android_kernel_samsung_msm8976/kernel/locking/qspinlock.c

557 lines
15 KiB
C

/*
* Queue spinlock
*
* This program is free software; you can redistribute it and/or modify
* it under the terms of the GNU General Public License as published by
* the Free Software Foundation; either version 2 of the License, or
* (at your option) any later version.
*
* This program is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU General Public License for more details.
*
* (C) Copyright 2013-2014 Hewlett-Packard Development Company, L.P.
*
* Authors: Waiman Long <waiman.long@hp.com>
* Peter Zijlstra <pzijlstr@redhat.com>
*/
#include <linux/smp.h>
#include <linux/bug.h>
#include <linux/cpumask.h>
#include <linux/percpu.h>
#include <linux/hardirq.h>
#include <linux/mutex.h>
#include <asm/byteorder.h>
#include <asm/qspinlock.h>
/*
* The basic principle of a queue-based spinlock can best be understood
* by studying a classic queue-based spinlock implementation called the
* MCS lock. The paper below provides a good description for this kind
* of lock.
*
* http://www.cise.ufl.edu/tr/DOC/REP-1992-71.pdf
*
* This queue spinlock implementation is based on the MCS lock, however to make
* it fit the 4 bytes we assume spinlock_t to be, and preserve its existing
* API, we must modify it somehow.
*
* In particular; where the traditional MCS lock consists of a tail pointer
* (8 bytes) and needs the next pointer (another 8 bytes) of its own node to
* unlock the next pending (next->locked), we compress both these: {tail,
* next->locked} into a single u32 value.
*
* Since a spinlock disables recursion of its own context and there is a limit
* to the contexts that can nest; namely: task, softirq, hardirq, nmi. As there
* are at most 4 nesting levels, it can be encoded by a 2-bit number. Now
* we can encode the tail by combining the 2-bit nesting level with the cpu
* number. With one byte for the lock value and 3 bytes for the tail, only a
* 32-bit word is now needed. Even though we only need 1 bit for the lock,
* we extend it to a full byte to achieve better performance for architectures
* that support atomic byte write.
*
* We also change the first spinner to spin on the lock bit instead of its
* node; whereby avoiding the need to carry a node from lock to unlock, and
* preserving existing lock API. This also makes the unlock code simpler and
* faster.
*
* N.B. The current implementation only supports architectures that allow
* atomic operations on smaller 8-bit and 16-bit data types.
*
*/
#include "mcs_spinlock.h"
/*
* Per-CPU queue node structures; we can never have more than 4 nested
* contexts: task, softirq, hardirq, nmi.
*
* Exactly fits one 64-byte cacheline on a 64-bit architecture.
*/
static DEFINE_PER_CPU_ALIGNED(struct mcs_spinlock, mcs_nodes[4]);
/*
* We must be able to distinguish between no-tail and the tail at 0:0,
* therefore increment the cpu number by one.
* BluesMan: The idx = atomic counter (nesting depth)
*/
static inline u32 encode_tail(int cpu, int idx)
{
u32 tail;
#ifdef CONFIG_DEBUG_SPINLOCK
BUG_ON(idx > 3);
#endif
tail = (cpu + 1) << _Q_TAIL_CPU_OFFSET;
tail |= idx << _Q_TAIL_IDX_OFFSET; /* assume < 4 */
return tail;
}
static inline struct mcs_spinlock *decode_tail(u32 tail)
{
int cpu = (tail >> _Q_TAIL_CPU_OFFSET) - 1;
int idx = (tail & _Q_TAIL_IDX_MASK) >> _Q_TAIL_IDX_OFFSET;
return per_cpu_ptr(&mcs_nodes[idx], cpu);
}
#ifdef CONFIG_DEBUG_QUEUE_SPINLOCK
/*
* BluesMan: These functions are highly verbose and may cause
* system lock up or undefined behaviour if used in hot-paths
* You've been warned.
* Keep these disabled unless we need to debug something runtime.
*
*/
#if 0
static void trace_dump_this_mcs_node(struct mcs_spinlock *node)
{
if(NULL == node) {
trace_printk("This Node is NULL, nothing to go on\n");
return;
}
trace_printk("node = %pS\n",node);
trace_printk("node->next = %pS, node->locked = %d,\n",
node->next, node->locked);
trace_printk("node->count(nesting) = %d\n, node->tail = %d\n",
node->count, node->tail);
}
static void trace_dump_per_cpu_mcs_nodes(void)
{
unsigned int cpu, j;
struct mcs_spinlock *node, *tmp;
/* iterate through all cpus, regardless of their state */
for(cpu = 0; cpu < CONFIG_NR_CPUS; cpu++) {
for(j = 0; j < 4; j++) {
if(!cpu_online(cpu))
trace_printk("cpu %d is down\n", cpu);
node = per_cpu_ptr(&mcs_nodes[j], cpu);
trace_dump_this_mcs_node(node);
if(node->next != NULL) {
/* We have contenders */
tmp = node->next;
trace_printk("node[%d] has children\n",j);
do {
trace_dump_this_mcs_node(tmp);
}while((tmp = tmp->next) != NULL);
}
}
}
}
#endif
#endif
#define _Q_LOCKED_PENDING_MASK (_Q_LOCKED_MASK | _Q_PENDING_MASK)
/*
* By using the whole 2nd least significant byte for the pending bit, we
* can allow better optimization of the lock acquisition for the pending
* bit holder.
*
* This internal structure is also used by the set_locked function which
* is not restricted to _Q_PENDING_BITS == 8.
*/
struct __qspinlock {
union {
atomic_t val;
#ifdef __LITTLE_ENDIAN
u8 locked;
struct {
u16 locked_pending;
u16 tail;
};
#else
struct {
u16 tail;
u16 locked_pending;
};
struct {
u8 reserved[3];
u8 locked;
};
#endif
};
};
#if _Q_PENDING_BITS == 8
/**
* clear_pending_set_locked - take ownership and clear the pending bit.
* @lock: Pointer to queue spinlock structure
* @val : Current value of the queue spinlock 32-bit word
*
* *,1,0 -> *,0,1
*
* Lock stealing is not allowed if this function is used.
*/
static __always_inline void
clear_pending_set_locked(struct qspinlock *lock, u32 val)
{
struct __qspinlock *l = (void *)lock;
WRITE_ONCE(l->locked_pending, _Q_LOCKED_VAL);
}
/*
* xchg_tail - Put in the new queue tail code word & retrieve previous one
* @lock : Pointer to queue spinlock structure
* @tail : The new queue tail code word
* Return: The previous queue tail code word
*
* xchg(lock, tail)
*
* p,*,* -> n,*,* ; prev = xchg(lock, node)
*/
static __always_inline u32 xchg_tail(struct qspinlock *lock, u32 tail)
{
struct __qspinlock *l = (void *)lock;
/* BluesMan: Debug print. */
/* pr_emerg("xchg_tail: sizeof l->tail = 0x%x, l->tail = 0x%x, tail = 0x%x\n",
sizeof(l->tail), l->tail, tail); */
return (u32)xchg(&l->tail, tail >> _Q_TAIL_OFFSET) << _Q_TAIL_OFFSET;
}
#else /* _Q_PENDING_BITS == 8 */
/**
* clear_pending_set_locked - take ownership and clear the pending bit.
* @lock: Pointer to queue spinlock structure
* @val : Current value of the queue spinlock 32-bit word
*
* *,1,0 -> *,0,1
*/
static __always_inline void
clear_pending_set_locked(struct qspinlock *lock, u32 val)
{
u32 new, old;
for (;;) {
new = (val & ~_Q_PENDING_MASK) | _Q_LOCKED_VAL;
old = atomic_cmpxchg(&lock->val, val, new);
if (old == val)
break;
val = old;
}
}
/**
* xchg_tail - Put in the new queue tail code word & retrieve previous one
* @lock : Pointer to queue spinlock structure
* @tail : The new queue tail code word
* Return: The previous queue tail code word
*
* xchg(lock, tail)
*
* p,*,* -> n,*,* ; prev = xchg(lock, node)
*/
static __always_inline u32 xchg_tail(struct qspinlock *lock, u32 tail)
{
u32 old, new, val = atomic_read(&lock->val);
for (;;) {
new = (val & _Q_LOCKED_PENDING_MASK) | tail;
old = atomic_cmpxchg(&lock->val, val, new);
if (old == val)
break;
val = old;
}
return old;
}
#endif /* _Q_PENDING_BITS == 8 */
/**
* set_locked - Set the lock bit and own the lock
* @lock: Pointer to queue spinlock structure
*
* *,*,0 -> *,0,1
*/
static __always_inline void set_locked(struct qspinlock *lock)
{
struct __qspinlock *l = (void *)lock;
WRITE_ONCE(l->locked, _Q_LOCKED_VAL);
}
/**
* queue_spin_lock_slowpath - acquire the queue spinlock
* @lock: Pointer to queue spinlock structure
* @val: Current value of the queue spinlock 32-bit word
*
* (queue tail, pending bit, lock value)
*
* fast : slow : unlock
* : :
* uncontended (0,0,0) -:--> (0,0,1) ------------------------------:--> (*,*,0)
* : | ^--------.------. / :
* : v \ \ | :
* pending : (0,1,1) +--> (0,1,0) \ | :
* : | ^--' | | :
* : v | | :
* uncontended : (n,x,y) +--> (n,0,0) --' | :
* queue : | ^--' | :
* : v | :
* contended : (*,x,y) +--> (*,0,0) ---> (*,0,1) -' :
* queue : ^--' :
*/
void queue_spin_lock_slowpath(struct qspinlock *lock, u32 val)
{
struct mcs_spinlock *prev, *next, *node;
u32 new, old, tail;
int idx;
BUILD_BUG_ON(CONFIG_NR_CPUS >= (1U << _Q_TAIL_CPU_BITS));
/*
* wait for in-progress pending->locked hand-overs
*
* 0,1,0 -> 0,0,1
* BluesMan: _Q_PENDING_VAL = 1 << 8 = 256
*/
if (val == _Q_PENDING_VAL) { /* check for the 8th bit set */
while ((val = atomic_read(&lock->val)) == _Q_PENDING_VAL) {
#ifdef CONFIG_DEBUG_QUEUE_SPINLOCK
trace_printk("Waiting for pending->locked handover\n");
#endif
cpu_relax();
}
}
/*
* trylock || pending
*
* 0,0,0 -> 0,0,1 ; trylock
* 0,0,1 -> 0,1,1 ; pending
*/
for (;;) {
/*
* If we observe any contention; queue.
*
* This iS
* 1111 1111 1111 1111 1111 1111 0000 0000 state:
*
* bit fields are:
* 0 - 7 :: locked byte
* 8 - 15 :: pending bit (now, extended to a byte)
* 16 - 17 :: tail index
* 18 - 31 :: tail cpu (+1)
*
* If we have a cpu waiting + pending bit set already,
* just queue.
*
*/
if (val & ~_Q_LOCKED_MASK) {
#ifdef CONFIG_DEBUG_QUEUE_SPINLOCK
trace_printk("contention detected, will make MCS Q\n");
#endif
goto queue;
}
/*
* BluesMan: Here at this state, there is no contention for the
* lock but it is unavailable. So, we will be the first ones to
* get it once it is released. Hence, instead of building the
* queue, we will set the pending bit.
*/
new = _Q_LOCKED_VAL;
if (val == new) {
#ifdef CONFIG_DEBUG_QUEUE_SPINLOCK
trace_printk("we're 1st waiter, setting pending bit\n");
#endif
new |= _Q_PENDING_VAL;
}
/*
* BluesMan: Test code to dump
* if we're having 16 bits
*/
/* if(sizeof(&lock->val) == 2)
WARN_ON(1); */
/*
* BluesMan:
* actual setting of locked + pending bit to val
*/
old = atomic_cmpxchg(&lock->val, val, new);
if (old == val)
break;
val = old;
}
/*
* we won the trylock
*/
if (new == _Q_LOCKED_VAL) {
#ifdef CONFIG_DEBUG_QUEUE_SPINLOCK
trace_printk("we won the trylock, returning from slowpath\n");
#endif
return;
}
/*
* we're pending, wait for the owner to go away.
*
* *,1,1 -> *,1,0
*
* this wait loop must be a load-acquire such that we match the
* store-release that clears the locked bit and create lock
* sequentiality; this is because not all clear_pending_set_locked()
* implementations imply full barriers.
*
* BluesMan: Here the first waiter spins on the main lock!
*/
while ((val = smp_load_acquire(&lock->val.counter)) &
_Q_LOCKED_MASK)
cpu_relax();
/*
* take ownership and clear the pending bit.
*
* *,1,0 -> *,0,1
*
* BluesMan: We were the first waiter and we've got the lock,
* rejoice!
*
*/
clear_pending_set_locked(lock, val);
#ifdef CONFIG_DEBUG_QUEUE_SPINLOCK
trace_printk("pending bit cleared, locked, returning from slowpath\n");
#endif
return;
/*
* End of pending bit optimistic spinning and beginning of MCS
* queuing.
*/
queue:
node = this_cpu_ptr(&mcs_nodes[0]);
/*
* BluesMan: increment nesting count
* However, idx still gets
* old count value though.
*/
idx = node->count++;
tail = encode_tail(smp_processor_id(), idx);
node += idx;
node->locked = 0;
node->next = NULL;
/*
* We touched a (possibly) cold cacheline in the per-cpu queue node;
* attempt the trylock once more in the hope someone let go while we
* weren't watching.
*/
if (queue_spin_trylock(lock)) {
#ifdef CONFIG_DEBUG_QUEUE_SPINLOCK
trace_printk("before MCS, we won trylock; going to release\n");
#endif
goto release;
}
/*
* We have already touched the queueing cacheline; don't bother with
* pending stuff.
*
* p,*,* -> n,*,*
*/
old = xchg_tail(lock, tail);
/*
* if there was a previous node; link it and wait until reaching the
* head of the waitqueue.
*/
if (old & _Q_TAIL_MASK) {
prev = decode_tail(old);
WRITE_ONCE(prev->next, node);
#ifdef CONFIG_DEBUG_QUEUE_SPINLOCK
trace_printk("we've cpu %d ahead of us\n",
(old >> _Q_TAIL_CPU_OFFSET) - 1);
/* trace_dump_this_mcs_node(prev); */
#endif
/*
* BluesMan: Read atomically the locked variable while
* == 0 and wfe
*/
arch_mcs_spin_lock_contended(&node->locked);
}
/*
* we're at the head of the waitqueue, wait for the owner & pending to
* go away.
* Load-acquired is used here because the set_locked()
* function below may not be a full memory barrier.
*
* *,x,y -> *,0,0
*/
while ((val = smp_load_acquire(&lock->val.counter)) &
_Q_LOCKED_PENDING_MASK)
cpu_relax();
/*
* claim the lock:
*
* n,0,0 -> 0,0,1 : lock, uncontended
* *,0,0 -> *,0,1 : lock, contended
*
* If the queue head is the only one in the queue (lock value == tail),
* clear the tail code and grab the lock. Otherwise, we only need
* to grab the lock.
*/
for (;;) {
if (val != tail) {
#ifdef CONFIG_DEBUG_QUEUE_SPINLOCK
trace_printk("contended Q, owner has released\n");
trace_printk("just grab the lock and wait for next\n");
#endif
set_locked(lock);
break;
}
/* val == tail */
old = atomic_cmpxchg(&lock->val, val, _Q_LOCKED_VAL);
if (old == val ) {
#ifdef CONFIG_DEBUG_QUEUE_SPINLOCK
trace_printk("uncontented Q, owner has released\n");
trace_printk("grab the lock, clear the tail and goto release\n");
#endif
goto release; /* No contention */
}
val = old;
}
#ifdef CONFIG_DEBUG_QUEUE_SPINLOCK
trace_printk("contented path\n");
/* trace_dump_this_mcs_node(node->next); */
#endif
/*
* contended path; wait for next, release.
*/
while (!(next = READ_ONCE(node->next)))
cpu_relax();
arch_mcs_spin_unlock_contended(&next->locked);
release:
/*
* BluesMan: Strictly debug code
*/
#ifdef CONFIG_DEBUG_QUEUE_SPINLOCK
trace_printk("we're at release, will dec the percpu ptr\n");
/* trace_dump_per_cpu_mcs_nodes(); */
#endif
/*
* release the node
*/
this_cpu_dec(mcs_nodes[0].count);
}
EXPORT_SYMBOL(queue_spin_lock_slowpath);