given the fact that I had only a couple of failures from tests on step 6, I'm fairly confident about the coverage of this conversion patch. If there is a breakage, it's likely to be something in one of the arch headers which should be easily discoverable easily on most builds of the specific arch. Signed-off-by: Tejun Heo <tj@kernel.org> Guess-its-ok-by: Christoph Lameter <cl@linux-foundation.org> Cc: Ingo Molnar <mingo@redhat.com> Cc: Lee Schermerhorn <Lee.Schermerhorn@hp.com>
2010-03-24 16:04:11 +08:00
#include <linux/slab.h>
#include <linux/module.h>
#include <net/drop_monitor.h>
#include <net/genetlink.h>
#include <net/netevent.h>
#include <trace/events/skb.h>
#include <trace/events/napi.h>
#include <asm/unaligned.h>
#define TRACE_ON 1
#define TRACE_OFF 0
/*
* Globals, our netlink socket pointer
* and the work handle that will send up
* netlink alerts
*/
static int trace_state = TRACE_OFF;
static bool monitor_hw;
/* net_dm_mutex
*
* An overall lock guarding every operation coming from userspace.
* It also guards the global 'hw_stats_list' list.
*/
static DEFINE_MUTEX(net_dm_mutex);
struct net_dm_stats {
u64 dropped;
struct u64_stats_sync syncp;
};
#define NET_DM_MAX_HW_TRAP_NAME_LEN 40
struct net_dm_hw_entry {
char trap_name[NET_DM_MAX_HW_TRAP_NAME_LEN];
u32 count;
};
struct net_dm_hw_entries {
u32 num_entries;
struct net_dm_hw_entry entries[0];
};
struct per_cpu_dm_data {
spinlock_t lock; /* Protects 'skb', 'hw_entries' and
* 'send_timer'
*/
union {
struct sk_buff *skb;
struct net_dm_hw_entries *hw_entries;
};
struct sk_buff_head drop_queue;
drop_monitor: dont sleep in atomic context drop_monitor calls several sleeping functions while in atomic context. BUG: sleeping function called from invalid context at mm/slub.c:943 in_atomic(): 1, irqs_disabled(): 0, pid: 2103, name: kworker/0:2 Pid: 2103, comm: kworker/0:2 Not tainted 3.5.0-rc1+ #55 Call Trace: [<ffffffff810697ca>] __might_sleep+0xca/0xf0 [<ffffffff811345a3>] kmem_cache_alloc_node+0x1b3/0x1c0 [<ffffffff8105578c>] ? queue_delayed_work_on+0x11c/0x130 [<ffffffff815343fb>] __alloc_skb+0x4b/0x230 [<ffffffffa00b0360>] ? reset_per_cpu_data+0x160/0x160 [drop_monitor] [<ffffffffa00b022f>] reset_per_cpu_data+0x2f/0x160 [drop_monitor] [<ffffffffa00b03ab>] send_dm_alert+0x4b/0xb0 [drop_monitor] [<ffffffff810568e0>] process_one_work+0x130/0x4c0 [<ffffffff81058249>] worker_thread+0x159/0x360 [<ffffffff810580f0>] ? manage_workers.isra.27+0x240/0x240 [<ffffffff8105d403>] kthread+0x93/0xa0 [<ffffffff816be6d4>] kernel_thread_helper+0x4/0x10 [<ffffffff8105d370>] ? kthread_freezable_should_stop+0x80/0x80 [<ffffffff816be6d0>] ? gs_change+0xb/0xb Rework the logic to call the sleeping functions in right context. Use standard timer/workqueue api to let system chose any cpu to perform the allocation and netlink send. Also avoid a loop if reset_per_cpu_data() cannot allocate memory : use mod_timer() to wait 1/10 second before next try. Signed-off-by: Eric Dumazet <edumazet@google.com> Cc: Neil Horman <nhorman@tuxdriver.com> Reviewed-by: Neil Horman <nhorman@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-06-04 08:18:19 +08:00
struct work_struct dm_alert_work;
struct timer_list send_timer;
struct net_dm_stats stats;
};
struct dm_hw_stat_delta {
struct net_device *dev;
unsigned long last_rx;
struct list_head list;
struct rcu_head rcu;
unsigned long last_drop_val;
};
static struct genl_family net_drop_monitor_family;
static DEFINE_PER_CPU(struct per_cpu_dm_data, dm_cpu_data);
static DEFINE_PER_CPU(struct per_cpu_dm_data, dm_hw_cpu_data);
static int dm_hit_limit = 64;
static int dm_delay = 1;
static unsigned long dm_hw_check_delta = 2*HZ;
static LIST_HEAD(hw_stats_list);
static enum net_dm_alert_mode net_dm_alert_mode = NET_DM_ALERT_MODE_SUMMARY;
static u32 net_dm_trunc_len;
static u32 net_dm_queue_len = 1000;
struct net_dm_alert_ops {
void (*kfree_skb_probe)(void *ignore, struct sk_buff *skb,
void *location);
void (*napi_poll_probe)(void *ignore, struct napi_struct *napi,
int work, int budget);
void (*work_item_func)(struct work_struct *work);
void (*hw_work_item_func)(struct work_struct *work);
void (*hw_probe)(struct sk_buff *skb,
const struct net_dm_hw_metadata *hw_metadata);
};
struct net_dm_skb_cb {
union {
struct net_dm_hw_metadata *hw_metadata;
void *pc;
};
};
#define NET_DM_SKB_CB(__skb) ((struct net_dm_skb_cb *)&((__skb)->cb[0]))
drop_monitor: dont sleep in atomic context drop_monitor calls several sleeping functions while in atomic context. BUG: sleeping function called from invalid context at mm/slub.c:943 in_atomic(): 1, irqs_disabled(): 0, pid: 2103, name: kworker/0:2 Pid: 2103, comm: kworker/0:2 Not tainted 3.5.0-rc1+ #55 Call Trace: [<ffffffff810697ca>] __might_sleep+0xca/0xf0 [<ffffffff811345a3>] kmem_cache_alloc_node+0x1b3/0x1c0 [<ffffffff8105578c>] ? queue_delayed_work_on+0x11c/0x130 [<ffffffff815343fb>] __alloc_skb+0x4b/0x230 [<ffffffffa00b0360>] ? reset_per_cpu_data+0x160/0x160 [drop_monitor] [<ffffffffa00b022f>] reset_per_cpu_data+0x2f/0x160 [drop_monitor] [<ffffffffa00b03ab>] send_dm_alert+0x4b/0xb0 [drop_monitor] [<ffffffff810568e0>] process_one_work+0x130/0x4c0 [<ffffffff81058249>] worker_thread+0x159/0x360 [<ffffffff810580f0>] ? manage_workers.isra.27+0x240/0x240 [<ffffffff8105d403>] kthread+0x93/0xa0 [<ffffffff816be6d4>] kernel_thread_helper+0x4/0x10 [<ffffffff8105d370>] ? kthread_freezable_should_stop+0x80/0x80 [<ffffffff816be6d0>] ? gs_change+0xb/0xb Rework the logic to call the sleeping functions in right context. Use standard timer/workqueue api to let system chose any cpu to perform the allocation and netlink send. Also avoid a loop if reset_per_cpu_data() cannot allocate memory : use mod_timer() to wait 1/10 second before next try. Signed-off-by: Eric Dumazet <edumazet@google.com> Cc: Neil Horman <nhorman@tuxdriver.com> Reviewed-by: Neil Horman <nhorman@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-06-04 08:18:19 +08:00
static struct sk_buff *reset_per_cpu_data(struct per_cpu_dm_data *data)
{
size_t al;
struct net_dm_alert_msg *msg;
struct nlattr *nla;
struct sk_buff *skb;
drop_monitor: dont sleep in atomic context drop_monitor calls several sleeping functions while in atomic context. BUG: sleeping function called from invalid context at mm/slub.c:943 in_atomic(): 1, irqs_disabled(): 0, pid: 2103, name: kworker/0:2 Pid: 2103, comm: kworker/0:2 Not tainted 3.5.0-rc1+ #55 Call Trace: [<ffffffff810697ca>] __might_sleep+0xca/0xf0 [<ffffffff811345a3>] kmem_cache_alloc_node+0x1b3/0x1c0 [<ffffffff8105578c>] ? queue_delayed_work_on+0x11c/0x130 [<ffffffff815343fb>] __alloc_skb+0x4b/0x230 [<ffffffffa00b0360>] ? reset_per_cpu_data+0x160/0x160 [drop_monitor] [<ffffffffa00b022f>] reset_per_cpu_data+0x2f/0x160 [drop_monitor] [<ffffffffa00b03ab>] send_dm_alert+0x4b/0xb0 [drop_monitor] [<ffffffff810568e0>] process_one_work+0x130/0x4c0 [<ffffffff81058249>] worker_thread+0x159/0x360 [<ffffffff810580f0>] ? manage_workers.isra.27+0x240/0x240 [<ffffffff8105d403>] kthread+0x93/0xa0 [<ffffffff816be6d4>] kernel_thread_helper+0x4/0x10 [<ffffffff8105d370>] ? kthread_freezable_should_stop+0x80/0x80 [<ffffffff816be6d0>] ? gs_change+0xb/0xb Rework the logic to call the sleeping functions in right context. Use standard timer/workqueue api to let system chose any cpu to perform the allocation and netlink send. Also avoid a loop if reset_per_cpu_data() cannot allocate memory : use mod_timer() to wait 1/10 second before next try. Signed-off-by: Eric Dumazet <edumazet@google.com> Cc: Neil Horman <nhorman@tuxdriver.com> Reviewed-by: Neil Horman <nhorman@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-06-04 08:18:19 +08:00
unsigned long flags;
void *msg_header;
al = sizeof(struct net_dm_alert_msg);
al += dm_hit_limit * sizeof(struct net_dm_drop_point);
al += sizeof(struct nlattr);
skb = genlmsg_new(al, GFP_KERNEL);
if (!skb)
goto err;
msg_header = genlmsg_put(skb, 0, 0, &net_drop_monitor_family,
0, NET_DM_CMD_ALERT);
if (!msg_header) {
nlmsg_free(skb);
skb = NULL;
goto err;
}
nla = nla_reserve(skb, NLA_UNSPEC,
sizeof(struct net_dm_alert_msg));
if (!nla) {
nlmsg_free(skb);
skb = NULL;
goto err;
}
msg = nla_data(nla);
memset(msg, 0, al);
goto out;
err:
mod_timer(&data->send_timer, jiffies + HZ / 10);
out:
drop_monitor: dont sleep in atomic context drop_monitor calls several sleeping functions while in atomic context. BUG: sleeping function called from invalid context at mm/slub.c:943 in_atomic(): 1, irqs_disabled(): 0, pid: 2103, name: kworker/0:2 Pid: 2103, comm: kworker/0:2 Not tainted 3.5.0-rc1+ #55 Call Trace: [<ffffffff810697ca>] __might_sleep+0xca/0xf0 [<ffffffff811345a3>] kmem_cache_alloc_node+0x1b3/0x1c0 [<ffffffff8105578c>] ? queue_delayed_work_on+0x11c/0x130 [<ffffffff815343fb>] __alloc_skb+0x4b/0x230 [<ffffffffa00b0360>] ? reset_per_cpu_data+0x160/0x160 [drop_monitor] [<ffffffffa00b022f>] reset_per_cpu_data+0x2f/0x160 [drop_monitor] [<ffffffffa00b03ab>] send_dm_alert+0x4b/0xb0 [drop_monitor] [<ffffffff810568e0>] process_one_work+0x130/0x4c0 [<ffffffff81058249>] worker_thread+0x159/0x360 [<ffffffff810580f0>] ? manage_workers.isra.27+0x240/0x240 [<ffffffff8105d403>] kthread+0x93/0xa0 [<ffffffff816be6d4>] kernel_thread_helper+0x4/0x10 [<ffffffff8105d370>] ? kthread_freezable_should_stop+0x80/0x80 [<ffffffff816be6d0>] ? gs_change+0xb/0xb Rework the logic to call the sleeping functions in right context. Use standard timer/workqueue api to let system chose any cpu to perform the allocation and netlink send. Also avoid a loop if reset_per_cpu_data() cannot allocate memory : use mod_timer() to wait 1/10 second before next try. Signed-off-by: Eric Dumazet <edumazet@google.com> Cc: Neil Horman <nhorman@tuxdriver.com> Reviewed-by: Neil Horman <nhorman@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-06-04 08:18:19 +08:00
spin_lock_irqsave(&data->lock, flags);
swap(data->skb, skb);
spin_unlock_irqrestore(&data->lock, flags);
if (skb) {
struct nlmsghdr *nlh = (struct nlmsghdr *)skb->data;
struct genlmsghdr *gnlh = (struct genlmsghdr *)nlmsg_data(nlh);
genlmsg_end(skb, genlmsg_data(gnlh));
}
drop_monitor: dont sleep in atomic context drop_monitor calls several sleeping functions while in atomic context. BUG: sleeping function called from invalid context at mm/slub.c:943 in_atomic(): 1, irqs_disabled(): 0, pid: 2103, name: kworker/0:2 Pid: 2103, comm: kworker/0:2 Not tainted 3.5.0-rc1+ #55 Call Trace: [<ffffffff810697ca>] __might_sleep+0xca/0xf0 [<ffffffff811345a3>] kmem_cache_alloc_node+0x1b3/0x1c0 [<ffffffff8105578c>] ? queue_delayed_work_on+0x11c/0x130 [<ffffffff815343fb>] __alloc_skb+0x4b/0x230 [<ffffffffa00b0360>] ? reset_per_cpu_data+0x160/0x160 [drop_monitor] [<ffffffffa00b022f>] reset_per_cpu_data+0x2f/0x160 [drop_monitor] [<ffffffffa00b03ab>] send_dm_alert+0x4b/0xb0 [drop_monitor] [<ffffffff810568e0>] process_one_work+0x130/0x4c0 [<ffffffff81058249>] worker_thread+0x159/0x360 [<ffffffff810580f0>] ? manage_workers.isra.27+0x240/0x240 [<ffffffff8105d403>] kthread+0x93/0xa0 [<ffffffff816be6d4>] kernel_thread_helper+0x4/0x10 [<ffffffff8105d370>] ? kthread_freezable_should_stop+0x80/0x80 [<ffffffff816be6d0>] ? gs_change+0xb/0xb Rework the logic to call the sleeping functions in right context. Use standard timer/workqueue api to let system chose any cpu to perform the allocation and netlink send. Also avoid a loop if reset_per_cpu_data() cannot allocate memory : use mod_timer() to wait 1/10 second before next try. Signed-off-by: Eric Dumazet <edumazet@google.com> Cc: Neil Horman <nhorman@tuxdriver.com> Reviewed-by: Neil Horman <nhorman@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-06-04 08:18:19 +08:00
return skb;
}
static const struct genl_multicast_group dropmon_mcgrps[] = {
{ .name = "events", },
};
drop_monitor: dont sleep in atomic context drop_monitor calls several sleeping functions while in atomic context. BUG: sleeping function called from invalid context at mm/slub.c:943 in_atomic(): 1, irqs_disabled(): 0, pid: 2103, name: kworker/0:2 Pid: 2103, comm: kworker/0:2 Not tainted 3.5.0-rc1+ #55 Call Trace: [<ffffffff810697ca>] __might_sleep+0xca/0xf0 [<ffffffff811345a3>] kmem_cache_alloc_node+0x1b3/0x1c0 [<ffffffff8105578c>] ? queue_delayed_work_on+0x11c/0x130 [<ffffffff815343fb>] __alloc_skb+0x4b/0x230 [<ffffffffa00b0360>] ? reset_per_cpu_data+0x160/0x160 [drop_monitor] [<ffffffffa00b022f>] reset_per_cpu_data+0x2f/0x160 [drop_monitor] [<ffffffffa00b03ab>] send_dm_alert+0x4b/0xb0 [drop_monitor] [<ffffffff810568e0>] process_one_work+0x130/0x4c0 [<ffffffff81058249>] worker_thread+0x159/0x360 [<ffffffff810580f0>] ? manage_workers.isra.27+0x240/0x240 [<ffffffff8105d403>] kthread+0x93/0xa0 [<ffffffff816be6d4>] kernel_thread_helper+0x4/0x10 [<ffffffff8105d370>] ? kthread_freezable_should_stop+0x80/0x80 [<ffffffff816be6d0>] ? gs_change+0xb/0xb Rework the logic to call the sleeping functions in right context. Use standard timer/workqueue api to let system chose any cpu to perform the allocation and netlink send. Also avoid a loop if reset_per_cpu_data() cannot allocate memory : use mod_timer() to wait 1/10 second before next try. Signed-off-by: Eric Dumazet <edumazet@google.com> Cc: Neil Horman <nhorman@tuxdriver.com> Reviewed-by: Neil Horman <nhorman@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-06-04 08:18:19 +08:00
static void send_dm_alert(struct work_struct *work)
{
struct sk_buff *skb;
drop_monitor: dont sleep in atomic context drop_monitor calls several sleeping functions while in atomic context. BUG: sleeping function called from invalid context at mm/slub.c:943 in_atomic(): 1, irqs_disabled(): 0, pid: 2103, name: kworker/0:2 Pid: 2103, comm: kworker/0:2 Not tainted 3.5.0-rc1+ #55 Call Trace: [<ffffffff810697ca>] __might_sleep+0xca/0xf0 [<ffffffff811345a3>] kmem_cache_alloc_node+0x1b3/0x1c0 [<ffffffff8105578c>] ? queue_delayed_work_on+0x11c/0x130 [<ffffffff815343fb>] __alloc_skb+0x4b/0x230 [<ffffffffa00b0360>] ? reset_per_cpu_data+0x160/0x160 [drop_monitor] [<ffffffffa00b022f>] reset_per_cpu_data+0x2f/0x160 [drop_monitor] [<ffffffffa00b03ab>] send_dm_alert+0x4b/0xb0 [drop_monitor] [<ffffffff810568e0>] process_one_work+0x130/0x4c0 [<ffffffff81058249>] worker_thread+0x159/0x360 [<ffffffff810580f0>] ? manage_workers.isra.27+0x240/0x240 [<ffffffff8105d403>] kthread+0x93/0xa0 [<ffffffff816be6d4>] kernel_thread_helper+0x4/0x10 [<ffffffff8105d370>] ? kthread_freezable_should_stop+0x80/0x80 [<ffffffff816be6d0>] ? gs_change+0xb/0xb Rework the logic to call the sleeping functions in right context. Use standard timer/workqueue api to let system chose any cpu to perform the allocation and netlink send. Also avoid a loop if reset_per_cpu_data() cannot allocate memory : use mod_timer() to wait 1/10 second before next try. Signed-off-by: Eric Dumazet <edumazet@google.com> Cc: Neil Horman <nhorman@tuxdriver.com> Reviewed-by: Neil Horman <nhorman@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-06-04 08:18:19 +08:00
struct per_cpu_dm_data *data;
drop_monitor: dont sleep in atomic context drop_monitor calls several sleeping functions while in atomic context. BUG: sleeping function called from invalid context at mm/slub.c:943 in_atomic(): 1, irqs_disabled(): 0, pid: 2103, name: kworker/0:2 Pid: 2103, comm: kworker/0:2 Not tainted 3.5.0-rc1+ #55 Call Trace: [<ffffffff810697ca>] __might_sleep+0xca/0xf0 [<ffffffff811345a3>] kmem_cache_alloc_node+0x1b3/0x1c0 [<ffffffff8105578c>] ? queue_delayed_work_on+0x11c/0x130 [<ffffffff815343fb>] __alloc_skb+0x4b/0x230 [<ffffffffa00b0360>] ? reset_per_cpu_data+0x160/0x160 [drop_monitor] [<ffffffffa00b022f>] reset_per_cpu_data+0x2f/0x160 [drop_monitor] [<ffffffffa00b03ab>] send_dm_alert+0x4b/0xb0 [drop_monitor] [<ffffffff810568e0>] process_one_work+0x130/0x4c0 [<ffffffff81058249>] worker_thread+0x159/0x360 [<ffffffff810580f0>] ? manage_workers.isra.27+0x240/0x240 [<ffffffff8105d403>] kthread+0x93/0xa0 [<ffffffff816be6d4>] kernel_thread_helper+0x4/0x10 [<ffffffff8105d370>] ? kthread_freezable_should_stop+0x80/0x80 [<ffffffff816be6d0>] ? gs_change+0xb/0xb Rework the logic to call the sleeping functions in right context. Use standard timer/workqueue api to let system chose any cpu to perform the allocation and netlink send. Also avoid a loop if reset_per_cpu_data() cannot allocate memory : use mod_timer() to wait 1/10 second before next try. Signed-off-by: Eric Dumazet <edumazet@google.com> Cc: Neil Horman <nhorman@tuxdriver.com> Reviewed-by: Neil Horman <nhorman@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-06-04 08:18:19 +08:00
data = container_of(work, struct per_cpu_dm_data, dm_alert_work);
drop_monitor: dont sleep in atomic context drop_monitor calls several sleeping functions while in atomic context. BUG: sleeping function called from invalid context at mm/slub.c:943 in_atomic(): 1, irqs_disabled(): 0, pid: 2103, name: kworker/0:2 Pid: 2103, comm: kworker/0:2 Not tainted 3.5.0-rc1+ #55 Call Trace: [<ffffffff810697ca>] __might_sleep+0xca/0xf0 [<ffffffff811345a3>] kmem_cache_alloc_node+0x1b3/0x1c0 [<ffffffff8105578c>] ? queue_delayed_work_on+0x11c/0x130 [<ffffffff815343fb>] __alloc_skb+0x4b/0x230 [<ffffffffa00b0360>] ? reset_per_cpu_data+0x160/0x160 [drop_monitor] [<ffffffffa00b022f>] reset_per_cpu_data+0x2f/0x160 [drop_monitor] [<ffffffffa00b03ab>] send_dm_alert+0x4b/0xb0 [drop_monitor] [<ffffffff810568e0>] process_one_work+0x130/0x4c0 [<ffffffff81058249>] worker_thread+0x159/0x360 [<ffffffff810580f0>] ? manage_workers.isra.27+0x240/0x240 [<ffffffff8105d403>] kthread+0x93/0xa0 [<ffffffff816be6d4>] kernel_thread_helper+0x4/0x10 [<ffffffff8105d370>] ? kthread_freezable_should_stop+0x80/0x80 [<ffffffff816be6d0>] ? gs_change+0xb/0xb Rework the logic to call the sleeping functions in right context. Use standard timer/workqueue api to let system chose any cpu to perform the allocation and netlink send. Also avoid a loop if reset_per_cpu_data() cannot allocate memory : use mod_timer() to wait 1/10 second before next try. Signed-off-by: Eric Dumazet <edumazet@google.com> Cc: Neil Horman <nhorman@tuxdriver.com> Reviewed-by: Neil Horman <nhorman@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-06-04 08:18:19 +08:00
skb = reset_per_cpu_data(data);
if (skb)
genlmsg_multicast(&net_drop_monitor_family, skb, 0,
0, GFP_KERNEL);
}
/*
* This is the timer function to delay the sending of an alert
* in the event that more drops will arrive during the
drop_monitor: dont sleep in atomic context drop_monitor calls several sleeping functions while in atomic context. BUG: sleeping function called from invalid context at mm/slub.c:943 in_atomic(): 1, irqs_disabled(): 0, pid: 2103, name: kworker/0:2 Pid: 2103, comm: kworker/0:2 Not tainted 3.5.0-rc1+ #55 Call Trace: [<ffffffff810697ca>] __might_sleep+0xca/0xf0 [<ffffffff811345a3>] kmem_cache_alloc_node+0x1b3/0x1c0 [<ffffffff8105578c>] ? queue_delayed_work_on+0x11c/0x130 [<ffffffff815343fb>] __alloc_skb+0x4b/0x230 [<ffffffffa00b0360>] ? reset_per_cpu_data+0x160/0x160 [drop_monitor] [<ffffffffa00b022f>] reset_per_cpu_data+0x2f/0x160 [drop_monitor] [<ffffffffa00b03ab>] send_dm_alert+0x4b/0xb0 [drop_monitor] [<ffffffff810568e0>] process_one_work+0x130/0x4c0 [<ffffffff81058249>] worker_thread+0x159/0x360 [<ffffffff810580f0>] ? manage_workers.isra.27+0x240/0x240 [<ffffffff8105d403>] kthread+0x93/0xa0 [<ffffffff816be6d4>] kernel_thread_helper+0x4/0x10 [<ffffffff8105d370>] ? kthread_freezable_should_stop+0x80/0x80 [<ffffffff816be6d0>] ? gs_change+0xb/0xb Rework the logic to call the sleeping functions in right context. Use standard timer/workqueue api to let system chose any cpu to perform the allocation and netlink send. Also avoid a loop if reset_per_cpu_data() cannot allocate memory : use mod_timer() to wait 1/10 second before next try. Signed-off-by: Eric Dumazet <edumazet@google.com> Cc: Neil Horman <nhorman@tuxdriver.com> Reviewed-by: Neil Horman <nhorman@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-06-04 08:18:19 +08:00
* hysteresis period.
*/
treewide: setup_timer() -> timer_setup() This converts all remaining cases of the old setup_timer() API into using timer_setup(), where the callback argument is the structure already holding the struct timer_list. These should have no behavioral changes, since they just change which pointer is passed into the callback with the same available pointers after conversion. It handles the following examples, in addition to some other variations. Casting from unsigned long: void my_callback(unsigned long data) { struct something *ptr = (struct something *)data; ... } ... setup_timer(&ptr->my_timer, my_callback, ptr); and forced object casts: void my_callback(struct something *ptr) { ... } ... setup_timer(&ptr->my_timer, my_callback, (unsigned long)ptr); become: void my_callback(struct timer_list *t) { struct something *ptr = from_timer(ptr, t, my_timer); ... } ... timer_setup(&ptr->my_timer, my_callback, 0); Direct function assignments: void my_callback(unsigned long data) { struct something *ptr = (struct something *)data; ... } ... ptr->my_timer.function = my_callback; have a temporary cast added, along with converting the args: void my_callback(struct timer_list *t) { struct something *ptr = from_timer(ptr, t, my_timer); ... } ... ptr->my_timer.function = (TIMER_FUNC_TYPE)my_callback; And finally, callbacks without a data assignment: void my_callback(unsigned long data) { ... } ... setup_timer(&ptr->my_timer, my_callback, 0); have their argument renamed to verify they're unused during conversion: void my_callback(struct timer_list *unused) { ... } ... timer_setup(&ptr->my_timer, my_callback, 0); The conversion is done with the following Coccinelle script: spatch --very-quiet --all-includes --include-headers \ -I ./arch/x86/include -I ./arch/x86/include/generated \ -I ./include -I ./arch/x86/include/uapi \ -I ./arch/x86/include/generated/uapi -I ./include/uapi \ -I ./include/generated/uapi --include ./include/linux/kconfig.h \ --dir . \ --cocci-file ~/src/data/timer_setup.cocci @fix_address_of@ expression e; @@ setup_timer( -&(e) +&e , ...) // Update any raw setup_timer() usages that have a NULL callback, but // would otherwise match change_timer_function_usage, since the latter // will update all function assignments done in the face of a NULL // function initialization in setup_timer(). @change_timer_function_usage_NULL@ expression _E; identifier _timer; type _cast_data; @@ ( -setup_timer(&_E->_timer, NULL, _E); +timer_setup(&_E->_timer, NULL, 0); | -setup_timer(&_E->_timer, NULL, (_cast_data)_E); +timer_setup(&_E->_timer, NULL, 0); | -setup_timer(&_E._timer, NULL, &_E); +timer_setup(&_E._timer, NULL, 0); | -setup_timer(&_E._timer, NULL, (_cast_data)&_E); +timer_setup(&_E._timer, NULL, 0); ) @change_timer_function_usage@ expression _E; identifier _timer; struct timer_list _stl; identifier _callback; type _cast_func, _cast_data; @@ ( -setup_timer(&_E->_timer, _callback, _E); +timer_setup(&_E->_timer, _callback, 0); | -setup_timer(&_E->_timer, &_callback, _E); +timer_setup(&_E->_timer, _callback, 0); | -setup_timer(&_E->_timer, _callback, (_cast_data)_E); +timer_setup(&_E->_timer, _callback, 0); | -setup_timer(&_E->_timer, &_callback, (_cast_data)_E); +timer_setup(&_E->_timer, _callback, 0); | -setup_timer(&_E->_timer, (_cast_func)_callback, _E); +timer_setup(&_E->_timer, _callback, 0); | -setup_timer(&_E->_timer, (_cast_func)&_callback, _E); +timer_setup(&_E->_timer, _callback, 0); | -setup_timer(&_E->_timer, (_cast_func)_callback, (_cast_data)_E); +timer_setup(&_E->_timer, _callback, 0); | -setup_timer(&_E->_timer, (_cast_func)&_callback, (_cast_data)_E); +timer_setup(&_E->_timer, _callback, 0); | -setup_timer(&_E._timer, _callback, (_cast_data)_E); +timer_setup(&_E._timer, _callback, 0); | -setup_timer(&_E._timer, _callback, (_cast_data)&_E); +timer_setup(&_E._timer, _callback, 0); | -setup_timer(&_E._timer, &_callback, (_cast_data)_E); +timer_setup(&_E._timer, _callback, 0); | -setup_timer(&_E._timer, &_callback, (_cast_data)&_E); +timer_setup(&_E._timer, _callback, 0); | -setup_timer(&_E._timer, (_cast_func)_callback, (_cast_data)_E); +timer_setup(&_E._timer, _callback, 0); | -setup_timer(&_E._timer, (_cast_func)_callback, (_cast_data)&_E); +timer_setup(&_E._timer, _callback, 0); | -setup_timer(&_E._timer, (_cast_func)&_callback, (_cast_data)_E); +timer_setup(&_E._timer, _callback, 0); | -setup_timer(&_E._timer, (_cast_func)&_callback, (_cast_data)&_E); +timer_setup(&_E._timer, _callback, 0); | _E->_timer@_stl.function = _callback; | _E->_timer@_stl.function = &_callback; | _E->_timer@_stl.function = (_cast_func)_callback; | _E->_timer@_stl.function = (_cast_func)&_callback; | _E._timer@_stl.function = _callback; | _E._timer@_stl.function = &_callback; | _E._timer@_stl.function = (_cast_func)_callback; | _E._timer@_stl.function = (_cast_func)&_callback; ) // callback(unsigned long arg) @change_callback_handle_cast depends on change_timer_function_usage@ identifier change_timer_function_usage._callback; identifier change_timer_function_usage._timer; type _origtype; identifier _origarg; type _handletype; identifier _handle; @@ void _callback( -_origtype _origarg +struct timer_list *t ) { ( ... when != _origarg _handletype *_handle = -(_handletype *)_origarg; +from_timer(_handle, t, _timer); ... when != _origarg | ... when != _origarg _handletype *_handle = -(void *)_origarg; +from_timer(_handle, t, _timer); ... when != _origarg | ... when != _origarg _handletype *_handle; ... when != _handle _handle = -(_handletype *)_origarg; +from_timer(_handle, t, _timer); ... when != _origarg | ... when != _origarg _handletype *_handle; ... when != _handle _handle = -(void *)_origarg; +from_timer(_handle, t, _timer); ... when != _origarg ) } // callback(unsigned long arg) without existing variable @change_callback_handle_cast_no_arg depends on change_timer_function_usage && !change_callback_handle_cast@ identifier change_timer_function_usage._callback; identifier change_timer_function_usage._timer; type _origtype; identifier _origarg; type _handletype; @@ void _callback( -_origtype _origarg +struct timer_list *t ) { + _handletype *_origarg = from_timer(_origarg, t, _timer); + ... when != _origarg - (_handletype *)_origarg + _origarg ... when != _origarg } // Avoid already converted callbacks. @match_callback_converted depends on change_timer_function_usage && !change_callback_handle_cast && !change_callback_handle_cast_no_arg@ identifier change_timer_function_usage._callback; identifier t; @@ void _callback(struct timer_list *t) { ... } // callback(struct something *handle) @change_callback_handle_arg depends on change_timer_function_usage && !match_callback_converted && !change_callback_handle_cast && !change_callback_handle_cast_no_arg@ identifier change_timer_function_usage._callback; identifier change_timer_function_usage._timer; type _handletype; identifier _handle; @@ void _callback( -_handletype *_handle +struct timer_list *t ) { + _handletype *_handle = from_timer(_handle, t, _timer); ... } // If change_callback_handle_arg ran on an empty function, remove // the added handler. @unchange_callback_handle_arg depends on change_timer_function_usage && change_callback_handle_arg@ identifier change_timer_function_usage._callback; identifier change_timer_function_usage._timer; type _handletype; identifier _handle; identifier t; @@ void _callback(struct timer_list *t) { - _handletype *_handle = from_timer(_handle, t, _timer); } // We only want to refactor the setup_timer() data argument if we've found // the matching callback. This undoes changes in change_timer_function_usage. @unchange_timer_function_usage depends on change_timer_function_usage && !change_callback_handle_cast && !change_callback_handle_cast_no_arg && !change_callback_handle_arg@ expression change_timer_function_usage._E; identifier change_timer_function_usage._timer; identifier change_timer_function_usage._callback; type change_timer_function_usage._cast_data; @@ ( -timer_setup(&_E->_timer, _callback, 0); +setup_timer(&_E->_timer, _callback, (_cast_data)_E); | -timer_setup(&_E._timer, _callback, 0); +setup_timer(&_E._timer, _callback, (_cast_data)&_E); ) // If we fixed a callback from a .function assignment, fix the // assignment cast now. @change_timer_function_assignment depends on change_timer_function_usage && (change_callback_handle_cast || change_callback_handle_cast_no_arg || change_callback_handle_arg)@ expression change_timer_function_usage._E; identifier change_timer_function_usage._timer; identifier change_timer_function_usage._callback; type _cast_func; typedef TIMER_FUNC_TYPE; @@ ( _E->_timer.function = -_callback +(TIMER_FUNC_TYPE)_callback ; | _E->_timer.function = -&_callback +(TIMER_FUNC_TYPE)_callback ; | _E->_timer.function = -(_cast_func)_callback; +(TIMER_FUNC_TYPE)_callback ; | _E->_timer.function = -(_cast_func)&_callback +(TIMER_FUNC_TYPE)_callback ; | _E._timer.function = -_callback +(TIMER_FUNC_TYPE)_callback ; | _E._timer.function = -&_callback; +(TIMER_FUNC_TYPE)_callback ; | _E._timer.function = -(_cast_func)_callback +(TIMER_FUNC_TYPE)_callback ; | _E._timer.function = -(_cast_func)&_callback +(TIMER_FUNC_TYPE)_callback ; ) // Sometimes timer functions are called directly. Replace matched args. @change_timer_function_calls depends on change_timer_function_usage && (change_callback_handle_cast || change_callback_handle_cast_no_arg || change_callback_handle_arg)@ expression _E; identifier change_timer_function_usage._timer; identifier change_timer_function_usage._callback; type _cast_data; @@ _callback( ( -(_cast_data)_E +&_E->_timer | -(_cast_data)&_E +&_E._timer | -_E +&_E->_timer ) ) // If a timer has been configured without a data argument, it can be // converted without regard to the callback argument, since it is unused. @match_timer_function_unused_data@ expression _E; identifier _timer; identifier _callback; @@ ( -setup_timer(&_E->_timer, _callback, 0); +timer_setup(&_E->_timer, _callback, 0); | -setup_timer(&_E->_timer, _callback, 0L); +timer_setup(&_E->_timer, _callback, 0); | -setup_timer(&_E->_timer, _callback, 0UL); +timer_setup(&_E->_timer, _callback, 0); | -setup_timer(&_E._timer, _callback, 0); +timer_setup(&_E._timer, _callback, 0); | -setup_timer(&_E._timer, _callback, 0L); +timer_setup(&_E._timer, _callback, 0); | -setup_timer(&_E._timer, _callback, 0UL); +timer_setup(&_E._timer, _callback, 0); | -setup_timer(&_timer, _callback, 0); +timer_setup(&_timer, _callback, 0); | -setup_timer(&_timer, _callback, 0L); +timer_setup(&_timer, _callback, 0); | -setup_timer(&_timer, _callback, 0UL); +timer_setup(&_timer, _callback, 0); | -setup_timer(_timer, _callback, 0); +timer_setup(_timer, _callback, 0); | -setup_timer(_timer, _callback, 0L); +timer_setup(_timer, _callback, 0); | -setup_timer(_timer, _callback, 0UL); +timer_setup(_timer, _callback, 0); ) @change_callback_unused_data depends on match_timer_function_unused_data@ identifier match_timer_function_unused_data._callback; type _origtype; identifier _origarg; @@ void _callback( -_origtype _origarg +struct timer_list *unused ) { ... when != _origarg } Signed-off-by: Kees Cook <keescook@chromium.org>
2017-10-17 05:43:17 +08:00
static void sched_send_work(struct timer_list *t)
{
treewide: setup_timer() -> timer_setup() This converts all remaining cases of the old setup_timer() API into using timer_setup(), where the callback argument is the structure already holding the struct timer_list. These should have no behavioral changes, since they just change which pointer is passed into the callback with the same available pointers after conversion. It handles the following examples, in addition to some other variations. Casting from unsigned long: void my_callback(unsigned long data) { struct something *ptr = (struct something *)data; ... } ... setup_timer(&ptr->my_timer, my_callback, ptr); and forced object casts: void my_callback(struct something *ptr) { ... } ... setup_timer(&ptr->my_timer, my_callback, (unsigned long)ptr); become: void my_callback(struct timer_list *t) { struct something *ptr = from_timer(ptr, t, my_timer); ... } ... timer_setup(&ptr->my_timer, my_callback, 0); Direct function assignments: void my_callback(unsigned long data) { struct something *ptr = (struct something *)data; ... } ... ptr->my_timer.function = my_callback; have a temporary cast added, along with converting the args: void my_callback(struct timer_list *t) { struct something *ptr = from_timer(ptr, t, my_timer); ... } ... ptr->my_timer.function = (TIMER_FUNC_TYPE)my_callback; And finally, callbacks without a data assignment: void my_callback(unsigned long data) { ... } ... setup_timer(&ptr->my_timer, my_callback, 0); have their argument renamed to verify they're unused during conversion: void my_callback(struct timer_list *unused) { ... } ... timer_setup(&ptr->my_timer, my_callback, 0); The conversion is done with the following Coccinelle script: spatch --very-quiet --all-includes --include-headers \ -I ./arch/x86/include -I ./arch/x86/include/generated \ -I ./include -I ./arch/x86/include/uapi \ -I ./arch/x86/include/generated/uapi -I ./include/uapi \ -I ./include/generated/uapi --include ./include/linux/kconfig.h \ --dir . \ --cocci-file ~/src/data/timer_setup.cocci @fix_address_of@ expression e; @@ setup_timer( -&(e) +&e , ...) // Update any raw setup_timer() usages that have a NULL callback, but // would otherwise match change_timer_function_usage, since the latter // will update all function assignments done in the face of a NULL // function initialization in setup_timer(). @change_timer_function_usage_NULL@ expression _E; identifier _timer; type _cast_data; @@ ( -setup_timer(&_E->_timer, NULL, _E); +timer_setup(&_E->_timer, NULL, 0); | -setup_timer(&_E->_timer, NULL, (_cast_data)_E); +timer_setup(&_E->_timer, NULL, 0); | -setup_timer(&_E._timer, NULL, &_E); +timer_setup(&_E._timer, NULL, 0); | -setup_timer(&_E._timer, NULL, (_cast_data)&_E); +timer_setup(&_E._timer, NULL, 0); ) @change_timer_function_usage@ expression _E; identifier _timer; struct timer_list _stl; identifier _callback; type _cast_func, _cast_data; @@ ( -setup_timer(&_E->_timer, _callback, _E); +timer_setup(&_E->_timer, _callback, 0); | -setup_timer(&_E->_timer, &_callback, _E); +timer_setup(&_E->_timer, _callback, 0); | -setup_timer(&_E->_timer, _callback, (_cast_data)_E); +timer_setup(&_E->_timer, _callback, 0); | -setup_timer(&_E->_timer, &_callback, (_cast_data)_E); +timer_setup(&_E->_timer, _callback, 0); | -setup_timer(&_E->_timer, (_cast_func)_callback, _E); +timer_setup(&_E->_timer, _callback, 0); | -setup_timer(&_E->_timer, (_cast_func)&_callback, _E); +timer_setup(&_E->_timer, _callback, 0); | -setup_timer(&_E->_timer, (_cast_func)_callback, (_cast_data)_E); +timer_setup(&_E->_timer, _callback, 0); | -setup_timer(&_E->_timer, (_cast_func)&_callback, (_cast_data)_E); +timer_setup(&_E->_timer, _callback, 0); | -setup_timer(&_E._timer, _callback, (_cast_data)_E); +timer_setup(&_E._timer, _callback, 0); | -setup_timer(&_E._timer, _callback, (_cast_data)&_E); +timer_setup(&_E._timer, _callback, 0); | -setup_timer(&_E._timer, &_callback, (_cast_data)_E); +timer_setup(&_E._timer, _callback, 0); | -setup_timer(&_E._timer, &_callback, (_cast_data)&_E); +timer_setup(&_E._timer, _callback, 0); | -setup_timer(&_E._timer, (_cast_func)_callback, (_cast_data)_E); +timer_setup(&_E._timer, _callback, 0); | -setup_timer(&_E._timer, (_cast_func)_callback, (_cast_data)&_E); +timer_setup(&_E._timer, _callback, 0); | -setup_timer(&_E._timer, (_cast_func)&_callback, (_cast_data)_E); +timer_setup(&_E._timer, _callback, 0); | -setup_timer(&_E._timer, (_cast_func)&_callback, (_cast_data)&_E); +timer_setup(&_E._timer, _callback, 0); | _E->_timer@_stl.function = _callback; | _E->_timer@_stl.function = &_callback; | _E->_timer@_stl.function = (_cast_func)_callback; | _E->_timer@_stl.function = (_cast_func)&_callback; | _E._timer@_stl.function = _callback; | _E._timer@_stl.function = &_callback; | _E._timer@_stl.function = (_cast_func)_callback; | _E._timer@_stl.function = (_cast_func)&_callback; ) // callback(unsigned long arg) @change_callback_handle_cast depends on change_timer_function_usage@ identifier change_timer_function_usage._callback; identifier change_timer_function_usage._timer; type _origtype; identifier _origarg; type _handletype; identifier _handle; @@ void _callback( -_origtype _origarg +struct timer_list *t ) { ( ... when != _origarg _handletype *_handle = -(_handletype *)_origarg; +from_timer(_handle, t, _timer); ... when != _origarg | ... when != _origarg _handletype *_handle = -(void *)_origarg; +from_timer(_handle, t, _timer); ... when != _origarg | ... when != _origarg _handletype *_handle; ... when != _handle _handle = -(_handletype *)_origarg; +from_timer(_handle, t, _timer); ... when != _origarg | ... when != _origarg _handletype *_handle; ... when != _handle _handle = -(void *)_origarg; +from_timer(_handle, t, _timer); ... when != _origarg ) } // callback(unsigned long arg) without existing variable @change_callback_handle_cast_no_arg depends on change_timer_function_usage && !change_callback_handle_cast@ identifier change_timer_function_usage._callback; identifier change_timer_function_usage._timer; type _origtype; identifier _origarg; type _handletype; @@ void _callback( -_origtype _origarg +struct timer_list *t ) { + _handletype *_origarg = from_timer(_origarg, t, _timer); + ... when != _origarg - (_handletype *)_origarg + _origarg ... when != _origarg } // Avoid already converted callbacks. @match_callback_converted depends on change_timer_function_usage && !change_callback_handle_cast && !change_callback_handle_cast_no_arg@ identifier change_timer_function_usage._callback; identifier t; @@ void _callback(struct timer_list *t) { ... } // callback(struct something *handle) @change_callback_handle_arg depends on change_timer_function_usage && !match_callback_converted && !change_callback_handle_cast && !change_callback_handle_cast_no_arg@ identifier change_timer_function_usage._callback; identifier change_timer_function_usage._timer; type _handletype; identifier _handle; @@ void _callback( -_handletype *_handle +struct timer_list *t ) { + _handletype *_handle = from_timer(_handle, t, _timer); ... } // If change_callback_handle_arg ran on an empty function, remove // the added handler. @unchange_callback_handle_arg depends on change_timer_function_usage && change_callback_handle_arg@ identifier change_timer_function_usage._callback; identifier change_timer_function_usage._timer; type _handletype; identifier _handle; identifier t; @@ void _callback(struct timer_list *t) { - _handletype *_handle = from_timer(_handle, t, _timer); } // We only want to refactor the setup_timer() data argument if we've found // the matching callback. This undoes changes in change_timer_function_usage. @unchange_timer_function_usage depends on change_timer_function_usage && !change_callback_handle_cast && !change_callback_handle_cast_no_arg && !change_callback_handle_arg@ expression change_timer_function_usage._E; identifier change_timer_function_usage._timer; identifier change_timer_function_usage._callback; type change_timer_function_usage._cast_data; @@ ( -timer_setup(&_E->_timer, _callback, 0); +setup_timer(&_E->_timer, _callback, (_cast_data)_E); | -timer_setup(&_E._timer, _callback, 0); +setup_timer(&_E._timer, _callback, (_cast_data)&_E); ) // If we fixed a callback from a .function assignment, fix the // assignment cast now. @change_timer_function_assignment depends on change_timer_function_usage && (change_callback_handle_cast || change_callback_handle_cast_no_arg || change_callback_handle_arg)@ expression change_timer_function_usage._E; identifier change_timer_function_usage._timer; identifier change_timer_function_usage._callback; type _cast_func; typedef TIMER_FUNC_TYPE; @@ ( _E->_timer.function = -_callback +(TIMER_FUNC_TYPE)_callback ; | _E->_timer.function = -&_callback +(TIMER_FUNC_TYPE)_callback ; | _E->_timer.function = -(_cast_func)_callback; +(TIMER_FUNC_TYPE)_callback ; | _E->_timer.function = -(_cast_func)&_callback +(TIMER_FUNC_TYPE)_callback ; | _E._timer.function = -_callback +(TIMER_FUNC_TYPE)_callback ; | _E._timer.function = -&_callback; +(TIMER_FUNC_TYPE)_callback ; | _E._timer.function = -(_cast_func)_callback +(TIMER_FUNC_TYPE)_callback ; | _E._timer.function = -(_cast_func)&_callback +(TIMER_FUNC_TYPE)_callback ; ) // Sometimes timer functions are called directly. Replace matched args. @change_timer_function_calls depends on change_timer_function_usage && (change_callback_handle_cast || change_callback_handle_cast_no_arg || change_callback_handle_arg)@ expression _E; identifier change_timer_function_usage._timer; identifier change_timer_function_usage._callback; type _cast_data; @@ _callback( ( -(_cast_data)_E +&_E->_timer | -(_cast_data)&_E +&_E._timer | -_E +&_E->_timer ) ) // If a timer has been configured without a data argument, it can be // converted without regard to the callback argument, since it is unused. @match_timer_function_unused_data@ expression _E; identifier _timer; identifier _callback; @@ ( -setup_timer(&_E->_timer, _callback, 0); +timer_setup(&_E->_timer, _callback, 0); | -setup_timer(&_E->_timer, _callback, 0L); +timer_setup(&_E->_timer, _callback, 0); | -setup_timer(&_E->_timer, _callback, 0UL); +timer_setup(&_E->_timer, _callback, 0); | -setup_timer(&_E._timer, _callback, 0); +timer_setup(&_E._timer, _callback, 0); | -setup_timer(&_E._timer, _callback, 0L); +timer_setup(&_E._timer, _callback, 0); | -setup_timer(&_E._timer, _callback, 0UL); +timer_setup(&_E._timer, _callback, 0); | -setup_timer(&_timer, _callback, 0); +timer_setup(&_timer, _callback, 0); | -setup_timer(&_timer, _callback, 0L); +timer_setup(&_timer, _callback, 0); | -setup_timer(&_timer, _callback, 0UL); +timer_setup(&_timer, _callback, 0); | -setup_timer(_timer, _callback, 0); +timer_setup(_timer, _callback, 0); | -setup_timer(_timer, _callback, 0L); +timer_setup(_timer, _callback, 0); | -setup_timer(_timer, _callback, 0UL); +timer_setup(_timer, _callback, 0); ) @change_callback_unused_data depends on match_timer_function_unused_data@ identifier match_timer_function_unused_data._callback; type _origtype; identifier _origarg; @@ void _callback( -_origtype _origarg +struct timer_list *unused ) { ... when != _origarg } Signed-off-by: Kees Cook <keescook@chromium.org>
2017-10-17 05:43:17 +08:00
struct per_cpu_dm_data *data = from_timer(data, t, send_timer);
drop_monitor: dont sleep in atomic context drop_monitor calls several sleeping functions while in atomic context. BUG: sleeping function called from invalid context at mm/slub.c:943 in_atomic(): 1, irqs_disabled(): 0, pid: 2103, name: kworker/0:2 Pid: 2103, comm: kworker/0:2 Not tainted 3.5.0-rc1+ #55 Call Trace: [<ffffffff810697ca>] __might_sleep+0xca/0xf0 [<ffffffff811345a3>] kmem_cache_alloc_node+0x1b3/0x1c0 [<ffffffff8105578c>] ? queue_delayed_work_on+0x11c/0x130 [<ffffffff815343fb>] __alloc_skb+0x4b/0x230 [<ffffffffa00b0360>] ? reset_per_cpu_data+0x160/0x160 [drop_monitor] [<ffffffffa00b022f>] reset_per_cpu_data+0x2f/0x160 [drop_monitor] [<ffffffffa00b03ab>] send_dm_alert+0x4b/0xb0 [drop_monitor] [<ffffffff810568e0>] process_one_work+0x130/0x4c0 [<ffffffff81058249>] worker_thread+0x159/0x360 [<ffffffff810580f0>] ? manage_workers.isra.27+0x240/0x240 [<ffffffff8105d403>] kthread+0x93/0xa0 [<ffffffff816be6d4>] kernel_thread_helper+0x4/0x10 [<ffffffff8105d370>] ? kthread_freezable_should_stop+0x80/0x80 [<ffffffff816be6d0>] ? gs_change+0xb/0xb Rework the logic to call the sleeping functions in right context. Use standard timer/workqueue api to let system chose any cpu to perform the allocation and netlink send. Also avoid a loop if reset_per_cpu_data() cannot allocate memory : use mod_timer() to wait 1/10 second before next try. Signed-off-by: Eric Dumazet <edumazet@google.com> Cc: Neil Horman <nhorman@tuxdriver.com> Reviewed-by: Neil Horman <nhorman@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-06-04 08:18:19 +08:00
schedule_work(&data->dm_alert_work);
}
static void trace_drop_common(struct sk_buff *skb, void *location)
{
struct net_dm_alert_msg *msg;
struct nlmsghdr *nlh;
struct nlattr *nla;
int i;
struct sk_buff *dskb;
drop_monitor: dont sleep in atomic context drop_monitor calls several sleeping functions while in atomic context. BUG: sleeping function called from invalid context at mm/slub.c:943 in_atomic(): 1, irqs_disabled(): 0, pid: 2103, name: kworker/0:2 Pid: 2103, comm: kworker/0:2 Not tainted 3.5.0-rc1+ #55 Call Trace: [<ffffffff810697ca>] __might_sleep+0xca/0xf0 [<ffffffff811345a3>] kmem_cache_alloc_node+0x1b3/0x1c0 [<ffffffff8105578c>] ? queue_delayed_work_on+0x11c/0x130 [<ffffffff815343fb>] __alloc_skb+0x4b/0x230 [<ffffffffa00b0360>] ? reset_per_cpu_data+0x160/0x160 [drop_monitor] [<ffffffffa00b022f>] reset_per_cpu_data+0x2f/0x160 [drop_monitor] [<ffffffffa00b03ab>] send_dm_alert+0x4b/0xb0 [drop_monitor] [<ffffffff810568e0>] process_one_work+0x130/0x4c0 [<ffffffff81058249>] worker_thread+0x159/0x360 [<ffffffff810580f0>] ? manage_workers.isra.27+0x240/0x240 [<ffffffff8105d403>] kthread+0x93/0xa0 [<ffffffff816be6d4>] kernel_thread_helper+0x4/0x10 [<ffffffff8105d370>] ? kthread_freezable_should_stop+0x80/0x80 [<ffffffff816be6d0>] ? gs_change+0xb/0xb Rework the logic to call the sleeping functions in right context. Use standard timer/workqueue api to let system chose any cpu to perform the allocation and netlink send. Also avoid a loop if reset_per_cpu_data() cannot allocate memory : use mod_timer() to wait 1/10 second before next try. Signed-off-by: Eric Dumazet <edumazet@google.com> Cc: Neil Horman <nhorman@tuxdriver.com> Reviewed-by: Neil Horman <nhorman@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-06-04 08:18:19 +08:00
struct per_cpu_dm_data *data;
unsigned long flags;
drop_monitor: dont sleep in atomic context drop_monitor calls several sleeping functions while in atomic context. BUG: sleeping function called from invalid context at mm/slub.c:943 in_atomic(): 1, irqs_disabled(): 0, pid: 2103, name: kworker/0:2 Pid: 2103, comm: kworker/0:2 Not tainted 3.5.0-rc1+ #55 Call Trace: [<ffffffff810697ca>] __might_sleep+0xca/0xf0 [<ffffffff811345a3>] kmem_cache_alloc_node+0x1b3/0x1c0 [<ffffffff8105578c>] ? queue_delayed_work_on+0x11c/0x130 [<ffffffff815343fb>] __alloc_skb+0x4b/0x230 [<ffffffffa00b0360>] ? reset_per_cpu_data+0x160/0x160 [drop_monitor] [<ffffffffa00b022f>] reset_per_cpu_data+0x2f/0x160 [drop_monitor] [<ffffffffa00b03ab>] send_dm_alert+0x4b/0xb0 [drop_monitor] [<ffffffff810568e0>] process_one_work+0x130/0x4c0 [<ffffffff81058249>] worker_thread+0x159/0x360 [<ffffffff810580f0>] ? manage_workers.isra.27+0x240/0x240 [<ffffffff8105d403>] kthread+0x93/0xa0 [<ffffffff816be6d4>] kernel_thread_helper+0x4/0x10 [<ffffffff8105d370>] ? kthread_freezable_should_stop+0x80/0x80 [<ffffffff816be6d0>] ? gs_change+0xb/0xb Rework the logic to call the sleeping functions in right context. Use standard timer/workqueue api to let system chose any cpu to perform the allocation and netlink send. Also avoid a loop if reset_per_cpu_data() cannot allocate memory : use mod_timer() to wait 1/10 second before next try. Signed-off-by: Eric Dumazet <edumazet@google.com> Cc: Neil Horman <nhorman@tuxdriver.com> Reviewed-by: Neil Horman <nhorman@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-06-04 08:18:19 +08:00
local_irq_save(flags);
data = this_cpu_ptr(&dm_cpu_data);
drop_monitor: dont sleep in atomic context drop_monitor calls several sleeping functions while in atomic context. BUG: sleeping function called from invalid context at mm/slub.c:943 in_atomic(): 1, irqs_disabled(): 0, pid: 2103, name: kworker/0:2 Pid: 2103, comm: kworker/0:2 Not tainted 3.5.0-rc1+ #55 Call Trace: [<ffffffff810697ca>] __might_sleep+0xca/0xf0 [<ffffffff811345a3>] kmem_cache_alloc_node+0x1b3/0x1c0 [<ffffffff8105578c>] ? queue_delayed_work_on+0x11c/0x130 [<ffffffff815343fb>] __alloc_skb+0x4b/0x230 [<ffffffffa00b0360>] ? reset_per_cpu_data+0x160/0x160 [drop_monitor] [<ffffffffa00b022f>] reset_per_cpu_data+0x2f/0x160 [drop_monitor] [<ffffffffa00b03ab>] send_dm_alert+0x4b/0xb0 [drop_monitor] [<ffffffff810568e0>] process_one_work+0x130/0x4c0 [<ffffffff81058249>] worker_thread+0x159/0x360 [<ffffffff810580f0>] ? manage_workers.isra.27+0x240/0x240 [<ffffffff8105d403>] kthread+0x93/0xa0 [<ffffffff816be6d4>] kernel_thread_helper+0x4/0x10 [<ffffffff8105d370>] ? kthread_freezable_should_stop+0x80/0x80 [<ffffffff816be6d0>] ? gs_change+0xb/0xb Rework the logic to call the sleeping functions in right context. Use standard timer/workqueue api to let system chose any cpu to perform the allocation and netlink send. Also avoid a loop if reset_per_cpu_data() cannot allocate memory : use mod_timer() to wait 1/10 second before next try. Signed-off-by: Eric Dumazet <edumazet@google.com> Cc: Neil Horman <nhorman@tuxdriver.com> Reviewed-by: Neil Horman <nhorman@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-06-04 08:18:19 +08:00
spin_lock(&data->lock);
dskb = data->skb;
if (!dskb)
goto out;
nlh = (struct nlmsghdr *)dskb->data;
nla = genlmsg_data(nlmsg_data(nlh));
msg = nla_data(nla);
for (i = 0; i < msg->entries; i++) {
if (!memcmp(&location, msg->points[i].pc, sizeof(void *))) {
msg->points[i].count++;
goto out;
}
}
drop_monitor: dont sleep in atomic context drop_monitor calls several sleeping functions while in atomic context. BUG: sleeping function called from invalid context at mm/slub.c:943 in_atomic(): 1, irqs_disabled(): 0, pid: 2103, name: kworker/0:2 Pid: 2103, comm: kworker/0:2 Not tainted 3.5.0-rc1+ #55 Call Trace: [<ffffffff810697ca>] __might_sleep+0xca/0xf0 [<ffffffff811345a3>] kmem_cache_alloc_node+0x1b3/0x1c0 [<ffffffff8105578c>] ? queue_delayed_work_on+0x11c/0x130 [<ffffffff815343fb>] __alloc_skb+0x4b/0x230 [<ffffffffa00b0360>] ? reset_per_cpu_data+0x160/0x160 [drop_monitor] [<ffffffffa00b022f>] reset_per_cpu_data+0x2f/0x160 [drop_monitor] [<ffffffffa00b03ab>] send_dm_alert+0x4b/0xb0 [drop_monitor] [<ffffffff810568e0>] process_one_work+0x130/0x4c0 [<ffffffff81058249>] worker_thread+0x159/0x360 [<ffffffff810580f0>] ? manage_workers.isra.27+0x240/0x240 [<ffffffff8105d403>] kthread+0x93/0xa0 [<ffffffff816be6d4>] kernel_thread_helper+0x4/0x10 [<ffffffff8105d370>] ? kthread_freezable_should_stop+0x80/0x80 [<ffffffff816be6d0>] ? gs_change+0xb/0xb Rework the logic to call the sleeping functions in right context. Use standard timer/workqueue api to let system chose any cpu to perform the allocation and netlink send. Also avoid a loop if reset_per_cpu_data() cannot allocate memory : use mod_timer() to wait 1/10 second before next try. Signed-off-by: Eric Dumazet <edumazet@google.com> Cc: Neil Horman <nhorman@tuxdriver.com> Reviewed-by: Neil Horman <nhorman@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-06-04 08:18:19 +08:00
if (msg->entries == dm_hit_limit)
goto out;
/*
* We need to create a new entry
*/
__nla_reserve_nohdr(dskb, sizeof(struct net_dm_drop_point));
nla->nla_len += NLA_ALIGN(sizeof(struct net_dm_drop_point));
memcpy(msg->points[msg->entries].pc, &location, sizeof(void *));
msg->points[msg->entries].count = 1;
msg->entries++;
if (!timer_pending(&data->send_timer)) {
data->send_timer.expires = jiffies + dm_delay * HZ;
drop_monitor: dont sleep in atomic context drop_monitor calls several sleeping functions while in atomic context. BUG: sleeping function called from invalid context at mm/slub.c:943 in_atomic(): 1, irqs_disabled(): 0, pid: 2103, name: kworker/0:2 Pid: 2103, comm: kworker/0:2 Not tainted 3.5.0-rc1+ #55 Call Trace: [<ffffffff810697ca>] __might_sleep+0xca/0xf0 [<ffffffff811345a3>] kmem_cache_alloc_node+0x1b3/0x1c0 [<ffffffff8105578c>] ? queue_delayed_work_on+0x11c/0x130 [<ffffffff815343fb>] __alloc_skb+0x4b/0x230 [<ffffffffa00b0360>] ? reset_per_cpu_data+0x160/0x160 [drop_monitor] [<ffffffffa00b022f>] reset_per_cpu_data+0x2f/0x160 [drop_monitor] [<ffffffffa00b03ab>] send_dm_alert+0x4b/0xb0 [drop_monitor] [<ffffffff810568e0>] process_one_work+0x130/0x4c0 [<ffffffff81058249>] worker_thread+0x159/0x360 [<ffffffff810580f0>] ? manage_workers.isra.27+0x240/0x240 [<ffffffff8105d403>] kthread+0x93/0xa0 [<ffffffff816be6d4>] kernel_thread_helper+0x4/0x10 [<ffffffff8105d370>] ? kthread_freezable_should_stop+0x80/0x80 [<ffffffff816be6d0>] ? gs_change+0xb/0xb Rework the logic to call the sleeping functions in right context. Use standard timer/workqueue api to let system chose any cpu to perform the allocation and netlink send. Also avoid a loop if reset_per_cpu_data() cannot allocate memory : use mod_timer() to wait 1/10 second before next try. Signed-off-by: Eric Dumazet <edumazet@google.com> Cc: Neil Horman <nhorman@tuxdriver.com> Reviewed-by: Neil Horman <nhorman@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-06-04 08:18:19 +08:00
add_timer(&data->send_timer);
}
out:
drop_monitor: dont sleep in atomic context drop_monitor calls several sleeping functions while in atomic context. BUG: sleeping function called from invalid context at mm/slub.c:943 in_atomic(): 1, irqs_disabled(): 0, pid: 2103, name: kworker/0:2 Pid: 2103, comm: kworker/0:2 Not tainted 3.5.0-rc1+ #55 Call Trace: [<ffffffff810697ca>] __might_sleep+0xca/0xf0 [<ffffffff811345a3>] kmem_cache_alloc_node+0x1b3/0x1c0 [<ffffffff8105578c>] ? queue_delayed_work_on+0x11c/0x130 [<ffffffff815343fb>] __alloc_skb+0x4b/0x230 [<ffffffffa00b0360>] ? reset_per_cpu_data+0x160/0x160 [drop_monitor] [<ffffffffa00b022f>] reset_per_cpu_data+0x2f/0x160 [drop_monitor] [<ffffffffa00b03ab>] send_dm_alert+0x4b/0xb0 [drop_monitor] [<ffffffff810568e0>] process_one_work+0x130/0x4c0 [<ffffffff81058249>] worker_thread+0x159/0x360 [<ffffffff810580f0>] ? manage_workers.isra.27+0x240/0x240 [<ffffffff8105d403>] kthread+0x93/0xa0 [<ffffffff816be6d4>] kernel_thread_helper+0x4/0x10 [<ffffffff8105d370>] ? kthread_freezable_should_stop+0x80/0x80 [<ffffffff816be6d0>] ? gs_change+0xb/0xb Rework the logic to call the sleeping functions in right context. Use standard timer/workqueue api to let system chose any cpu to perform the allocation and netlink send. Also avoid a loop if reset_per_cpu_data() cannot allocate memory : use mod_timer() to wait 1/10 second before next try. Signed-off-by: Eric Dumazet <edumazet@google.com> Cc: Neil Horman <nhorman@tuxdriver.com> Reviewed-by: Neil Horman <nhorman@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-06-04 08:18:19 +08:00
spin_unlock_irqrestore(&data->lock, flags);
}
tracing: Let tracepoints have data passed to tracepoint callbacks This patch adds data to be passed to tracepoint callbacks. The created functions from DECLARE_TRACE() now need a mandatory data parameter. For example: DECLARE_TRACE(mytracepoint, int value, value) Will create the register function: int register_trace_mytracepoint((void(*)(void *data, int value))probe, void *data); As the first argument, all callbacks (probes) must take a (void *data) parameter. So a callback for the above tracepoint will look like: void myprobe(void *data, int value) { } The callback may choose to ignore the data parameter. This change allows callbacks to register a private data pointer along with the function probe. void mycallback(void *data, int value); register_trace_mytracepoint(mycallback, mydata); Then the mycallback() will receive the "mydata" as the first parameter before the args. A more detailed example: DECLARE_TRACE(mytracepoint, TP_PROTO(int status), TP_ARGS(status)); /* In the C file */ DEFINE_TRACE(mytracepoint, TP_PROTO(int status), TP_ARGS(status)); [...] trace_mytracepoint(status); /* In a file registering this tracepoint */ int my_callback(void *data, int status) { struct my_struct my_data = data; [...] } [...] my_data = kmalloc(sizeof(*my_data), GFP_KERNEL); init_my_data(my_data); register_trace_mytracepoint(my_callback, my_data); The same callback can also be registered to the same tracepoint as long as the data registered is different. Note, the data must also be used to unregister the callback: unregister_trace_mytracepoint(my_callback, my_data); Because of the data parameter, tracepoints declared this way can not have no args. That is: DECLARE_TRACE(mytracepoint, TP_PROTO(void), TP_ARGS()); will cause an error. If no arguments are needed, a new macro can be used instead: DECLARE_TRACE_NOARGS(mytracepoint); Since there are no arguments, the proto and args fields are left out. This is part of a series to make the tracepoint footprint smaller: text data bss dec hex filename 4913961 1088356 861512 6863829 68bbd5 vmlinux.orig 4914025 1088868 861512 6864405 68be15 vmlinux.class 4918492 1084612 861512 6864616 68bee8 vmlinux.tracepoint Again, this patch also increases the size of the kernel, but lays the ground work for decreasing it. v5: Fixed net/core/drop_monitor.c to handle these updates. v4: Moved the DECLARE_TRACE() DECLARE_TRACE_NOARGS out of the #ifdef CONFIG_TRACE_POINTS, since the two are the same in both cases. The __DECLARE_TRACE() is what changes. Thanks to Frederic Weisbecker for pointing this out. v3: Made all register_* functions require data to be passed and all callbacks to take a void * parameter as its first argument. This makes the calling functions comply with C standards. Also added more comments to the modifications of DECLARE_TRACE(). v2: Made the DECLARE_TRACE() have the ability to pass arguments and added a new DECLARE_TRACE_NOARGS() for tracepoints that do not need any arguments. Acked-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com> Acked-by: Masami Hiramatsu <mhiramat@redhat.com> Acked-by: Frederic Weisbecker <fweisbec@gmail.com> Cc: Neil Horman <nhorman@tuxdriver.com> Cc: David S. Miller <davem@davemloft.net> Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
2010-04-21 05:04:50 +08:00
static void trace_kfree_skb_hit(void *ignore, struct sk_buff *skb, void *location)
{
trace_drop_common(skb, location);
}
static void trace_napi_poll_hit(void *ignore, struct napi_struct *napi,
int work, int budget)
{
struct dm_hw_stat_delta *new_stat;
/*
* Don't check napi structures with no associated device
*/
if (!napi->dev)
return;
rcu_read_lock();
list_for_each_entry_rcu(new_stat, &hw_stats_list, list) {
/*
* only add a note to our monitor buffer if:
* 1) this is the dev we received on
* 2) its after the last_rx delta
* 3) our rx_dropped count has gone up
*/
if ((new_stat->dev == napi->dev) &&
(time_after(jiffies, new_stat->last_rx + dm_hw_check_delta)) &&
(napi->dev->stats.rx_dropped != new_stat->last_drop_val)) {
trace_drop_common(NULL, NULL);
new_stat->last_drop_val = napi->dev->stats.rx_dropped;
new_stat->last_rx = jiffies;
break;
}
}
rcu_read_unlock();
}
static struct net_dm_hw_entries *
net_dm_hw_reset_per_cpu_data(struct per_cpu_dm_data *hw_data)
{
struct net_dm_hw_entries *hw_entries;
unsigned long flags;
hw_entries = kzalloc(struct_size(hw_entries, entries, dm_hit_limit),
GFP_KERNEL);
if (!hw_entries) {
/* If the memory allocation failed, we try to perform another
* allocation in 1/10 second. Otherwise, the probe function
* will constantly bail out.
*/
mod_timer(&hw_data->send_timer, jiffies + HZ / 10);
}
spin_lock_irqsave(&hw_data->lock, flags);
swap(hw_data->hw_entries, hw_entries);
spin_unlock_irqrestore(&hw_data->lock, flags);
return hw_entries;
}
static int net_dm_hw_entry_put(struct sk_buff *msg,
const struct net_dm_hw_entry *hw_entry)
{
struct nlattr *attr;
attr = nla_nest_start(msg, NET_DM_ATTR_HW_ENTRY);
if (!attr)
return -EMSGSIZE;
if (nla_put_string(msg, NET_DM_ATTR_HW_TRAP_NAME, hw_entry->trap_name))
goto nla_put_failure;
if (nla_put_u32(msg, NET_DM_ATTR_HW_TRAP_COUNT, hw_entry->count))
goto nla_put_failure;
nla_nest_end(msg, attr);
return 0;
nla_put_failure:
nla_nest_cancel(msg, attr);
return -EMSGSIZE;
}
static int net_dm_hw_entries_put(struct sk_buff *msg,
const struct net_dm_hw_entries *hw_entries)
{
struct nlattr *attr;
int i;
attr = nla_nest_start(msg, NET_DM_ATTR_HW_ENTRIES);
if (!attr)
return -EMSGSIZE;
for (i = 0; i < hw_entries->num_entries; i++) {
int rc;
rc = net_dm_hw_entry_put(msg, &hw_entries->entries[i]);
if (rc)
goto nla_put_failure;
}
nla_nest_end(msg, attr);
return 0;
nla_put_failure:
nla_nest_cancel(msg, attr);
return -EMSGSIZE;
}
static int
net_dm_hw_summary_report_fill(struct sk_buff *msg,
const struct net_dm_hw_entries *hw_entries)
{
struct net_dm_alert_msg anc_hdr = { 0 };
void *hdr;
int rc;
hdr = genlmsg_put(msg, 0, 0, &net_drop_monitor_family, 0,
NET_DM_CMD_ALERT);
if (!hdr)
return -EMSGSIZE;
/* We need to put the ancillary header in order not to break user
* space.
*/
if (nla_put(msg, NLA_UNSPEC, sizeof(anc_hdr), &anc_hdr))
goto nla_put_failure;
rc = net_dm_hw_entries_put(msg, hw_entries);
if (rc)
goto nla_put_failure;
genlmsg_end(msg, hdr);
return 0;
nla_put_failure:
genlmsg_cancel(msg, hdr);
return -EMSGSIZE;
}
static void net_dm_hw_summary_work(struct work_struct *work)
{
struct net_dm_hw_entries *hw_entries;
struct per_cpu_dm_data *hw_data;
struct sk_buff *msg;
int rc;
hw_data = container_of(work, struct per_cpu_dm_data, dm_alert_work);
hw_entries = net_dm_hw_reset_per_cpu_data(hw_data);
if (!hw_entries)
return;
msg = nlmsg_new(NLMSG_DEFAULT_SIZE, GFP_KERNEL);
if (!msg)
goto out;
rc = net_dm_hw_summary_report_fill(msg, hw_entries);
if (rc) {
nlmsg_free(msg);
goto out;
}
genlmsg_multicast(&net_drop_monitor_family, msg, 0, 0, GFP_KERNEL);
out:
kfree(hw_entries);
}
static void
net_dm_hw_summary_probe(struct sk_buff *skb,
const struct net_dm_hw_metadata *hw_metadata)
{
struct net_dm_hw_entries *hw_entries;
struct net_dm_hw_entry *hw_entry;
struct per_cpu_dm_data *hw_data;
unsigned long flags;
int i;
hw_data = this_cpu_ptr(&dm_hw_cpu_data);
spin_lock_irqsave(&hw_data->lock, flags);
hw_entries = hw_data->hw_entries;
if (!hw_entries)
goto out;
for (i = 0; i < hw_entries->num_entries; i++) {
hw_entry = &hw_entries->entries[i];
if (!strncmp(hw_entry->trap_name, hw_metadata->trap_name,
NET_DM_MAX_HW_TRAP_NAME_LEN - 1)) {
hw_entry->count++;
goto out;
}
}
if (WARN_ON_ONCE(hw_entries->num_entries == dm_hit_limit))
goto out;
hw_entry = &hw_entries->entries[hw_entries->num_entries];
strlcpy(hw_entry->trap_name, hw_metadata->trap_name,
NET_DM_MAX_HW_TRAP_NAME_LEN - 1);
hw_entry->count = 1;
hw_entries->num_entries++;
if (!timer_pending(&hw_data->send_timer)) {
hw_data->send_timer.expires = jiffies + dm_delay * HZ;
add_timer(&hw_data->send_timer);
}
out:
spin_unlock_irqrestore(&hw_data->lock, flags);
}
static const struct net_dm_alert_ops net_dm_alert_summary_ops = {
.kfree_skb_probe = trace_kfree_skb_hit,
.napi_poll_probe = trace_napi_poll_hit,
.work_item_func = send_dm_alert,
.hw_work_item_func = net_dm_hw_summary_work,
.hw_probe = net_dm_hw_summary_probe,
};
static void net_dm_packet_trace_kfree_skb_hit(void *ignore,
struct sk_buff *skb,
void *location)
{
ktime_t tstamp = ktime_get_real();
struct per_cpu_dm_data *data;
struct sk_buff *nskb;
unsigned long flags;
if (!skb_mac_header_was_set(skb))
return;
nskb = skb_clone(skb, GFP_ATOMIC);
if (!nskb)
return;
NET_DM_SKB_CB(nskb)->pc = location;
/* Override the timestamp because we care about the time when the
* packet was dropped.
*/
nskb->tstamp = tstamp;
data = this_cpu_ptr(&dm_cpu_data);
spin_lock_irqsave(&data->drop_queue.lock, flags);
if (skb_queue_len(&data->drop_queue) < net_dm_queue_len)
__skb_queue_tail(&data->drop_queue, nskb);
else
goto unlock_free;
spin_unlock_irqrestore(&data->drop_queue.lock, flags);
schedule_work(&data->dm_alert_work);
return;
unlock_free:
spin_unlock_irqrestore(&data->drop_queue.lock, flags);
u64_stats_update_begin(&data->stats.syncp);
data->stats.dropped++;
u64_stats_update_end(&data->stats.syncp);
consume_skb(nskb);
}
static void net_dm_packet_trace_napi_poll_hit(void *ignore,
struct napi_struct *napi,
int work, int budget)
{
}
static size_t net_dm_in_port_size(void)
{
/* NET_DM_ATTR_IN_PORT nest */
return nla_total_size(0) +
/* NET_DM_ATTR_PORT_NETDEV_IFINDEX */
nla_total_size(sizeof(u32)) +
/* NET_DM_ATTR_PORT_NETDEV_NAME */
nla_total_size(IFNAMSIZ + 1);
}
#define NET_DM_MAX_SYMBOL_LEN 40
static size_t net_dm_packet_report_size(size_t payload_len)
{
size_t size;
size = nlmsg_msg_size(GENL_HDRLEN + net_drop_monitor_family.hdrsize);
return NLMSG_ALIGN(size) +
/* NET_DM_ATTR_ORIGIN */
nla_total_size(sizeof(u16)) +
/* NET_DM_ATTR_PC */
nla_total_size(sizeof(u64)) +
/* NET_DM_ATTR_SYMBOL */
nla_total_size(NET_DM_MAX_SYMBOL_LEN + 1) +
/* NET_DM_ATTR_IN_PORT */
net_dm_in_port_size() +
/* NET_DM_ATTR_TIMESTAMP */
nla_total_size(sizeof(u64)) +
/* NET_DM_ATTR_ORIG_LEN */
nla_total_size(sizeof(u32)) +
/* NET_DM_ATTR_PROTO */
nla_total_size(sizeof(u16)) +
/* NET_DM_ATTR_PAYLOAD */
nla_total_size(payload_len);
}
static int net_dm_packet_report_in_port_put(struct sk_buff *msg, int ifindex,
const char *name)
{
struct nlattr *attr;
attr = nla_nest_start(msg, NET_DM_ATTR_IN_PORT);
if (!attr)
return -EMSGSIZE;
if (ifindex &&
nla_put_u32(msg, NET_DM_ATTR_PORT_NETDEV_IFINDEX, ifindex))
goto nla_put_failure;
if (name && nla_put_string(msg, NET_DM_ATTR_PORT_NETDEV_NAME, name))
goto nla_put_failure;
nla_nest_end(msg, attr);
return 0;
nla_put_failure:
nla_nest_cancel(msg, attr);
return -EMSGSIZE;
}
static int net_dm_packet_report_fill(struct sk_buff *msg, struct sk_buff *skb,
size_t payload_len)
{
u64 pc = (u64)(uintptr_t) NET_DM_SKB_CB(skb)->pc;
char buf[NET_DM_MAX_SYMBOL_LEN];
struct nlattr *attr;
void *hdr;
int rc;
hdr = genlmsg_put(msg, 0, 0, &net_drop_monitor_family, 0,
NET_DM_CMD_PACKET_ALERT);
if (!hdr)
return -EMSGSIZE;
if (nla_put_u16(msg, NET_DM_ATTR_ORIGIN, NET_DM_ORIGIN_SW))
goto nla_put_failure;
if (nla_put_u64_64bit(msg, NET_DM_ATTR_PC, pc, NET_DM_ATTR_PAD))
goto nla_put_failure;
snprintf(buf, sizeof(buf), "%pS", NET_DM_SKB_CB(skb)->pc);
if (nla_put_string(msg, NET_DM_ATTR_SYMBOL, buf))
goto nla_put_failure;
rc = net_dm_packet_report_in_port_put(msg, skb->skb_iif, NULL);
if (rc)
goto nla_put_failure;
if (nla_put_u64_64bit(msg, NET_DM_ATTR_TIMESTAMP,
ktime_to_ns(skb->tstamp), NET_DM_ATTR_PAD))
goto nla_put_failure;
if (nla_put_u32(msg, NET_DM_ATTR_ORIG_LEN, skb->len))
goto nla_put_failure;
if (!payload_len)
goto out;
if (nla_put_u16(msg, NET_DM_ATTR_PROTO, be16_to_cpu(skb->protocol)))
goto nla_put_failure;
attr = skb_put(msg, nla_total_size(payload_len));
attr->nla_type = NET_DM_ATTR_PAYLOAD;
attr->nla_len = nla_attr_size(payload_len);
if (skb_copy_bits(skb, 0, nla_data(attr), payload_len))
goto nla_put_failure;
out:
genlmsg_end(msg, hdr);
return 0;
nla_put_failure:
genlmsg_cancel(msg, hdr);
return -EMSGSIZE;
}
#define NET_DM_MAX_PACKET_SIZE (0xffff - NLA_HDRLEN - NLA_ALIGNTO)
static void net_dm_packet_report(struct sk_buff *skb)
{
struct sk_buff *msg;
size_t payload_len;
int rc;
/* Make sure we start copying the packet from the MAC header */
if (skb->data > skb_mac_header(skb))
skb_push(skb, skb->data - skb_mac_header(skb));
else
skb_pull(skb, skb_mac_header(skb) - skb->data);
/* Ensure packet fits inside a single netlink attribute */
payload_len = min_t(size_t, skb->len, NET_DM_MAX_PACKET_SIZE);
if (net_dm_trunc_len)
payload_len = min_t(size_t, net_dm_trunc_len, payload_len);
msg = nlmsg_new(net_dm_packet_report_size(payload_len), GFP_KERNEL);
if (!msg)
goto out;
rc = net_dm_packet_report_fill(msg, skb, payload_len);
if (rc) {
nlmsg_free(msg);
goto out;
}
genlmsg_multicast(&net_drop_monitor_family, msg, 0, 0, GFP_KERNEL);
out:
consume_skb(skb);
}
static void net_dm_packet_work(struct work_struct *work)
{
struct per_cpu_dm_data *data;
struct sk_buff_head list;
struct sk_buff *skb;
unsigned long flags;
data = container_of(work, struct per_cpu_dm_data, dm_alert_work);
__skb_queue_head_init(&list);
spin_lock_irqsave(&data->drop_queue.lock, flags);
skb_queue_splice_tail_init(&data->drop_queue, &list);
spin_unlock_irqrestore(&data->drop_queue.lock, flags);
while ((skb = __skb_dequeue(&list)))
net_dm_packet_report(skb);
}
static size_t
net_dm_hw_packet_report_size(size_t payload_len,
const struct net_dm_hw_metadata *hw_metadata)
{
size_t size;
size = nlmsg_msg_size(GENL_HDRLEN + net_drop_monitor_family.hdrsize);
return NLMSG_ALIGN(size) +
/* NET_DM_ATTR_ORIGIN */
nla_total_size(sizeof(u16)) +
/* NET_DM_ATTR_HW_TRAP_GROUP_NAME */
nla_total_size(strlen(hw_metadata->trap_group_name) + 1) +
/* NET_DM_ATTR_HW_TRAP_NAME */
nla_total_size(strlen(hw_metadata->trap_name) + 1) +
/* NET_DM_ATTR_IN_PORT */
net_dm_in_port_size() +
/* NET_DM_ATTR_TIMESTAMP */
nla_total_size(sizeof(u64)) +
/* NET_DM_ATTR_ORIG_LEN */
nla_total_size(sizeof(u32)) +
/* NET_DM_ATTR_PROTO */
nla_total_size(sizeof(u16)) +
/* NET_DM_ATTR_PAYLOAD */
nla_total_size(payload_len);
}
static int net_dm_hw_packet_report_fill(struct sk_buff *msg,
struct sk_buff *skb, size_t payload_len)
{
struct net_dm_hw_metadata *hw_metadata;
struct nlattr *attr;
void *hdr;
hw_metadata = NET_DM_SKB_CB(skb)->hw_metadata;
hdr = genlmsg_put(msg, 0, 0, &net_drop_monitor_family, 0,
NET_DM_CMD_PACKET_ALERT);
if (!hdr)
return -EMSGSIZE;
if (nla_put_u16(msg, NET_DM_ATTR_ORIGIN, NET_DM_ORIGIN_HW))
goto nla_put_failure;
if (nla_put_string(msg, NET_DM_ATTR_HW_TRAP_GROUP_NAME,
hw_metadata->trap_group_name))
goto nla_put_failure;
if (nla_put_string(msg, NET_DM_ATTR_HW_TRAP_NAME,
hw_metadata->trap_name))
goto nla_put_failure;
if (hw_metadata->input_dev) {
struct net_device *dev = hw_metadata->input_dev;
int rc;
rc = net_dm_packet_report_in_port_put(msg, dev->ifindex,
dev->name);
if (rc)
goto nla_put_failure;
}
if (nla_put_u64_64bit(msg, NET_DM_ATTR_TIMESTAMP,
ktime_to_ns(skb->tstamp), NET_DM_ATTR_PAD))
goto nla_put_failure;
if (nla_put_u32(msg, NET_DM_ATTR_ORIG_LEN, skb->len))
goto nla_put_failure;
if (!payload_len)
goto out;
if (nla_put_u16(msg, NET_DM_ATTR_PROTO, be16_to_cpu(skb->protocol)))
goto nla_put_failure;
attr = skb_put(msg, nla_total_size(payload_len));
attr->nla_type = NET_DM_ATTR_PAYLOAD;
attr->nla_len = nla_attr_size(payload_len);
if (skb_copy_bits(skb, 0, nla_data(attr), payload_len))
goto nla_put_failure;
out:
genlmsg_end(msg, hdr);
return 0;
nla_put_failure:
genlmsg_cancel(msg, hdr);
return -EMSGSIZE;
}
static struct net_dm_hw_metadata *
net_dm_hw_metadata_clone(const struct net_dm_hw_metadata *hw_metadata)
{
struct net_dm_hw_metadata *n_hw_metadata;
const char *trap_group_name;
const char *trap_name;
n_hw_metadata = kmalloc(sizeof(*hw_metadata), GFP_ATOMIC);
if (!n_hw_metadata)
return NULL;
trap_group_name = kmemdup(hw_metadata->trap_group_name,
strlen(hw_metadata->trap_group_name) + 1,
GFP_ATOMIC | __GFP_ZERO);
if (!trap_group_name)
goto free_hw_metadata;
n_hw_metadata->trap_group_name = trap_group_name;
trap_name = kmemdup(hw_metadata->trap_name,
strlen(hw_metadata->trap_name) + 1,
GFP_ATOMIC | __GFP_ZERO);
if (!trap_name)
goto free_trap_group;
n_hw_metadata->trap_name = trap_name;
n_hw_metadata->input_dev = hw_metadata->input_dev;
if (n_hw_metadata->input_dev)
dev_hold(n_hw_metadata->input_dev);
return n_hw_metadata;
free_trap_group:
kfree(trap_group_name);
free_hw_metadata:
kfree(n_hw_metadata);
return NULL;
}
static void
net_dm_hw_metadata_free(const struct net_dm_hw_metadata *hw_metadata)
{
if (hw_metadata->input_dev)
dev_put(hw_metadata->input_dev);
kfree(hw_metadata->trap_name);
kfree(hw_metadata->trap_group_name);
kfree(hw_metadata);
}
static void net_dm_hw_packet_report(struct sk_buff *skb)
{
struct net_dm_hw_metadata *hw_metadata;
struct sk_buff *msg;
size_t payload_len;
int rc;
if (skb->data > skb_mac_header(skb))
skb_push(skb, skb->data - skb_mac_header(skb));
else
skb_pull(skb, skb_mac_header(skb) - skb->data);
payload_len = min_t(size_t, skb->len, NET_DM_MAX_PACKET_SIZE);
if (net_dm_trunc_len)
payload_len = min_t(size_t, net_dm_trunc_len, payload_len);
hw_metadata = NET_DM_SKB_CB(skb)->hw_metadata;
msg = nlmsg_new(net_dm_hw_packet_report_size(payload_len, hw_metadata),
GFP_KERNEL);
if (!msg)
goto out;
rc = net_dm_hw_packet_report_fill(msg, skb, payload_len);
if (rc) {
nlmsg_free(msg);
goto out;
}
genlmsg_multicast(&net_drop_monitor_family, msg, 0, 0, GFP_KERNEL);
out:
net_dm_hw_metadata_free(NET_DM_SKB_CB(skb)->hw_metadata);
consume_skb(skb);
}
static void net_dm_hw_packet_work(struct work_struct *work)
{
struct per_cpu_dm_data *hw_data;
struct sk_buff_head list;
struct sk_buff *skb;
unsigned long flags;
hw_data = container_of(work, struct per_cpu_dm_data, dm_alert_work);
__skb_queue_head_init(&list);
spin_lock_irqsave(&hw_data->drop_queue.lock, flags);
skb_queue_splice_tail_init(&hw_data->drop_queue, &list);
spin_unlock_irqrestore(&hw_data->drop_queue.lock, flags);
while ((skb = __skb_dequeue(&list)))
net_dm_hw_packet_report(skb);
}
static void
net_dm_hw_packet_probe(struct sk_buff *skb,
const struct net_dm_hw_metadata *hw_metadata)
{
struct net_dm_hw_metadata *n_hw_metadata;
ktime_t tstamp = ktime_get_real();
struct per_cpu_dm_data *hw_data;
struct sk_buff *nskb;
unsigned long flags;
if (!skb_mac_header_was_set(skb))
return;
nskb = skb_clone(skb, GFP_ATOMIC);
if (!nskb)
return;
n_hw_metadata = net_dm_hw_metadata_clone(hw_metadata);
if (!n_hw_metadata)
goto free;
NET_DM_SKB_CB(nskb)->hw_metadata = n_hw_metadata;
nskb->tstamp = tstamp;
hw_data = this_cpu_ptr(&dm_hw_cpu_data);
spin_lock_irqsave(&hw_data->drop_queue.lock, flags);
if (skb_queue_len(&hw_data->drop_queue) < net_dm_queue_len)
__skb_queue_tail(&hw_data->drop_queue, nskb);
else
goto unlock_free;
spin_unlock_irqrestore(&hw_data->drop_queue.lock, flags);
schedule_work(&hw_data->dm_alert_work);
return;
unlock_free:
spin_unlock_irqrestore(&hw_data->drop_queue.lock, flags);
u64_stats_update_begin(&hw_data->stats.syncp);
hw_data->stats.dropped++;
u64_stats_update_end(&hw_data->stats.syncp);
net_dm_hw_metadata_free(n_hw_metadata);
free:
consume_skb(nskb);
}
static const struct net_dm_alert_ops net_dm_alert_packet_ops = {
.kfree_skb_probe = net_dm_packet_trace_kfree_skb_hit,
.napi_poll_probe = net_dm_packet_trace_napi_poll_hit,
.work_item_func = net_dm_packet_work,
.hw_work_item_func = net_dm_hw_packet_work,
.hw_probe = net_dm_hw_packet_probe,
};
static const struct net_dm_alert_ops *net_dm_alert_ops_arr[] = {
[NET_DM_ALERT_MODE_SUMMARY] = &net_dm_alert_summary_ops,
[NET_DM_ALERT_MODE_PACKET] = &net_dm_alert_packet_ops,
};
void net_dm_hw_report(struct sk_buff *skb,
const struct net_dm_hw_metadata *hw_metadata)
{
rcu_read_lock();
if (!monitor_hw)
goto out;
net_dm_alert_ops_arr[net_dm_alert_mode]->hw_probe(skb, hw_metadata);
out:
rcu_read_unlock();
}
EXPORT_SYMBOL_GPL(net_dm_hw_report);
static int net_dm_hw_monitor_start(struct netlink_ext_ack *extack)
{
const struct net_dm_alert_ops *ops;
int cpu;
if (monitor_hw) {
NL_SET_ERR_MSG_MOD(extack, "Hardware monitoring already enabled");
return -EAGAIN;
}
ops = net_dm_alert_ops_arr[net_dm_alert_mode];
if (!try_module_get(THIS_MODULE)) {
NL_SET_ERR_MSG_MOD(extack, "Failed to take reference on module");
return -ENODEV;
}
for_each_possible_cpu(cpu) {
struct per_cpu_dm_data *hw_data = &per_cpu(dm_hw_cpu_data, cpu);
struct net_dm_hw_entries *hw_entries;
INIT_WORK(&hw_data->dm_alert_work, ops->hw_work_item_func);
timer_setup(&hw_data->send_timer, sched_send_work, 0);
hw_entries = net_dm_hw_reset_per_cpu_data(hw_data);
kfree(hw_entries);
}
monitor_hw = true;
return 0;
}
static void net_dm_hw_monitor_stop(struct netlink_ext_ack *extack)
{
int cpu;
if (!monitor_hw)
NL_SET_ERR_MSG_MOD(extack, "Hardware monitoring already disabled");
monitor_hw = false;
/* After this call returns we are guaranteed that no CPU is processing
* any hardware drops.
*/
synchronize_rcu();
for_each_possible_cpu(cpu) {
struct per_cpu_dm_data *hw_data = &per_cpu(dm_hw_cpu_data, cpu);
struct sk_buff *skb;
del_timer_sync(&hw_data->send_timer);
cancel_work_sync(&hw_data->dm_alert_work);
while ((skb = __skb_dequeue(&hw_data->drop_queue))) {
struct net_dm_hw_metadata *hw_metadata;
hw_metadata = NET_DM_SKB_CB(skb)->hw_metadata;
net_dm_hw_metadata_free(hw_metadata);
consume_skb(skb);
}
}
module_put(THIS_MODULE);
}
static int net_dm_trace_on_set(struct netlink_ext_ack *extack)
{
const struct net_dm_alert_ops *ops;
int cpu, rc;
ops = net_dm_alert_ops_arr[net_dm_alert_mode];
if (!try_module_get(THIS_MODULE)) {
NL_SET_ERR_MSG_MOD(extack, "Failed to take reference on module");
return -ENODEV;
}
for_each_possible_cpu(cpu) {
struct per_cpu_dm_data *data = &per_cpu(dm_cpu_data, cpu);
struct sk_buff *skb;
INIT_WORK(&data->dm_alert_work, ops->work_item_func);
timer_setup(&data->send_timer, sched_send_work, 0);
/* Allocate a new per-CPU skb for the summary alert message and
* free the old one which might contain stale data from
* previous tracing.
*/
skb = reset_per_cpu_data(data);
consume_skb(skb);
}
rc = register_trace_kfree_skb(ops->kfree_skb_probe, NULL);
if (rc) {
NL_SET_ERR_MSG_MOD(extack, "Failed to connect probe to kfree_skb() tracepoint");
goto err_module_put;
}
rc = register_trace_napi_poll(ops->napi_poll_probe, NULL);
if (rc) {
NL_SET_ERR_MSG_MOD(extack, "Failed to connect probe to napi_poll() tracepoint");
goto err_unregister_trace;
}
return 0;
err_unregister_trace:
unregister_trace_kfree_skb(ops->kfree_skb_probe, NULL);
err_module_put:
module_put(THIS_MODULE);
return rc;
}
static void net_dm_trace_off_set(void)
{
struct dm_hw_stat_delta *new_stat, *temp;
const struct net_dm_alert_ops *ops;
int cpu;
ops = net_dm_alert_ops_arr[net_dm_alert_mode];
unregister_trace_napi_poll(ops->napi_poll_probe, NULL);
unregister_trace_kfree_skb(ops->kfree_skb_probe, NULL);
tracepoint_synchronize_unregister();
/* Make sure we do not send notifications to user space after request
* to stop tracing returns.
*/
for_each_possible_cpu(cpu) {
struct per_cpu_dm_data *data = &per_cpu(dm_cpu_data, cpu);
struct sk_buff *skb;
del_timer_sync(&data->send_timer);
cancel_work_sync(&data->dm_alert_work);
while ((skb = __skb_dequeue(&data->drop_queue)))
consume_skb(skb);
}
list_for_each_entry_safe(new_stat, temp, &hw_stats_list, list) {
if (new_stat->dev == NULL) {
list_del_rcu(&new_stat->list);
kfree_rcu(new_stat, rcu);
}
}
module_put(THIS_MODULE);
}
static int set_all_monitor_traces(int state, struct netlink_ext_ack *extack)
{
int rc = 0;
if (state == trace_state) {
NL_SET_ERR_MSG_MOD(extack, "Trace state already set to requested state");
return -EAGAIN;
}
switch (state) {
case TRACE_ON:
rc = net_dm_trace_on_set(extack);
break;
case TRACE_OFF:
net_dm_trace_off_set();
break;
default:
rc = 1;
break;
}
if (!rc)
trace_state = state;
else
rc = -EINPROGRESS;
return rc;
}
static bool net_dm_is_monitoring(void)
{
return trace_state == TRACE_ON || monitor_hw;
}
static int net_dm_alert_mode_get_from_info(struct genl_info *info,
enum net_dm_alert_mode *p_alert_mode)
{
u8 val;
val = nla_get_u8(info->attrs[NET_DM_ATTR_ALERT_MODE]);
switch (val) {
case NET_DM_ALERT_MODE_SUMMARY: /* fall-through */
case NET_DM_ALERT_MODE_PACKET:
*p_alert_mode = val;
break;
default:
return -EINVAL;
}
return 0;
}
static int net_dm_alert_mode_set(struct genl_info *info)
{
struct netlink_ext_ack *extack = info->extack;
enum net_dm_alert_mode alert_mode;
int rc;
if (!info->attrs[NET_DM_ATTR_ALERT_MODE])
return 0;
rc = net_dm_alert_mode_get_from_info(info, &alert_mode);
if (rc) {
NL_SET_ERR_MSG_MOD(extack, "Invalid alert mode");
return -EINVAL;
}
net_dm_alert_mode = alert_mode;
return 0;
}
static void net_dm_trunc_len_set(struct genl_info *info)
{
if (!info->attrs[NET_DM_ATTR_TRUNC_LEN])
return;
net_dm_trunc_len = nla_get_u32(info->attrs[NET_DM_ATTR_TRUNC_LEN]);
}
static void net_dm_queue_len_set(struct genl_info *info)
{
if (!info->attrs[NET_DM_ATTR_QUEUE_LEN])
return;
net_dm_queue_len = nla_get_u32(info->attrs[NET_DM_ATTR_QUEUE_LEN]);
}
static int net_dm_cmd_config(struct sk_buff *skb,
struct genl_info *info)
{
struct netlink_ext_ack *extack = info->extack;
int rc;
if (net_dm_is_monitoring()) {
NL_SET_ERR_MSG_MOD(extack, "Cannot configure drop monitor during monitoring");
return -EBUSY;
}
rc = net_dm_alert_mode_set(info);
if (rc)
return rc;
net_dm_trunc_len_set(info);
net_dm_queue_len_set(info);
return 0;
}
static int net_dm_monitor_start(bool set_sw, bool set_hw,
struct netlink_ext_ack *extack)
{
bool sw_set = false;
int rc;
if (set_sw) {
rc = set_all_monitor_traces(TRACE_ON, extack);
if (rc)
return rc;
sw_set = true;
}
if (set_hw) {
rc = net_dm_hw_monitor_start(extack);
if (rc)
goto err_monitor_hw;
}
return 0;
err_monitor_hw:
if (sw_set)
set_all_monitor_traces(TRACE_OFF, extack);
return rc;
}
static void net_dm_monitor_stop(bool set_sw, bool set_hw,
struct netlink_ext_ack *extack)
{
if (set_hw)
net_dm_hw_monitor_stop(extack);
if (set_sw)
set_all_monitor_traces(TRACE_OFF, extack);
}
static int net_dm_cmd_trace(struct sk_buff *skb,
struct genl_info *info)
{
bool set_sw = !!info->attrs[NET_DM_ATTR_SW_DROPS];
bool set_hw = !!info->attrs[NET_DM_ATTR_HW_DROPS];
struct netlink_ext_ack *extack = info->extack;
/* To maintain backward compatibility, we start / stop monitoring of
* software drops if no flag is specified.
*/
if (!set_sw && !set_hw)
set_sw = true;
switch (info->genlhdr->cmd) {
case NET_DM_CMD_START:
return net_dm_monitor_start(set_sw, set_hw, extack);
case NET_DM_CMD_STOP:
net_dm_monitor_stop(set_sw, set_hw, extack);
return 0;
}
return -EOPNOTSUPP;
}
static int net_dm_config_fill(struct sk_buff *msg, struct genl_info *info)
{
void *hdr;
hdr = genlmsg_put(msg, info->snd_portid, info->snd_seq,
&net_drop_monitor_family, 0, NET_DM_CMD_CONFIG_NEW);
if (!hdr)
return -EMSGSIZE;
if (nla_put_u8(msg, NET_DM_ATTR_ALERT_MODE, net_dm_alert_mode))
goto nla_put_failure;
if (nla_put_u32(msg, NET_DM_ATTR_TRUNC_LEN, net_dm_trunc_len))
goto nla_put_failure;
if (nla_put_u32(msg, NET_DM_ATTR_QUEUE_LEN, net_dm_queue_len))
goto nla_put_failure;
genlmsg_end(msg, hdr);
return 0;
nla_put_failure:
genlmsg_cancel(msg, hdr);
return -EMSGSIZE;
}
static int net_dm_cmd_config_get(struct sk_buff *skb, struct genl_info *info)
{
struct sk_buff *msg;
int rc;
msg = nlmsg_new(NLMSG_DEFAULT_SIZE, GFP_KERNEL);
if (!msg)
return -ENOMEM;
rc = net_dm_config_fill(msg, info);
if (rc)
goto free_msg;
return genlmsg_reply(msg, info);
free_msg:
nlmsg_free(msg);
return rc;
}
static void net_dm_stats_read(struct net_dm_stats *stats)
{
int cpu;
memset(stats, 0, sizeof(*stats));
for_each_possible_cpu(cpu) {
struct per_cpu_dm_data *data = &per_cpu(dm_cpu_data, cpu);
struct net_dm_stats *cpu_stats = &data->stats;
unsigned int start;
u64 dropped;
do {
start = u64_stats_fetch_begin_irq(&cpu_stats->syncp);
dropped = cpu_stats->dropped;
} while (u64_stats_fetch_retry_irq(&cpu_stats->syncp, start));
stats->dropped += dropped;
}
}
static int net_dm_stats_put(struct sk_buff *msg)
{
struct net_dm_stats stats;
struct nlattr *attr;
net_dm_stats_read(&stats);
attr = nla_nest_start(msg, NET_DM_ATTR_STATS);
if (!attr)
return -EMSGSIZE;
if (nla_put_u64_64bit(msg, NET_DM_ATTR_STATS_DROPPED,
stats.dropped, NET_DM_ATTR_PAD))
goto nla_put_failure;
nla_nest_end(msg, attr);
return 0;
nla_put_failure:
nla_nest_cancel(msg, attr);
return -EMSGSIZE;
}
static void net_dm_hw_stats_read(struct net_dm_stats *stats)
{
int cpu;
memset(stats, 0, sizeof(*stats));
for_each_possible_cpu(cpu) {
struct per_cpu_dm_data *hw_data = &per_cpu(dm_hw_cpu_data, cpu);
struct net_dm_stats *cpu_stats = &hw_data->stats;
unsigned int start;
u64 dropped;
do {
start = u64_stats_fetch_begin_irq(&cpu_stats->syncp);
dropped = cpu_stats->dropped;
} while (u64_stats_fetch_retry_irq(&cpu_stats->syncp, start));
stats->dropped += dropped;
}
}
static int net_dm_hw_stats_put(struct sk_buff *msg)
{
struct net_dm_stats stats;
struct nlattr *attr;
net_dm_hw_stats_read(&stats);
attr = nla_nest_start(msg, NET_DM_ATTR_HW_STATS);
if (!attr)
return -EMSGSIZE;
if (nla_put_u64_64bit(msg, NET_DM_ATTR_STATS_DROPPED,
stats.dropped, NET_DM_ATTR_PAD))
goto nla_put_failure;
nla_nest_end(msg, attr);
return 0;
nla_put_failure:
nla_nest_cancel(msg, attr);
return -EMSGSIZE;
}
static int net_dm_stats_fill(struct sk_buff *msg, struct genl_info *info)
{
void *hdr;
int rc;
hdr = genlmsg_put(msg, info->snd_portid, info->snd_seq,
&net_drop_monitor_family, 0, NET_DM_CMD_STATS_NEW);
if (!hdr)
return -EMSGSIZE;
rc = net_dm_stats_put(msg);
if (rc)
goto nla_put_failure;
rc = net_dm_hw_stats_put(msg);
if (rc)
goto nla_put_failure;
genlmsg_end(msg, hdr);
return 0;
nla_put_failure:
genlmsg_cancel(msg, hdr);
return -EMSGSIZE;
}
static int net_dm_cmd_stats_get(struct sk_buff *skb, struct genl_info *info)
{
struct sk_buff *msg;
int rc;
msg = nlmsg_new(NLMSG_DEFAULT_SIZE, GFP_KERNEL);
if (!msg)
return -ENOMEM;
rc = net_dm_stats_fill(msg, info);
if (rc)
goto free_msg;
return genlmsg_reply(msg, info);
free_msg:
nlmsg_free(msg);
return rc;
}
static int dropmon_net_event(struct notifier_block *ev_block,
unsigned long event, void *ptr)
{
struct net_device *dev = netdev_notifier_info_to_dev(ptr);
struct dm_hw_stat_delta *new_stat = NULL;
struct dm_hw_stat_delta *tmp;
switch (event) {
case NETDEV_REGISTER:
new_stat = kzalloc(sizeof(struct dm_hw_stat_delta), GFP_KERNEL);
if (!new_stat)
goto out;
new_stat->dev = dev;
new_stat->last_rx = jiffies;
mutex_lock(&net_dm_mutex);
list_add_rcu(&new_stat->list, &hw_stats_list);
mutex_unlock(&net_dm_mutex);
break;
case NETDEV_UNREGISTER:
mutex_lock(&net_dm_mutex);
list_for_each_entry_safe(new_stat, tmp, &hw_stats_list, list) {
if (new_stat->dev == dev) {
new_stat->dev = NULL;
if (trace_state == TRACE_OFF) {
list_del_rcu(&new_stat->list);
kfree_rcu(new_stat, rcu);
break;
}
}
}
mutex_unlock(&net_dm_mutex);
break;
}
out:
return NOTIFY_DONE;
}
static const struct nla_policy net_dm_nl_policy[NET_DM_ATTR_MAX + 1] = {
[NET_DM_ATTR_UNSPEC] = { .strict_start_type = NET_DM_ATTR_UNSPEC + 1 },
[NET_DM_ATTR_ALERT_MODE] = { .type = NLA_U8 },
[NET_DM_ATTR_TRUNC_LEN] = { .type = NLA_U32 },
[NET_DM_ATTR_QUEUE_LEN] = { .type = NLA_U32 },
[NET_DM_ATTR_SW_DROPS] = {. type = NLA_FLAG },
[NET_DM_ATTR_HW_DROPS] = {. type = NLA_FLAG },
};
static const struct genl_ops dropmon_ops[] = {
{
.cmd = NET_DM_CMD_CONFIG,
.validate = GENL_DONT_VALIDATE_STRICT | GENL_DONT_VALIDATE_DUMP,
.doit = net_dm_cmd_config,
.flags = GENL_ADMIN_PERM,
},
{
.cmd = NET_DM_CMD_START,
.validate = GENL_DONT_VALIDATE_STRICT | GENL_DONT_VALIDATE_DUMP,
.doit = net_dm_cmd_trace,
},
{
.cmd = NET_DM_CMD_STOP,
.validate = GENL_DONT_VALIDATE_STRICT | GENL_DONT_VALIDATE_DUMP,
.doit = net_dm_cmd_trace,
},
{
.cmd = NET_DM_CMD_CONFIG_GET,
.doit = net_dm_cmd_config_get,
},
{
.cmd = NET_DM_CMD_STATS_GET,
.doit = net_dm_cmd_stats_get,
},
};
static int net_dm_nl_pre_doit(const struct genl_ops *ops,
struct sk_buff *skb, struct genl_info *info)
{
mutex_lock(&net_dm_mutex);
return 0;
}
static void net_dm_nl_post_doit(const struct genl_ops *ops,
struct sk_buff *skb, struct genl_info *info)
{
mutex_unlock(&net_dm_mutex);
}
static struct genl_family net_drop_monitor_family __ro_after_init = {
.hdrsize = 0,
.name = "NET_DM",
.version = 2,
.maxattr = NET_DM_ATTR_MAX,
.policy = net_dm_nl_policy,
.pre_doit = net_dm_nl_pre_doit,
.post_doit = net_dm_nl_post_doit,
.module = THIS_MODULE,
.ops = dropmon_ops,
.n_ops = ARRAY_SIZE(dropmon_ops),
.mcgrps = dropmon_mcgrps,
.n_mcgrps = ARRAY_SIZE(dropmon_mcgrps),
};
static struct notifier_block dropmon_net_notifier = {
.notifier_call = dropmon_net_event
};
static void __net_dm_cpu_data_init(struct per_cpu_dm_data *data)
{
spin_lock_init(&data->lock);
skb_queue_head_init(&data->drop_queue);
u64_stats_init(&data->stats.syncp);
}
static void __net_dm_cpu_data_fini(struct per_cpu_dm_data *data)
{
WARN_ON(!skb_queue_empty(&data->drop_queue));
}
static void net_dm_cpu_data_init(int cpu)
{
struct per_cpu_dm_data *data;
data = &per_cpu(dm_cpu_data, cpu);
__net_dm_cpu_data_init(data);
}
static void net_dm_cpu_data_fini(int cpu)
{
struct per_cpu_dm_data *data;
data = &per_cpu(dm_cpu_data, cpu);
/* At this point, we should have exclusive access
* to this struct and can free the skb inside it.
*/
consume_skb(data->skb);
__net_dm_cpu_data_fini(data);
}
static void net_dm_hw_cpu_data_init(int cpu)
{
struct per_cpu_dm_data *hw_data;
hw_data = &per_cpu(dm_hw_cpu_data, cpu);
__net_dm_cpu_data_init(hw_data);
}
static void net_dm_hw_cpu_data_fini(int cpu)
{
struct per_cpu_dm_data *hw_data;
hw_data = &per_cpu(dm_hw_cpu_data, cpu);
kfree(hw_data->hw_entries);
__net_dm_cpu_data_fini(hw_data);
}
static int __init init_net_drop_monitor(void)
{
int cpu, rc;
pr_info("Initializing network drop monitor service\n");
if (sizeof(void *) > 8) {
pr_err("Unable to store program counters on this arch, Drop monitor failed\n");
return -ENOSPC;
}
rc = genl_register_family(&net_drop_monitor_family);
if (rc) {
pr_err("Could not create drop monitor netlink family\n");
return rc;
}
WARN_ON(net_drop_monitor_family.mcgrp_offset != NET_DM_GRP_ALERT);
rc = register_netdevice_notifier(&dropmon_net_notifier);
if (rc < 0) {
pr_crit("Failed to register netdevice notifier\n");
goto out_unreg;
}
rc = 0;
for_each_possible_cpu(cpu) {
net_dm_cpu_data_init(cpu);
net_dm_hw_cpu_data_init(cpu);
}
goto out;
out_unreg:
genl_unregister_family(&net_drop_monitor_family);
out:
return rc;
}
static void exit_net_drop_monitor(void)
{
int cpu;
BUG_ON(unregister_netdevice_notifier(&dropmon_net_notifier));
/*
* Because of the module_get/put we do in the trace state change path
* we are guarnateed not to have any current users when we get here
*/
for_each_possible_cpu(cpu) {
net_dm_hw_cpu_data_fini(cpu);
net_dm_cpu_data_fini(cpu);
}
BUG_ON(genl_unregister_family(&net_drop_monitor_family));
}
module_init(init_net_drop_monitor);
module_exit(exit_net_drop_monitor);
MODULE_LICENSE("GPL v2");
MODULE_AUTHOR("Neil Horman <nhorman@tuxdriver.com>");
MODULE_ALIAS_GENL_FAMILY("NET_DM");