linux/net/core/gen_stats.c

486 lines
13 KiB
C
Raw Normal View History

// SPDX-License-Identifier: GPL-2.0-or-later
/*
* net/core/gen_stats.c
*
* Authors: Thomas Graf <tgraf@suug.ch>
* Jamal Hadi Salim
* Alexey Kuznetsov, <kuznet@ms2.inr.ac.ru>
*
* See Documentation/networking/gen_stats.rst
*/
#include <linux/types.h>
#include <linux/kernel.h>
#include <linux/module.h>
#include <linux/interrupt.h>
#include <linux/socket.h>
#include <linux/rtnetlink.h>
#include <linux/gen_stats.h>
#include <net/netlink.h>
#include <net/gen_stats.h>
net: sched: Protect Qdisc::bstats with u64_stats The not-per-CPU variant of qdisc tc (traffic control) statistics, Qdisc::gnet_stats_basic_packed bstats, is protected with Qdisc::running sequence counter. This sequence counter is used for reliably protecting bstats reads from parallel writes. Meanwhile, the seqcount's write section covers a much wider area than bstats update: qdisc_run_begin() => qdisc_run_end(). That read/write section asymmetry can lead to needless retries of the read section. To prepare for removing the Qdisc::running sequence counter altogether, introduce a u64_stats sync point inside bstats instead. Modify _bstats_update() to start/end the bstats u64_stats write section. For bisectability, and finer commits granularity, the bstats read section is still protected with a Qdisc::running read/retry loop and qdisc_run_begin/end() still starts/ends that seqcount write section. Once all call sites are modified to use _bstats_update(), the Qdisc::running seqcount will be removed and bstats read/retry loop will be modified to utilize the internal u64_stats sync point. Note, using u64_stats implies no sequence counter protection for 64-bit architectures. This can lead to the statistics "packets" vs. "bytes" values getting out of sync on rare occasions. The individual values will still be valid. [bigeasy: Minor commit message edits, init all gnet_stats_basic_packed.] Signed-off-by: Ahmed S. Darwish <a.darwish@linutronix.de> Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de> Signed-off-by: David S. Miller <davem@davemloft.net>
2021-10-16 16:49:07 +08:00
#include <net/sch_generic.h>
static inline int
gnet_stats_copy(struct gnet_dump *d, int type, void *buf, int size, int padattr)
{
if (nla_put_64bit(d->skb, type, size, buf, padattr))
goto nla_put_failure;
return 0;
nla_put_failure:
if (d->lock)
spin_unlock_bh(d->lock);
kfree(d->xstats);
d->xstats = NULL;
d->xstats_len = 0;
return -1;
}
/**
* gnet_stats_start_copy_compat - start dumping procedure in compatibility mode
* @skb: socket buffer to put statistics TLVs into
* @type: TLV type for top level statistic TLV
* @tc_stats_type: TLV type for backward compatibility struct tc_stats TLV
* @xstats_type: TLV type for backward compatibility xstats TLV
* @lock: statistics lock
* @d: dumping handle
* @padattr: padding attribute
*
* Initializes the dumping handle, grabs the statistic lock and appends
* an empty TLV header to the socket buffer for use a container for all
* other statistic TLVS.
*
* The dumping handle is marked to be in backward compatibility mode telling
* all gnet_stats_copy_XXX() functions to fill a local copy of struct tc_stats.
*
* Returns 0 on success or -1 if the room in the socket buffer was not sufficient.
*/
int
gnet_stats_start_copy_compat(struct sk_buff *skb, int type, int tc_stats_type,
int xstats_type, spinlock_t *lock,
struct gnet_dump *d, int padattr)
__acquires(lock)
{
memset(d, 0, sizeof(*d));
if (type)
d->tail = (struct nlattr *)skb_tail_pointer(skb);
d->skb = skb;
d->compat_tc_stats = tc_stats_type;
d->compat_xstats = xstats_type;
d->padattr = padattr;
if (lock) {
d->lock = lock;
spin_lock_bh(lock);
}
if (d->tail) {
int ret = gnet_stats_copy(d, type, NULL, 0, padattr);
/* The initial attribute added in gnet_stats_copy() may be
* preceded by a padding attribute, in which case d->tail will
* end up pointing at the padding instead of the real attribute.
* Fix this so gnet_stats_finish_copy() adjusts the length of
* the right attribute.
*/
if (ret == 0 && d->tail->nla_type == padattr)
d->tail = (struct nlattr *)((char *)d->tail +
NLA_ALIGN(d->tail->nla_len));
return ret;
}
return 0;
}
EXPORT_SYMBOL(gnet_stats_start_copy_compat);
/**
* gnet_stats_start_copy - start dumping procedure in compatibility mode
* @skb: socket buffer to put statistics TLVs into
* @type: TLV type for top level statistic TLV
* @lock: statistics lock
* @d: dumping handle
* @padattr: padding attribute
*
* Initializes the dumping handle, grabs the statistic lock and appends
* an empty TLV header to the socket buffer for use a container for all
* other statistic TLVS.
*
* Returns 0 on success or -1 if the room in the socket buffer was not sufficient.
*/
int
gnet_stats_start_copy(struct sk_buff *skb, int type, spinlock_t *lock,
struct gnet_dump *d, int padattr)
{
return gnet_stats_start_copy_compat(skb, type, 0, 0, lock, d, padattr);
}
EXPORT_SYMBOL(gnet_stats_start_copy);
net: sched: Protect Qdisc::bstats with u64_stats The not-per-CPU variant of qdisc tc (traffic control) statistics, Qdisc::gnet_stats_basic_packed bstats, is protected with Qdisc::running sequence counter. This sequence counter is used for reliably protecting bstats reads from parallel writes. Meanwhile, the seqcount's write section covers a much wider area than bstats update: qdisc_run_begin() => qdisc_run_end(). That read/write section asymmetry can lead to needless retries of the read section. To prepare for removing the Qdisc::running sequence counter altogether, introduce a u64_stats sync point inside bstats instead. Modify _bstats_update() to start/end the bstats u64_stats write section. For bisectability, and finer commits granularity, the bstats read section is still protected with a Qdisc::running read/retry loop and qdisc_run_begin/end() still starts/ends that seqcount write section. Once all call sites are modified to use _bstats_update(), the Qdisc::running seqcount will be removed and bstats read/retry loop will be modified to utilize the internal u64_stats sync point. Note, using u64_stats implies no sequence counter protection for 64-bit architectures. This can lead to the statistics "packets" vs. "bytes" values getting out of sync on rare occasions. The individual values will still be valid. [bigeasy: Minor commit message edits, init all gnet_stats_basic_packed.] Signed-off-by: Ahmed S. Darwish <a.darwish@linutronix.de> Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de> Signed-off-by: David S. Miller <davem@davemloft.net>
2021-10-16 16:49:07 +08:00
/* Must not be inlined, due to u64_stats seqcount_t lockdep key */
void gnet_stats_basic_sync_init(struct gnet_stats_basic_sync *b)
net: sched: Protect Qdisc::bstats with u64_stats The not-per-CPU variant of qdisc tc (traffic control) statistics, Qdisc::gnet_stats_basic_packed bstats, is protected with Qdisc::running sequence counter. This sequence counter is used for reliably protecting bstats reads from parallel writes. Meanwhile, the seqcount's write section covers a much wider area than bstats update: qdisc_run_begin() => qdisc_run_end(). That read/write section asymmetry can lead to needless retries of the read section. To prepare for removing the Qdisc::running sequence counter altogether, introduce a u64_stats sync point inside bstats instead. Modify _bstats_update() to start/end the bstats u64_stats write section. For bisectability, and finer commits granularity, the bstats read section is still protected with a Qdisc::running read/retry loop and qdisc_run_begin/end() still starts/ends that seqcount write section. Once all call sites are modified to use _bstats_update(), the Qdisc::running seqcount will be removed and bstats read/retry loop will be modified to utilize the internal u64_stats sync point. Note, using u64_stats implies no sequence counter protection for 64-bit architectures. This can lead to the statistics "packets" vs. "bytes" values getting out of sync on rare occasions. The individual values will still be valid. [bigeasy: Minor commit message edits, init all gnet_stats_basic_packed.] Signed-off-by: Ahmed S. Darwish <a.darwish@linutronix.de> Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de> Signed-off-by: David S. Miller <davem@davemloft.net>
2021-10-16 16:49:07 +08:00
{
u64_stats_set(&b->bytes, 0);
u64_stats_set(&b->packets, 0);
net: sched: Protect Qdisc::bstats with u64_stats The not-per-CPU variant of qdisc tc (traffic control) statistics, Qdisc::gnet_stats_basic_packed bstats, is protected with Qdisc::running sequence counter. This sequence counter is used for reliably protecting bstats reads from parallel writes. Meanwhile, the seqcount's write section covers a much wider area than bstats update: qdisc_run_begin() => qdisc_run_end(). That read/write section asymmetry can lead to needless retries of the read section. To prepare for removing the Qdisc::running sequence counter altogether, introduce a u64_stats sync point inside bstats instead. Modify _bstats_update() to start/end the bstats u64_stats write section. For bisectability, and finer commits granularity, the bstats read section is still protected with a Qdisc::running read/retry loop and qdisc_run_begin/end() still starts/ends that seqcount write section. Once all call sites are modified to use _bstats_update(), the Qdisc::running seqcount will be removed and bstats read/retry loop will be modified to utilize the internal u64_stats sync point. Note, using u64_stats implies no sequence counter protection for 64-bit architectures. This can lead to the statistics "packets" vs. "bytes" values getting out of sync on rare occasions. The individual values will still be valid. [bigeasy: Minor commit message edits, init all gnet_stats_basic_packed.] Signed-off-by: Ahmed S. Darwish <a.darwish@linutronix.de> Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de> Signed-off-by: David S. Miller <davem@davemloft.net>
2021-10-16 16:49:07 +08:00
u64_stats_init(&b->syncp);
}
EXPORT_SYMBOL(gnet_stats_basic_sync_init);
net: sched: Protect Qdisc::bstats with u64_stats The not-per-CPU variant of qdisc tc (traffic control) statistics, Qdisc::gnet_stats_basic_packed bstats, is protected with Qdisc::running sequence counter. This sequence counter is used for reliably protecting bstats reads from parallel writes. Meanwhile, the seqcount's write section covers a much wider area than bstats update: qdisc_run_begin() => qdisc_run_end(). That read/write section asymmetry can lead to needless retries of the read section. To prepare for removing the Qdisc::running sequence counter altogether, introduce a u64_stats sync point inside bstats instead. Modify _bstats_update() to start/end the bstats u64_stats write section. For bisectability, and finer commits granularity, the bstats read section is still protected with a Qdisc::running read/retry loop and qdisc_run_begin/end() still starts/ends that seqcount write section. Once all call sites are modified to use _bstats_update(), the Qdisc::running seqcount will be removed and bstats read/retry loop will be modified to utilize the internal u64_stats sync point. Note, using u64_stats implies no sequence counter protection for 64-bit architectures. This can lead to the statistics "packets" vs. "bytes" values getting out of sync on rare occasions. The individual values will still be valid. [bigeasy: Minor commit message edits, init all gnet_stats_basic_packed.] Signed-off-by: Ahmed S. Darwish <a.darwish@linutronix.de> Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de> Signed-off-by: David S. Miller <davem@davemloft.net>
2021-10-16 16:49:07 +08:00
static void gnet_stats_add_basic_cpu(struct gnet_stats_basic_sync *bstats,
struct gnet_stats_basic_sync __percpu *cpu)
{
u64 t_bytes = 0, t_packets = 0;
int i;
for_each_possible_cpu(i) {
struct gnet_stats_basic_sync *bcpu = per_cpu_ptr(cpu, i);
unsigned int start;
u64 bytes, packets;
do {
start = u64_stats_fetch_begin(&bcpu->syncp);
bytes = u64_stats_read(&bcpu->bytes);
packets = u64_stats_read(&bcpu->packets);
} while (u64_stats_fetch_retry(&bcpu->syncp, start));
t_bytes += bytes;
t_packets += packets;
}
_bstats_update(bstats, t_bytes, t_packets);
}
net: sched: Remove Qdisc::running sequence counter The Qdisc::running sequence counter has two uses: 1. Reliably reading qdisc's tc statistics while the qdisc is running (a seqcount read/retry loop at gnet_stats_add_basic()). 2. As a flag, indicating whether the qdisc in question is running (without any retry loops). For the first usage, the Qdisc::running sequence counter write section, qdisc_run_begin() => qdisc_run_end(), covers a much wider area than what is actually needed: the raw qdisc's bstats update. A u64_stats sync point was thus introduced (in previous commits) inside the bstats structure itself. A local u64_stats write section is then started and stopped for the bstats updates. Use that u64_stats sync point mechanism for the bstats read/retry loop at gnet_stats_add_basic(). For the second qdisc->running usage, a __QDISC_STATE_RUNNING bit flag, accessed with atomic bitops, is sufficient. Using a bit flag instead of a sequence counter at qdisc_run_begin/end() and qdisc_is_running() leads to the SMP barriers implicitly added through raw_read_seqcount() and write_seqcount_begin/end() getting removed. All call sites have been surveyed though, and no required ordering was identified. Now that the qdisc->running sequence counter is no longer used, remove it. Note, using u64_stats implies no sequence counter protection for 64-bit architectures. This can lead to the qdisc tc statistics "packets" vs. "bytes" values getting out of sync on rare occasions. The individual values will still be valid. Signed-off-by: Ahmed S. Darwish <a.darwish@linutronix.de> Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de> Signed-off-by: David S. Miller <davem@davemloft.net>
2021-10-16 16:49:10 +08:00
void gnet_stats_add_basic(struct gnet_stats_basic_sync *bstats,
struct gnet_stats_basic_sync __percpu *cpu,
net: sched: Remove Qdisc::running sequence counter The Qdisc::running sequence counter has two uses: 1. Reliably reading qdisc's tc statistics while the qdisc is running (a seqcount read/retry loop at gnet_stats_add_basic()). 2. As a flag, indicating whether the qdisc in question is running (without any retry loops). For the first usage, the Qdisc::running sequence counter write section, qdisc_run_begin() => qdisc_run_end(), covers a much wider area than what is actually needed: the raw qdisc's bstats update. A u64_stats sync point was thus introduced (in previous commits) inside the bstats structure itself. A local u64_stats write section is then started and stopped for the bstats updates. Use that u64_stats sync point mechanism for the bstats read/retry loop at gnet_stats_add_basic(). For the second qdisc->running usage, a __QDISC_STATE_RUNNING bit flag, accessed with atomic bitops, is sufficient. Using a bit flag instead of a sequence counter at qdisc_run_begin/end() and qdisc_is_running() leads to the SMP barriers implicitly added through raw_read_seqcount() and write_seqcount_begin/end() getting removed. All call sites have been surveyed though, and no required ordering was identified. Now that the qdisc->running sequence counter is no longer used, remove it. Note, using u64_stats implies no sequence counter protection for 64-bit architectures. This can lead to the qdisc tc statistics "packets" vs. "bytes" values getting out of sync on rare occasions. The individual values will still be valid. Signed-off-by: Ahmed S. Darwish <a.darwish@linutronix.de> Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de> Signed-off-by: David S. Miller <davem@davemloft.net>
2021-10-16 16:49:10 +08:00
struct gnet_stats_basic_sync *b, bool running)
{
net: sched: Remove Qdisc::running sequence counter The Qdisc::running sequence counter has two uses: 1. Reliably reading qdisc's tc statistics while the qdisc is running (a seqcount read/retry loop at gnet_stats_add_basic()). 2. As a flag, indicating whether the qdisc in question is running (without any retry loops). For the first usage, the Qdisc::running sequence counter write section, qdisc_run_begin() => qdisc_run_end(), covers a much wider area than what is actually needed: the raw qdisc's bstats update. A u64_stats sync point was thus introduced (in previous commits) inside the bstats structure itself. A local u64_stats write section is then started and stopped for the bstats updates. Use that u64_stats sync point mechanism for the bstats read/retry loop at gnet_stats_add_basic(). For the second qdisc->running usage, a __QDISC_STATE_RUNNING bit flag, accessed with atomic bitops, is sufficient. Using a bit flag instead of a sequence counter at qdisc_run_begin/end() and qdisc_is_running() leads to the SMP barriers implicitly added through raw_read_seqcount() and write_seqcount_begin/end() getting removed. All call sites have been surveyed though, and no required ordering was identified. Now that the qdisc->running sequence counter is no longer used, remove it. Note, using u64_stats implies no sequence counter protection for 64-bit architectures. This can lead to the qdisc tc statistics "packets" vs. "bytes" values getting out of sync on rare occasions. The individual values will still be valid. Signed-off-by: Ahmed S. Darwish <a.darwish@linutronix.de> Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de> Signed-off-by: David S. Miller <davem@davemloft.net>
2021-10-16 16:49:10 +08:00
unsigned int start;
u64 bytes = 0;
u64 packets = 0;
WARN_ON_ONCE((cpu || running) && in_hardirq());
net: sched: Remove Qdisc::running sequence counter The Qdisc::running sequence counter has two uses: 1. Reliably reading qdisc's tc statistics while the qdisc is running (a seqcount read/retry loop at gnet_stats_add_basic()). 2. As a flag, indicating whether the qdisc in question is running (without any retry loops). For the first usage, the Qdisc::running sequence counter write section, qdisc_run_begin() => qdisc_run_end(), covers a much wider area than what is actually needed: the raw qdisc's bstats update. A u64_stats sync point was thus introduced (in previous commits) inside the bstats structure itself. A local u64_stats write section is then started and stopped for the bstats updates. Use that u64_stats sync point mechanism for the bstats read/retry loop at gnet_stats_add_basic(). For the second qdisc->running usage, a __QDISC_STATE_RUNNING bit flag, accessed with atomic bitops, is sufficient. Using a bit flag instead of a sequence counter at qdisc_run_begin/end() and qdisc_is_running() leads to the SMP barriers implicitly added through raw_read_seqcount() and write_seqcount_begin/end() getting removed. All call sites have been surveyed though, and no required ordering was identified. Now that the qdisc->running sequence counter is no longer used, remove it. Note, using u64_stats implies no sequence counter protection for 64-bit architectures. This can lead to the qdisc tc statistics "packets" vs. "bytes" values getting out of sync on rare occasions. The individual values will still be valid. Signed-off-by: Ahmed S. Darwish <a.darwish@linutronix.de> Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de> Signed-off-by: David S. Miller <davem@davemloft.net>
2021-10-16 16:49:10 +08:00
if (cpu) {
gnet_stats_add_basic_cpu(bstats, cpu);
return;
}
do {
if (running)
start = u64_stats_fetch_begin(&b->syncp);
bytes = u64_stats_read(&b->bytes);
packets = u64_stats_read(&b->packets);
} while (running && u64_stats_fetch_retry(&b->syncp, start));
_bstats_update(bstats, bytes, packets);
}
EXPORT_SYMBOL(gnet_stats_add_basic);
static void gnet_stats_read_basic(u64 *ret_bytes, u64 *ret_packets,
struct gnet_stats_basic_sync __percpu *cpu,
struct gnet_stats_basic_sync *b, bool running)
{
unsigned int start;
if (cpu) {
u64 t_bytes = 0, t_packets = 0;
int i;
for_each_possible_cpu(i) {
struct gnet_stats_basic_sync *bcpu = per_cpu_ptr(cpu, i);
unsigned int start;
u64 bytes, packets;
do {
start = u64_stats_fetch_begin(&bcpu->syncp);
bytes = u64_stats_read(&bcpu->bytes);
packets = u64_stats_read(&bcpu->packets);
} while (u64_stats_fetch_retry(&bcpu->syncp, start));
t_bytes += bytes;
t_packets += packets;
}
*ret_bytes = t_bytes;
*ret_packets = t_packets;
return;
}
do {
if (running)
start = u64_stats_fetch_begin(&b->syncp);
*ret_bytes = u64_stats_read(&b->bytes);
*ret_packets = u64_stats_read(&b->packets);
} while (running && u64_stats_fetch_retry(&b->syncp, start));
}
static int
net: sched: Remove Qdisc::running sequence counter The Qdisc::running sequence counter has two uses: 1. Reliably reading qdisc's tc statistics while the qdisc is running (a seqcount read/retry loop at gnet_stats_add_basic()). 2. As a flag, indicating whether the qdisc in question is running (without any retry loops). For the first usage, the Qdisc::running sequence counter write section, qdisc_run_begin() => qdisc_run_end(), covers a much wider area than what is actually needed: the raw qdisc's bstats update. A u64_stats sync point was thus introduced (in previous commits) inside the bstats structure itself. A local u64_stats write section is then started and stopped for the bstats updates. Use that u64_stats sync point mechanism for the bstats read/retry loop at gnet_stats_add_basic(). For the second qdisc->running usage, a __QDISC_STATE_RUNNING bit flag, accessed with atomic bitops, is sufficient. Using a bit flag instead of a sequence counter at qdisc_run_begin/end() and qdisc_is_running() leads to the SMP barriers implicitly added through raw_read_seqcount() and write_seqcount_begin/end() getting removed. All call sites have been surveyed though, and no required ordering was identified. Now that the qdisc->running sequence counter is no longer used, remove it. Note, using u64_stats implies no sequence counter protection for 64-bit architectures. This can lead to the qdisc tc statistics "packets" vs. "bytes" values getting out of sync on rare occasions. The individual values will still be valid. Signed-off-by: Ahmed S. Darwish <a.darwish@linutronix.de> Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de> Signed-off-by: David S. Miller <davem@davemloft.net>
2021-10-16 16:49:10 +08:00
___gnet_stats_copy_basic(struct gnet_dump *d,
struct gnet_stats_basic_sync __percpu *cpu,
struct gnet_stats_basic_sync *b,
net: sched: Remove Qdisc::running sequence counter The Qdisc::running sequence counter has two uses: 1. Reliably reading qdisc's tc statistics while the qdisc is running (a seqcount read/retry loop at gnet_stats_add_basic()). 2. As a flag, indicating whether the qdisc in question is running (without any retry loops). For the first usage, the Qdisc::running sequence counter write section, qdisc_run_begin() => qdisc_run_end(), covers a much wider area than what is actually needed: the raw qdisc's bstats update. A u64_stats sync point was thus introduced (in previous commits) inside the bstats structure itself. A local u64_stats write section is then started and stopped for the bstats updates. Use that u64_stats sync point mechanism for the bstats read/retry loop at gnet_stats_add_basic(). For the second qdisc->running usage, a __QDISC_STATE_RUNNING bit flag, accessed with atomic bitops, is sufficient. Using a bit flag instead of a sequence counter at qdisc_run_begin/end() and qdisc_is_running() leads to the SMP barriers implicitly added through raw_read_seqcount() and write_seqcount_begin/end() getting removed. All call sites have been surveyed though, and no required ordering was identified. Now that the qdisc->running sequence counter is no longer used, remove it. Note, using u64_stats implies no sequence counter protection for 64-bit architectures. This can lead to the qdisc tc statistics "packets" vs. "bytes" values getting out of sync on rare occasions. The individual values will still be valid. Signed-off-by: Ahmed S. Darwish <a.darwish@linutronix.de> Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de> Signed-off-by: David S. Miller <davem@davemloft.net>
2021-10-16 16:49:10 +08:00
int type, bool running)
{
u64 bstats_bytes, bstats_packets;
gnet_stats_read_basic(&bstats_bytes, &bstats_packets, cpu, b, running);
if (d->compat_tc_stats && type == TCA_STATS_BASIC) {
d->tc_stats.bytes = bstats_bytes;
d->tc_stats.packets = bstats_packets;
}
if (d->tail) {
struct gnet_stats_basic sb;
int res;
memset(&sb, 0, sizeof(sb));
sb.bytes = bstats_bytes;
sb.packets = bstats_packets;
res = gnet_stats_copy(d, type, &sb, sizeof(sb), TCA_STATS_PAD);
if (res < 0 || sb.packets == bstats_packets)
return res;
/* emit 64bit stats only if needed */
return gnet_stats_copy(d, TCA_STATS_PKT64, &bstats_packets,
sizeof(bstats_packets), TCA_STATS_PAD);
}
return 0;
}
/**
* gnet_stats_copy_basic - copy basic statistics into statistic TLV
* @d: dumping handle
* @cpu: copy statistic per cpu
* @b: basic statistics
net: sched: Remove Qdisc::running sequence counter The Qdisc::running sequence counter has two uses: 1. Reliably reading qdisc's tc statistics while the qdisc is running (a seqcount read/retry loop at gnet_stats_add_basic()). 2. As a flag, indicating whether the qdisc in question is running (without any retry loops). For the first usage, the Qdisc::running sequence counter write section, qdisc_run_begin() => qdisc_run_end(), covers a much wider area than what is actually needed: the raw qdisc's bstats update. A u64_stats sync point was thus introduced (in previous commits) inside the bstats structure itself. A local u64_stats write section is then started and stopped for the bstats updates. Use that u64_stats sync point mechanism for the bstats read/retry loop at gnet_stats_add_basic(). For the second qdisc->running usage, a __QDISC_STATE_RUNNING bit flag, accessed with atomic bitops, is sufficient. Using a bit flag instead of a sequence counter at qdisc_run_begin/end() and qdisc_is_running() leads to the SMP barriers implicitly added through raw_read_seqcount() and write_seqcount_begin/end() getting removed. All call sites have been surveyed though, and no required ordering was identified. Now that the qdisc->running sequence counter is no longer used, remove it. Note, using u64_stats implies no sequence counter protection for 64-bit architectures. This can lead to the qdisc tc statistics "packets" vs. "bytes" values getting out of sync on rare occasions. The individual values will still be valid. Signed-off-by: Ahmed S. Darwish <a.darwish@linutronix.de> Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de> Signed-off-by: David S. Miller <davem@davemloft.net>
2021-10-16 16:49:10 +08:00
* @running: true if @b represents a running qdisc, thus @b's
* internal values might change during basic reads.
* Only used if @cpu is NULL
*
* Context: task; must not be run from IRQ or BH contexts
*
* Appends the basic statistics to the top level TLV created by
* gnet_stats_start_copy().
*
* Returns 0 on success or -1 with the statistic lock released
* if the room in the socket buffer was not sufficient.
*/
int
net: sched: Remove Qdisc::running sequence counter The Qdisc::running sequence counter has two uses: 1. Reliably reading qdisc's tc statistics while the qdisc is running (a seqcount read/retry loop at gnet_stats_add_basic()). 2. As a flag, indicating whether the qdisc in question is running (without any retry loops). For the first usage, the Qdisc::running sequence counter write section, qdisc_run_begin() => qdisc_run_end(), covers a much wider area than what is actually needed: the raw qdisc's bstats update. A u64_stats sync point was thus introduced (in previous commits) inside the bstats structure itself. A local u64_stats write section is then started and stopped for the bstats updates. Use that u64_stats sync point mechanism for the bstats read/retry loop at gnet_stats_add_basic(). For the second qdisc->running usage, a __QDISC_STATE_RUNNING bit flag, accessed with atomic bitops, is sufficient. Using a bit flag instead of a sequence counter at qdisc_run_begin/end() and qdisc_is_running() leads to the SMP barriers implicitly added through raw_read_seqcount() and write_seqcount_begin/end() getting removed. All call sites have been surveyed though, and no required ordering was identified. Now that the qdisc->running sequence counter is no longer used, remove it. Note, using u64_stats implies no sequence counter protection for 64-bit architectures. This can lead to the qdisc tc statistics "packets" vs. "bytes" values getting out of sync on rare occasions. The individual values will still be valid. Signed-off-by: Ahmed S. Darwish <a.darwish@linutronix.de> Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de> Signed-off-by: David S. Miller <davem@davemloft.net>
2021-10-16 16:49:10 +08:00
gnet_stats_copy_basic(struct gnet_dump *d,
struct gnet_stats_basic_sync __percpu *cpu,
net: sched: Remove Qdisc::running sequence counter The Qdisc::running sequence counter has two uses: 1. Reliably reading qdisc's tc statistics while the qdisc is running (a seqcount read/retry loop at gnet_stats_add_basic()). 2. As a flag, indicating whether the qdisc in question is running (without any retry loops). For the first usage, the Qdisc::running sequence counter write section, qdisc_run_begin() => qdisc_run_end(), covers a much wider area than what is actually needed: the raw qdisc's bstats update. A u64_stats sync point was thus introduced (in previous commits) inside the bstats structure itself. A local u64_stats write section is then started and stopped for the bstats updates. Use that u64_stats sync point mechanism for the bstats read/retry loop at gnet_stats_add_basic(). For the second qdisc->running usage, a __QDISC_STATE_RUNNING bit flag, accessed with atomic bitops, is sufficient. Using a bit flag instead of a sequence counter at qdisc_run_begin/end() and qdisc_is_running() leads to the SMP barriers implicitly added through raw_read_seqcount() and write_seqcount_begin/end() getting removed. All call sites have been surveyed though, and no required ordering was identified. Now that the qdisc->running sequence counter is no longer used, remove it. Note, using u64_stats implies no sequence counter protection for 64-bit architectures. This can lead to the qdisc tc statistics "packets" vs. "bytes" values getting out of sync on rare occasions. The individual values will still be valid. Signed-off-by: Ahmed S. Darwish <a.darwish@linutronix.de> Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de> Signed-off-by: David S. Miller <davem@davemloft.net>
2021-10-16 16:49:10 +08:00
struct gnet_stats_basic_sync *b,
bool running)
{
net: sched: Remove Qdisc::running sequence counter The Qdisc::running sequence counter has two uses: 1. Reliably reading qdisc's tc statistics while the qdisc is running (a seqcount read/retry loop at gnet_stats_add_basic()). 2. As a flag, indicating whether the qdisc in question is running (without any retry loops). For the first usage, the Qdisc::running sequence counter write section, qdisc_run_begin() => qdisc_run_end(), covers a much wider area than what is actually needed: the raw qdisc's bstats update. A u64_stats sync point was thus introduced (in previous commits) inside the bstats structure itself. A local u64_stats write section is then started and stopped for the bstats updates. Use that u64_stats sync point mechanism for the bstats read/retry loop at gnet_stats_add_basic(). For the second qdisc->running usage, a __QDISC_STATE_RUNNING bit flag, accessed with atomic bitops, is sufficient. Using a bit flag instead of a sequence counter at qdisc_run_begin/end() and qdisc_is_running() leads to the SMP barriers implicitly added through raw_read_seqcount() and write_seqcount_begin/end() getting removed. All call sites have been surveyed though, and no required ordering was identified. Now that the qdisc->running sequence counter is no longer used, remove it. Note, using u64_stats implies no sequence counter protection for 64-bit architectures. This can lead to the qdisc tc statistics "packets" vs. "bytes" values getting out of sync on rare occasions. The individual values will still be valid. Signed-off-by: Ahmed S. Darwish <a.darwish@linutronix.de> Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de> Signed-off-by: David S. Miller <davem@davemloft.net>
2021-10-16 16:49:10 +08:00
return ___gnet_stats_copy_basic(d, cpu, b, TCA_STATS_BASIC, running);
}
EXPORT_SYMBOL(gnet_stats_copy_basic);
/**
* gnet_stats_copy_basic_hw - copy basic hw statistics into statistic TLV
* @d: dumping handle
* @cpu: copy statistic per cpu
* @b: basic statistics
net: sched: Remove Qdisc::running sequence counter The Qdisc::running sequence counter has two uses: 1. Reliably reading qdisc's tc statistics while the qdisc is running (a seqcount read/retry loop at gnet_stats_add_basic()). 2. As a flag, indicating whether the qdisc in question is running (without any retry loops). For the first usage, the Qdisc::running sequence counter write section, qdisc_run_begin() => qdisc_run_end(), covers a much wider area than what is actually needed: the raw qdisc's bstats update. A u64_stats sync point was thus introduced (in previous commits) inside the bstats structure itself. A local u64_stats write section is then started and stopped for the bstats updates. Use that u64_stats sync point mechanism for the bstats read/retry loop at gnet_stats_add_basic(). For the second qdisc->running usage, a __QDISC_STATE_RUNNING bit flag, accessed with atomic bitops, is sufficient. Using a bit flag instead of a sequence counter at qdisc_run_begin/end() and qdisc_is_running() leads to the SMP barriers implicitly added through raw_read_seqcount() and write_seqcount_begin/end() getting removed. All call sites have been surveyed though, and no required ordering was identified. Now that the qdisc->running sequence counter is no longer used, remove it. Note, using u64_stats implies no sequence counter protection for 64-bit architectures. This can lead to the qdisc tc statistics "packets" vs. "bytes" values getting out of sync on rare occasions. The individual values will still be valid. Signed-off-by: Ahmed S. Darwish <a.darwish@linutronix.de> Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de> Signed-off-by: David S. Miller <davem@davemloft.net>
2021-10-16 16:49:10 +08:00
* @running: true if @b represents a running qdisc, thus @b's
* internal values might change during basic reads.
* Only used if @cpu is NULL
*
* Context: task; must not be run from IRQ or BH contexts
*
* Appends the basic statistics to the top level TLV created by
* gnet_stats_start_copy().
*
* Returns 0 on success or -1 with the statistic lock released
* if the room in the socket buffer was not sufficient.
*/
int
net: sched: Remove Qdisc::running sequence counter The Qdisc::running sequence counter has two uses: 1. Reliably reading qdisc's tc statistics while the qdisc is running (a seqcount read/retry loop at gnet_stats_add_basic()). 2. As a flag, indicating whether the qdisc in question is running (without any retry loops). For the first usage, the Qdisc::running sequence counter write section, qdisc_run_begin() => qdisc_run_end(), covers a much wider area than what is actually needed: the raw qdisc's bstats update. A u64_stats sync point was thus introduced (in previous commits) inside the bstats structure itself. A local u64_stats write section is then started and stopped for the bstats updates. Use that u64_stats sync point mechanism for the bstats read/retry loop at gnet_stats_add_basic(). For the second qdisc->running usage, a __QDISC_STATE_RUNNING bit flag, accessed with atomic bitops, is sufficient. Using a bit flag instead of a sequence counter at qdisc_run_begin/end() and qdisc_is_running() leads to the SMP barriers implicitly added through raw_read_seqcount() and write_seqcount_begin/end() getting removed. All call sites have been surveyed though, and no required ordering was identified. Now that the qdisc->running sequence counter is no longer used, remove it. Note, using u64_stats implies no sequence counter protection for 64-bit architectures. This can lead to the qdisc tc statistics "packets" vs. "bytes" values getting out of sync on rare occasions. The individual values will still be valid. Signed-off-by: Ahmed S. Darwish <a.darwish@linutronix.de> Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de> Signed-off-by: David S. Miller <davem@davemloft.net>
2021-10-16 16:49:10 +08:00
gnet_stats_copy_basic_hw(struct gnet_dump *d,
struct gnet_stats_basic_sync __percpu *cpu,
net: sched: Remove Qdisc::running sequence counter The Qdisc::running sequence counter has two uses: 1. Reliably reading qdisc's tc statistics while the qdisc is running (a seqcount read/retry loop at gnet_stats_add_basic()). 2. As a flag, indicating whether the qdisc in question is running (without any retry loops). For the first usage, the Qdisc::running sequence counter write section, qdisc_run_begin() => qdisc_run_end(), covers a much wider area than what is actually needed: the raw qdisc's bstats update. A u64_stats sync point was thus introduced (in previous commits) inside the bstats structure itself. A local u64_stats write section is then started and stopped for the bstats updates. Use that u64_stats sync point mechanism for the bstats read/retry loop at gnet_stats_add_basic(). For the second qdisc->running usage, a __QDISC_STATE_RUNNING bit flag, accessed with atomic bitops, is sufficient. Using a bit flag instead of a sequence counter at qdisc_run_begin/end() and qdisc_is_running() leads to the SMP barriers implicitly added through raw_read_seqcount() and write_seqcount_begin/end() getting removed. All call sites have been surveyed though, and no required ordering was identified. Now that the qdisc->running sequence counter is no longer used, remove it. Note, using u64_stats implies no sequence counter protection for 64-bit architectures. This can lead to the qdisc tc statistics "packets" vs. "bytes" values getting out of sync on rare occasions. The individual values will still be valid. Signed-off-by: Ahmed S. Darwish <a.darwish@linutronix.de> Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de> Signed-off-by: David S. Miller <davem@davemloft.net>
2021-10-16 16:49:10 +08:00
struct gnet_stats_basic_sync *b,
bool running)
{
net: sched: Remove Qdisc::running sequence counter The Qdisc::running sequence counter has two uses: 1. Reliably reading qdisc's tc statistics while the qdisc is running (a seqcount read/retry loop at gnet_stats_add_basic()). 2. As a flag, indicating whether the qdisc in question is running (without any retry loops). For the first usage, the Qdisc::running sequence counter write section, qdisc_run_begin() => qdisc_run_end(), covers a much wider area than what is actually needed: the raw qdisc's bstats update. A u64_stats sync point was thus introduced (in previous commits) inside the bstats structure itself. A local u64_stats write section is then started and stopped for the bstats updates. Use that u64_stats sync point mechanism for the bstats read/retry loop at gnet_stats_add_basic(). For the second qdisc->running usage, a __QDISC_STATE_RUNNING bit flag, accessed with atomic bitops, is sufficient. Using a bit flag instead of a sequence counter at qdisc_run_begin/end() and qdisc_is_running() leads to the SMP barriers implicitly added through raw_read_seqcount() and write_seqcount_begin/end() getting removed. All call sites have been surveyed though, and no required ordering was identified. Now that the qdisc->running sequence counter is no longer used, remove it. Note, using u64_stats implies no sequence counter protection for 64-bit architectures. This can lead to the qdisc tc statistics "packets" vs. "bytes" values getting out of sync on rare occasions. The individual values will still be valid. Signed-off-by: Ahmed S. Darwish <a.darwish@linutronix.de> Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de> Signed-off-by: David S. Miller <davem@davemloft.net>
2021-10-16 16:49:10 +08:00
return ___gnet_stats_copy_basic(d, cpu, b, TCA_STATS_BASIC_HW, running);
}
EXPORT_SYMBOL(gnet_stats_copy_basic_hw);
/**
* gnet_stats_copy_rate_est - copy rate estimator statistics into statistics TLV
* @d: dumping handle
* @rate_est: rate estimator
*
* Appends the rate estimator statistics to the top level TLV created by
* gnet_stats_start_copy().
*
* Returns 0 on success or -1 with the statistic lock released
* if the room in the socket buffer was not sufficient.
*/
int
pkt_sched: gen_estimator: Dont report fake rate estimators Jarek Poplawski a écrit : > > > Hmm... So you made me to do some "real" work here, and guess what?: > there is one serious checkpatch warning! ;-) Plus, this new parameter > should be added to the function description. Otherwise: > Signed-off-by: Jarek Poplawski <jarkao2@gmail.com> > > Thanks, > Jarek P. > > PS: I guess full "Don't" would show we really mean it... Okay :) Here is the last round, before the night ! Thanks again [RFC] pkt_sched: gen_estimator: Don't report fake rate estimators We currently send TCA_STATS_RATE_EST elements to netlink users, even if no estimator is running. # tc -s -d qdisc qdisc pfifo_fast 0: dev eth0 root bands 3 priomap 1 2 2 2 1 2 0 0 1 1 1 1 1 1 1 1 Sent 112833764978 bytes 1495081739 pkt (dropped 0, overlimits 0 requeues 0) rate 0bit 0pps backlog 0b 0p requeues 0 User has no way to tell if the "rate 0bit 0pps" is a real estimation, or a fake one (because no estimator is active) After this patch, tc command output is : $ tc -s -d qdisc qdisc pfifo_fast 0: dev eth0 root bands 3 priomap 1 2 2 2 1 2 0 0 1 1 1 1 1 1 1 1 Sent 561075 bytes 1196 pkt (dropped 0, overlimits 0 requeues 0) backlog 0b 0p requeues 0 We add a parameter to gnet_stats_copy_rate_est() function so that it can use gen_estimator_active(bstats, r), as suggested by Jarek. This parameter can be NULL if check is not necessary, (htb for example has a mandatory rate estimator) Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com> Signed-off-by: Jarek Poplawski <jarkao2@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-02 18:32:18 +08:00
gnet_stats_copy_rate_est(struct gnet_dump *d,
struct net_rate_estimator __rcu **rate_est)
{
struct gnet_stats_rate_est64 sample;
struct gnet_stats_rate_est est;
int res;
if (!gen_estimator_read(rate_est, &sample))
pkt_sched: gen_estimator: Dont report fake rate estimators Jarek Poplawski a écrit : > > > Hmm... So you made me to do some "real" work here, and guess what?: > there is one serious checkpatch warning! ;-) Plus, this new parameter > should be added to the function description. Otherwise: > Signed-off-by: Jarek Poplawski <jarkao2@gmail.com> > > Thanks, > Jarek P. > > PS: I guess full "Don't" would show we really mean it... Okay :) Here is the last round, before the night ! Thanks again [RFC] pkt_sched: gen_estimator: Don't report fake rate estimators We currently send TCA_STATS_RATE_EST elements to netlink users, even if no estimator is running. # tc -s -d qdisc qdisc pfifo_fast 0: dev eth0 root bands 3 priomap 1 2 2 2 1 2 0 0 1 1 1 1 1 1 1 1 Sent 112833764978 bytes 1495081739 pkt (dropped 0, overlimits 0 requeues 0) rate 0bit 0pps backlog 0b 0p requeues 0 User has no way to tell if the "rate 0bit 0pps" is a real estimation, or a fake one (because no estimator is active) After this patch, tc command output is : $ tc -s -d qdisc qdisc pfifo_fast 0: dev eth0 root bands 3 priomap 1 2 2 2 1 2 0 0 1 1 1 1 1 1 1 1 Sent 561075 bytes 1196 pkt (dropped 0, overlimits 0 requeues 0) backlog 0b 0p requeues 0 We add a parameter to gnet_stats_copy_rate_est() function so that it can use gen_estimator_active(bstats, r), as suggested by Jarek. This parameter can be NULL if check is not necessary, (htb for example has a mandatory rate estimator) Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com> Signed-off-by: Jarek Poplawski <jarkao2@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-10-02 18:32:18 +08:00
return 0;
est.bps = min_t(u64, UINT_MAX, sample.bps);
/* we have some time before reaching 2^32 packets per second */
est.pps = sample.pps;
if (d->compat_tc_stats) {
d->tc_stats.bps = est.bps;
d->tc_stats.pps = est.pps;
}
if (d->tail) {
res = gnet_stats_copy(d, TCA_STATS_RATE_EST, &est, sizeof(est),
TCA_STATS_PAD);
if (res < 0 || est.bps == sample.bps)
return res;
/* emit 64bit stats only if needed */
return gnet_stats_copy(d, TCA_STATS_RATE_EST64, &sample,
sizeof(sample), TCA_STATS_PAD);
}
return 0;
}
EXPORT_SYMBOL(gnet_stats_copy_rate_est);
static void gnet_stats_add_queue_cpu(struct gnet_stats_queue *qstats,
const struct gnet_stats_queue __percpu *q)
{
int i;
for_each_possible_cpu(i) {
const struct gnet_stats_queue *qcpu = per_cpu_ptr(q, i);
qstats->qlen += qcpu->qlen;
qstats->backlog += qcpu->backlog;
qstats->drops += qcpu->drops;
qstats->requeues += qcpu->requeues;
qstats->overlimits += qcpu->overlimits;
}
}
void gnet_stats_add_queue(struct gnet_stats_queue *qstats,
const struct gnet_stats_queue __percpu *cpu,
const struct gnet_stats_queue *q)
{
if (cpu) {
gnet_stats_add_queue_cpu(qstats, cpu);
} else {
qstats->qlen += q->qlen;
qstats->backlog += q->backlog;
qstats->drops += q->drops;
qstats->requeues += q->requeues;
qstats->overlimits += q->overlimits;
}
}
EXPORT_SYMBOL(gnet_stats_add_queue);
/**
* gnet_stats_copy_queue - copy queue statistics into statistics TLV
* @d: dumping handle
* @cpu_q: per cpu queue statistics
* @q: queue statistics
* @qlen: queue length statistics
*
* Appends the queue statistics to the top level TLV created by
* gnet_stats_start_copy(). Using per cpu queue statistics if
* they are available.
*
* Returns 0 on success or -1 with the statistic lock released
* if the room in the socket buffer was not sufficient.
*/
int
gnet_stats_copy_queue(struct gnet_dump *d,
struct gnet_stats_queue __percpu *cpu_q,
struct gnet_stats_queue *q, __u32 qlen)
{
struct gnet_stats_queue qstats = {0};
gnet_stats_add_queue(&qstats, cpu_q, q);
qstats.qlen = qlen;
if (d->compat_tc_stats) {
d->tc_stats.drops = qstats.drops;
d->tc_stats.qlen = qstats.qlen;
d->tc_stats.backlog = qstats.backlog;
d->tc_stats.overlimits = qstats.overlimits;
}
if (d->tail)
return gnet_stats_copy(d, TCA_STATS_QUEUE,
&qstats, sizeof(qstats),
TCA_STATS_PAD);
return 0;
}
EXPORT_SYMBOL(gnet_stats_copy_queue);
/**
* gnet_stats_copy_app - copy application specific statistics into statistics TLV
* @d: dumping handle
* @st: application specific statistics data
* @len: length of data
*
* Appends the application specific statistics to the top level TLV created by
* gnet_stats_start_copy() and remembers the data for XSTATS if the dumping
* handle is in backward compatibility mode.
*
* Returns 0 on success or -1 with the statistic lock released
* if the room in the socket buffer was not sufficient.
*/
int
gnet_stats_copy_app(struct gnet_dump *d, void *st, int len)
{
if (d->compat_xstats) {
d->xstats = kmemdup(st, len, GFP_ATOMIC);
if (!d->xstats)
goto err_out;
d->xstats_len = len;
}
if (d->tail)
return gnet_stats_copy(d, TCA_STATS_APP, st, len,
TCA_STATS_PAD);
return 0;
err_out:
if (d->lock)
spin_unlock_bh(d->lock);
d->xstats_len = 0;
return -1;
}
EXPORT_SYMBOL(gnet_stats_copy_app);
/**
* gnet_stats_finish_copy - finish dumping procedure
* @d: dumping handle
*
* Corrects the length of the top level TLV to include all TLVs added
* by gnet_stats_copy_XXX() calls. Adds the backward compatibility TLVs
* if gnet_stats_start_copy_compat() was used and releases the statistics
* lock.
*
* Returns 0 on success or -1 with the statistic lock released
* if the room in the socket buffer was not sufficient.
*/
int
gnet_stats_finish_copy(struct gnet_dump *d)
{
if (d->tail)
d->tail->nla_len = skb_tail_pointer(d->skb) - (u8 *)d->tail;
if (d->compat_tc_stats)
if (gnet_stats_copy(d, d->compat_tc_stats, &d->tc_stats,
sizeof(d->tc_stats), d->padattr) < 0)
return -1;
if (d->compat_xstats && d->xstats) {
if (gnet_stats_copy(d, d->compat_xstats, d->xstats,
d->xstats_len, d->padattr) < 0)
return -1;
}
if (d->lock)
spin_unlock_bh(d->lock);
kfree(d->xstats);
d->xstats = NULL;
d->xstats_len = 0;
return 0;
}
EXPORT_SYMBOL(gnet_stats_finish_copy);