LMDB
 All Data Structures Files Functions Variables Typedefs Enumerations Enumerator Macros Groups Pages
Data Structures | Macros
Reader Lock Table
LMDB Internals

Data Structures

struct  MDB_rxbody
 
struct  MDB_reader
 
struct  MDB_txbody
 
struct  MDB_txninfo
 

Macros

#define DEFAULT_READERS   126
 
#define CACHELINE   64
 
#define MDB_LOCK_FORMAT
 

Detailed Description

Readers don't acquire any locks for their data access. Instead, they simply record their transaction ID in the reader table. The reader mutex is needed just to find an empty slot in the reader table. The slot's address is saved in thread-specific data so that subsequent read transactions started by the same thread need no further locking to proceed.

If MDB_NOTLS is set, the slot address is not saved in thread-specific data.

No reader table is used if the database is on a read-only filesystem, or if MDB_NOLOCK is set.

Since the database uses multi-version concurrency control, readers don't actually need any locking. This table is used to keep track of which readers are using data from which old transactions, so that we'll know when a particular old transaction is no longer in use. Old transactions that have discarded any data pages can then have those pages reclaimed for use by a later write transaction.

The lock table is constructed such that reader slots are aligned with the processor's cache line size. Any slot is only ever used by one thread. This alignment guarantees that there will be no contention or cache thrashing as threads update their own slot info, and also eliminates any need for locking when accessing a slot.

A writer thread will scan every slot in the table to determine the oldest outstanding reader transaction. Any freed pages older than this will be reclaimed by the writer. The writer doesn't use any locks when scanning this table. This means that there's no guarantee that the writer will see the most up-to-date reader info, but that's not required for correct operation - all we need is to know the upper bound on the oldest reader, we don't care at all about the newest reader. So the only consequence of reading stale information here is that old pages might hang around a while longer before being reclaimed. That's actually good anyway, because the longer we delay reclaiming old pages, the more likely it is that a string of contiguous pages can be found after coalescing old pages from many old transactions together.


Data Structure Documentation

struct MDB_rxbody

The information we store in a single slot of the reader table. In addition to a transaction ID, we also record the process and thread ID that owns a slot, so that we can detect stale information, e.g. threads or processes that went away without cleaning up.

Note
We currently don't check for stale records. We simply re-init the table when we know that we're the only process opening the lock file.

Data Fields

volatile txnid_t mrb_txnid
 
volatile MDB_PID_T mrb_pid
 
volatile MDB_THR_T mrb_tid
 

Field Documentation

volatile txnid_t MDB_rxbody::mrb_txnid

Current Transaction ID when this transaction began, or (txnid_t)-1. Multiple readers that start at the same time will probably have the same ID here. Again, it's not important to exclude them from anything; all we need to know is which version of the DB they started from so we can avoid overwriting any data used in that particular version.

volatile MDB_PID_T MDB_rxbody::mrb_pid

The process ID of the process owning this reader txn.

volatile MDB_THR_T MDB_rxbody::mrb_tid

The thread ID of the thread owning this txn.

struct MDB_reader

The actual reader record, with cacheline padding.

Data Fields

union {
   MDB_rxbody   mrx
 
   char   pad [(sizeof(MDB_rxbody)+CACHELINE-1)&~(CACHELINE-1)]
 
mru
 

Field Documentation

char MDB_reader::pad[(sizeof(MDB_rxbody)+CACHELINE-1)&~(CACHELINE-1)]

cache line alignment

struct MDB_txbody

The header for the reader table. The table resides in a memory-mapped file. (This is a different file than is used for the main database.)

For POSIX the actual mutexes reside in the shared memory of this mapped file. On Windows, mutexes are named objects allocated by the kernel; we store the mutex names in this mapped file so that other processes can grab them. This same approach is also used on MacOSX/Darwin (using named semaphores) since MacOSX doesn't support process-shared POSIX mutexes. For these cases where a named object is used, the object name is derived from a 64 bit FNV hash of the environment pathname. As such, naming collisions are extremely unlikely. If a collision occurs, the results are unpredictable.

Data Fields

uint32_t mtb_magic
 
uint32_t mtb_format
 
mdb_mutex_t mtb_rmutex
 
volatile txnid_t mtb_txnid
 
volatile unsigned mtb_numreaders
 

Field Documentation

uint32_t MDB_txbody::mtb_magic

Stamp identifying this as an LMDB file. It must be set to MDB_MAGIC.

uint32_t MDB_txbody::mtb_format

Format of this lock file. Must be set to MDB_LOCK_FORMAT.

mdb_mutex_t MDB_txbody::mtb_rmutex

Mutex protecting access to this table. This is the reader table lock used with LOCK_MUTEX().

volatile txnid_t MDB_txbody::mtb_txnid

The ID of the last transaction committed to the database. This is recorded here only for convenience; the value can always be determined by reading the main database meta pages.

volatile unsigned MDB_txbody::mtb_numreaders

The number of slots that have been used in the reader table. This always records the maximum count, it is not decremented when readers release their slots.

struct MDB_txninfo

The actual reader table definition.

Data Fields

union {
   MDB_txbody   mtb
 
   char   pad [(sizeof(MDB_txbody)+CACHELINE-1)&~(CACHELINE-1)]
 
mt1
 
union {
   mdb_mutex_t   mt2_wmutex
 
   char   pad [(MNAME_LEN+CACHELINE-1)&~(CACHELINE-1)]
 
mt2
 
MDB_reader mti_readers [1]
 

Macro Definition Documentation

#define DEFAULT_READERS   126

Number of slots in the reader table. This value was chosen somewhat arbitrarily. 126 readers plus a couple mutexes fit exactly into 8KB on my development machine. Applications should set the table size using mdb_env_set_maxreaders().

#define CACHELINE   64

The size of a CPU cache line in bytes. We want our lock structures aligned to this size to avoid false cache line sharing in the lock table. This value works for most CPUs. For Itanium this should be 128.

#define MDB_LOCK_FORMAT
Value:
((uint32_t) \
/* Flags which describe functionality */ \
+ (((MDB_PIDLOCK) != 0) << 16)))

Lockfile format signature: version, features and field layout