/*
* Allocate and initialize walsender-related shared memory.
*/
void
ReplicationSlotsShmemInit(void)
{
bool found;
if (max_replication_slots == 0)
return;
ReplicationSlotCtl = (ReplicationSlotCtlData *)
ShmemInitStruct("ReplicationSlot Ctl", ReplicationSlotsShmemSize(),
&found);
LWLockRegisterTranche(LWTRANCHE_REPLICATION_SLOT_IO_IN_PROGRESS,
"replication_slot_io");
if (!found)
{
int i;
/* First time through, so initialize */
MemSet(ReplicationSlotCtl, 0, ReplicationSlotsShmemSize());
for (i = 0; i < max_replication_slots; i++)
{
ReplicationSlot *slot = &ReplicationSlotCtl->replication_slots[i];
/* everything else is zeroed by the memset above */
SpinLockInit(&slot->mutex);
LWLockInitialize(&slot->io_in_progress_lock, LWTRANCHE_REPLICATION_SLOT_IO_IN_PROGRESS);
ConditionVariableInit(&slot->active_cv);
}
}
}
/*
* Allocate shmem space for the main LWLock array and initialize it. We also
* register the main tranch here.
*/
void
CreateLWLocks(void)
{
if (!IsUnderPostmaster)
{
int numLocks = NumLWLocks();
Size spaceLocks = LWLockShmemSize();
LWLockPadded *lock;
int *LWLockCounter;
char *ptr;
int id;
/* Allocate space */
ptr = (char *) ShmemAlloc(spaceLocks);
/* Leave room for dynamic allocation of locks and tranches */
ptr += 3 * sizeof(int);
/* Ensure desired alignment of LWLock array */
ptr += LWLOCK_PADDED_SIZE - ((uintptr_t) ptr) % LWLOCK_PADDED_SIZE;
MainLWLockArray = (LWLockPadded *) ptr;
/* Initialize all LWLocks in main array */
for (id = 0, lock = MainLWLockArray; id < numLocks; id++, lock++)
LWLockInitialize(&lock->lock, 0);
/*
* Initialize the dynamic-allocation counters, which are stored just
* before the first LWLock. LWLockCounter[0] is the allocation
* counter for lwlocks, LWLockCounter[1] is the maximum number that
* can be allocated from the main array, and LWLockCounter[2] is the
* allocation counter for tranches.
*/
LWLockCounter = (int *) ((char *) MainLWLockArray - 3 * sizeof(int));
LWLockCounter[0] = NUM_FIXED_LWLOCKS;
LWLockCounter[1] = numLocks;
LWLockCounter[2] = 1; /* 0 is the main array */
}
if (LWLockTrancheArray == NULL)
{
LWLockTranchesAllocated = 16;
LWLockTrancheArray = (LWLockTranche **)
MemoryContextAlloc(TopMemoryContext,
LWLockTranchesAllocated * sizeof(LWLockTranche *));
}
MainLWLockTranche.name = "main";
MainLWLockTranche.array_base = MainLWLockArray;
MainLWLockTranche.array_stride = sizeof(LWLockPadded);
LWLockRegisterTranche(0, &MainLWLockTranche);
}
/*
* IPCMessageBrokerShmemInit
*/
void
IPCMessageBrokerShmemInit(void)
{
bool found;
Size size = IPCMessageBrokerShmemSize();
LWLockAcquire(IPCMessageBrokerIndexLock, LW_EXCLUSIVE);
broker_meta = ShmemInitStruct("BrokerMeta", size, &found);
if (!found)
{
HASHCTL ctl;
int i;
MemSet(broker_meta, 0, size);
pg_atomic_init_flag(&broker_meta->waiting);
MemSet(&ctl, 0, sizeof(HASHCTL));
ctl.keysize = sizeof(Oid);
ctl.entrysize = broker_db_meta_size;
ctl.hash = oid_hash;
broker_meta->db_meta_hash = ShmemInitHash("BrokerDBMetaHash", 4, 16, &ctl, HASH_ELEM | HASH_FUNCTION);
/* Initialize LWLocks that we'll use for all IPC queues */
broker_meta->tranche_id = LWLockNewTrancheId();
broker_meta->tranche.name = "BrokerMetaLWLocks";
broker_meta->tranche.array_base = (void *) broker_meta->locks;
broker_meta->tranche.array_stride = sizeof(lw_lock_slot);
for (i = 0; i < max_worker_processes; i++)
{
lw_lock_slot *slot = &broker_meta->locks[i];
LWLockInitialize(&slot->lock, broker_meta->tranche_id);
slot->dbid = InvalidOid;
}
}
LWLockRelease(IPCMessageBrokerIndexLock);
LWLockRegisterTranche(broker_meta->tranche_id, &broker_meta->tranche);
}
/*
* Allocate and initialize walsender-related shared memory.
*/
void
ReplicationSlotsShmemInit(void)
{
bool found;
if (max_replication_slots == 0)
return;
ReplicationSlotCtl = (ReplicationSlotCtlData *)
ShmemInitStruct("ReplicationSlot Ctl", ReplicationSlotsShmemSize(),
&found);
ReplSlotIOLWLockTranche.name = "Replication Slot IO Locks";
ReplSlotIOLWLockTranche.array_base =
((char *) ReplicationSlotCtl) + offsetof(ReplicationSlotCtlData, replication_slots) +offsetof(ReplicationSlot, io_in_progress_lock);
ReplSlotIOLWLockTranche.array_stride = sizeof(ReplicationSlot);
LWLockRegisterTranche(LWTRANCHE_REPLICATION_SLOT_IO_IN_PROGRESS,
&ReplSlotIOLWLockTranche);
if (!found)
{
int i;
/* First time through, so initialize */
MemSet(ReplicationSlotCtl, 0, ReplicationSlotsShmemSize());
for (i = 0; i < max_replication_slots; i++)
{
ReplicationSlot *slot = &ReplicationSlotCtl->replication_slots[i];
/* everything else is zeroed by the memset above */
SpinLockInit(&slot->mutex);
LWLockInitialize(&slot->io_in_progress_lock, LWTRANCHE_REPLICATION_SLOT_IO_IN_PROGRESS);
}
}
}
/*
* Initialize shared buffer pool
*
* This is called once during shared-memory initialization (either in the
* postmaster, or in a standalone backend).
*/
void
InitBufferPool(void)
{
bool foundBufs,
foundDescs,
foundIOLocks,
foundBufCkpt;
/* Align descriptors to a cacheline boundary. */
BufferDescriptors = (BufferDescPadded *)
ShmemInitStruct("Buffer Descriptors",
NBuffers * sizeof(BufferDescPadded),
&foundDescs);
BufferBlocks = (char *)
ShmemInitStruct("Buffer Blocks",
NBuffers * (Size) BLCKSZ, &foundBufs);
/* Align lwlocks to cacheline boundary */
BufferIOLWLockArray = (LWLockMinimallyPadded *)
ShmemInitStruct("Buffer IO Locks",
NBuffers * (Size) sizeof(LWLockMinimallyPadded),
&foundIOLocks);
LWLockRegisterTranche(LWTRANCHE_BUFFER_IO_IN_PROGRESS, "buffer_io");
LWLockRegisterTranche(LWTRANCHE_BUFFER_CONTENT, "buffer_content");
/*
* The array used to sort to-be-checkpointed buffer ids is located in
* shared memory, to avoid having to allocate significant amounts of
* memory at runtime. As that'd be in the middle of a checkpoint, or when
* the checkpointer is restarted, memory allocation failures would be
* painful.
*/
CkptBufferIds = (CkptSortItem *)
ShmemInitStruct("Checkpoint BufferIds",
NBuffers * sizeof(CkptSortItem), &foundBufCkpt);
if (foundDescs || foundBufs || foundIOLocks || foundBufCkpt)
{
/* should find all of these, or none of them */
Assert(foundDescs && foundBufs && foundIOLocks && foundBufCkpt);
/* note: this path is only taken in EXEC_BACKEND case */
}
else
{
int i;
/*
* Initialize all the buffer headers.
*/
for (i = 0; i < NBuffers; i++)
{
BufferDesc *buf = GetBufferDescriptor(i);
CLEAR_BUFFERTAG(buf->tag);
pg_atomic_init_u32(&buf->state, 0);
buf->wait_backend_pid = 0;
buf->buf_id = i;
/*
* Initially link all the buffers together as unused. Subsequent
* management of this list is done by freelist.c.
*/
buf->freeNext = i + 1;
LWLockInitialize(BufferDescriptorGetContentLock(buf),
LWTRANCHE_BUFFER_CONTENT);
LWLockInitialize(BufferDescriptorGetIOLock(buf),
LWTRANCHE_BUFFER_IO_IN_PROGRESS);
}
/* Correct last entry of linked list */
GetBufferDescriptor(NBuffers - 1)->freeNext = FREENEXT_END_OF_LIST;
}
/* Init other shared buffer-management stuff */
StrategyInitialize(!foundDescs);
/* Initialize per-backend file flush context */
WritebackContextInit(&BackendWritebackContext,
&backend_flush_after);
}
/*
* Initialize shared buffer pool
*
* This is called once during shared-memory initialization (either in the
* postmaster, or in a standalone backend).
*/
void
InitBufferPool(void)
{
bool foundBufs,
foundDescs,
foundIOLocks;
/* Align descriptors to a cacheline boundary. */
BufferDescriptors = (BufferDescPadded *)
CACHELINEALIGN(
ShmemInitStruct("Buffer Descriptors",
NBuffers * sizeof(BufferDescPadded)
+ PG_CACHE_LINE_SIZE,
&foundDescs));
BufferBlocks = (char *)
ShmemInitStruct("Buffer Blocks",
NBuffers * (Size) BLCKSZ, &foundBufs);
/* Align lwlocks to cacheline boundary */
BufferIOLWLockArray = (LWLockMinimallyPadded *)
CACHELINEALIGN(ShmemInitStruct("Buffer IO Locks",
NBuffers * (Size) sizeof(LWLockMinimallyPadded)
+ PG_CACHE_LINE_SIZE,
&foundIOLocks));
BufferIOLWLockTranche.name = "buffer_io";
BufferIOLWLockTranche.array_base = BufferIOLWLockArray;
BufferIOLWLockTranche.array_stride = sizeof(LWLockMinimallyPadded);
LWLockRegisterTranche(LWTRANCHE_BUFFER_IO_IN_PROGRESS,
&BufferIOLWLockTranche);
BufferContentLWLockTranche.name = "buffer_content";
BufferContentLWLockTranche.array_base =
((char *) BufferDescriptors) + offsetof(BufferDesc, content_lock);
BufferContentLWLockTranche.array_stride = sizeof(BufferDescPadded);
LWLockRegisterTranche(LWTRANCHE_BUFFER_CONTENT,
&BufferContentLWLockTranche);
if (foundDescs || foundBufs || foundIOLocks)
{
/* should find all of these, or none of them */
Assert(foundDescs && foundBufs && foundIOLocks);
/* note: this path is only taken in EXEC_BACKEND case */
}
else
{
int i;
/*
* Initialize all the buffer headers.
*/
for (i = 0; i < NBuffers; i++)
{
BufferDesc *buf = GetBufferDescriptor(i);
CLEAR_BUFFERTAG(buf->tag);
buf->flags = 0;
buf->usage_count = 0;
buf->refcount = 0;
buf->wait_backend_pid = 0;
SpinLockInit(&buf->buf_hdr_lock);
buf->buf_id = i;
/*
* Initially link all the buffers together as unused. Subsequent
* management of this list is done by freelist.c.
*/
buf->freeNext = i + 1;
LWLockInitialize(BufferDescriptorGetContentLock(buf),
LWTRANCHE_BUFFER_CONTENT);
LWLockInitialize(BufferDescriptorGetIOLock(buf),
LWTRANCHE_BUFFER_IO_IN_PROGRESS);
}
/* Correct last entry of linked list */
GetBufferDescriptor(NBuffers - 1)->freeNext = FREENEXT_END_OF_LIST;
}
/* Init other shared buffer-management stuff */
StrategyInitialize(!foundDescs);
}
void
SimpleLruInit(SlruCtl ctl, const char *name, int nslots, int nlsns,
LWLock *ctllock, const char *subdir, int tranche_id)
{
SlruShared shared;
bool found;
shared = (SlruShared) ShmemInitStruct(name,
SimpleLruShmemSize(nslots, nlsns),
&found);
if (!IsUnderPostmaster)
{
/* Initialize locks and shared memory area */
char *ptr;
Size offset;
int slotno;
Assert(!found);
memset(shared, 0, sizeof(SlruSharedData));
shared->ControlLock = ctllock;
shared->num_slots = nslots;
shared->lsn_groups_per_page = nlsns;
shared->cur_lru_count = 0;
/* shared->latest_page_number will be set later */
ptr = (char *) shared;
offset = MAXALIGN(sizeof(SlruSharedData));
shared->page_buffer = (char **) (ptr + offset);
offset += MAXALIGN(nslots * sizeof(char *));
shared->page_status = (SlruPageStatus *) (ptr + offset);
offset += MAXALIGN(nslots * sizeof(SlruPageStatus));
shared->page_dirty = (bool *) (ptr + offset);
offset += MAXALIGN(nslots * sizeof(bool));
shared->page_number = (int *) (ptr + offset);
offset += MAXALIGN(nslots * sizeof(int));
shared->page_lru_count = (int *) (ptr + offset);
offset += MAXALIGN(nslots * sizeof(int));
if (nlsns > 0)
{
shared->group_lsn = (XLogRecPtr *) (ptr + offset);
offset += MAXALIGN(nslots * nlsns * sizeof(XLogRecPtr));
}
/* Initialize LWLocks */
shared->buffer_locks = (LWLockPadded *) ShmemAlloc(sizeof(LWLockPadded) * nslots);
Assert(strlen(name) + 1 < SLRU_MAX_NAME_LENGTH);
strlcpy(shared->lwlock_tranche_name, name, SLRU_MAX_NAME_LENGTH);
shared->lwlock_tranche_id = tranche_id;
ptr += BUFFERALIGN(offset);
for (slotno = 0; slotno < nslots; slotno++)
{
LWLockInitialize(&shared->buffer_locks[slotno].lock,
shared->lwlock_tranche_id);
shared->page_buffer[slotno] = ptr;
shared->page_status[slotno] = SLRU_PAGE_EMPTY;
shared->page_dirty[slotno] = false;
shared->page_lru_count[slotno] = 0;
ptr += BLCKSZ;
}
}
else
Assert(found);
/* Register SLRU tranche in the main tranches array */
LWLockRegisterTranche(shared->lwlock_tranche_id,
shared->lwlock_tranche_name);
/*
* Initialize the unshared control struct, including directory path. We
* assume caller set PagePrecedes.
*/
ctl->shared = shared;
ctl->do_fsync = true; /* default behavior */
StrNCpy(ctl->Dir, subdir, sizeof(ctl->Dir));
}
/*
* BackendManagementShmemInit is the callback that is to be called on shared
* memory startup hook. The function sets up the necessary shared memory
* segment for the backend manager.
*/
static void
BackendManagementShmemInit(void)
{
bool alreadyInitialized = false;
/* we may update the shmem, acquire lock exclusively */
LWLockAcquire(AddinShmemInitLock, LW_EXCLUSIVE);
backendManagementShmemData =
(BackendManagementShmemData *) ShmemInitStruct(
"Backend Management Shmem",
BackendManagementShmemSize(),
&alreadyInitialized);
if (!alreadyInitialized)
{
int backendIndex = 0;
char *trancheName = "Backend Management Tranche";
#if (PG_VERSION_NUM >= 100000)
NamedLWLockTranche *namedLockTranche =
&backendManagementShmemData->namedLockTranche;
#else
LWLockTranche *lockTranche = &backendManagementShmemData->lockTranche;
#endif
/* start by zeroing out all the memory */
memset(backendManagementShmemData, 0,
BackendManagementShmemSize());
#if (PG_VERSION_NUM >= 100000)
namedLockTranche->trancheId = LWLockNewTrancheId();
LWLockRegisterTranche(namedLockTranche->trancheId, trancheName);
LWLockInitialize(&backendManagementShmemData->lock,
namedLockTranche->trancheId);
#else
backendManagementShmemData->trancheId = LWLockNewTrancheId();
/* we only need a single lock */
lockTranche->array_base = &backendManagementShmemData->lock;
lockTranche->array_stride = sizeof(LWLock);
lockTranche->name = trancheName;
LWLockRegisterTranche(backendManagementShmemData->trancheId, lockTranche);
LWLockInitialize(&backendManagementShmemData->lock,
backendManagementShmemData->trancheId);
#endif
/* start the distributed transaction ids from 1 */
pg_atomic_init_u64(&backendManagementShmemData->nextTransactionNumber, 1);
/*
* We need to init per backend's spinlock before any backend
* starts its execution.
*/
for (backendIndex = 0; backendIndex < MaxBackends; ++backendIndex)
{
SpinLockInit(&backendManagementShmemData->backends[backendIndex].mutex);
}
}
LWLockRelease(AddinShmemInitLock);
if (prev_shmem_startup_hook != NULL)
{
prev_shmem_startup_hook();
}
}
bool
ora_lock_shmem(size_t size, int max_pipes, int max_events, int max_locks, bool reset)
{
int i;
bool found;
sh_memory *sh_mem;
if (pipes == NULL)
{
sh_mem = ShmemInitStruct("dbms_pipe", size, &found);
if (sh_mem == NULL)
ereport(FATAL,
(errcode(ERRCODE_OUT_OF_MEMORY),
errmsg("out of memory"),
errdetail("Failed while allocation block %lu bytes in shared memory.", (unsigned long) size)));
if (!found)
{
#if PG_VERSION_NUM >= 90600
sh_mem->tranche_id = LWLockNewTrancheId();
LWLockInitialize(&sh_mem->shmem_lock, sh_mem->tranche_id);
{
#if PG_VERSION_NUM >= 100000
LWLockRegisterTranche(sh_mem->tranche_id, "orafce");
#else
static LWLockTranche tranche;
tranche.name = "orafce";
tranche.array_base = &sh_mem->shmem_lock;
tranche.array_stride = sizeof(LWLock);
LWLockRegisterTranche(sh_mem->tranche_id, &tranche);
#endif
shmem_lockid = &sh_mem->shmem_lock;
}
#else
shmem_lockid = sh_mem->shmem_lockid = LWLockAssign();
#endif
LWLockAcquire(shmem_lockid, LW_EXCLUSIVE);
sh_mem->size = size - sh_memory_size;
ora_sinit(sh_mem->data, size, true);
pipes = sh_mem->pipes = ora_salloc(max_pipes*sizeof(pipe));
sid = sh_mem->sid = 1;
for (i = 0; i < max_pipes; i++)
pipes[i].is_valid = false;
events = sh_mem->events = ora_salloc(max_events*sizeof(alert_event));
locks = sh_mem->locks = ora_salloc(max_locks*sizeof(alert_lock));
for (i = 0; i < max_events; i++)
{
events[i].event_name = NULL;
events[i].max_receivers = 0;
events[i].receivers = NULL;
events[i].messages = NULL;
}
for (i = 0; i < max_locks; i++)
{
locks[i].sid = -1;
locks[i].echo = NULL;
}
}
else if (pipes == NULL)
{
#if PG_VERSION_NUM >= 90600
#if PG_VERSION_NUM >= 100000
LWLockRegisterTranche(sh_mem->tranche_id, "orafce");
#else
static LWLockTranche tranche;
tranche.name = "orafce";
tranche.array_base = &sh_mem->shmem_lock;
tranche.array_stride = sizeof(LWLock);
LWLockRegisterTranche(sh_mem->tranche_id, &tranche);
#endif
shmem_lockid = &sh_mem->shmem_lock;
#else
shmem_lockid = sh_mem->shmem_lockid;
#endif
pipes = sh_mem->pipes;
LWLockAcquire(shmem_lockid, LW_EXCLUSIVE);
ora_sinit(sh_mem->data, sh_mem->size, reset);
sid = ++(sh_mem->sid);
events = sh_mem->events;
locks = sh_mem->locks;
}
}
else
{
LWLockAcquire(shmem_lockid, LW_EXCLUSIVE);
}
return pipes != NULL;
}
