summaryrefslogtreecommitdiff
path: root/lib/percpu_counter.c
blob: aeaa6d7344475518a3b73b3a7062106e77cadcea (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
/*
 * Fast batching percpu counters.
 */

#include <linux/percpu_counter.h>
#include <linux/notifier.h>
#include <linux/mutex.h>
#include <linux/init.h>
#include <linux/cpu.h>
#include <linux/module.h>

static LIST_HEAD(percpu_counters);
static DEFINE_MUTEX(percpu_counters_lock);

void percpu_counter_set(struct percpu_counter *fbc, s64 amount)
{
	int cpu;

	spin_lock(&fbc->lock);
	for_each_possible_cpu(cpu) {
		s32 *pcount = per_cpu_ptr(fbc->counters, cpu);
		*pcount = 0;
	}
	fbc->count = amount;
	spin_unlock(&fbc->lock);
}
EXPORT_SYMBOL(percpu_counter_set);

void __percpu_counter_add(struct percpu_counter *fbc, s64 amount, s32 batch)
{
	s64 count;
	s32 *pcount;
	int cpu = get_cpu();

	pcount = per_cpu_ptr(fbc->counters, cpu);
	count = *pcount + amount;
	if (count >= batch || count <= -batch) {
		spin_lock(&fbc->lock);
		fbc->count += count;
		*pcount = 0;
		spin_unlock(&fbc->lock);
	} else {
		*pcount = count;
	}
	put_cpu();
}
EXPORT_SYMBOL(__percpu_counter_add);

/*
 * Add up all the per-cpu counts, return the result.  This is a more accurate
 * but much slower version of percpu_counter_read_positive()
 */
s64 __percpu_counter_sum(struct percpu_counter *fbc)
{
	s64 ret;
	int cpu;

	spin_lock(&fbc->lock);
	ret = fbc->count;
	for_each_online_cpu(cpu) {
		s32 *pcount = per_cpu_ptr(fbc->counters, cpu);
		ret += *pcount;
	}
	spin_unlock(&fbc->lock);
	return ret;
}
EXPORT_SYMBOL(__percpu_counter_sum);

int __percpu_counter_init(struct percpu_counter *fbc, s64 amount,
			  struct lock_class_key *key)
{
	spin_lock_init(&fbc->lock);
	lockdep_set_class(&fbc->lock, key);
	fbc->count = amount;
	fbc->counters = alloc_percpu(s32);
	if (!fbc->counters)
		return -ENOMEM;
#ifdef CONFIG_HOTPLUG_CPU
	mutex_lock(&percpu_counters_lock);
	list_add(&fbc->list, &percpu_counters);
	mutex_unlock(&percpu_counters_lock);
#endif
	return 0;
}
EXPORT_SYMBOL(__percpu_counter_init);

void percpu_counter_destroy(struct percpu_counter *fbc)
{
	if (!fbc->counters)
		return;

#ifdef CONFIG_HOTPLUG_CPU
	mutex_lock(&percpu_counters_lock);
	list_del(&fbc->list);
	mutex_unlock(&percpu_counters_lock);
#endif
	free_percpu(fbc->counters);
	fbc->counters = NULL;
}
EXPORT_SYMBOL(percpu_counter_destroy);

int percpu_counter_batch __read_mostly = 32;
EXPORT_SYMBOL(percpu_counter_batch);

static void compute_batch_value(void)
{
	int nr = num_online_cpus();

	percpu_counter_batch = max(32, nr*2);
}

static int __cpuinit percpu_counter_hotcpu_callback(struct notifier_block *nb,
					unsigned long action, void *hcpu)
{
#ifdef CONFIG_HOTPLUG_CPU
	unsigned int cpu;
	struct percpu_counter *fbc;

	compute_batch_value();
	if (action != CPU_DEAD)
		return NOTIFY_OK;

	cpu = (unsigned long)hcpu;
	mutex_lock(&percpu_counters_lock);
	list_for_each_entry(fbc, &percpu_counters, list) {
		s32 *pcount;
		unsigned long flags;

		spin_lock_irqsave(&fbc->lock, flags);
		pcount = per_cpu_ptr(fbc->counters, cpu);
		fbc->count += *pcount;
		*pcount = 0;
		spin_unlock_irqrestore(&fbc->lock, flags);
	}
	mutex_unlock(&percpu_counters_lock);
#endif
	return NOTIFY_OK;
}

static int __init percpu_counter_startup(void)
{
	compute_batch_value();
	hotcpu_notifier(percpu_counter_hotcpu_callback, 0);
	return 0;
}
module_init(percpu_counter_startup);
l Goutham <sgoutham@marvell.com> Signed-off-by: David S. Miller <davem@davemloft.net> 2020-01-27octeontx2-pf: Attach NIX and NPA block LFsSunil Goutham For a PF to function as a NIC, NPA (for Rx buffers, Tx descriptors etc) and NIX (for rcv, send and completion queues) are the minimum resources needed. So request admin function (AF) to attach one each of NIX and NPA block LFs (local functions). Only AF can configure a LF's contexts, so request AF to allocate memory for NPA aura/pool and NIX RQ/SQ/CQ HW contexts. Upon receiving response, save some of the HW constants like number of pointers per stack page, size of send queue buffer (SQBs, where SQEs are queued by HW) e.t.c which are later used to initialize queues. A HW context here is like a state machine maintained for a descriptor queue. eg size, head/tail pointers, irq etc etc. HW maintains this in memory. Signed-off-by: Sunil Goutham <sgoutham@marvell.com> Signed-off-by: David S. Miller <davem@davemloft.net>