/*
* Find a page in a shared buffer, reading it in if necessary.
* The page number must correspond to an already-initialized page.
*
* If write_ok is true then it is OK to return a page that is in
* WRITE_IN_PROGRESS state; it is the caller's responsibility to be sure
* that modification of the page is safe. If write_ok is false then we
* will not return the page until it is not undergoing active I/O.
*
* The passed-in xid is used only for error reporting, and may be
* InvalidTransactionId if no specific xid is associated with the action.
*
* Return value is the shared-buffer slot number now holding the page.
* The buffer's LRU access info is updated.
*
* Control lock must be held at entry, and will be held at exit.
*/
int
SimpleLruReadPage(SlruCtl ctl, int pageno, bool write_ok,
TransactionId xid)
{
SlruShared shared = ctl->shared;
/* Outer loop handles restart if we must wait for someone else's I/O */
for (;;)
{
int slotno;
bool ok;
/* See if page already is in memory; if not, pick victim slot */
slotno = SlruSelectLRUPage(ctl, pageno);
/* Did we find the page in memory? */
if (shared->page_number[slotno] == pageno &&
shared->page_status[slotno] != SLRU_PAGE_EMPTY)
{
/*
* If page is still being read in, we must wait for I/O. Likewise
* if the page is being written and the caller said that's not OK.
*/
if (shared->page_status[slotno] == SLRU_PAGE_READ_IN_PROGRESS ||
(shared->page_status[slotno] == SLRU_PAGE_WRITE_IN_PROGRESS &&
!write_ok))
{
SimpleLruWaitIO(ctl, slotno);
/* Now we must recheck state from the top */
continue;
}
/* Otherwise, it's ready to use */
SlruRecentlyUsed(shared, slotno);
return slotno;
}
/* We found no match; assert we selected a freeable slot */
Assert(shared->page_status[slotno] == SLRU_PAGE_EMPTY ||
(shared->page_status[slotno] == SLRU_PAGE_VALID &&
!shared->page_dirty[slotno]));
/* Mark the slot read-busy */
shared->page_number[slotno] = pageno;
shared->page_status[slotno] = SLRU_PAGE_READ_IN_PROGRESS;
shared->page_dirty[slotno] = false;
/* Acquire per-buffer lock (cannot deadlock, see notes at top) */
LWLockAcquire(shared->buffer_locks[slotno], LW_EXCLUSIVE);
/* Release control lock while doing I/O */
LWLockRelease(shared->ControlLock);
/* Do the read */
ok = SlruPhysicalReadPage(ctl, pageno, slotno);
/* Set the LSNs for this newly read-in page to zero */
SimpleLruZeroLSNs(ctl, slotno);
/* Re-acquire control lock and update page state */
LWLockAcquire(shared->ControlLock, LW_EXCLUSIVE);
Assert(shared->page_number[slotno] == pageno &&
shared->page_status[slotno] == SLRU_PAGE_READ_IN_PROGRESS &&
!shared->page_dirty[slotno]);
shared->page_status[slotno] = ok ? SLRU_PAGE_VALID : SLRU_PAGE_EMPTY;
LWLockRelease(shared->buffer_locks[slotno]);
/* Now it's okay to ereport if we failed */
if (!ok)
SlruReportIOError(ctl, pageno, xid);
SlruRecentlyUsed(shared, slotno);
return slotno;
}
}
/*
* SIInsertDataEntries
* Add new invalidation message(s) to the buffer.
*/
void
SIInsertDataEntries(const SharedInvalidationMessage *data, int n)
{
SISeg *segP = shmInvalBuffer;
/*
* N can be arbitrarily large. We divide the work into groups of no more
* than WRITE_QUANTUM messages, to be sure that we don't hold the lock for
* an unreasonably long time. (This is not so much because we care about
* letting in other writers, as that some just-caught-up backend might be
* trying to do SICleanupQueue to pass on its signal, and we don't want it
* to have to wait a long time.) Also, we need to consider calling
* SICleanupQueue every so often.
*/
while (n > 0)
{
int nthistime = Min(n, WRITE_QUANTUM);
int numMsgs;
int max;
n -= nthistime;
LWLockAcquire(SInvalWriteLock, LW_EXCLUSIVE);
/*
* If the buffer is full, we *must* acquire some space. Clean the
* queue and reset anyone who is preventing space from being freed.
* Otherwise, clean the queue only when it's exceeded the next
* fullness threshold. We have to loop and recheck the buffer state
* after any call of SICleanupQueue.
*/
for (;;)
{
numMsgs = segP->maxMsgNum - segP->minMsgNum;
if (numMsgs + nthistime > MAXNUMMESSAGES ||
numMsgs >= segP->nextThreshold)
SICleanupQueue(true, nthistime);
else
break;
}
/*
* Insert new message(s) into proper slot of circular buffer
*/
max = segP->maxMsgNum;
while (nthistime-- > 0)
{
segP->buffer[max % MAXNUMMESSAGES] = *data++;
max++;
}
/* Update current value of maxMsgNum using spinlock */
{
/* use volatile pointer to prevent code rearrangement */
volatile SISeg *vsegP = segP;
SpinLockAcquire(&vsegP->msgnumLock);
vsegP->maxMsgNum = max;
SpinLockRelease(&vsegP->msgnumLock);
}
LWLockRelease(SInvalWriteLock);
}
}
/*
* SICleanupQueue
* Remove messages that have been consumed by all active backends
*
* callerHasWriteLock is TRUE if caller is holding SInvalWriteLock.
* minFree is the minimum number of message slots to make free.
*
* Possible side effects of this routine include marking one or more
* backends as "reset" in the array, and sending PROCSIG_CATCHUP_INTERRUPT
* to some backend that seems to be getting too far behind. We signal at
* most one backend at a time, for reasons explained at the top of the file.
*
* Caution: because we transiently release write lock when we have to signal
* some other backend, it is NOT guaranteed that there are still minFree
* free message slots at exit. Caller must recheck and perhaps retry.
*/
void
SICleanupQueue(bool callerHasWriteLock, int minFree)
{
SISeg *segP = shmInvalBuffer;
int min,
minsig,
lowbound,
numMsgs,
i;
ProcState *needSig = NULL;
/* Lock out all writers and readers */
if (!callerHasWriteLock)
LWLockAcquire(SInvalWriteLock, LW_EXCLUSIVE);
LWLockAcquire(SInvalReadLock, LW_EXCLUSIVE);
/*
* Recompute minMsgNum = minimum of all backends' nextMsgNum, identify the
* furthest-back backend that needs signaling (if any), and reset any
* backends that are too far back. Note that because we ignore sendOnly
* backends here it is possible for them to keep sending messages without
* a problem even when they are the only active backend.
*/
min = segP->maxMsgNum;
minsig = min - SIG_THRESHOLD;
lowbound = min - MAXNUMMESSAGES + minFree;
for (i = 0; i < segP->lastBackend; i++)
{
ProcState *stateP = &segP->procState[i];
int n = stateP->nextMsgNum;
/* Ignore if inactive or already in reset state */
if (stateP->procPid == 0 || stateP->resetState || stateP->sendOnly)
continue;
/*
* If we must free some space and this backend is preventing it, force
* him into reset state and then ignore until he catches up.
*/
if (n < lowbound)
{
stateP->resetState = true;
/* no point in signaling him ... */
continue;
}
/* Track the global minimum nextMsgNum */
if (n < min)
min = n;
/* Also see who's furthest back of the unsignaled backends */
if (n < minsig && !stateP->signaled)
{
minsig = n;
needSig = stateP;
}
}
segP->minMsgNum = min;
/*
* When minMsgNum gets really large, decrement all message counters so as
* to forestall overflow of the counters. This happens seldom enough that
* folding it into the previous loop would be a loser.
*/
if (min >= MSGNUMWRAPAROUND)
{
segP->minMsgNum -= MSGNUMWRAPAROUND;
segP->maxMsgNum -= MSGNUMWRAPAROUND;
for (i = 0; i < segP->lastBackend; i++)
{
/* we don't bother skipping inactive entries here */
segP->procState[i].nextMsgNum -= MSGNUMWRAPAROUND;
}
}
/*
* Determine how many messages are still in the queue, and set the
* threshold at which we should repeat SICleanupQueue().
*/
numMsgs = segP->maxMsgNum - segP->minMsgNum;
if (numMsgs < CLEANUP_MIN)
segP->nextThreshold = CLEANUP_MIN;
else
segP->nextThreshold = (numMsgs / CLEANUP_QUANTUM + 1) * CLEANUP_QUANTUM;
/*
* Lastly, signal anyone who needs a catchup interrupt. Since
* SendProcSignal() might not be fast, we don't want to hold locks while
* executing it.
*/
if (needSig)
{
pid_t his_pid = needSig->procPid;
BackendId his_backendId = (needSig - &segP->procState[0]) + 1;
needSig->signaled = true;
LWLockRelease(SInvalReadLock);
LWLockRelease(SInvalWriteLock);
elog(DEBUG4, "sending sinval catchup signal to PID %d", (int) his_pid);
SendProcSignal(his_pid, PROCSIG_CATCHUP_INTERRUPT, his_backendId);
if (callerHasWriteLock)
LWLockAcquire(SInvalWriteLock, LW_EXCLUSIVE);
}
else
{
LWLockRelease(SInvalReadLock);
if (!callerHasWriteLock)
LWLockRelease(SInvalWriteLock);
}
}
/*
* CommitTS resource manager's routines
*/
void
commit_ts_redo(XLogReaderState *record)
{
uint8 info = XLogRecGetInfo(record) & ~XLR_INFO_MASK;
/* Backup blocks are not used in commit_ts records */
Assert(!XLogRecHasAnyBlockRefs(record));
if (info == COMMIT_TS_ZEROPAGE)
{
int pageno;
int slotno;
memcpy(&pageno, XLogRecGetData(record), sizeof(int));
LWLockAcquire(CommitTsControlLock, LW_EXCLUSIVE);
slotno = ZeroCommitTsPage(pageno, false);
SimpleLruWritePage(CommitTsCtl, slotno);
Assert(!CommitTsCtl->shared->page_dirty[slotno]);
LWLockRelease(CommitTsControlLock);
}
else if (info == COMMIT_TS_TRUNCATE)
{
int pageno;
memcpy(&pageno, XLogRecGetData(record), sizeof(int));
/*
* During XLOG replay, latest_page_number isn't set up yet; insert a
* suitable value to bypass the sanity test in SimpleLruTruncate.
*/
CommitTsCtl->shared->latest_page_number = pageno;
SimpleLruTruncate(CommitTsCtl, pageno);
}
else if (info == COMMIT_TS_SETTS)
{
xl_commit_ts_set *setts = (xl_commit_ts_set *) XLogRecGetData(record);
int nsubxids;
TransactionId *subxids;
nsubxids = ((XLogRecGetDataLen(record) - SizeOfCommitTsSet) /
sizeof(TransactionId));
if (nsubxids > 0)
{
subxids = palloc(sizeof(TransactionId) * nsubxids);
memcpy(subxids,
XLogRecGetData(record) + SizeOfCommitTsSet,
sizeof(TransactionId) * nsubxids);
}
else
subxids = NULL;
TransactionTreeSetCommitTsData(setts->mainxid, nsubxids, subxids,
setts->timestamp, setts->nodeid, true);
if (subxids)
pfree(subxids);
}
else
elog(PANIC, "commit_ts_redo: unknown op code %u", info);
}
/*
* task_tracker_assign_task creates a new task in the shared hash or updates an
* already existing task. The function also creates a schema for the job if it
* doesn't already exist.
*/
Datum
task_tracker_assign_task(PG_FUNCTION_ARGS)
{
uint64 jobId = PG_GETARG_INT64(0);
uint32 taskId = PG_GETARG_UINT32(1);
text *taskCallStringText = PG_GETARG_TEXT_P(2);
StringInfo jobSchemaName = JobSchemaName(jobId);
bool schemaExists = false;
WorkerTask *workerTask = NULL;
char *taskCallString = text_to_cstring(taskCallStringText);
uint32 taskCallStringLength = strlen(taskCallString);
/* check that we have a running task tracker on this host */
bool taskTrackerRunning = TaskTrackerRunning();
if (!taskTrackerRunning)
{
ereport(ERROR, (errcode(ERRCODE_CANNOT_CONNECT_NOW),
errmsg("the task tracker has been disabled or shut down")));
}
/* check that we have enough space in our shared hash for this string */
if (taskCallStringLength >= TASK_CALL_STRING_SIZE)
{
ereport(ERROR, (errcode(ERRCODE_INVALID_PARAMETER_VALUE),
errmsg("task call string exceeds maximum assignable length")));
}
/*
* If the schema does not exist, we create it. However, the schema does not
* become visible to other processes until the transaction commits, and we
* therefore do not release the resource lock in this case. Otherwise, the
* schema is already visible, and we immediately release the resource lock.
*/
LockJobResource(jobId, AccessExclusiveLock);
schemaExists = JobSchemaExists(jobSchemaName);
if (!schemaExists)
{
/* lock gets automatically released upon return from this function */
CreateJobSchema(jobSchemaName);
}
else
{
UnlockJobResource(jobId, AccessExclusiveLock);
}
LWLockAcquire(&WorkerTasksSharedState->taskHashLock, LW_EXCLUSIVE);
/* check if we already have the task in our shared hash */
workerTask = WorkerTasksHashFind(jobId, taskId);
if (workerTask == NULL)
{
CreateTask(jobId, taskId, taskCallString);
}
else
{
UpdateTask(workerTask, taskCallString);
}
LWLockRelease(&WorkerTasksSharedState->taskHashLock);
PG_RETURN_VOID();
}
/*
* shmem_startup hook: allocate or attach to shared memory,
* then load any pre-existing statistics from file.
*/
static void
pgss_shmem_startup(void)
{
bool found;
HASHCTL info;
FILE *file;
uint32 header;
int32 num;
int32 i;
int query_size;
int buffer_size;
char *buffer = NULL;
if (prev_shmem_startup_hook)
prev_shmem_startup_hook();
/* reset in case this is a restart within the postmaster */
pgss = NULL;
pgss_hash = NULL;
/*
* Create or attach to the shared memory state, including hash table
*/
LWLockAcquire(AddinShmemInitLock, LW_EXCLUSIVE);
pgss = ShmemInitStruct("pg_stat_statements",
sizeof(pgssSharedState),
&found);
if (!found)
{
/* First time through ... */
pgss->lock = LWLockAssign();
pgss->query_size = pgstat_track_activity_query_size;
}
/* Be sure everyone agrees on the hash table entry size */
query_size = pgss->query_size;
memset(&info, 0, sizeof(info));
info.keysize = sizeof(pgssHashKey);
info.entrysize = offsetof(pgssEntry, query) +query_size;
info.hash = pgss_hash_fn;
info.match = pgss_match_fn;
pgss_hash = ShmemInitHash("pg_stat_statements hash",
pgss_max, pgss_max,
&info,
HASH_ELEM | HASH_FUNCTION | HASH_COMPARE);
LWLockRelease(AddinShmemInitLock);
/*
* If we're in the postmaster (or a standalone backend...), set up a shmem
* exit hook to dump the statistics to disk.
*/
if (!IsUnderPostmaster)
on_shmem_exit(pgss_shmem_shutdown, (Datum) 0);
/*
* Attempt to load old statistics from the dump file, if this is the first
* time through and we weren't told not to.
*/
if (found || !pgss_save)
return;
/*
* Note: we don't bother with locks here, because there should be no other
* processes running when this code is reached.
*/
file = AllocateFile(PGSS_DUMP_FILE, PG_BINARY_R);
if (file == NULL)
{
if (errno == ENOENT)
return; /* ignore not-found error */
goto error;
}
buffer_size = query_size;
buffer = (char *) palloc(buffer_size);
if (fread(&header, sizeof(uint32), 1, file) != 1 ||
header != PGSS_FILE_HEADER ||
fread(&num, sizeof(int32), 1, file) != 1)
goto error;
for (i = 0; i < num; i++)
{
pgssEntry temp;
pgssEntry *entry;
if (fread(&temp, offsetof(pgssEntry, mutex), 1, file) != 1)
goto error;
/* Encoding is the only field we can easily sanity-check */
if (!PG_VALID_BE_ENCODING(temp.key.encoding))
goto error;
/* Previous incarnation might have had a larger query_size */
if (temp.key.query_len >= buffer_size)
{
buffer = (char *) repalloc(buffer, temp.key.query_len + 1);
buffer_size = temp.key.query_len + 1;
}
if (fread(buffer, 1, temp.key.query_len, file) != temp.key.query_len)
goto error;
buffer[temp.key.query_len] = '\0';
/* Clip to available length if needed */
if (temp.key.query_len >= query_size)
temp.key.query_len = pg_encoding_mbcliplen(temp.key.encoding,
buffer,
temp.key.query_len,
query_size - 1);
temp.key.query_ptr = buffer;
/* make the hashtable entry (discards old entries if too many) */
entry = entry_alloc(&temp.key);
/* copy in the actual stats */
entry->counters = temp.counters;
}
pfree(buffer);
FreeFile(file);
return;
error:
ereport(LOG,
(errcode_for_file_access(),
errmsg("could not read pg_stat_statement file \"%s\": %m",
PGSS_DUMP_FILE)));
if (buffer)
pfree(buffer);
if (file)
FreeFile(file);
/* If possible, throw away the bogus file; ignore any error */
unlink(PGSS_DUMP_FILE);
}
/*
* Retrieve statement statistics.
*/
Datum
pg_stat_statements(PG_FUNCTION_ARGS)
{
ReturnSetInfo *rsinfo = (ReturnSetInfo *) fcinfo->resultinfo;
TupleDesc tupdesc;
Tuplestorestate *tupstore;
MemoryContext per_query_ctx;
MemoryContext oldcontext;
Oid userid = GetUserId();
bool is_superuser = superuser();
HASH_SEQ_STATUS hash_seq;
pgssEntry *entry;
if (!pgss || !pgss_hash)
ereport(ERROR,
(errcode(ERRCODE_OBJECT_NOT_IN_PREREQUISITE_STATE),
errmsg("pg_stat_statements must be loaded via shared_preload_libraries")));
/* check to see if caller supports us returning a tuplestore */
if (rsinfo == NULL || !IsA(rsinfo, ReturnSetInfo))
ereport(ERROR,
(errcode(ERRCODE_FEATURE_NOT_SUPPORTED),
errmsg("set-valued function called in context that cannot accept a set")));
if (!(rsinfo->allowedModes & SFRM_Materialize))
ereport(ERROR,
(errcode(ERRCODE_FEATURE_NOT_SUPPORTED),
errmsg("materialize mode required, but it is not " \
"allowed in this context")));
/* Build a tuple descriptor for our result type */
if (get_call_result_type(fcinfo, NULL, &tupdesc) != TYPEFUNC_COMPOSITE)
elog(ERROR, "return type must be a row type");
per_query_ctx = rsinfo->econtext->ecxt_per_query_memory;
oldcontext = MemoryContextSwitchTo(per_query_ctx);
tupstore = tuplestore_begin_heap(true, false, work_mem);
rsinfo->returnMode = SFRM_Materialize;
rsinfo->setResult = tupstore;
rsinfo->setDesc = tupdesc;
MemoryContextSwitchTo(oldcontext);
LWLockAcquire(pgss->lock, LW_SHARED);
hash_seq_init(&hash_seq, pgss_hash);
while ((entry = hash_seq_search(&hash_seq)) != NULL)
{
Datum values[PG_STAT_STATEMENTS_COLS];
bool nulls[PG_STAT_STATEMENTS_COLS];
int i = 0;
Counters tmp;
memset(values, 0, sizeof(values));
memset(nulls, 0, sizeof(nulls));
values[i++] = ObjectIdGetDatum(entry->key.userid);
values[i++] = ObjectIdGetDatum(entry->key.dbid);
if (is_superuser || entry->key.userid == userid)
{
char *qstr;
qstr = (char *)
pg_do_encoding_conversion((unsigned char *) entry->query,
entry->key.query_len,
entry->key.encoding,
GetDatabaseEncoding());
values[i++] = CStringGetTextDatum(qstr);
if (qstr != entry->query)
pfree(qstr);
}
else
values[i++] = CStringGetTextDatum("<insufficient privilege>");
/* copy counters to a local variable to keep locking time short */
{
volatile pgssEntry *e = (volatile pgssEntry *) entry;
SpinLockAcquire(&e->mutex);
tmp = e->counters;
SpinLockRelease(&e->mutex);
}
values[i++] = Int64GetDatumFast(tmp.calls);
values[i++] = Float8GetDatumFast(tmp.total_time);
values[i++] = Int64GetDatumFast(tmp.rows);
values[i++] = Int64GetDatumFast(tmp.shared_blks_hit);
values[i++] = Int64GetDatumFast(tmp.shared_blks_read);
values[i++] = Int64GetDatumFast(tmp.shared_blks_written);
values[i++] = Int64GetDatumFast(tmp.local_blks_hit);
values[i++] = Int64GetDatumFast(tmp.local_blks_read);
values[i++] = Int64GetDatumFast(tmp.local_blks_written);
values[i++] = Int64GetDatumFast(tmp.temp_blks_read);
values[i++] = Int64GetDatumFast(tmp.temp_blks_written);
Assert(i == PG_STAT_STATEMENTS_COLS);
tuplestore_putvalues(tupstore, tupdesc, values, nulls);
}
LWLockRelease(pgss->lock);
/* clean up and return the tuplestore */
tuplestore_donestoring(tupstore);
return (Datum) 0;
}
/*
* Each database using a table space is isolated into its own name space
* by a subdirectory named for the database OID. On first creation of an
* object in the tablespace, create the subdirectory. If the subdirectory
* already exists, fall through quietly.
*
* isRedo indicates that we are creating an object during WAL replay.
* In this case we will cope with the possibility of the tablespace
* directory not being there either --- this could happen if we are
* replaying an operation on a table in a subsequently-dropped tablespace.
* We handle this by making a directory in the place where the tablespace
* symlink would normally be. This isn't an exact replay of course, but
* it's the best we can do given the available information.
*
* If tablespaces are not supported, we still need it in case we have to
* re-create a database subdirectory (of $PGDATA/base) during WAL replay.
*/
void
TablespaceCreateDbspace(Oid spcNode, Oid dbNode, bool isRedo)
{
struct stat st;
char *dir;
/*
* The global tablespace doesn't have per-database subdirectories, so
* nothing to do for it.
*/
if (spcNode == GLOBALTABLESPACE_OID)
return;
Assert(OidIsValid(spcNode));
Assert(OidIsValid(dbNode));
dir = GetDatabasePath(dbNode, spcNode);
if (stat(dir, &st) < 0)
{
/* Directory does not exist? */
if (errno == ENOENT)
{
/*
* Acquire TablespaceCreateLock to ensure that no DROP TABLESPACE
* or TablespaceCreateDbspace is running concurrently.
*/
LWLockAcquire(TablespaceCreateLock, LW_EXCLUSIVE);
/*
* Recheck to see if someone created the directory while we were
* waiting for lock.
*/
if (stat(dir, &st) == 0 && S_ISDIR(st.st_mode))
{
/* Directory was created */
}
else
{
/* Directory creation failed? */
if (mkdir(dir, S_IRWXU) < 0)
{
char *parentdir;
/* Failure other than not exists or not in WAL replay? */
if (errno != ENOENT || !isRedo)
ereport(ERROR,
(errcode_for_file_access(),
errmsg("could not create directory \"%s\": %m",
dir)));
/*
* Parent directories are missing during WAL replay, so
* continue by creating simple parent directories rather
* than a symlink.
*/
/* create two parents up if not exist */
parentdir = pstrdup(dir);
get_parent_directory(parentdir);
get_parent_directory(parentdir);
/* Can't create parent and it doesn't already exist? */
if (mkdir(parentdir, S_IRWXU) < 0 && errno != EEXIST)
ereport(ERROR,
(errcode_for_file_access(),
errmsg("could not create directory \"%s\": %m",
parentdir)));
pfree(parentdir);
/* create one parent up if not exist */
parentdir = pstrdup(dir);
get_parent_directory(parentdir);
/* Can't create parent and it doesn't already exist? */
if (mkdir(parentdir, S_IRWXU) < 0 && errno != EEXIST)
ereport(ERROR,
(errcode_for_file_access(),
errmsg("could not create directory \"%s\": %m",
parentdir)));
pfree(parentdir);
/* Create database directory */
if (mkdir(dir, S_IRWXU) < 0)
ereport(ERROR,
(errcode_for_file_access(),
errmsg("could not create directory \"%s\": %m",
dir)));
}
}
LWLockRelease(TablespaceCreateLock);
}
else
{
ereport(ERROR,
(errcode_for_file_access(),
errmsg("could not stat directory \"%s\": %m", dir)));
}
}
else
{
/* Is it not a directory? */
if (!S_ISDIR(st.st_mode))
ereport(ERROR,
(errcode(ERRCODE_WRONG_OBJECT_TYPE),
errmsg("\"%s\" exists but is not a directory",
dir)));
}
pfree(dir);
}
void
smgrDoAppendOnlyResyncEofs(bool forCommit)
{
HASH_SEQ_STATUS iterateStatus;
AppendOnlyMirrorResyncEofs *entry;
AppendOnlyMirrorResyncEofs *entryExample = NULL;
int appendOnlyMirrorResyncEofsCount;
if (AppendOnlyMirrorResyncEofsTable == NULL)
return;
if (Debug_persistent_print ||
Debug_persistent_appendonly_commit_count_print)
elog(Persistent_DebugPrintLevel(),
"Storage Manager: Enter Append-Only mirror resync eofs list entries (Append-Only commit work count %d)",
FileRepPrimary_GetAppendOnlyCommitWorkCount());
hash_seq_init(
&iterateStatus,
AppendOnlyMirrorResyncEofsTable);
appendOnlyMirrorResyncEofsCount = 0;
while ((entry = hash_seq_search(&iterateStatus)) != NULL)
{
if (entryExample == NULL)
{
entryExample = entry;
}
if (forCommit)
{
PersistentFileSysObj_UpdateAppendOnlyMirrorResyncEofs(&entry->key.relFileNode,
entry->key.segmentFileNum,
&entry->persistentTid,
entry->persistentSerialNum,
entry->mirrorCatchupRequired,
entry->mirrorNewEof,
/* recovery */ false,
/* flushToXLog */ false);
}
else
{
/*
* Abort case.
*/
if (entry->didIncrementCommitCount)
{
int systemAppendOnlyCommitWorkCount;
LWLockAcquire(FileRepAppendOnlyCommitCountLock, LW_EXCLUSIVE);
systemAppendOnlyCommitWorkCount =
FileRepPrimary_FinishedAppendOnlyCommitWork(1);
if (entry->isDistributedTransaction)
{
PrepareDecrAppendOnlyCommitWork(entry->gid);
}
if (Debug_persistent_print ||
Debug_persistent_appendonly_commit_count_print)
elog(Persistent_DebugPrintLevel(),
"Storage Manager: Append-Only Mirror Resync EOFs decrementing commit work for aborted transaction "
"(system count %d). "
"Relation %u/%u/%u, segment file #%d (persistent serial num " INT64_FORMAT ", TID %s) ",
systemAppendOnlyCommitWorkCount,
entry->key.relFileNode.spcNode,
entry->key.relFileNode.dbNode,
entry->key.relFileNode.relNode,
entry->key.segmentFileNum,
entry->persistentSerialNum,
ItemPointerToString(&entry->persistentTid));
pendingAppendOnlyMirrorResyncIntentCount--;
LWLockRelease(FileRepAppendOnlyCommitCountLock);
}
}
if (Debug_persistent_print ||
Debug_persistent_appendonly_commit_count_print)
elog(Persistent_DebugPrintLevel(),
"Storage Manager: Append-Only mirror resync eofs list entry #%d: %u/%u/%u, segment file #%d, relation name '%s' "
"(forCommit %s, persistent TID %s, persistent serial number " INT64_FORMAT ", mirror catchup required %s, mirror new EOF " INT64_FORMAT ")",
appendOnlyMirrorResyncEofsCount,
entry->key.relFileNode.spcNode,
entry->key.relFileNode.dbNode,
entry->key.relFileNode.relNode,
entry->key.segmentFileNum,
(entry->relationName == NULL ? "<null>" : entry->relationName),
(forCommit ? "true" : "false"),
ItemPointerToString(&entry->persistentTid),
entry->persistentSerialNum,
(entry->mirrorCatchupRequired ? "true" : "false"),
entry->mirrorNewEof);
appendOnlyMirrorResyncEofsCount++;
}
/*
* If we collected Append-Only mirror resync EOFs and bumped the intent
* count, we need to decrement the counts as part of our end transaction
* work here.
*/
if (pendingAppendOnlyMirrorResyncIntentCount > 0)
{
MIRRORED_LOCK_DECLARE;
int oldSystemAppendOnlyCommitWorkCount;
int newSystemAppendOnlyCommitWorkCount;
int resultSystemAppendOnlyCommitWorkCount;
if (appendOnlyMirrorResyncEofsCount != pendingAppendOnlyMirrorResyncIntentCount)
elog(ERROR, "Pending Append-Only Mirror Resync EOFs intent count mismatch (pending %d, table count %d)",
pendingAppendOnlyMirrorResyncIntentCount,
appendOnlyMirrorResyncEofsCount);
if (entryExample == NULL)
elog(ERROR, "Not expecting an empty Append-Only Mirror Resync hash table when the local intent count is non-zero (%d)",
pendingAppendOnlyMirrorResyncIntentCount);
MIRRORED_LOCK;
/*NOTE: When we use the MirroredLock for the whole routine, this can go. */
LWLockAcquire(FileRepAppendOnlyCommitCountLock, LW_EXCLUSIVE);
oldSystemAppendOnlyCommitWorkCount = FileRepPrimary_GetAppendOnlyCommitWorkCount();
newSystemAppendOnlyCommitWorkCount =
oldSystemAppendOnlyCommitWorkCount -
pendingAppendOnlyMirrorResyncIntentCount;
if (newSystemAppendOnlyCommitWorkCount < 0)
elog(ERROR,
"Append-Only Mirror Resync EOFs intent count would go negative "
"(system count %d, entry count %d). "
"Example relation %u/%u/%u, segment file #%d (persistent serial num " INT64_FORMAT ", TID %s)",
oldSystemAppendOnlyCommitWorkCount,
pendingAppendOnlyMirrorResyncIntentCount,
entryExample->key.relFileNode.spcNode,
entryExample->key.relFileNode.dbNode,
entryExample->key.relFileNode.relNode,
entryExample->key.segmentFileNum,
entryExample->persistentSerialNum,
ItemPointerToString(&entryExample->persistentTid));
resultSystemAppendOnlyCommitWorkCount =
FileRepPrimary_FinishedAppendOnlyCommitWork(pendingAppendOnlyMirrorResyncIntentCount);
/*
* Should match since we are under FileRepAppendOnlyCommitCountLock
* EXCLUSIVE.
*/
Assert(newSystemAppendOnlyCommitWorkCount == resultSystemAppendOnlyCommitWorkCount);
pendingAppendOnlyMirrorResyncIntentCount = 0;
if (Debug_persistent_print ||
Debug_persistent_appendonly_commit_count_print)
elog(Persistent_DebugPrintLevel(),
"Storage Manager: Append-Only Mirror Resync EOFs intent count finishing %s work with system count %d remaining "
"(enter system count %d, entry count %d, result system count %d). "
"Example relation %u/%u/%u, segment file #%d (persistent serial num " INT64_FORMAT ", TID %s)",
(forCommit ? "commit" : "abort"),
newSystemAppendOnlyCommitWorkCount,
oldSystemAppendOnlyCommitWorkCount,
pendingAppendOnlyMirrorResyncIntentCount,
resultSystemAppendOnlyCommitWorkCount,
entryExample->key.relFileNode.spcNode,
entryExample->key.relFileNode.dbNode,
entryExample->key.relFileNode.relNode,
entryExample->key.segmentFileNum,
entryExample->persistentSerialNum,
ItemPointerToString(&entryExample->persistentTid));
LWLockRelease(FileRepAppendOnlyCommitCountLock);
MIRRORED_UNLOCK;
}
}
/*
* Wait for synchronous replication, if requested by user.
*
* Initially backends start in state SYNC_REP_NOT_WAITING and then
* change that state to SYNC_REP_WAITING before adding ourselves
* to the wait queue. During SyncRepWakeQueue() a WALSender changes
* the state to SYNC_REP_WAIT_COMPLETE once replication is confirmed.
* This backend then resets its state to SYNC_REP_NOT_WAITING.
*/
void
SyncRepWaitForLSN(XLogRecPtr XactCommitLSN)
{
char *new_status = NULL;
const char *old_status;
/*
* Fast exit if user has not requested sync replication, or
* there are no sync replication standby names defined.
* Note that those standbys don't need to be connected.
*/
if (!SyncRepRequested() || !SyncStandbysDefined())
return;
Assert(SHMQueueIsDetached(&(MyProc->syncRepLinks)));
Assert(WalSndCtl != NULL);
/* Reset the latch before adding ourselves to the queue. */
ResetLatch(&MyProc->waitLatch);
/*
* Set our waitLSN so WALSender will know when to wake us, and add
* ourselves to the queue.
*/
LWLockAcquire(SyncRepLock, LW_EXCLUSIVE);
Assert(MyProc->syncRepState == SYNC_REP_NOT_WAITING);
if (!WalSndCtl->sync_standbys_defined)
{
/*
* We don't wait for sync rep if WalSndCtl->sync_standbys_defined is
* not set. See SyncRepUpdateSyncStandbysDefined.
*/
LWLockRelease(SyncRepLock);
return;
}
MyProc->waitLSN = XactCommitLSN;
MyProc->syncRepState = SYNC_REP_WAITING;
SyncRepQueueInsert();
Assert(SyncRepQueueIsOrderedByLSN());
LWLockRelease(SyncRepLock);
/* Alter ps display to show waiting for sync rep. */
if (update_process_title)
{
int len;
old_status = get_ps_display(&len);
new_status = (char *) palloc(len + 32 + 1);
memcpy(new_status, old_status, len);
sprintf(new_status + len, " waiting for %X/%X",
XactCommitLSN.xlogid, XactCommitLSN.xrecoff);
set_ps_display(new_status, false);
new_status[len] = '\0'; /* truncate off " waiting ..." */
}
/*
* Wait for specified LSN to be confirmed.
*
* Each proc has its own wait latch, so we perform a normal latch
* check/wait loop here.
*/
for (;;)
{
int syncRepState;
/*
* Wait on latch for up to 60 seconds. This allows us to
* check for postmaster death regularly while waiting.
* Note that timeout here does not necessarily release from loop.
*/
WaitLatch(&MyProc->waitLatch, 60000000L);
/* Must reset the latch before testing state. */
ResetLatch(&MyProc->waitLatch);
/*
* Try checking the state without the lock first. There's no guarantee
* that we'll read the most up-to-date value, so if it looks like we're
* still waiting, recheck while holding the lock. But if it looks like
* we're done, we must really be done, because once walsender changes
* the state to SYNC_REP_WAIT_COMPLETE, it will never update it again,
* so we can't be seeing a stale value in that case.
*/
syncRepState = MyProc->syncRepState;
if (syncRepState == SYNC_REP_WAITING)
{
LWLockAcquire(SyncRepLock, LW_SHARED);
syncRepState = MyProc->syncRepState;
LWLockRelease(SyncRepLock);
}
if (syncRepState == SYNC_REP_WAIT_COMPLETE)
break;
/*
* If a wait for synchronous replication is pending, we can neither
* acknowledge the commit nor raise ERROR or FATAL. The latter
* would lead the client to believe that that the transaction
* aborted, which is not true: it's already committed locally.
* The former is no good either: the client has requested
* synchronous replication, and is entitled to assume that an
* acknowledged commit is also replicated, which may not be true.
* So in this case we issue a WARNING (which some clients may
* be able to interpret) and shut off further output. We do NOT
* reset ProcDiePending, so that the process will die after the
* commit is cleaned up.
*/
if (ProcDiePending)
{
ereport(WARNING,
(errcode(ERRCODE_ADMIN_SHUTDOWN),
errmsg("canceling the wait for synchronous replication and terminating connection due to administrator command"),
errdetail("The transaction has already committed locally, but may not have been replicated to the standby.")));
whereToSendOutput = DestNone;
SyncRepCancelWait();
break;
}
/*
* It's unclear what to do if a query cancel interrupt arrives. We
* can't actually abort at this point, but ignoring the interrupt
* altogether is not helpful, so we just terminate the wait with
* a suitable warning.
*/
if (QueryCancelPending)
{
QueryCancelPending = false;
ereport(WARNING,
(errmsg("canceling wait for synchronous replication due to user request"),
errdetail("The transaction has already committed locally, but may not have been replicated to the standby.")));
SyncRepCancelWait();
break;
}
/*
* If the postmaster dies, we'll probably never get an acknowledgement,
* because all the wal sender processes will exit. So just bail out.
*/
if (!PostmasterIsAlive(true))
{
ProcDiePending = true;
whereToSendOutput = DestNone;
SyncRepCancelWait();
break;
}
}
/*
* WalSender has checked our LSN and has removed us from queue. Clean up
* state and leave. It's OK to reset these shared memory fields without
* holding SyncRepLock, because any walsenders will ignore us anyway when
* we're not on the queue.
*/
Assert(SHMQueueIsDetached(&(MyProc->syncRepLinks)));
MyProc->syncRepState = SYNC_REP_NOT_WAITING;
MyProc->waitLSN.xlogid = 0;
MyProc->waitLSN.xrecoff = 0;
if (new_status)
{
/* Reset ps display */
set_ps_display(new_status, false);
pfree(new_status);
}
}
Datum
dbms_alert_defered_signal(PG_FUNCTION_ARGS)
{
TriggerData *trigdata = (TriggerData *) fcinfo->context;
TupleDesc tupdesc;
HeapTuple rettuple;
char *relname;
text *name;
text *message;
int event_col;
int message_col;
Datum datum;
bool isnull;
int cycle = 0;
float8 endtime;
float8 timeout = 2;
if (!CALLED_AS_TRIGGER(fcinfo))
ereport(ERROR,
(errcode(ERRCODE_TRIGGERED_ACTION_EXCEPTION),
errmsg("not called by trigger manager")));
if (!TRIGGER_FIRED_BY_INSERT(trigdata->tg_event))
ereport(ERROR,
(errcode(ERRCODE_TRIGGERED_ACTION_EXCEPTION),
errmsg("not called on valid event")));
if (SPI_connect() < 0)
ereport(ERROR,
(errcode(ERRCODE_TRIGGERED_ACTION_EXCEPTION),
errmsg("SPI_connect failed")));
if (strcmp((relname = SPI_getrelname(trigdata->tg_relation)), "ora_alerts") != 0)
ereport(ERROR,
(errcode(ERRCODE_TRIGGERED_ACTION_EXCEPTION),
errmsg("not called with valid relation")));
rettuple = trigdata->tg_trigtuple;
tupdesc = trigdata->tg_relation->rd_att;
if (SPI_ERROR_NOATTRIBUTE == (event_col = SPI_fnumber(tupdesc, "event")))
ereport(ERROR,
(errcode(ERRCODE_TRIGGERED_ACTION_EXCEPTION),
errmsg("attribute event not found")));
if (SPI_ERROR_NOATTRIBUTE == (message_col = SPI_fnumber(tupdesc, "message")))
ereport(ERROR,
(errcode(ERRCODE_TRIGGERED_ACTION_EXCEPTION),
errmsg("attribute message not found")));
datum = SPI_getbinval(rettuple, tupdesc, event_col, &isnull);
if (isnull)
ereport(ERROR,
(errcode(ERRCODE_NULL_VALUE_NOT_ALLOWED),
errmsg("event name is NULL"),
errdetail("Eventname may not be NULL.")));
name = DatumGetTextP(datum);
datum = SPI_getbinval(rettuple, tupdesc, message_col, &isnull);
if (isnull)
message = NULL;
else
message = DatumGetTextP(datum);
WATCH_PRE(timeout, endtime, cycle);
if (ora_lock_shmem(SHMEMMSGSZ, MAX_PIPES, MAX_EVENTS, MAX_LOCKS, false))
{
ItemPointer tid;
Oid argtypes[1] = {TIDOID};
char nulls[1] = {' '};
Datum values[1];
void *plan;
create_message(name, message);
LWLockRelease(shmem_lock);
tid = &rettuple->t_data->t_ctid;
if (!(plan = SPI_prepare("DELETE FROM ora_alerts WHERE ctid = $1", 1, argtypes)))
ereport(ERROR,
(errcode(ERRCODE_TRIGGERED_ACTION_EXCEPTION),
errmsg("SPI_prepare failed")));
values[0] = ItemPointerGetDatum(tid);
if (SPI_OK_DELETE != SPI_execute_plan(plan, values, nulls, false, 1))
ereport(ERROR,
(errcode(ERRCODE_TRIGGERED_ACTION_EXCEPTION),
errmsg("can't execute sql")));
SPI_finish();
return PointerGetDatum(rettuple);
}
WATCH_POST(timeout, endtime, cycle);
LOCK_ERROR();
PG_RETURN_NULL();
}
/*
* Update the LSNs on each queue based upon our latest state. This
* implements a simple policy of first-valid-standby-releases-waiter.
*
* Other policies are possible, which would change what we do here and what
* perhaps also which information we store as well.
*/
void
SyncRepReleaseWaiters(void)
{
volatile WalSndCtlData *walsndctl = WalSndCtl;
volatile WalSnd *syncWalSnd = NULL;
int numprocs = 0;
int priority = 0;
int i;
/*
* If this WALSender is serving a standby that is not on the list of
* potential standbys then we have nothing to do. If we are still
* starting up or still running base backup, then leave quickly also.
*/
if (MyWalSnd->sync_standby_priority == 0 ||
MyWalSnd->state < WALSNDSTATE_STREAMING)
return;
/*
* We're a potential sync standby. Release waiters if we are the
* highest priority standby. If there are multiple standbys with
* same priorities then we use the first mentioned standby.
* If you change this, also change pg_stat_get_wal_senders().
*/
LWLockAcquire(SyncRepLock, LW_EXCLUSIVE);
for (i = 0; i < max_wal_senders; i++)
{
/* use volatile pointer to prevent code rearrangement */
volatile WalSnd *walsnd = &walsndctl->walsnds[i];
if (walsnd->pid != 0 &&
walsnd->sync_standby_priority > 0 &&
(priority == 0 ||
priority > walsnd->sync_standby_priority))
{
priority = walsnd->sync_standby_priority;
syncWalSnd = walsnd;
}
}
/*
* We should have found ourselves at least.
*/
Assert(syncWalSnd);
/*
* If we aren't managing the highest priority standby then just leave.
*/
if (syncWalSnd != MyWalSnd)
{
LWLockRelease(SyncRepLock);
announce_next_takeover = true;
return;
}
if (XLByteLT(walsndctl->lsn, MyWalSnd->flush))
{
/*
* Set the lsn first so that when we wake backends they will
* release up to this location.
*/
walsndctl->lsn = MyWalSnd->flush;
numprocs = SyncRepWakeQueue(false);
}
LWLockRelease(SyncRepLock);
elog(DEBUG3, "released %d procs up to %X/%X",
numprocs,
MyWalSnd->flush.xlogid,
MyWalSnd->flush.xrecoff);
/*
* If we are managing the highest priority standby, though we weren't
* prior to this, then announce we are now the sync standby.
*/
if (announce_next_takeover)
{
announce_next_takeover = false;
ereport(LOG,
(errmsg("standby \"%s\" is now the synchronous standby with priority %u",
application_name, MyWalSnd->sync_standby_priority)));
}
}
/*
* ShmemInitStruct -- Create/attach to a structure in shared memory.
*
* This is called during initialization to find or allocate
* a data structure in shared memory. If no other process
* has created the structure, this routine allocates space
* for it. If it exists already, a pointer to the existing
* structure is returned.
*
* Returns: pointer to the object. *foundPtr is set TRUE if the object was
* already in the shmem index (hence, already initialized).
*
* Note: before Postgres 9.0, this function returned NULL for some failure
* cases. Now, it always throws error instead, so callers need not check
* for NULL.
*/
void *
ShmemInitStruct(const char *name, Size size, bool *foundPtr)
{
ShmemIndexEnt *result;
void *structPtr;
LWLockAcquire(ShmemIndexLock, LW_EXCLUSIVE);
if (!ShmemIndex)
{
PGShmemHeader *shmemseghdr = ShmemSegHdr;
/* Must be trying to create/attach to ShmemIndex itself */
Assert(strcmp(name, "ShmemIndex") == 0);
if (IsUnderPostmaster)
{
/* Must be initializing a (non-standalone) backend */
Assert(shmemseghdr->index != NULL);
structPtr = shmemseghdr->index;
*foundPtr = TRUE;
}
else
{
/*
* If the shmem index doesn't exist, we are bootstrapping: we must
* be trying to init the shmem index itself.
*
* Notice that the ShmemIndexLock is released before the shmem
* index has been initialized. This should be OK because no other
* process can be accessing shared memory yet.
*/
Assert(shmemseghdr->index == NULL);
structPtr = ShmemAlloc(size);
shmemseghdr->index = structPtr;
*foundPtr = FALSE;
}
LWLockRelease(ShmemIndexLock);
return structPtr;
}
/* look it up in the shmem index */
result = (ShmemIndexEnt *)
hash_search(ShmemIndex, name, HASH_ENTER_NULL, foundPtr);
if (!result)
{
LWLockRelease(ShmemIndexLock);
ereport(ERROR,
(errcode(ERRCODE_OUT_OF_MEMORY),
errmsg("could not create ShmemIndex entry for data structure \"%s\"",
name)));
}
if (*foundPtr)
{
/*
* Structure is in the shmem index so someone else has allocated it
* already. The size better be the same as the size we are trying to
* initialize to, or there is a name conflict (or worse).
*/
if (result->size != size)
{
LWLockRelease(ShmemIndexLock);
ereport(ERROR,
(errmsg("ShmemIndex entry size is wrong for data structure"
" \"%s\": expected %zu, actual %zu",
name, size, result->size)));
}
structPtr = result->location;
}
else
{
/* It isn't in the table yet. allocate and initialize it */
structPtr = ShmemAllocNoError(size);
if (structPtr == NULL)
{
/* out of memory; remove the failed ShmemIndex entry */
hash_search(ShmemIndex, name, HASH_REMOVE, NULL);
LWLockRelease(ShmemIndexLock);
ereport(ERROR,
(errcode(ERRCODE_OUT_OF_MEMORY),
errmsg("not enough shared memory for data structure"
" \"%s\" (%zu bytes requested)",
name, size)));
}
result->size = size;
result->location = structPtr;
}
LWLockRelease(ShmemIndexLock);
Assert(ShmemAddrIsValid(structPtr));
Assert(structPtr == (void *) CACHELINEALIGN(structPtr));
return structPtr;
}
/*
* Write a page from a shared buffer, if necessary.
* Does nothing if the specified slot is not dirty.
*
* NOTE: only one write attempt is made here. Hence, it is possible that
* the page is still dirty at exit (if someone else re-dirtied it during
* the write). However, we *do* attempt a fresh write even if the page
* is already being written; this is for checkpoints.
*
* Control lock must be held at entry, and will be held at exit.
*/
static void
SlruInternalWritePage(SlruCtl ctl, int slotno, SlruFlush fdata)
{
SlruShared shared = ctl->shared;
int pageno = shared->page_number[slotno];
bool ok;
/* If a write is in progress, wait for it to finish */
while (shared->page_status[slotno] == SLRU_PAGE_WRITE_IN_PROGRESS &&
shared->page_number[slotno] == pageno)
{
SimpleLruWaitIO(ctl, slotno);
}
/*
* Do nothing if page is not dirty, or if buffer no longer contains the
* same page we were called for.
*/
if (!shared->page_dirty[slotno] ||
shared->page_status[slotno] != SLRU_PAGE_VALID ||
shared->page_number[slotno] != pageno)
return;
/*
* Mark the slot write-busy, and clear the dirtybit. After this point, a
* transaction status update on this page will mark it dirty again.
*/
shared->page_status[slotno] = SLRU_PAGE_WRITE_IN_PROGRESS;
shared->page_dirty[slotno] = false;
/* Acquire per-buffer lock (cannot deadlock, see notes at top) */
LWLockAcquire(shared->buffer_locks[slotno], LW_EXCLUSIVE);
/* Release control lock while doing I/O */
LWLockRelease(shared->ControlLock);
/* Do the write */
ok = SlruPhysicalWritePage(ctl, pageno, slotno, fdata);
/* If we failed, and we're in a flush, better close the files */
if (!ok && fdata)
{
int i;
for (i = 0; i < fdata->num_files; i++)
close(fdata->fd[i]);
}
/* Re-acquire control lock and update page state */
LWLockAcquire(shared->ControlLock, LW_EXCLUSIVE);
Assert(shared->page_number[slotno] == pageno &&
shared->page_status[slotno] == SLRU_PAGE_WRITE_IN_PROGRESS);
/* If we failed to write, mark the page dirty again */
if (!ok)
shared->page_dirty[slotno] = true;
shared->page_status[slotno] = SLRU_PAGE_VALID;
LWLockRelease(shared->buffer_locks[slotno]);
/* Now it's okay to ereport if we failed */
if (!ok)
SlruReportIOError(ctl, pageno, InvalidTransactionId);
}
/*
* Returns activity of walsenders, including pids and xlog locations sent to
* standby servers.
*/
Datum
pg_stat_get_wal_senders(PG_FUNCTION_ARGS)
{
#define PG_STAT_GET_WAL_SENDERS_COLS 8
ReturnSetInfo *rsinfo = (ReturnSetInfo *) fcinfo->resultinfo;
TupleDesc tupdesc;
Tuplestorestate *tupstore;
MemoryContext per_query_ctx;
MemoryContext oldcontext;
int *sync_priority;
int priority = 0;
int sync_standby = -1;
int i;
/* check to see if caller supports us returning a tuplestore */
if (rsinfo == NULL || !IsA(rsinfo, ReturnSetInfo))
ereport(ERROR,
(errcode(ERRCODE_FEATURE_NOT_SUPPORTED),
errmsg("set-valued function called in context that cannot accept a set")));
if (!(rsinfo->allowedModes & SFRM_Materialize))
ereport(ERROR,
(errcode(ERRCODE_FEATURE_NOT_SUPPORTED),
errmsg("materialize mode required, but it is not " \
"allowed in this context")));
/* Build a tuple descriptor for our result type */
if (get_call_result_type(fcinfo, NULL, &tupdesc) != TYPEFUNC_COMPOSITE)
elog(ERROR, "return type must be a row type");
per_query_ctx = rsinfo->econtext->ecxt_per_query_memory;
oldcontext = MemoryContextSwitchTo(per_query_ctx);
tupstore = tuplestore_begin_heap(true, false, work_mem);
rsinfo->returnMode = SFRM_Materialize;
rsinfo->setResult = tupstore;
rsinfo->setDesc = tupdesc;
MemoryContextSwitchTo(oldcontext);
/*
* Get the priorities of sync standbys all in one go, to minimise lock
* acquisitions and to allow us to evaluate who is the current sync
* standby. This code must match the code in SyncRepReleaseWaiters().
*/
sync_priority = palloc(sizeof(int) * max_wal_senders);
LWLockAcquire(SyncRepLock, LW_SHARED);
for (i = 0; i < max_wal_senders; i++)
{
/* use volatile pointer to prevent code rearrangement */
volatile WalSnd *walsnd = &WalSndCtl->walsnds[i];
if (walsnd->pid != 0)
{
sync_priority[i] = walsnd->sync_standby_priority;
if (walsnd->state == WALSNDSTATE_STREAMING &&
walsnd->sync_standby_priority > 0 &&
(priority == 0 ||
priority > walsnd->sync_standby_priority))
{
priority = walsnd->sync_standby_priority;
sync_standby = i;
}
}
}
LWLockRelease(SyncRepLock);
for (i = 0; i < max_wal_senders; i++)
{
/* use volatile pointer to prevent code rearrangement */
volatile WalSnd *walsnd = &WalSndCtl->walsnds[i];
char location[MAXFNAMELEN];
XLogRecPtr sentPtr;
XLogRecPtr write;
XLogRecPtr flush;
XLogRecPtr apply;
WalSndState state;
Datum values[PG_STAT_GET_WAL_SENDERS_COLS];
bool nulls[PG_STAT_GET_WAL_SENDERS_COLS];
if (walsnd->pid == 0)
continue;
SpinLockAcquire(&walsnd->mutex);
sentPtr = walsnd->sentPtr;
state = walsnd->state;
write = walsnd->write;
flush = walsnd->flush;
apply = walsnd->apply;
SpinLockRelease(&walsnd->mutex);
memset(nulls, 0, sizeof(nulls));
values[0] = Int32GetDatum(walsnd->pid);
if (!superuser())
{
/*
* Only superusers can see details. Other users only get the pid
* value to know it's a walsender, but no details.
*/
MemSet(&nulls[1], true, PG_STAT_GET_WAL_SENDERS_COLS - 1);
}
else
{
values[1] = CStringGetTextDatum(WalSndGetStateString(state));
snprintf(location, sizeof(location), "%X/%X",
sentPtr.xlogid, sentPtr.xrecoff);
values[2] = CStringGetTextDatum(location);
if (write.xlogid == 0 && write.xrecoff == 0)
nulls[3] = true;
snprintf(location, sizeof(location), "%X/%X",
write.xlogid, write.xrecoff);
values[3] = CStringGetTextDatum(location);
if (flush.xlogid == 0 && flush.xrecoff == 0)
nulls[4] = true;
snprintf(location, sizeof(location), "%X/%X",
flush.xlogid, flush.xrecoff);
values[4] = CStringGetTextDatum(location);
if (apply.xlogid == 0 && apply.xrecoff == 0)
nulls[5] = true;
snprintf(location, sizeof(location), "%X/%X",
apply.xlogid, apply.xrecoff);
values[5] = CStringGetTextDatum(location);
values[6] = Int32GetDatum(sync_priority[i]);
/*
* More easily understood version of standby state. This is purely
* informational, not different from priority.
*/
if (sync_priority[i] == 0)
values[7] = CStringGetTextDatum("async");
else if (i == sync_standby)
values[7] = CStringGetTextDatum("sync");
else
values[7] = CStringGetTextDatum("potential");
}
tuplestore_putvalues(tupstore, tupdesc, values, nulls);
}
pfree(sync_priority);
/* clean up and return the tuplestore */
tuplestore_donestoring(tupstore);
return (Datum) 0;
}
int
smgrGetAppendOnlyMirrorResyncEofs(EndXactRecKind endXactRecKind,
PersistentEndXactAppendOnlyMirrorResyncEofs **ptr)
{
int nestLevel = GetCurrentTransactionNestLevel();
int nentries;
PersistentEndXactAppendOnlyMirrorResyncEofs *rptr;
HASH_SEQ_STATUS iterateStatus;
AppendOnlyMirrorResyncEofs *entry;
int entryIndex;
if (endXactRecKind == EndXactRecKind_Abort)
{
/*
* No Append-Only Mirror Resync EOF information needed on abort.
*/
*ptr = NULL;
return 0;
}
nentries = 0;
if (AppendOnlyMirrorResyncEofsTable != NULL)
{
hash_seq_init(&iterateStatus,
AppendOnlyMirrorResyncEofsTable);
while ((entry = hash_seq_search(&iterateStatus)) != NULL)
{
if (entry->key.nestLevel >= nestLevel)
nentries++;
}
}
if (nentries == 0)
{
*ptr = NULL;
return 0;
}
if (Debug_persistent_print ||
Debug_persistent_appendonly_commit_count_print)
elog(Persistent_DebugPrintLevel(),
"Storage Manager: Get Append-Only mirror resync eofs list entries (current transaction nest level %d, Append-Only commit work system count %d)",
nestLevel,
FileRepPrimary_GetAppendOnlyCommitWorkCount());
rptr = (PersistentEndXactAppendOnlyMirrorResyncEofs *)
palloc(nentries * sizeof(PersistentEndXactAppendOnlyMirrorResyncEofs));
*ptr = rptr;
entryIndex = 0;
hash_seq_init(&iterateStatus, AppendOnlyMirrorResyncEofsTable);
while ((entry = hash_seq_search(&iterateStatus)) != NULL)
{
MIRRORED_LOCK_DECLARE;
bool returned;
int resultSystemAppendOnlyCommitCount;
returned = false;
if (entry->key.nestLevel >= nestLevel)
{
MIRRORED_LOCK;
MirroredAppendOnly_EndXactCatchup(entryIndex,
&entry->key.relFileNode,
entry->key.segmentFileNum,
entry->key.nestLevel,
entry->relationName,
&entry->persistentTid,
entry->persistentSerialNum,
&mirroredLockLocalVars,
entry->mirrorCatchupRequired,
entry->mirrorDataLossTrackingState,
entry->mirrorDataLossTrackingSessionNum,
entry->mirrorNewEof);
/*
* See if the mirror situation for this Append-Only segment file
* has changed since we flushed it to disk.
*/
rptr->relFileNode = entry->key.relFileNode;
rptr->segmentFileNum = entry->key.segmentFileNum;
rptr->persistentTid = entry->persistentTid;
rptr->persistentSerialNum = entry->persistentSerialNum;
if (entry->mirrorCatchupRequired)
{
rptr->mirrorLossEof = INT64CONST(-1);
}
else
{
rptr->mirrorLossEof = entry->mirrorNewEof;
}
rptr->mirrorNewEof = entry->mirrorNewEof;
rptr++;
returned = true;
START_CRIT_SECTION();
LWLockAcquire(FileRepAppendOnlyCommitCountLock, LW_EXCLUSIVE);
resultSystemAppendOnlyCommitCount =
FileRepPrimary_IntentAppendOnlyCommitWork();
/* Set this inside the Critical Section. */
entry->didIncrementCommitCount = true;
if (endXactRecKind == EndXactRecKind_Prepare)
{
char gid[TMGIDSIZE];
if (!getDistributedTransactionIdentifier(gid))
elog(ERROR, "Unable to obtain gid during prepare");
PrepareIntentAppendOnlyCommitWork(gid);
entry->isDistributedTransaction = true;
memcpy(entry->gid, gid, TMGIDSIZE);
}
pendingAppendOnlyMirrorResyncIntentCount++;
}
else
{
MIRRORED_LOCK;
START_CRIT_SECTION();
LWLockAcquire(FileRepAppendOnlyCommitCountLock, LW_EXCLUSIVE);
resultSystemAppendOnlyCommitCount =
FileRepPrimary_GetAppendOnlyCommitWorkCount();
}
if (Debug_persistent_print ||
Debug_persistent_appendonly_commit_count_print)
{
if (entry->relationName == NULL)
elog(Persistent_DebugPrintLevel(),
"Storage Manager: Get Append-Only mirror resync eofs list entry #%d: %u/%u/%u, segment file #%d "
"(returned %s, result system Append-Only commit count %d, transaction nest level %d, persistent TID %s, persistent serial number " INT64_FORMAT ", mirror catchup required %s, mirror new EOF " INT64_FORMAT ")",
entryIndex,
entry->key.relFileNode.spcNode,
entry->key.relFileNode.dbNode,
entry->key.relFileNode.relNode,
entry->key.segmentFileNum,
(returned ? "true" : "false"),
resultSystemAppendOnlyCommitCount,
entry->key.nestLevel,
ItemPointerToString(&entry->persistentTid),
entry->persistentSerialNum,
(entry->mirrorCatchupRequired ? "true" : "false"),
entry->mirrorNewEof);
else
elog(Persistent_DebugPrintLevel(),
"Storage Manager: Get Append-Only mirror resync eofs list entry #%d: %u/%u/%u, segment file #%d, relation name '%s' "
"(returned %s, result system Append-Only commit count %d, transaction nest level %d, persistent TID %s, persistent serial number " INT64_FORMAT ", mirror catchup required %s, mirror new EOF " INT64_FORMAT ")",
entryIndex,
entry->key.relFileNode.spcNode,
entry->key.relFileNode.dbNode,
entry->key.relFileNode.relNode,
entry->key.segmentFileNum,
entry->relationName,
(returned ? "true" : "false"),
resultSystemAppendOnlyCommitCount,
entry->key.nestLevel,
ItemPointerToString(&entry->persistentTid),
entry->persistentSerialNum,
(entry->mirrorCatchupRequired ? "true" : "false"),
entry->mirrorNewEof);
}
LWLockRelease(FileRepAppendOnlyCommitCountLock);
END_CRIT_SECTION();
MIRRORED_UNLOCK;
entryIndex++;
}
return nentries;
}
/*
* Hot Standby feedback
*/
static void
ProcessStandbyHSFeedbackMessage(void)
{
StandbyHSFeedbackMessage reply;
TransactionId newxmin = InvalidTransactionId;
pq_copymsgbytes(&reply_message, (char *) &reply, sizeof(StandbyHSFeedbackMessage));
elog(DEBUG2, "hot standby feedback xmin %u epoch %u",
reply.xmin,
reply.epoch);
/*
* Update the WalSender's proc xmin to allow it to be visible to
* snapshots. This will hold back the removal of dead rows and thereby
* prevent the generation of cleanup conflicts on the standby server.
*/
if (TransactionIdIsValid(reply.xmin))
{
TransactionId nextXid;
uint32 nextEpoch;
bool epochOK = false;
GetNextXidAndEpoch(&nextXid, &nextEpoch);
/*
* Epoch of oldestXmin should be same as standby or if the counter has
* wrapped, then one less than reply.
*/
if (reply.xmin <= nextXid)
{
if (reply.epoch == nextEpoch)
epochOK = true;
}
else
{
if (nextEpoch > 0 && reply.epoch == nextEpoch - 1)
epochOK = true;
}
/*
* Feedback from standby must not go backwards, nor should it go
* forwards further than our most recent xid.
*/
if (epochOK && TransactionIdPrecedesOrEquals(reply.xmin, nextXid))
{
if (!TransactionIdIsValid(MyProc->xmin))
{
TransactionId oldestXmin = GetOldestXmin(true, true);
if (TransactionIdPrecedes(oldestXmin, reply.xmin))
newxmin = reply.xmin;
else
newxmin = oldestXmin;
}
else
{
if (TransactionIdPrecedes(MyProc->xmin, reply.xmin))
newxmin = reply.xmin;
else
newxmin = MyProc->xmin; /* stay the same */
}
}
}
/*
* Grab the ProcArrayLock to set xmin, or invalidate for bad reply
*/
if (MyProc->xmin != newxmin)
{
LWLockAcquire(ProcArrayLock, LW_SHARED);
MyProc->xmin = newxmin;
LWLockRelease(ProcArrayLock);
}
}
/*
* CheckPointTwoPhase -- handle 2PC component of checkpointing.
*
* We must fsync the state file of any GXACT that is valid and has a PREPARE
* LSN <= the checkpoint's redo horizon. (If the gxact isn't valid yet or
* has a later LSN, this checkpoint is not responsible for fsyncing it.)
*
* This is deliberately run as late as possible in the checkpoint sequence,
* because GXACTs ordinarily have short lifespans, and so it is quite
* possible that GXACTs that were valid at checkpoint start will no longer
* exist if we wait a little bit.
*
* If a GXACT remains valid across multiple checkpoints, it'll be fsynced
* each time. This is considered unusual enough that we don't bother to
* expend any extra code to avoid the redundant fsyncs. (They should be
* reasonably cheap anyway, since they won't cause I/O.)
*/
void
CheckPointTwoPhase(XLogRecPtr redo_horizon)
{
TransactionId *xids;
int nxids;
char path[MAXPGPATH];
int i;
/*
* We don't want to hold the TwoPhaseStateLock while doing I/O, so we grab
* it just long enough to make a list of the XIDs that require fsyncing,
* and then do the I/O afterwards.
*
* This approach creates a race condition: someone else could delete a
* GXACT between the time we release TwoPhaseStateLock and the time we try
* to open its state file. We handle this by special-casing ENOENT
* failures: if we see that, we verify that the GXACT is no longer valid,
* and if so ignore the failure.
*/
if (max_prepared_xacts <= 0)
return; /* nothing to do */
TRACE_POSTGRESQL_TWOPHASE_CHECKPOINT_START();
xids = (TransactionId *) palloc(max_prepared_xacts * sizeof(TransactionId));
nxids = 0;
LWLockAcquire(TwoPhaseStateLock, LW_SHARED);
for (i = 0; i < TwoPhaseState->numPrepXacts; i++)
{
GlobalTransaction gxact = TwoPhaseState->prepXacts[i];
if (gxact->valid &&
XLByteLE(gxact->prepare_lsn, redo_horizon))
xids[nxids++] = gxact->proc.xid;
}
LWLockRelease(TwoPhaseStateLock);
for (i = 0; i < nxids; i++)
{
TransactionId xid = xids[i];
int fd;
TwoPhaseFilePath(path, xid);
fd = BasicOpenFile(path, O_RDWR | PG_BINARY, 0);
if (fd < 0)
{
if (errno == ENOENT)
{
/* OK if gxact is no longer valid */
if (!TransactionIdIsPrepared(xid))
continue;
/* Restore errno in case it was changed */
errno = ENOENT;
}
ereport(ERROR,
(errcode_for_file_access(),
errmsg("could not open two-phase state file \"%s\": %m",
path)));
}
if (pg_fsync(fd) != 0)
{
close(fd);
ereport(ERROR,
(errcode_for_file_access(),
errmsg("could not fsync two-phase state file \"%s\": %m",
path)));
}
if (close(fd) != 0)
ereport(ERROR,
(errcode_for_file_access(),
errmsg("could not close two-phase state file \"%s\": %m",
path)));
}
pfree(xids);
TRACE_POSTGRESQL_TWOPHASE_CHECKPOINT_DONE();
}
/*
* Store some statistics for a statement.
*/
static void
pgss_store(const char *query, double total_time, uint64 rows,
const BufferUsage *bufusage)
{
pgssHashKey key;
double usage;
pgssEntry *entry;
Assert(query != NULL);
/* Safety check... */
if (!pgss || !pgss_hash)
return;
/* Set up key for hashtable search */
key.userid = GetUserId();
key.dbid = MyDatabaseId;
key.encoding = GetDatabaseEncoding();
key.query_len = strlen(query);
if (key.query_len >= pgss->query_size)
key.query_len = pg_encoding_mbcliplen(key.encoding,
query,
key.query_len,
pgss->query_size - 1);
key.query_ptr = query;
usage = USAGE_EXEC(duration);
/* Lookup the hash table entry with shared lock. */
LWLockAcquire(pgss->lock, LW_SHARED);
entry = (pgssEntry *) hash_search(pgss_hash, &key, HASH_FIND, NULL);
if (!entry)
{
/* Must acquire exclusive lock to add a new entry. */
LWLockRelease(pgss->lock);
LWLockAcquire(pgss->lock, LW_EXCLUSIVE);
entry = entry_alloc(&key);
}
/* Grab the spinlock while updating the counters. */
{
volatile pgssEntry *e = (volatile pgssEntry *) entry;
SpinLockAcquire(&e->mutex);
e->counters.calls += 1;
e->counters.total_time += total_time;
e->counters.rows += rows;
e->counters.shared_blks_hit += bufusage->shared_blks_hit;
e->counters.shared_blks_read += bufusage->shared_blks_read;
e->counters.shared_blks_written += bufusage->shared_blks_written;
e->counters.local_blks_hit += bufusage->local_blks_hit;
e->counters.local_blks_read += bufusage->local_blks_read;
e->counters.local_blks_written += bufusage->local_blks_written;
e->counters.temp_blks_read += bufusage->temp_blks_read;
e->counters.temp_blks_written += bufusage->temp_blks_written;
e->counters.usage += usage;
SpinLockRelease(&e->mutex);
}
LWLockRelease(pgss->lock);
}
/*
* MarkAsPreparing
* Reserve the GID for the given transaction.
*
* Internally, this creates a gxact struct and puts it into the active array.
* NOTE: this is also used when reloading a gxact after a crash; so avoid
* assuming that we can use very much backend context.
*/
GlobalTransaction
MarkAsPreparing(TransactionId xid, const char *gid,
TimestampTz prepared_at, Oid owner, Oid databaseid)
{
GlobalTransaction gxact;
int i;
if (strlen(gid) >= GIDSIZE)
ereport(ERROR,
(errcode(ERRCODE_INVALID_PARAMETER_VALUE),
errmsg("transaction identifier \"%s\" is too long",
gid)));
/* fail immediately if feature is disabled */
if (max_prepared_xacts == 0)
ereport(ERROR,
(errcode(ERRCODE_OBJECT_NOT_IN_PREREQUISITE_STATE),
errmsg("prepared transactions are disabled"),
errhint("Set max_prepared_transactions to a nonzero value.")));
LWLockAcquire(TwoPhaseStateLock, LW_EXCLUSIVE);
/*
* First, find and recycle any gxacts that failed during prepare. We do
* this partly to ensure we don't mistakenly say their GIDs are still
* reserved, and partly so we don't fail on out-of-slots unnecessarily.
*/
for (i = 0; i < TwoPhaseState->numPrepXacts; i++)
{
gxact = TwoPhaseState->prepXacts[i];
if (!gxact->valid && !TransactionIdIsActive(gxact->locking_xid))
{
/* It's dead Jim ... remove from the active array */
TwoPhaseState->numPrepXacts--;
TwoPhaseState->prepXacts[i] = TwoPhaseState->prepXacts[TwoPhaseState->numPrepXacts];
/* and put it back in the freelist */
gxact->proc.links.next = (SHM_QUEUE *) TwoPhaseState->freeGXacts;
TwoPhaseState->freeGXacts = gxact;
/* Back up index count too, so we don't miss scanning one */
i--;
}
}
/* Check for conflicting GID */
for (i = 0; i < TwoPhaseState->numPrepXacts; i++)
{
gxact = TwoPhaseState->prepXacts[i];
if (strcmp(gxact->gid, gid) == 0)
{
ereport(ERROR,
(errcode(ERRCODE_DUPLICATE_OBJECT),
errmsg("transaction identifier \"%s\" is already in use",
gid)));
}
}
/* Get a free gxact from the freelist */
if (TwoPhaseState->freeGXacts == NULL)
ereport(ERROR,
(errcode(ERRCODE_OUT_OF_MEMORY),
errmsg("maximum number of prepared transactions reached"),
errhint("Increase max_prepared_transactions (currently %d).",
max_prepared_xacts)));
gxact = TwoPhaseState->freeGXacts;
TwoPhaseState->freeGXacts = (GlobalTransaction) gxact->proc.links.next;
/* Initialize it */
MemSet(&gxact->proc, 0, sizeof(PGPROC));
SHMQueueElemInit(&(gxact->proc.links));
gxact->proc.waitStatus = STATUS_OK;
/* We set up the gxact's VXID as InvalidBackendId/XID */
gxact->proc.lxid = (LocalTransactionId) xid;
gxact->proc.xid = xid;
gxact->proc.xmin = InvalidTransactionId;
gxact->proc.pid = 0;
gxact->proc.backendId = InvalidBackendId;
gxact->proc.databaseId = databaseid;
gxact->proc.roleId = owner;
gxact->proc.inCommit = false;
gxact->proc.vacuumFlags = 0;
gxact->proc.lwWaiting = false;
gxact->proc.lwExclusive = false;
gxact->proc.lwWaitLink = NULL;
gxact->proc.waitLock = NULL;
gxact->proc.waitProcLock = NULL;
for (i = 0; i < NUM_LOCK_PARTITIONS; i++)
SHMQueueInit(&(gxact->proc.myProcLocks[i]));
/* subxid data must be filled later by GXactLoadSubxactData */
gxact->proc.subxids.overflowed = false;
gxact->proc.subxids.nxids = 0;
gxact->prepared_at = prepared_at;
/* initialize LSN to 0 (start of WAL) */
gxact->prepare_lsn.xlogid = 0;
gxact->prepare_lsn.xrecoff = 0;
gxact->owner = owner;
gxact->locking_xid = xid;
gxact->valid = false;
strcpy(gxact->gid, gid);
/* And insert it into the active array */
Assert(TwoPhaseState->numPrepXacts < max_prepared_xacts);
TwoPhaseState->prepXacts[TwoPhaseState->numPrepXacts++] = gxact;
LWLockRelease(TwoPhaseStateLock);
return gxact;
}
/*
* Interrogate the commit timestamp of a transaction.
*
* The return value indicates whether a commit timestamp record was found for
* the given xid. The timestamp value is returned in *ts (which may not be
* null), and the origin node for the Xid is returned in *nodeid, if it's not
* null.
*/
bool
TransactionIdGetCommitTsData(TransactionId xid, TimestampTz *ts,
RepOriginId *nodeid)
{
int pageno = TransactionIdToCTsPage(xid);
int entryno = TransactionIdToCTsEntry(xid);
int slotno;
CommitTimestampEntry entry;
TransactionId oldestCommitTs;
TransactionId newestCommitTs;
/* error if the given Xid doesn't normally commit */
if (!TransactionIdIsNormal(xid))
ereport(ERROR,
(errcode(ERRCODE_INVALID_PARAMETER_VALUE),
errmsg("cannot retrieve commit timestamp for transaction %u", xid)));
LWLockAcquire(CommitTsLock, LW_SHARED);
/* Error if module not enabled */
if (!commitTsShared->commitTsActive)
error_commit_ts_disabled();
/*
* If we're asked for the cached value, return that. Otherwise, fall
* through to read from SLRU.
*/
if (commitTsShared->xidLastCommit == xid)
{
*ts = commitTsShared->dataLastCommit.time;
if (nodeid)
*nodeid = commitTsShared->dataLastCommit.nodeid;
LWLockRelease(CommitTsLock);
return *ts != 0;
}
oldestCommitTs = ShmemVariableCache->oldestCommitTs;
newestCommitTs = ShmemVariableCache->newestCommitTs;
/* neither is invalid, or both are */
Assert(TransactionIdIsValid(oldestCommitTs) == TransactionIdIsValid(newestCommitTs));
LWLockRelease(CommitTsLock);
/*
* Return empty if the requested value is outside our valid range.
*/
if (!TransactionIdIsValid(oldestCommitTs) ||
TransactionIdPrecedes(xid, oldestCommitTs) ||
TransactionIdPrecedes(newestCommitTs, xid))
{
*ts = 0;
if (nodeid)
*nodeid = InvalidRepOriginId;
return false;
}
/* lock is acquired by SimpleLruReadPage_ReadOnly */
slotno = SimpleLruReadPage_ReadOnly(CommitTsCtl, pageno, xid);
memcpy(&entry,
CommitTsCtl->shared->page_buffer[slotno] +
SizeOfCommitTimestampEntry * entryno,
SizeOfCommitTimestampEntry);
*ts = entry.time;
if (nodeid)
*nodeid = entry.nodeid;
LWLockRelease(CommitTsControlLock);
return *ts != 0;
}
/*
* LockGXact
* Locate the prepared transaction and mark it busy for COMMIT or PREPARE.
*/
static GlobalTransaction
LockGXact(const char *gid, Oid user)
{
int i;
LWLockAcquire(TwoPhaseStateLock, LW_EXCLUSIVE);
for (i = 0; i < TwoPhaseState->numPrepXacts; i++)
{
GlobalTransaction gxact = TwoPhaseState->prepXacts[i];
/* Ignore not-yet-valid GIDs */
if (!gxact->valid)
continue;
if (strcmp(gxact->gid, gid) != 0)
continue;
/* Found it, but has someone else got it locked? */
if (TransactionIdIsValid(gxact->locking_xid))
{
if (TransactionIdIsActive(gxact->locking_xid))
ereport(ERROR,
(errcode(ERRCODE_OBJECT_NOT_IN_PREREQUISITE_STATE),
errmsg("prepared transaction with identifier \"%s\" is busy",
gid)));
gxact->locking_xid = InvalidTransactionId;
}
if (user != gxact->owner && !superuser_arg(user))
ereport(ERROR,
(errcode(ERRCODE_INSUFFICIENT_PRIVILEGE),
errmsg("permission denied to finish prepared transaction"),
errhint("Must be superuser or the user that prepared the transaction.")));
/*
* Note: it probably would be possible to allow committing from
* another database; but at the moment NOTIFY is known not to work and
* there may be some other issues as well. Hence disallow until
* someone gets motivated to make it work.
*/
if (MyDatabaseId != gxact->proc.databaseId)
ereport(ERROR,
(errcode(ERRCODE_FEATURE_NOT_SUPPORTED),
errmsg("prepared transaction belongs to another database"),
errhint("Connect to the database where the transaction was prepared to finish it.")));
/* OK for me to lock it */
gxact->locking_xid = GetTopTransactionId();
LWLockRelease(TwoPhaseStateLock);
return gxact;
}
LWLockRelease(TwoPhaseStateLock);
ereport(ERROR,
(errcode(ERRCODE_UNDEFINED_OBJECT),
errmsg("prepared transaction with identifier \"%s\" does not exist",
gid)));
/* NOTREACHED */
return NULL;
}
void
PersistentTablespace_AddMirrorAll(
int16 pridbid,
int16 mirdbid)
{
TablespaceDirEntry tablespaceDirEntry;
HASH_SEQ_STATUS hstat;
WRITE_PERSISTENT_STATE_ORDERED_LOCK_DECLARE;
hash_seq_init(&hstat, persistentTablespaceSharedHashTable);
if (Persistent_BeforePersistenceWork())
elog(ERROR, "persistent table changes forbidden");
PersistentTablespace_VerifyInitScan();
WRITE_PERSISTENT_STATE_ORDERED_LOCK;
LWLockAcquire(TablespaceHashLock, LW_SHARED);
while ((tablespaceDirEntry = hash_seq_search(&hstat)) != NULL)
{
PersistentFileSysObjName fsObjName;
Oid tblspc = tablespaceDirEntry->key.tablespaceOid;
ItemPointerData persistentTid;
uint64 persistentSerialNum;
tablespaceDirEntry = PersistentTablespace_FindEntryUnderLock(tblspc);
if (tablespaceDirEntry == NULL)
elog(ERROR, "Did not find persistent tablespace entry %u",
tblspc);
persistentSerialNum = tablespaceDirEntry->persistentSerialNum;
ItemPointerCopy(&tablespaceDirEntry->persistentTid, &persistentTid);
/*
* We release TablespaceHashLock in the middle of the loop and
* re-acquire it after doing persistent table change. This is needed
* to prevent holding the lock for any purpose other than to protect
* the tablespace shared hash table. Not releasing this lock could
* result in file I/O and potential deadlock due to other LW locks
* being acquired in the process. Releasing the lock this way is safe
* because we are still holding PersistentObjLock in exclusive mode.
* Any change to the filespace shared hash table is also protected by
* PersistentObjLock.
*/
LWLockRelease(TablespaceHashLock);
PersistentFileSysObjName_SetTablespaceDir(&fsObjName, tblspc);
PersistentFileSysObj_AddMirror(&fsObjName,
&persistentTid,
persistentSerialNum,
pridbid,
mirdbid,
NULL,
true,
/* flushToXlog */ false);
LWLockAcquire(TablespaceHashLock, LW_SHARED);
}
LWLockRelease(TablespaceHashLock);
WRITE_PERSISTENT_STATE_ORDERED_UNLOCK;
}
/*
* StrategyGetBuffer
*
* Called by the bufmgr to get the next candidate buffer to use in
* BufferAlloc(). The only hard requirement BufferAlloc() has is that
* the selected buffer must not currently be pinned by anyone.
*
* strategy is a BufferAccessStrategy object, or NULL for default strategy.
*
* To ensure that no one else can pin the buffer before we do, we must
* return the buffer with the buffer header spinlock still held. If
* *lock_held is set on exit, we have returned with the BufFreelistLock
* still held, as well; the caller must release that lock once the spinlock
* is dropped. We do it that way because releasing the BufFreelistLock
* might awaken other processes, and it would be bad to do the associated
* kernel calls while holding the buffer header spinlock.
*/
volatile BufferDesc *
StrategyGetBuffer(BufferAccessStrategy strategy, bool *lock_held)
{
volatile BufferDesc *buf;
Latch *bgwriterLatch;
int trycounter;
/*
* If given a strategy object, see whether it can select a buffer. We
* assume strategy objects don't need the BufFreelistLock.
*/
if (strategy != NULL)
{
buf = GetBufferFromRing(strategy);
if (buf != NULL)
{
*lock_held = false;
return buf;
}
}
/* Nope, so lock the freelist */
*lock_held = true;
LWLockAcquire(BufFreelistLock, LW_EXCLUSIVE);
/*
* We count buffer allocation requests so that the bgwriter can estimate
* the rate of buffer consumption. Note that buffers recycled by a
* strategy object are intentionally not counted here.
*/
StrategyControl->numBufferAllocs++;
/*
* If bgwriterLatch is set, we need to waken the bgwriter, but we should
* not do so while holding BufFreelistLock; so release and re-grab. This
* is annoyingly tedious, but it happens at most once per bgwriter cycle,
* so the performance hit is minimal.
*/
bgwriterLatch = StrategyControl->bgwriterLatch;
if (bgwriterLatch)
{
StrategyControl->bgwriterLatch = NULL;
LWLockRelease(BufFreelistLock);
SetLatch(bgwriterLatch);
LWLockAcquire(BufFreelistLock, LW_EXCLUSIVE);
}
/*
* Try to get a buffer from the freelist. Note that the freeNext fields
* are considered to be protected by the BufFreelistLock not the
* individual buffer spinlocks, so it's OK to manipulate them without
* holding the spinlock.
*/
while (StrategyControl->firstFreeBuffer >= 0)
{
buf = &BufferDescriptors[StrategyControl->firstFreeBuffer];
Assert(buf->freeNext != FREENEXT_NOT_IN_LIST);
/* Unconditionally remove buffer from freelist */
StrategyControl->firstFreeBuffer = buf->freeNext;
buf->freeNext = FREENEXT_NOT_IN_LIST;
/*
* If the buffer is pinned or has a nonzero usage_count, we cannot use
* it; discard it and retry. (This can only happen if VACUUM put a
* valid buffer in the freelist and then someone else used it before
* we got to it. It's probably impossible altogether as of 8.3, but
* we'd better check anyway.)
*/
LockBufHdr(buf);
if (buf->refcount == 0 && buf->usage_count == 0)
{
if (strategy != NULL)
AddBufferToRing(strategy, buf);
return buf;
}
UnlockBufHdr(buf);
}
if(true)
{
/* Now using LRU buffer replacement policy */
int currMinTimeStamp = INT_MAX;
BufferStrategyControl *currTargetBuffer;
for (StrategyControl->nextVictimBuffer=0;StrategyControl->nextVictimBuffer<NBuffers;StrategyControl->nextVictimBuffer++)
{
buf = &BufferDescriptors[StrategyControl->nextVictimBuffer];
if(StrategyControl->nextVictimBuffer == NBuffers-1 && currMinTimeStamp == INT_MAX)
{
StrategyControl ->completePasses++;
elog(ERROR,"No available buffer frame");
}
LockBufHdr(buf);
if(buf->refcount==0 && buf->timer<currMinTimeStamp)
{
currMinTimeStamp = buf->timer;
currTargetBuffer = buf;
}
UnlockBufHdr(buf);
}
return currTargetBuffer;
}
else
{
/* Nothing on the freelist, so run the "clock sweep" algorithm */
trycounter = NBuffers;
for (;;)
{
buf = &BufferDescriptors[StrategyControl->nextVictimBuffer];
if (++StrategyControl->nextVictimBuffer >= NBuffers)
{
StrategyControl->nextVictimBuffer = 0;
StrategyControl->completePasses++;
}
/*
* If the buffer is pinned or has a nonzero usage_count, we cannot use
* it; decrement the usage_count (unless pinned) and keep scanning.
*/
LockBufHdr(buf);
if (buf->refcount == 0)
{
if (buf->usage_count > 0)
{
buf->usage_count--;
trycounter = NBuffers;
}
else
{
/* Found a usable buffer */
if (strategy != NULL)
AddBufferToRing(strategy, buf);
return buf;
}
}
else if (--trycounter == 0)
{
/*
* We've scanned all the buffers without making any state changes,
* so all the buffers are pinned (or were when we looked at them).
* We could hope that someone will free one eventually, but it's
* probably better to fail than to risk getting stuck in an
* infinite loop.
*/
UnlockBufHdr(buf);
elog(ERROR, "no unpinned buffers available");
}
UnlockBufHdr(buf);
}
/* not reached */
return NULL;
}
}
/*
* SharedInvalBackendInit
* Initialize a new backend to operate on the sinval buffer
*/
void
SharedInvalBackendInit(bool sendOnly)
{
int index;
ProcState *stateP = NULL;
SISeg *segP = shmInvalBuffer;
/*
* This can run in parallel with read operations, and for that matter with
* write operations; but not in parallel with additions and removals of
* backends, nor in parallel with SICleanupQueue. It doesn't seem worth
* having a third lock, so we choose to use SInvalWriteLock to serialize
* additions/removals.
*/
LWLockAcquire(SInvalWriteLock, LW_EXCLUSIVE);
/* Look for a free entry in the procState array */
for (index = 0; index < segP->lastBackend; index++)
{
if (segP->procState[index].procPid == 0) /* inactive slot? */
{
stateP = &segP->procState[index];
break;
}
}
if (stateP == NULL)
{
if (segP->lastBackend < segP->maxBackends)
{
stateP = &segP->procState[segP->lastBackend];
Assert(stateP->procPid == 0);
segP->lastBackend++;
}
else
{
/*
* out of procState slots: MaxBackends exceeded -- report normally
*/
MyBackendId = InvalidBackendId;
LWLockRelease(SInvalWriteLock);
ereport(FATAL,
(errcode(ERRCODE_TOO_MANY_CONNECTIONS),
errmsg("sorry, too many clients already")));
}
}
MyBackendId = (stateP - &segP->procState[0]) + 1;
/* Advertise assigned backend ID in MyProc */
MyProc->backendId = MyBackendId;
/* Fetch next local transaction ID into local memory */
nextLocalTransactionId = stateP->nextLXID;
/* mark myself active, with all extant messages already read */
stateP->procPid = MyProcPid;
stateP->nextMsgNum = segP->maxMsgNum;
stateP->resetState = false;
stateP->signaled = false;
stateP->sendOnly = sendOnly;
LWLockRelease(SInvalWriteLock);
/* register exit routine to mark my entry inactive at exit */
on_shmem_exit(CleanupInvalidationState, PointerGetDatum(segP));
elog(DEBUG4, "my backend id is %d", MyBackendId);
}
/*
* CheckDeadLock
*
* We only get to this routine, if DEADLOCK_TIMEOUT fired while waiting for a
* lock to be released by some other process. Check if there's a deadlock; if
* not, just return. (But signal ProcSleep to log a message, if
* log_lock_waits is true.) If we have a real deadlock, remove ourselves from
* the lock's wait queue and signal an error to ProcSleep.
*/
static void
CheckDeadLock(void)
{
int i;
/*
* Acquire exclusive lock on the entire shared lock data structures. Must
* grab LWLocks in partition-number order to avoid LWLock deadlock.
*
* Note that the deadlock check interrupt had better not be enabled
* anywhere that this process itself holds lock partition locks, else this
* will wait forever. Also note that LWLockAcquire creates a critical
* section, so that this routine cannot be interrupted by cancel/die
* interrupts.
*/
for (i = 0; i < NUM_LOCK_PARTITIONS; i++)
LWLockAcquire(LockHashPartitionLockByIndex(i), LW_EXCLUSIVE);
/*
* Check to see if we've been awoken by anyone in the interim.
*
* If we have, we can return and resume our transaction -- happy day.
* Before we are awoken the process releasing the lock grants it to us so
* we know that we don't have to wait anymore.
*
* We check by looking to see if we've been unlinked from the wait queue.
* This is quicker than checking our semaphore's state, since no kernel
* call is needed, and it is safe because we hold the lock partition lock.
*/
if (MyProc->links.prev == NULL ||
MyProc->links.next == NULL)
goto check_done;
#ifdef LOCK_DEBUG
if (Debug_deadlocks)
DumpAllLocks();
#endif
/* Run the deadlock check, and set deadlock_state for use by ProcSleep */
deadlock_state = DeadLockCheck(MyProc);
if (deadlock_state == DS_HARD_DEADLOCK)
{
/*
* Oops. We have a deadlock.
*
* Get this process out of wait state. (Note: we could do this more
* efficiently by relying on lockAwaited, but use this coding to
* preserve the flexibility to kill some other transaction than the
* one detecting the deadlock.)
*
* RemoveFromWaitQueue sets MyProc->waitStatus to STATUS_ERROR, so
* ProcSleep will report an error after we return from the signal
* handler.
*/
Assert(MyProc->waitLock != NULL);
RemoveFromWaitQueue(MyProc, LockTagHashCode(&(MyProc->waitLock->tag)));
/*
* We're done here. Transaction abort caused by the error that
* ProcSleep will raise will cause any other locks we hold to be
* released, thus allowing other processes to wake up; we don't need
* to do that here. NOTE: an exception is that releasing locks we
* hold doesn't consider the possibility of waiters that were blocked
* behind us on the lock we just failed to get, and might now be
* wakable because we're not in front of them anymore. However,
* RemoveFromWaitQueue took care of waking up any such processes.
*/
}
/*
* And release locks. We do this in reverse order for two reasons: (1)
* Anyone else who needs more than one of the locks will be trying to lock
* them in increasing order; we don't want to release the other process
* until it can get all the locks it needs. (2) This avoids O(N^2)
* behavior inside LWLockRelease.
*/
check_done:
for (i = NUM_LOCK_PARTITIONS; --i >= 0;)
LWLockRelease(LockHashPartitionLockByIndex(i));
}
/*
* SIGetDataEntries
* get next SI message(s) for current backend, if there are any
*
* Possible return values:
* 0: no SI message available
* n>0: next n SI messages have been extracted into data[]
* -1: SI reset message extracted
*
* If the return value is less than the array size "datasize", the caller
* can assume that there are no more SI messages after the one(s) returned.
* Otherwise, another call is needed to collect more messages.
*
* NB: this can run in parallel with other instances of SIGetDataEntries
* executing on behalf of other backends, since each instance will modify only
* fields of its own backend's ProcState, and no instance will look at fields
* of other backends' ProcStates. We express this by grabbing SInvalReadLock
* in shared mode. Note that this is not exactly the normal (read-only)
* interpretation of a shared lock! Look closely at the interactions before
* allowing SInvalReadLock to be grabbed in shared mode for any other reason!
*
* NB: this can also run in parallel with SIInsertDataEntries. It is not
* guaranteed that we will return any messages added after the routine is
* entered.
*
* Note: we assume that "datasize" is not so large that it might be important
* to break our hold on SInvalReadLock into segments.
*/
int
SIGetDataEntries(SharedInvalidationMessage *data, int datasize)
{
SISeg *segP;
ProcState *stateP;
int max;
int n;
LWLockAcquire(SInvalReadLock, LW_SHARED);
segP = shmInvalBuffer;
stateP = &segP->procState[MyBackendId - 1];
/* Fetch current value of maxMsgNum using spinlock */
{
/* use volatile pointer to prevent code rearrangement */
volatile SISeg *vsegP = segP;
SpinLockAcquire(&vsegP->msgnumLock);
max = vsegP->maxMsgNum;
SpinLockRelease(&vsegP->msgnumLock);
}
if (stateP->resetState)
{
/*
* Force reset. We can say we have dealt with any messages added
* since the reset, as well; and that means we should clear the
* signaled flag, too.
*/
stateP->nextMsgNum = max;
stateP->resetState = false;
stateP->signaled = false;
LWLockRelease(SInvalReadLock);
return -1;
}
/*
* Retrieve messages and advance backend's counter, until data array is
* full or there are no more messages.
*
* There may be other backends that haven't read the message(s), so we
* cannot delete them here. SICleanupQueue() will eventually remove them
* from the queue.
*/
n = 0;
while (n < datasize && stateP->nextMsgNum < max)
{
data[n++] = segP->buffer[stateP->nextMsgNum % MAXNUMMESSAGES];
stateP->nextMsgNum++;
}
/*
* Reset our "signaled" flag whenever we have caught up completely.
*/
if (stateP->nextMsgNum >= max)
stateP->signaled = false;
LWLockRelease(SInvalReadLock);
return n;
}
/*
* ProcSleep -- put a process to sleep on the specified lock
*
* Caller must have set MyProc->heldLocks to reflect locks already held
* on the lockable object by this process (under all XIDs).
*
* The lock table's partition lock must be held at entry, and will be held
* at exit.
*
* Result: STATUS_OK if we acquired the lock, STATUS_ERROR if not (deadlock).
*
* ASSUME: that no one will fiddle with the queue until after
* we release the partition lock.
*
* NOTES: The process queue is now a priority queue for locking.
*/
int
ProcSleep(LOCALLOCK *locallock, LockMethod lockMethodTable)
{
LOCKMODE lockmode = locallock->tag.mode;
LOCK *lock = locallock->lock;
PROCLOCK *proclock = locallock->proclock;
uint32 hashcode = locallock->hashcode;
LWLock *partitionLock = LockHashPartitionLock(hashcode);
PROC_QUEUE *waitQueue = &(lock->waitProcs);
LOCKMASK myHeldLocks = MyProc->heldLocks;
bool early_deadlock = false;
bool allow_autovacuum_cancel = true;
int myWaitStatus;
PGPROC *proc;
int i;
/*
* Determine where to add myself in the wait queue.
*
* Normally I should go at the end of the queue. However, if I already
* hold locks that conflict with the request of any previous waiter, put
* myself in the queue just in front of the first such waiter. This is not
* a necessary step, since deadlock detection would move me to before that
* waiter anyway; but it's relatively cheap to detect such a conflict
* immediately, and avoid delaying till deadlock timeout.
*
* Special case: if I find I should go in front of some waiter, check to
* see if I conflict with already-held locks or the requests before that
* waiter. If not, then just grant myself the requested lock immediately.
* This is the same as the test for immediate grant in LockAcquire, except
* we are only considering the part of the wait queue before my insertion
* point.
*/
if (myHeldLocks != 0)
{
LOCKMASK aheadRequests = 0;
proc = (PGPROC *) waitQueue->links.next;
for (i = 0; i < waitQueue->size; i++)
{
/* Must he wait for me? */
if (lockMethodTable->conflictTab[proc->waitLockMode] & myHeldLocks)
{
/* Must I wait for him ? */
if (lockMethodTable->conflictTab[lockmode] & proc->heldLocks)
{
/*
* Yes, so we have a deadlock. Easiest way to clean up
* correctly is to call RemoveFromWaitQueue(), but we
* can't do that until we are *on* the wait queue. So, set
* a flag to check below, and break out of loop. Also,
* record deadlock info for later message.
*/
RememberSimpleDeadLock(MyProc, lockmode, lock, proc);
early_deadlock = true;
break;
}
/* I must go before this waiter. Check special case. */
if ((lockMethodTable->conflictTab[lockmode] & aheadRequests) == 0 &&
LockCheckConflicts(lockMethodTable,
lockmode,
lock,
proclock) == STATUS_OK)
{
/* Skip the wait and just grant myself the lock. */
GrantLock(lock, proclock, lockmode);
GrantAwaitedLock();
return STATUS_OK;
}
/* Break out of loop to put myself before him */
break;
}
/* Nope, so advance to next waiter */
aheadRequests |= LOCKBIT_ON(proc->waitLockMode);
proc = (PGPROC *) proc->links.next;
}
/*
* If we fall out of loop normally, proc points to waitQueue head, so
* we will insert at tail of queue as desired.
*/
}
else
{
/* I hold no locks, so I can't push in front of anyone. */
proc = (PGPROC *) &(waitQueue->links);
}
/*
* Insert self into queue, ahead of the given proc (or at tail of queue).
*/
SHMQueueInsertBefore(&(proc->links), &(MyProc->links));
waitQueue->size++;
lock->waitMask |= LOCKBIT_ON(lockmode);
/* Set up wait information in PGPROC object, too */
MyProc->waitLock = lock;
MyProc->waitProcLock = proclock;
MyProc->waitLockMode = lockmode;
MyProc->waitStatus = STATUS_WAITING;
/*
* If we detected deadlock, give up without waiting. This must agree with
* CheckDeadLock's recovery code, except that we shouldn't release the
* semaphore since we haven't tried to lock it yet.
*/
if (early_deadlock)
{
RemoveFromWaitQueue(MyProc, hashcode);
return STATUS_ERROR;
}
/* mark that we are waiting for a lock */
lockAwaited = locallock;
/*
* Release the lock table's partition lock.
*
* NOTE: this may also cause us to exit critical-section state, possibly
* allowing a cancel/die interrupt to be accepted. This is OK because we
* have recorded the fact that we are waiting for a lock, and so
* LockErrorCleanup will clean up if cancel/die happens.
*/
LWLockRelease(partitionLock);
/*
* Also, now that we will successfully clean up after an ereport, it's
* safe to check to see if there's a buffer pin deadlock against the
* Startup process. Of course, that's only necessary if we're doing Hot
* Standby and are not the Startup process ourselves.
*/
if (RecoveryInProgress() && !InRecovery)
CheckRecoveryConflictDeadlock();
/* Reset deadlock_state before enabling the timeout handler */
deadlock_state = DS_NOT_YET_CHECKED;
got_deadlock_timeout = false;
/*
* Set timer so we can wake up after awhile and check for a deadlock. If a
* deadlock is detected, the handler releases the process's semaphore and
* sets MyProc->waitStatus = STATUS_ERROR, allowing us to know that we
* must report failure rather than success.
*
* By delaying the check until we've waited for a bit, we can avoid
* running the rather expensive deadlock-check code in most cases.
*
* If LockTimeout is set, also enable the timeout for that. We can save a
* few cycles by enabling both timeout sources in one call.
*/
if (LockTimeout > 0)
{
EnableTimeoutParams timeouts[2];
timeouts[0].id = DEADLOCK_TIMEOUT;
timeouts[0].type = TMPARAM_AFTER;
timeouts[0].delay_ms = DeadlockTimeout;
timeouts[1].id = LOCK_TIMEOUT;
timeouts[1].type = TMPARAM_AFTER;
timeouts[1].delay_ms = LockTimeout;
enable_timeouts(timeouts, 2);
}
else
enable_timeout_after(DEADLOCK_TIMEOUT, DeadlockTimeout);
/*
* If somebody wakes us between LWLockRelease and WaitLatch, the latch
* will not wait. But a set latch does not necessarily mean that the lock
* is free now, as there are many other sources for latch sets than
* somebody releasing the lock.
*
* We process interrupts whenever the latch has been set, so cancel/die
* interrupts are processed quickly. This means we must not mind losing
* control to a cancel/die interrupt here. We don't, because we have no
* shared-state-change work to do after being granted the lock (the
* grantor did it all). We do have to worry about canceling the deadlock
* timeout and updating the locallock table, but if we lose control to an
* error, LockErrorCleanup will fix that up.
*/
do
{
WaitLatch(MyLatch, WL_LATCH_SET, 0);
ResetLatch(MyLatch);
/* check for deadlocks first, as that's probably log-worthy */
if (got_deadlock_timeout)
{
CheckDeadLock();
got_deadlock_timeout = false;
}
CHECK_FOR_INTERRUPTS();
/*
* waitStatus could change from STATUS_WAITING to something else
* asynchronously. Read it just once per loop to prevent surprising
* behavior (such as missing log messages).
*/
myWaitStatus = *((volatile int *) &MyProc->waitStatus);
/*
* If we are not deadlocked, but are waiting on an autovacuum-induced
* task, send a signal to interrupt it.
*/
if (deadlock_state == DS_BLOCKED_BY_AUTOVACUUM && allow_autovacuum_cancel)
{
PGPROC *autovac = GetBlockingAutoVacuumPgproc();
PGXACT *autovac_pgxact = &ProcGlobal->allPgXact[autovac->pgprocno];
LWLockAcquire(ProcArrayLock, LW_EXCLUSIVE);
/*
* Only do it if the worker is not working to protect against Xid
* wraparound.
*/
if ((autovac_pgxact->vacuumFlags & PROC_IS_AUTOVACUUM) &&
!(autovac_pgxact->vacuumFlags & PROC_VACUUM_FOR_WRAPAROUND))
{
int pid = autovac->pid;
StringInfoData locktagbuf;
StringInfoData logbuf; /* errdetail for server log */
initStringInfo(&locktagbuf);
initStringInfo(&logbuf);
DescribeLockTag(&locktagbuf, &lock->tag);
appendStringInfo(&logbuf,
_("Process %d waits for %s on %s."),
MyProcPid,
GetLockmodeName(lock->tag.locktag_lockmethodid,
lockmode),
locktagbuf.data);
/* release lock as quickly as possible */
LWLockRelease(ProcArrayLock);
ereport(LOG,
(errmsg("sending cancel to blocking autovacuum PID %d",
pid),
errdetail_log("%s", logbuf.data)));
pfree(logbuf.data);
pfree(locktagbuf.data);
/* send the autovacuum worker Back to Old Kent Road */
if (kill(pid, SIGINT) < 0)
{
/* Just a warning to allow multiple callers */
ereport(WARNING,
(errmsg("could not send signal to process %d: %m",
pid)));
}
}
else
LWLockRelease(ProcArrayLock);
/* prevent signal from being resent more than once */
allow_autovacuum_cancel = false;
}
/*
* If awoken after the deadlock check interrupt has run, and
* log_lock_waits is on, then report about the wait.
*/
if (log_lock_waits && deadlock_state != DS_NOT_YET_CHECKED)
{
StringInfoData buf,
lock_waiters_sbuf,
lock_holders_sbuf;
const char *modename;
long secs;
int usecs;
long msecs;
SHM_QUEUE *procLocks;
PROCLOCK *proclock;
bool first_holder = true,
first_waiter = true;
int lockHoldersNum = 0;
initStringInfo(&buf);
initStringInfo(&lock_waiters_sbuf);
initStringInfo(&lock_holders_sbuf);
DescribeLockTag(&buf, &locallock->tag.lock);
modename = GetLockmodeName(locallock->tag.lock.locktag_lockmethodid,
lockmode);
TimestampDifference(get_timeout_start_time(DEADLOCK_TIMEOUT),
GetCurrentTimestamp(),
&secs, &usecs);
msecs = secs * 1000 + usecs / 1000;
usecs = usecs % 1000;
/*
* we loop over the lock's procLocks to gather a list of all
* holders and waiters. Thus we will be able to provide more
* detailed information for lock debugging purposes.
*
* lock->procLocks contains all processes which hold or wait for
* this lock.
*/
LWLockAcquire(partitionLock, LW_SHARED);
procLocks = &(lock->procLocks);
proclock = (PROCLOCK *) SHMQueueNext(procLocks, procLocks,
offsetof(PROCLOCK, lockLink));
while (proclock)
{
/*
* we are a waiter if myProc->waitProcLock == proclock; we are
* a holder if it is NULL or something different
*/
if (proclock->tag.myProc->waitProcLock == proclock)
{
if (first_waiter)
{
appendStringInfo(&lock_waiters_sbuf, "%d",
proclock->tag.myProc->pid);
first_waiter = false;
}
else
appendStringInfo(&lock_waiters_sbuf, ", %d",
proclock->tag.myProc->pid);
}
else
{
if (first_holder)
{
appendStringInfo(&lock_holders_sbuf, "%d",
proclock->tag.myProc->pid);
first_holder = false;
}
else
appendStringInfo(&lock_holders_sbuf, ", %d",
proclock->tag.myProc->pid);
lockHoldersNum++;
}
proclock = (PROCLOCK *) SHMQueueNext(procLocks, &proclock->lockLink,
offsetof(PROCLOCK, lockLink));
}
LWLockRelease(partitionLock);
if (deadlock_state == DS_SOFT_DEADLOCK)
ereport(LOG,
(errmsg("process %d avoided deadlock for %s on %s by rearranging queue order after %ld.%03d ms",
MyProcPid, modename, buf.data, msecs, usecs),
(errdetail_log_plural("Process holding the lock: %s. Wait queue: %s.",
"Processes holding the lock: %s. Wait queue: %s.",
lockHoldersNum, lock_holders_sbuf.data, lock_waiters_sbuf.data))));
else if (deadlock_state == DS_HARD_DEADLOCK)
{
/*
* This message is a bit redundant with the error that will be
* reported subsequently, but in some cases the error report
* might not make it to the log (eg, if it's caught by an
* exception handler), and we want to ensure all long-wait
* events get logged.
*/
ereport(LOG,
(errmsg("process %d detected deadlock while waiting for %s on %s after %ld.%03d ms",
MyProcPid, modename, buf.data, msecs, usecs),
(errdetail_log_plural("Process holding the lock: %s. Wait queue: %s.",
"Processes holding the lock: %s. Wait queue: %s.",
lockHoldersNum, lock_holders_sbuf.data, lock_waiters_sbuf.data))));
}
if (myWaitStatus == STATUS_WAITING)
ereport(LOG,
(errmsg("process %d still waiting for %s on %s after %ld.%03d ms",
MyProcPid, modename, buf.data, msecs, usecs),
(errdetail_log_plural("Process holding the lock: %s. Wait queue: %s.",
"Processes holding the lock: %s. Wait queue: %s.",
lockHoldersNum, lock_holders_sbuf.data, lock_waiters_sbuf.data))));
else if (myWaitStatus == STATUS_OK)
ereport(LOG,
(errmsg("process %d acquired %s on %s after %ld.%03d ms",
MyProcPid, modename, buf.data, msecs, usecs)));
else
{
Assert(myWaitStatus == STATUS_ERROR);
/*
* Currently, the deadlock checker always kicks its own
* process, which means that we'll only see STATUS_ERROR when
* deadlock_state == DS_HARD_DEADLOCK, and there's no need to
* print redundant messages. But for completeness and
* future-proofing, print a message if it looks like someone
* else kicked us off the lock.
*/
if (deadlock_state != DS_HARD_DEADLOCK)
ereport(LOG,
(errmsg("process %d failed to acquire %s on %s after %ld.%03d ms",
MyProcPid, modename, buf.data, msecs, usecs),
(errdetail_log_plural("Process holding the lock: %s. Wait queue: %s.",
"Processes holding the lock: %s. Wait queue: %s.",
lockHoldersNum, lock_holders_sbuf.data, lock_waiters_sbuf.data))));
}
/*
* At this point we might still need to wait for the lock. Reset
* state so we don't print the above messages again.
*/
deadlock_state = DS_NO_DEADLOCK;
pfree(buf.data);
pfree(lock_holders_sbuf.data);
pfree(lock_waiters_sbuf.data);
}
} while (myWaitStatus == STATUS_WAITING);
/*
* Disable the timers, if they are still running. As in LockErrorCleanup,
* we must preserve the LOCK_TIMEOUT indicator flag: if a lock timeout has
* already caused QueryCancelPending to become set, we want the cancel to
* be reported as a lock timeout, not a user cancel.
*/
if (LockTimeout > 0)
{
DisableTimeoutParams timeouts[2];
timeouts[0].id = DEADLOCK_TIMEOUT;
timeouts[0].keep_indicator = false;
timeouts[1].id = LOCK_TIMEOUT;
timeouts[1].keep_indicator = true;
disable_timeouts(timeouts, 2);
}
else
disable_timeout(DEADLOCK_TIMEOUT, false);
/*
* Re-acquire the lock table's partition lock. We have to do this to hold
* off cancel/die interrupts before we can mess with lockAwaited (else we
* might have a missed or duplicated locallock update).
*/
LWLockAcquire(partitionLock, LW_EXCLUSIVE);
/*
* We no longer want LockErrorCleanup to do anything.
*/
lockAwaited = NULL;
/*
* If we got the lock, be sure to remember it in the locallock table.
*/
if (MyProc->waitStatus == STATUS_OK)
GrantAwaitedLock();
/*
* We don't have to do anything else, because the awaker did all the
* necessary update of the lock table and MyProc.
*/
return MyProc->waitStatus;
}
/*
* TransactionTreeSetCommitTsData
*
* Record the final commit timestamp of transaction entries in the commit log
* for a transaction and its subtransaction tree, as efficiently as possible.
*
* xid is the top level transaction id.
*
* subxids is an array of xids of length nsubxids, representing subtransactions
* in the tree of xid. In various cases nsubxids may be zero.
* The reason why tracking just the parent xid commit timestamp is not enough
* is that the subtrans SLRU does not stay valid across crashes (it's not
* permanent) so we need to keep the information about them here. If the
* subtrans implementation changes in the future, we might want to revisit the
* decision of storing timestamp info for each subxid.
*
* The write_xlog parameter tells us whether to include an XLog record of this
* or not. Normally, this is called from transaction commit routines (both
* normal and prepared) and the information will be stored in the transaction
* commit XLog record, and so they should pass "false" for this. The XLog redo
* code should use "false" here as well. Other callers probably want to pass
* true, so that the given values persist in case of crashes.
*/
void
TransactionTreeSetCommitTsData(TransactionId xid, int nsubxids,
TransactionId *subxids, TimestampTz timestamp,
RepOriginId nodeid, bool write_xlog)
{
int i;
TransactionId headxid;
TransactionId newestXact;
/*
* No-op if the module is not active.
*
* An unlocked read here is fine, because in a standby (the only place
* where the flag can change in flight) this routine is only called by the
* recovery process, which is also the only process which can change the
* flag.
*/
if (!commitTsShared->commitTsActive)
return;
/*
* Comply with the WAL-before-data rule: if caller specified it wants this
* value to be recorded in WAL, do so before touching the data.
*/
if (write_xlog)
WriteSetTimestampXlogRec(xid, nsubxids, subxids, timestamp, nodeid);
/*
* Figure out the latest Xid in this batch: either the last subxid if
* there's any, otherwise the parent xid.
*/
if (nsubxids > 0)
newestXact = subxids[nsubxids - 1];
else
newestXact = xid;
/*
* We split the xids to set the timestamp to in groups belonging to the
* same SLRU page; the first element in each such set is its head. The
* first group has the main XID as the head; subsequent sets use the first
* subxid not on the previous page as head. This way, we only have to
* lock/modify each SLRU page once.
*/
for (i = 0, headxid = xid;;)
{
int pageno = TransactionIdToCTsPage(headxid);
int j;
for (j = i; j < nsubxids; j++)
{
if (TransactionIdToCTsPage(subxids[j]) != pageno)
break;
}
/* subxids[i..j] are on the same page as the head */
SetXidCommitTsInPage(headxid, j - i, subxids + i, timestamp, nodeid,
pageno);
/* if we wrote out all subxids, we're done. */
if (j + 1 >= nsubxids)
break;
/*
* Set the new head and skip over it, as well as over the subxids we
* just wrote.
*/
headxid = subxids[j];
i += j - i + 1;
}
/* update the cached value in shared memory */
LWLockAcquire(CommitTsLock, LW_EXCLUSIVE);
commitTsShared->xidLastCommit = xid;
commitTsShared->dataLastCommit.time = timestamp;
commitTsShared->dataLastCommit.nodeid = nodeid;
/* and move forwards our endpoint, if needed */
if (TransactionIdPrecedes(ShmemVariableCache->newestCommitTsXid, newestXact))
ShmemVariableCache->newestCommitTsXid = newestXact;
LWLockRelease(CommitTsLock);
}
/*
* Remove all segments before the one holding the passed page number
*/
void
SimpleLruTruncate(SlruCtl ctl, int cutoffPage)
{
SlruShared shared = ctl->shared;
int slotno;
/*
* The cutoff point is the start of the segment containing cutoffPage.
*/
cutoffPage -= cutoffPage % SLRU_PAGES_PER_SEGMENT;
/*
* Scan shared memory and remove any pages preceding the cutoff page, to
* ensure we won't rewrite them later. (Since this is normally called in
* or just after a checkpoint, any dirty pages should have been flushed
* already ... we're just being extra careful here.)
*/
LWLockAcquire(shared->ControlLock, LW_EXCLUSIVE);
restart:;
/*
* While we are holding the lock, make an important safety check: the
* planned cutoff point must be <= the current endpoint page. Otherwise we
* have already wrapped around, and proceeding with the truncation would
* risk removing the current segment.
*/
if (ctl->PagePrecedes(shared->latest_page_number, cutoffPage))
{
LWLockRelease(shared->ControlLock);
ereport(LOG,
(errmsg("could not truncate directory \"%s\": apparent wraparound",
ctl->Dir)));
return;
}
for (slotno = 0; slotno < shared->num_slots; slotno++)
{
if (shared->page_status[slotno] == SLRU_PAGE_EMPTY)
continue;
if (!ctl->PagePrecedes(shared->page_number[slotno], cutoffPage))
continue;
/*
* If page is clean, just change state to EMPTY (expected case).
*/
if (shared->page_status[slotno] == SLRU_PAGE_VALID &&
!shared->page_dirty[slotno])
{
shared->page_status[slotno] = SLRU_PAGE_EMPTY;
continue;
}
/*
* Hmm, we have (or may have) I/O operations acting on the page, so
* we've got to wait for them to finish and then start again. This is
* the same logic as in SlruSelectLRUPage. (XXX if page is dirty,
* wouldn't it be OK to just discard it without writing it? For now,
* keep the logic the same as it was.)
*/
if (shared->page_status[slotno] == SLRU_PAGE_VALID)
SlruInternalWritePage(ctl, slotno, NULL);
else
SimpleLruWaitIO(ctl, slotno);
goto restart;
}
LWLockRelease(shared->ControlLock);
/* Now we can remove the old segment(s) */
(void) SlruScanDirectory(ctl, SlruScanDirCbDeleteCutoff, &cutoffPage);
}
