cachepc-linux

Fork of AMDESE/linux with modifications for CachePC side-channel attack
git clone https://git.sinitax.com/sinitax/cachepc-linux
Log | Files | Refs | README | LICENSE | sfeed.txt

circular-buffers.rst (8364B)


      1================
      2Circular Buffers
      3================
      4
      5:Author: David Howells <dhowells@redhat.com>
      6:Author: Paul E. McKenney <paulmck@linux.ibm.com>
      7
      8
      9Linux provides a number of features that can be used to implement circular
     10buffering.  There are two sets of such features:
     11
     12 (1) Convenience functions for determining information about power-of-2 sized
     13     buffers.
     14
     15 (2) Memory barriers for when the producer and the consumer of objects in the
     16     buffer don't want to share a lock.
     17
     18To use these facilities, as discussed below, there needs to be just one
     19producer and just one consumer.  It is possible to handle multiple producers by
     20serialising them, and to handle multiple consumers by serialising them.
     21
     22
     23.. Contents:
     24
     25 (*) What is a circular buffer?
     26
     27 (*) Measuring power-of-2 buffers.
     28
     29 (*) Using memory barriers with circular buffers.
     30     - The producer.
     31     - The consumer.
     32
     33
     34
     35What is a circular buffer?
     36==========================
     37
     38First of all, what is a circular buffer?  A circular buffer is a buffer of
     39fixed, finite size into which there are two indices:
     40
     41 (1) A 'head' index - the point at which the producer inserts items into the
     42     buffer.
     43
     44 (2) A 'tail' index - the point at which the consumer finds the next item in
     45     the buffer.
     46
     47Typically when the tail pointer is equal to the head pointer, the buffer is
     48empty; and the buffer is full when the head pointer is one less than the tail
     49pointer.
     50
     51The head index is incremented when items are added, and the tail index when
     52items are removed.  The tail index should never jump the head index, and both
     53indices should be wrapped to 0 when they reach the end of the buffer, thus
     54allowing an infinite amount of data to flow through the buffer.
     55
     56Typically, items will all be of the same unit size, but this isn't strictly
     57required to use the techniques below.  The indices can be increased by more
     58than 1 if multiple items or variable-sized items are to be included in the
     59buffer, provided that neither index overtakes the other.  The implementer must
     60be careful, however, as a region more than one unit in size may wrap the end of
     61the buffer and be broken into two segments.
     62
     63Measuring power-of-2 buffers
     64============================
     65
     66Calculation of the occupancy or the remaining capacity of an arbitrarily sized
     67circular buffer would normally be a slow operation, requiring the use of a
     68modulus (divide) instruction.  However, if the buffer is of a power-of-2 size,
     69then a much quicker bitwise-AND instruction can be used instead.
     70
     71Linux provides a set of macros for handling power-of-2 circular buffers.  These
     72can be made use of by::
     73
     74	#include <linux/circ_buf.h>
     75
     76The macros are:
     77
     78 (#) Measure the remaining capacity of a buffer::
     79
     80	CIRC_SPACE(head_index, tail_index, buffer_size);
     81
     82     This returns the amount of space left in the buffer[1] into which items
     83     can be inserted.
     84
     85
     86 (#) Measure the maximum consecutive immediate space in a buffer::
     87
     88	CIRC_SPACE_TO_END(head_index, tail_index, buffer_size);
     89
     90     This returns the amount of consecutive space left in the buffer[1] into
     91     which items can be immediately inserted without having to wrap back to the
     92     beginning of the buffer.
     93
     94
     95 (#) Measure the occupancy of a buffer::
     96
     97	CIRC_CNT(head_index, tail_index, buffer_size);
     98
     99     This returns the number of items currently occupying a buffer[2].
    100
    101
    102 (#) Measure the non-wrapping occupancy of a buffer::
    103
    104	CIRC_CNT_TO_END(head_index, tail_index, buffer_size);
    105
    106     This returns the number of consecutive items[2] that can be extracted from
    107     the buffer without having to wrap back to the beginning of the buffer.
    108
    109
    110Each of these macros will nominally return a value between 0 and buffer_size-1,
    111however:
    112
    113 (1) CIRC_SPACE*() are intended to be used in the producer.  To the producer
    114     they will return a lower bound as the producer controls the head index,
    115     but the consumer may still be depleting the buffer on another CPU and
    116     moving the tail index.
    117
    118     To the consumer it will show an upper bound as the producer may be busy
    119     depleting the space.
    120
    121 (2) CIRC_CNT*() are intended to be used in the consumer.  To the consumer they
    122     will return a lower bound as the consumer controls the tail index, but the
    123     producer may still be filling the buffer on another CPU and moving the
    124     head index.
    125
    126     To the producer it will show an upper bound as the consumer may be busy
    127     emptying the buffer.
    128
    129 (3) To a third party, the order in which the writes to the indices by the
    130     producer and consumer become visible cannot be guaranteed as they are
    131     independent and may be made on different CPUs - so the result in such a
    132     situation will merely be a guess, and may even be negative.
    133
    134Using memory barriers with circular buffers
    135===========================================
    136
    137By using memory barriers in conjunction with circular buffers, you can avoid
    138the need to:
    139
    140 (1) use a single lock to govern access to both ends of the buffer, thus
    141     allowing the buffer to be filled and emptied at the same time; and
    142
    143 (2) use atomic counter operations.
    144
    145There are two sides to this: the producer that fills the buffer, and the
    146consumer that empties it.  Only one thing should be filling a buffer at any one
    147time, and only one thing should be emptying a buffer at any one time, but the
    148two sides can operate simultaneously.
    149
    150
    151The producer
    152------------
    153
    154The producer will look something like this::
    155
    156	spin_lock(&producer_lock);
    157
    158	unsigned long head = buffer->head;
    159	/* The spin_unlock() and next spin_lock() provide needed ordering. */
    160	unsigned long tail = READ_ONCE(buffer->tail);
    161
    162	if (CIRC_SPACE(head, tail, buffer->size) >= 1) {
    163		/* insert one item into the buffer */
    164		struct item *item = buffer[head];
    165
    166		produce_item(item);
    167
    168		smp_store_release(buffer->head,
    169				  (head + 1) & (buffer->size - 1));
    170
    171		/* wake_up() will make sure that the head is committed before
    172		 * waking anyone up */
    173		wake_up(consumer);
    174	}
    175
    176	spin_unlock(&producer_lock);
    177
    178This will instruct the CPU that the contents of the new item must be written
    179before the head index makes it available to the consumer and then instructs the
    180CPU that the revised head index must be written before the consumer is woken.
    181
    182Note that wake_up() does not guarantee any sort of barrier unless something
    183is actually awakened.  We therefore cannot rely on it for ordering.  However,
    184there is always one element of the array left empty.  Therefore, the
    185producer must produce two elements before it could possibly corrupt the
    186element currently being read by the consumer.  Therefore, the unlock-lock
    187pair between consecutive invocations of the consumer provides the necessary
    188ordering between the read of the index indicating that the consumer has
    189vacated a given element and the write by the producer to that same element.
    190
    191
    192The Consumer
    193------------
    194
    195The consumer will look something like this::
    196
    197	spin_lock(&consumer_lock);
    198
    199	/* Read index before reading contents at that index. */
    200	unsigned long head = smp_load_acquire(buffer->head);
    201	unsigned long tail = buffer->tail;
    202
    203	if (CIRC_CNT(head, tail, buffer->size) >= 1) {
    204
    205		/* extract one item from the buffer */
    206		struct item *item = buffer[tail];
    207
    208		consume_item(item);
    209
    210		/* Finish reading descriptor before incrementing tail. */
    211		smp_store_release(buffer->tail,
    212				  (tail + 1) & (buffer->size - 1));
    213	}
    214
    215	spin_unlock(&consumer_lock);
    216
    217This will instruct the CPU to make sure the index is up to date before reading
    218the new item, and then it shall make sure the CPU has finished reading the item
    219before it writes the new tail pointer, which will erase the item.
    220
    221Note the use of READ_ONCE() and smp_load_acquire() to read the
    222opposition index.  This prevents the compiler from discarding and
    223reloading its cached value.  This isn't strictly needed if you can
    224be sure that the opposition index will _only_ be used the once.
    225The smp_load_acquire() additionally forces the CPU to order against
    226subsequent memory references.  Similarly, smp_store_release() is used
    227in both algorithms to write the thread's index.  This documents the
    228fact that we are writing to something that can be read concurrently,
    229prevents the compiler from tearing the store, and enforces ordering
    230against previous accesses.
    231
    232
    233Further reading
    234===============
    235
    236See also Documentation/memory-barriers.txt for a description of Linux's memory
    237barrier facilities.